Toxicological Profile for


Draft for Public Comment

June 2018

 

Q.

U.S. Department of Health and Human Services

 

Agency for Toxic Substances and Disease Registry



PERFLUOROALKYLS

ii

FOREWORD
This toxicological profile is prepared in accordance with guidelines developed by the Agency for Toxic
Substances and Disease Registry (ATSDR) and the Environmental Protection Agency (EPA). The
original guidelines were published in the Federal Register on April 17, 1987. Each profile will be revised
and republished as necessary.
The ATSDR toxicological profile succinctly characterizes the toxicologic and adverse health effects
information for these toxic substances described therein. Each peer-reviewed profile identifies and
reviews the key literature that describes a substance's toxicologic properties. Other pertinent literature is
also presented, but is described in less detail than the key studies. The profile is not intended to be an
exhaustive document; however, more comprehensive sources of specialty information are referenced.
The focus of the profiles is on health and toxicologic information; therefore, each toxicological profile
begins with a relevance to public health discussion which would allow a public health professional to
make a real-time determination of whether the presence of a particular substance in the environment
poses a potential threat to human health. The adequacy of information to determine a substance's health
effects is described in a health effects summary. Data needs that are of significance to the protection of
public health are identified by ATSDR and EPA.
Each profile includes the following:
(A) The examination, summary, and interpretation of available toxicologic information and
epidemiologic evaluations on a toxic substance to ascertain the levels of significant human
exposure for the substance and the associated acute, intermediate, and chronic health effects;
(B) A determination of whether adequate information on the health effects of each substance is
available or in the process of development to determine the levels of exposure that present a
significant risk to human health due to acute, intermediate, and chronic duration exposures;
and
(C) Where appropriate, identification of toxicologic testing needed to identify the types or levels
of exposure that may present significant risk of adverse health effects in humans.
The principal audiences for the toxicological profiles are health professionals at the Federal, State, and
local levels; interested private sector organizations and groups; and members of the public. ATSDR plans
to revise these documents in response to public comments and as additional data become available.
Therefore, we encourage comments that will make the toxicological profile series of the greatest use.
Electronic comments may be submitted via: www.regulations.gov. Follow the on-line instructions for
submitting comments.
Written comments may also be sent to: Agency for Toxic Substances and Disease Registry
Division of Toxicology and Human Health Sciences
Environmental Toxicology Branch
Regular Mailing Address:
1600 Clifton Road, N.E.
Mail Stop F-57
Atlanta, Georgia 30329-4027

Physical Mailing Address:
4770 Buford Highway
Building 102, 1st floor, MS F-57
Chamblee, Georgia 30341

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

iii

The toxicological profiles are developed under the Comprehensive Environmental Response,
Compensation, and Liability Act of 1980, as amended (CERCLA or Superfund). CERCLA section
104(i)(1) directs the Administrator of ATSDR to “…effectuate and implement the health related
authorities” of the statute. This includes the preparation of toxicological profiles for hazardous
substances most commonly found at facilities on the CERCLA National Priorities List (NPL) and that
pose the most significant potential threat to human health, as determined by ATSDR and the EPA.
Section 104(i)(3) of CERCLA, as amended, directs the Administrator of ATSDR to prepare a
toxicological profile for each substance on the list. In addition, ATSDR has the authority to prepare
toxicological profiles for substances not found at sites on the NPL, in an effort to “…establish and
maintain inventory of literature, research, and studies on the health effects of toxic substances” under
CERCLA Section 104(i)(1)(B), to respond to requests for consultation under section 104(i)(4), and as
otherwise necessary to support the site-specific response actions conducted by ATSDR.
This profile reflects ATSDR’s assessment of all relevant toxicologic testing and information that has been
peer-reviewed. Staffs of the Centers for Disease Control and Prevention and other Federal scientists have
also reviewed the profile. In addition, this profile has been peer-reviewed by a nongovernmental panel
and is being made available for public review. Final responsibility for the contents and views expressed
in this toxicological profile resides with ATSDR.

Patrick N. Breysse, Ph.D., CIH
Director, National Center for Environmental Health and
Agency for Toxic Substances and Disease Registry
Centers for Disease Control and Prevention

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

iv

VERSION HISTORY
Date

June 2018
August 2015
May 2009

Description
Draft for public comment toxicological profile released
Draft for public comment toxicological profile released
Draft for public comment toxicological profile released

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

v

CONTRIBUTORS & REVIEWERS
CHEMICAL MANAGER TEAM
Selene Chou, Ph.D. (Lead)
Dennis Jones, DVM, Ph.D.
Hana R. Pohl, M.D., Ph.D.
Patricia Ruiz, Ph.D.
Franco Scinicariello, M.D., M.P.H.
Melanie Buser, M.P.H.

ATSDR, Division of Toxicology and Human Health
Sciences, Atlanta, GA

Lisa Ingerman, Ph.D., DABT
Lynn Barber, M.S.
Heather Carlson-Lynch, M.S., DABT
Mario Citra, Ph.D.
Gary L. Diamond, Ph.D.
Julie Klotzbach, Ph.D.
Fernando T. Llados, Ph.D.
Daniel J. Plewak, B.S.
SRC, Inc., North Syracuse, NY

REVIEWERS
Interagency Minimal Risk Level Workgroup:
Includes ATSDR; National Center for Environmental Health (NCEH); National Institute of
Occupational Health and Safety (NIOSH); U.S. Environmental Protection Agency (EPA); National
Toxicology Program (NTP).
Additional reviews for science and/or policy:
ATSDR, Division of Community Health Investigations; EPA; NCEH, Division of Laboratory Science.

PEER REVIEWERS
1.

David A. Savitz, Ph.D., Professor of Epidemiology, Professor of Obstetrics and Gynecology, Brown
University, Providence, Rhode Island

2.

Deborah A. Cory-Slechta, Ph.D., Professor of Environmental Medicine, Pediatrics and Public Health
Sciences, Acting Chair, Department of Environmental Medicine, PI, NIEHS Center of Excellence,
Department of Environmental Medicine, University of Rochester Medical Center, Rochester, New York

3.

Jamie DeWitt, Ph.D., Associate Professor, Department of Pharmacology & Toxicology, Brody School of
Medicine, East Carolina University, Greenville, North Carolina

4. Edward Emmett, M.D., Professor, Center of Excellence in Environmental Toxicology, University of
Pennsylvania, Philadelphia, Pennsylvania

These experts collectively have knowledge of toxicology, chemistry, and/or health effects. All reviewers
were selected in conformity with Section 104(I)(13) of the Comprehensive Environmental Response,
Compensation, and Liability Act, as amended.
ATSDR scientists review peer reviewers’ comments and determine whether changes will be made to the
profile based on comments. The peer reviewers’ comments and responses to these comments are part of
the administrative record for this compound.
The listing of peer reviewers should not be understood to imply their approval of the profile's final
content. The responsibility for the content of this profile lies with ATSDR.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

vi

CONTENTS
FOREWORD ................................................................................................................................................ ii
VERSION HISTORY .................................................................................................................................. iv
CONTRIBUTORS & REVIEWERS ............................................................................................................ v
CONTENTS................................................................................................................................................. vi
LIST OF FIGURES ..................................................................................................................................... ix
LIST OF TABLES ...................................................................................................................................... xii
CHAPTER 1. RELEVANCE TO PUBLIC HEALTH ................................................................................ 1
1.1 OVERVIEW AND U.S. EXPOSURES ......................................................................................... 1
1.2 SUMMARY OF HEALTH EFFECTS ........................................................................................... 4
1.3 MINIMAL RISK LEVELS (MRLs) ............................................................................................ 15
CHAPTER 2. HEALTH EFFECTS ........................................................................................................... 21
2.1 INTRODUCTION ........................................................................................................................ 21
2.2 DEATH ...................................................................................................................................... 106
2.3 BODY WEIGHT ........................................................................................................................ 109
2.4 RESPIRATORY ......................................................................................................................... 121
2.5 CARDIOVASCULAR ............................................................................................................... 123
2.6 GASTROINTESTINAL ............................................................................................................. 135
2.7 HEMATOLOGICAL ................................................................................................................. 137
2.8 MUSCULOSKELETAL ............................................................................................................ 141
2.9 HEPATIC ................................................................................................................................... 146
2.10 RENAL ....................................................................................................................................... 202
2.11 DERMAL ................................................................................................................................... 219
2.12 OCULAR.................................................................................................................................... 220
2.13 ENDOCRINE ............................................................................................................................. 221
2.14 IMMUNOLOGICAL ................................................................................................................. 244
2.15 NEUROLOGICAL ..................................................................................................................... 293
2.16 REPRODUCTIVE...................................................................................................................... 300
2.17 DEVELOPMENTAL ................................................................................................................. 340
2.18 OTHER NONCANCER ............................................................................................................. 406
2.19 CANCER .................................................................................................................................... 418
2.20 MECHANISM OF TOXICITY .................................................................................................. 433
2.20.1 Cellular Mechanisms of Toxicity ........................................................................................ 434
2.20.2 Hepatic Toxicity Mechanisms ............................................................................................. 441
2.20.3 Developmental Toxicity Mechanisms ................................................................................. 443
2.20.4 Immunotoxicity Mechanisms .............................................................................................. 444
2.20.5 Endocrine Mechanisms ....................................................................................................... 445
2.20.6 Cancer Mechanisms ............................................................................................................ 446
2.21 GENOTOXICITY ...................................................................................................................... 447
CHAPTER 3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS,
CHEMICAL INTERACTIONS ....................................................................................... 450
3.1 TOXICOKINETICS ................................................................................................................... 450
3.1.1
Absorption ........................................................................................................................... 451
3.1.2
Distribution ......................................................................................................................... 455
3.1.3
Metabolism.......................................................................................................................... 469
3.1.4
Excretion ............................................................................................................................. 469

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

vii

3.1.5
Physiologically Based Pharmacokinetic (PBPK)/Pharmacodynamic (PD) Models ........... 493
3.1.5.1
Loccisano et al. (2012a, 2012b) Rat Models ............................................................... 494
3.1.5.2
Loccisano et al. (2011, 2013) Monkey and Human Models ........................................ 500
3.1.5.3
Rodriguez et al. (2009) Mouse Model ......................................................................... 504
3.1.5.4.
Wambaugh et al. 2013 (Andersen et al. 2006) Model ................................................. 506
3.1.5.5
Harris and Barton (2008) Rat Model ........................................................................... 509
3.1.5.6
Worley and Fisher (2015a, 2015b) Rat Model ............................................................ 512
3.1.5.7
Worley et al. (2017b) Human Model ........................................................................... 513
3.1.5.8
Fàbrega et al. (2014, 2016) Human Model .................................................................. 513
3.1.6
Animal-to-Human Extrapolations ....................................................................................... 514
3.2 CHILDREN AND OTHER POPULATIONS THAT ARE UNUSUALLY SUSCEPTIBLE ... 515
3.3 BIOMARKERS OF EXPOSURE AND EFFECT ..................................................................... 518
3.3.1
Biomarkers of Exposure ...................................................................................................... 519
3.3.2
Biomarkers of Effect ........................................................................................................... 520
3.4 INTERACTIONS WITH OTHER CHEMICALS ..................................................................... 520
CHAPTER 4. CHEMICAL AND PHYSICAL INFORMATION .......................................................... 521
4.1 CHEMICAL IDENTITY............................................................................................................ 521
4.2 PHYSICAL AND CHEMICAL PROPERTIES......................................................................... 521
CHAPTER 5. POTENTIAL FOR HUMAN EXPOSURE ...................................................................... 533
5.1 OVERVIEW ............................................................................................................................... 533
5.2 PRODUCTION, IMPORT/EXPORT, USE, AND DISPOSAL ................................................ 537
5.2.1
Production ........................................................................................................................... 537
5.2.2
Import/Export ...................................................................................................................... 545
5.2.3
Use ...................................................................................................................................... 545
5.2.4
Disposal ............................................................................................................................... 546
5.3 RELEASES TO THE ENVIRONMENT ................................................................................... 546
5.3.1
Air ....................................................................................................................................... 551
5.3.2
Water ................................................................................................................................... 552
5.3.3
Soil ...................................................................................................................................... 554
5.4 ENVIRONMENTAL FATE ...................................................................................................... 555
5.4.1
Transport and Partitioning................................................................................................... 555
5.4.2
Transformation and Degradation ........................................................................................ 559
5.5 LEVELS IN THE ENVIRONMENT ......................................................................................... 561
5.5.1
Air ....................................................................................................................................... 564
5.5.2
Water ................................................................................................................................... 571
5.5.3
Sediment and Soil ............................................................................................................... 582
5.5.4
Other Media ........................................................................................................................ 587
5.6 GENERAL POPULATION EXPOSURE .................................................................................. 596
5.7 POPULATIONS WITH POTENTIALLY HIGH EXPOSURES .............................................. 623
CHAPTER 6. ADEQUACY OF THE DATABASE ............................................................................... 628
6.1 Existing Information on Health Effects ...................................................................................... 628
6.2 Identification of Data Needs ....................................................................................................... 632
6.3 Ongoing Studies ......................................................................................................................... 642
CHAPTER 7. REGULATIONS AND GUIDELINES ............................................................................ 644
CHAPTER 8. REFERENCES ................................................................................................................. 647

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

APPENDICES
APPENDIX A.
APPENDIX B.
APPENDIX C.
APPENDIX D.
APPENDIX E.
APPENDIX F.

viii

ATSDR MINIMAL RISK LEVELS AND WORKSHEETS......................................... A-1
LITERATURE SEARCH FRAMEWORK FOR PERFLUOROALKYLS ................... B-1
USER’S GUIDE ............................................................................................................. C-1
QUICK REFERENCE FOR HEALTH CARE PROVIDERS ....................................... D-1
GLOSSARY ................................................................................................................... E-1
ACRONYMS, ABBREVIATIONS, AND SYMBOLS .................................................. F-1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

ix

LIST OF FIGURES
1-1. Health Effects Found in Animals Following Oral Exposure to PFOA................................................. 7
1-2. Health Effects Found in Animals Following Oral Exposure to PFOS ................................................. 8
1-3. Health Effects Found in Animals Following Oral Exposure to Other Perfluoroalkyls ........................ 9
1-4. Summary of Sensitive Targets of PFOA – Oral ................................................................................. 17
1-5. Summary of Sensitive Targets of PFOS – Oral .................................................................................. 18
2-1. Overview of the Number of Studies Examining PFOA Health Effects.............................................. 27
2-2. Overview of the Number of Studies Examining PFOS Health Effects .............................................. 28
2-3. Overview of the Number of Studies Examining Other Perfluoroalkyls Health Effects ..................... 29
2-4. Levels of Significant Exposure to PFOA – Inhalation ....................................................................... 32
2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Inhalation ............................................... 34
2-6. Levels of Significant Exposure to PFOA – Oral ................................................................................ 58
2-7. Levels of Significant Exposure to PFOS – Oral ................................................................................. 81
2-8. Levels of Significant Exposure to Other Perfluoroalkyls – Oral ........................................................ 99
2-9. Serum Total Cholesterol Levels Relative to Serum PFOA Levels ................................................... 173
2-10. Risk of Abnormal Cholesterol Levels Relative to PFOA Levels (Presented as Adjusted
Ratios) ............................................................................................................................................ 174
2-11. Serum LDL Cholesterol Levels Relative to Serum PFOA Levels ................................................. 175
2-12. Risk of Abnormal LDL Cholesterol Levels Relative to PFOA Levels (Presented as
Adjusted Ratios) ............................................................................................................................. 176
2-13. Serum Total Cholesterol Levels Relative to Serum PFOS Levels ................................................. 189
2-14. Risk of Abnormal Cholesterol Levels Relative to PFOS Levels (Presented as Adjusted
Ratios) ............................................................................................................................................ 190
2-15. Serum LDL Cholesterol Levels Relative to Serum PFOS Levels .................................................. 191
2-16. Risk of Abnormal LDL Cholesterol Levels Relative to PFOS Levels (Presented as
Adjusted Ratios) ............................................................................................................................. 192
2-17. Risk of Hyperuricemia Relative to PFOA Levels (Presented as Adjusted Odds Ratios) ............... 213
2-18. Risk of Hyperuricemia Relative to PFOS Levels (Presented as Adjusted Odds Ratios)................ 215

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

x

2-19. Antibody Responses Relative to Serum PFOA Levels in Epidemiology Studies .......................... 270
2-20. Risk of Asthma Diagnosis Relative to PFOA Levels (Presented as Adjusted Odds Ratios).......... 272
2-21. Antibody Responses Relative to Serum PFOS Levels in Epidemiology Studies ........................... 278
2-22. Risk of Asthma Diagnosis Relative to PFOS Levels (Presented as Adjusted Odds Ratios) .......... 280
2-23. Antibody Responses Relative to Serum PFHxS Levels in Epidemiology Studies ......................... 283
2-24. Risk of Asthma Diagnosis Relative to PFHxS Levels (Presented as Adjusted Odds Ratios) ........ 285
2-25. Antibody Responses Relative to Serum PFNA Levels in Epidemiology Studies .......................... 287
2-26. Risk of Asthma Diagnosis Relative to PFNA Levels (Presented as Adjusted Odds Ratios).......... 288
2-27. Antibody Responses Relative to Serum PFDeA Levels in Epidemiology Studies ........................ 290
2-28. Risk of Asthma Diagnosis Relative to PFDeA Levels (Presented as Adjusted Odds Ratios) ........ 292
2-29. Fecundability Relative to PFOA Levels (Presented as Adjusted Fecundability Ratios) ................ 326
2-30. Infertility Relative to PFOA Levels (Presented as Adjusted Odds Ratios) .................................... 327
2-31. Fecundability Relative to PFOS Levels (Presented as Adjusted Fecundability Ratios) ................. 333
2-32. Infertility Relative to PFOS Levels (Presented as Adjusted Odds Ratios) ..................................... 334
2-33. Risk of Low Birth Weight Infant Relative to PFOA Levels (Presented as Adjusted Odds
Ratios) ............................................................................................................................................ 379
2-34. Risk of Small For Gestational Age Infant Relative to PFOA Levels (Presented as Adjusted
Odds Ratios) ................................................................................................................................... 380
2-35. Risk of Low Birth Weight Infant Relative to PFOS Levels (Presented as Adjusted Odds
Ratios) ............................................................................................................................................ 391
2-36. Risk of Small for Gestational Age Infant Relative to PFOS Levels (Presented as Adjusted
Odds Ratios) ................................................................................................................................... 392
2-37. Diabetes Risk Relative to Serum PFOA Levels (Presented as Adjusted Ratios) ........................... 415
2-38. Diabetes Risk Relative to Serum PFOS Levels (Presented as Adjusted Odds Ratios)................... 417
3-1. Tissue Concentrations of 14C in Male and Female Rats Following a Single Gavage Dose of
[14C]PFOA at 1, 5, or 25 mg/kg ........................................................................................................ 458
3-2. Tissue Concentrations of 14C in Male (Upper Panel) and Female (Lower Panel) Rats
Following Oral Doses of PFOA for 28 Days at Doses of 3, 10, or 30 mg/kg/day ........................... 459

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

xi

3-3. Structure of PBPK Model of PFOA and PFOS in the Rat................................................................ 495
3-4. PBPK Model Structure for Simulating PFOA and PFOS Exposure During Gestation in the
Rat (Dam, Left; Fetus, Right) ........................................................................................................... 496
3-5. PBPK Model Structure for Simulating PFOA/PFOS Exposure During Lactation in the Rat
(Dam, Left; Pup, Right) .................................................................................................................... 497
3-6. Structure of PBPK Model for PFOA and PFOS in Monkeys and Humans ...................................... 501
3-7. Renal Resorption Pharmacokinetic Model of Gestation and Lactation used in the Analysis of
CD-1 Mice ........................................................................................................................................ 505
3-8. Andersen et al. (2006) Pharmacokinetic Model with Oral Absorption ............................................ 507
3-9. Conceptual Representation of a Physiologically Based Pharmacokinetic Model for PFOS
Exposure in Rats ............................................................................................................................... 510
5-1. Number of NPL Sites with Perfluoroalkyls Contamination ............................................................. 533
5-2. Timeline of Important Events in the History of Polyfluorinated Compounds.................................. 539
6-1. Summary of Existing Health Effects Studies on PFOA By Route and Endpoint ............................. 629
6-2. Summary of Existing Health Effects Studies on PFOS By Route and Endpoint ............................. 630
6-3. Summary of Existing Health Effects Studies on Other Perfluoroalkyls By Route and
Endpoint ........................................................................................................................................... 631

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

xii

LIST OF TABLES
1-1. Summary of Estimated Elimination Half-lives for Select Perfluoroalkyls ........................................... 4
1-2. Overview of Provisional Minimal Risk Levels Derived for Perfluoroalkyl Compounds................... 15
1-3. Provisional Minimal Risk Levels (MRLs) for PFOA ......................................................................... 19
1-4. Provisional Minimal Risk Levels (MRLs) for PFOS ......................................................................... 19
1-5. Provisional Minimal Risk Levels (MRLs) for PFHxS ....................................................................... 20
1-6. Provisional Minimal Risk Levels (MRLs) for PFNA ......................................................................... 20
2-1. Levels of Significant Exposure to PFOA – Inhalation ....................................................................... 30
2-2. Levels of Significant Exposure to Other Perfluoroalkyls – Inhalation ............................................... 33
2-3. Levels of Significant Exposure to PFOA – Oral ................................................................................ 35
2-4. Levels of Significant Exposure to PFOS – Oral ................................................................................. 62
2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral ........................................................ 85
2-6. Levels of Significant Exposure to PFOA – Dermal ......................................................................... 104
2-7. Summary of Childhood Growth in Humans ..................................................................................... 110
2-8. Summary of Cardiovascular Outcomes in Humans.......................................................................... 125
2-9. Summary of Skeletal Outcomes in Humans ..................................................................................... 142
2-10. Summary of Liver Disease in Humans ........................................................................................... 148
2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humans ................. 150
2-12. Summary of Serum Lipid Outcomes in Humans ............................................................................ 156
2-13. Summary of Renal Outcomes in Humans....................................................................................... 203
2-14. Summary of Uric Acid Outcomes in Humans ................................................................................ 206
2-15. Summary of Thyroid Outcomes in Humans ................................................................................... 223
2-16. Summary of Immunological Outcomes in Humans ....................................................................... 245
2-17. Summary of Neurological Outcomes in Humans ........................................................................... 294
2-18. Summary of Alterations in Reproductive Hormone Levels in Humans ......................................... 301
2-19. Summary of Male Reproductive Outcomes in Humans ................................................................. 308

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

xiii

2-20. Summary of Female Reproductive Outcomes in Humans.............................................................. 313
2-21. Summary of Fertility Outcomes in Humans ................................................................................... 318
2-22. Summary of Pregnancy Outcomes in Humans ............................................................................... 341
2-23. Summary of Birth Outcomes in Humans........................................................................................ 344
2-24. Summary of Neurodevelopmental Outcomes in Humans .............................................................. 360
2-25. Summary of Effects on the Development of the Reproductive System in Humans ....................... 371
2-26. Summary of Outcomes Related to Diabetes in Humans................................................................. 407
2-27. Summary of Cancer Outcomes in Humans..................................................................................... 419
2-28. Transactivation of Human and Mouse PPARα in Transfected Cos-1 Cells Exposed to
Perfluoroalkyl Compounds (In Order of Decreasing C20max in the Mouse) .................................... 436
2-29. Gene Expression Changes Induced by Perfluoroalkyl Compounds ............................................... 439
2-30. Hepatic Effects of Perfluoroalkyl Compounds in Wild-Type and PPARα-Null Mice
Exposed Orally ............................................................................................................................... 442
3-1. Tissue Distribution and Excretion of 14C-Radioactivity from Both Sexes of Rats, Mice,
Hamsters, and Rabbits Dosed with 14C-Labeled APFO ................................................................... 461
3-2. Serum (or Plasma) Concentrations in Matched Human Maternal-Infant Pairs ................................ 463
3-3. Matched Serum (or Plasma) and Breast Milk Concentrations in Humans ....................................... 467
3-4. Excretory Clearance of PFOA and PFOS in Humans ...................................................................... 471
3-5. Summary Elimination Half-Lives for Perfluoroalkyls Estimated in Humans and
Experimental Animals ...................................................................................................................... 473
3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in Experimental Animals ................. 481
3-7. Estimated and Assumed Pharmacokinetic Parameters for the Modified Andersen et al.
(2006) Model for PFOA and PFOS .................................................................................................. 508
4-1. Chemical Identity of Perfluoroalkyls................................................................................................ 522
4-2. Physical and Chemical Properties of Perfluoroalkyls....................................................................... 527
5-1. Content (ppm) and Percent Reduction of PFOA, PFOA Homologues, or PFOA Precursors in
Products from 2006 and 2013 U.S. Operations of Fluoropolymer/Fluorotelomer Companies ........ 541
5-2. U.S. Production Volume Ranges for Perfluoroalkyls (1986–2002) Reported under the EPA
Inventory Update Rule...................................................................................................................... 545

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

xiv

5-3. Reported Emissions of PFOA, PFOA Homologues, or PFOA Precursors in Products from
the 2006 and 2013 U.S. Operations of Fluoropolymer/Fluorotelomer Companies .......................... 547
5-4. Global Historical PFCA Production and Emissions Estimates from 1951 to 2004 .......................... 550
5-5. Biological Monitoring of PFOA and PFOS in the Arctic ................................................................. 556
5-6. Lowest Limit of Detection Based on Standards ............................................................................... 562
5-7. Summary of Environmental Levels of Perfluoroalkyls .................................................................... 562
5-8. Perfluoroalkyls Levels in Water, Soil, and Air of National Priorities List (NPL) Sites ................... 563
5-9. Concentrations of Perfluoroalkyl in Outdoor Air ............................................................................. 565
5-10. Concentrations of Perfluoroalkyl in Indoor Air.............................................................................. 568
5-11. Concentration (ng/g) of Perfluoroalkyls in 39 Dust Samples ......................................................... 570
5-12. Concentrations of PFOA and PFOS in Surface Water (ng/L) ........................................................ 572
5-13. Concentrations of Other Perfluoroalkyls in Surface Water ............................................................ 574
5-14. Concentrations of Perfluoroalkyls in Surface Water and Groundwater at Fluorochemical
Industrial Facilities ......................................................................................................................... 576
5-15. Concentrations of PFOA and PFOS in Ocean Water ..................................................................... 578
5-16. Concentrations of Perfluoroalkyls in Soil and Sediment at Fluorochemical Industrial
Facilities ......................................................................................................................................... 583
5-17. Summary of Perfluoroalkyl Compounds Detected in Soil, Sediment, Surface Water, and
Groundwater at 10 Military Installations........................................................................................ 586
5-18. Detections of PFOA in 31 U.S. Food Items ................................................................................... 588
5-19. Detections of Perfluoroalkyls in Fish from U.S. Lakes and Rivers ................................................ 589
5-20. Fluorotelomer Alcohols Detected in Microwaveable Popcorn Bags Produced in China and
the United States ............................................................................................................................. 594
5-21. Concentrations of PFOA and PFOS in Human Serum Collected in the United States .................. 597
5-22. Concentrations of Other Perfluoroalkyls in Human Serum Collected in the United States ........... 600
5-23. Concentrations of PFOA, PFOS, and PFHxS in Human Serum for Occupationally Exposed
Individuals ...................................................................................................................................... 608
5-24. Percent Detection and Levels of PFOA and PFOS in Children’s Serum, Umbilical Cord
Blood, and Breast Milk................................................................................................................... 611

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

xv

5-25. Percent Detection and Levels of Other Perfluoroalkyls in Children’s Serum, Umbilical
Cord Blood, and Breast Milk.......................................................................................................... 616
5-26. Blood Serum Levels for 69,030 Current and Former Residents of Six Water Districts in the
Mid-Ohio Valley (2005–2006) ....................................................................................................... 625
6-1. Ongoing Studies on Perfluoroalkyls ................................................................................................. 642
7-1. Regulations and Guidelines Applicable to Perfluoroalkyls .............................................................. 644
7-2. Select State Drinking Water and Daily Intake Levels for Perfluoroalkyls ....................................... 646

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

1

CHAPTER 1. RELEVANCE TO PUBLIC HEALTH
This toxicological profile on perfluoroalkyls discusses information on 14 perfluoroalkyl compounds that
have been measured in the serum collected from a representative U.S. population 12 years of age and
older in the National Health and Nutrition Examination Survey (NHANES) 2003–2004 (Calafat et al.
2007b), as well as 2 compounds (PFBA and PFHxA) that have been identified in other monitoring
studies. These compounds include:

Perfluorobutyric acid (PFBA)
Perfluorohexanoic acid (PFHxA)
Perfluoroheptanoic acid (PFHpA)
Perfluorooctanoic acid (PFOA)
Perfluorononanoic acid (PFNA)
Perfluorodecanoic acid (PFDeA)
Perfluoroundecanoic acid (PFUA)
Perfluorobutane sulfonic acid (PFBuS)
Perfluorohexane sulfonic acid (PFHxS)
Perfluorooctane sulfonic acid (PFOS)
Perfluorododecanoic acid (PFDoA)
Perfluorooctane sulfonamide (PFOSA)
2-(N-Methyl-perfluorooctane sulfonamide) acetic acid (Me-PFOSA-AcOH)
2-(N-Ethyl-perfluorooctane sulfonamide) acetic acid (Et-PFOSA-AcOH)
The term “perfluoroalkyls” used throughout the toxicological profile is referring to these
14 compounds and the information may not be applicable to other perfluoroalkyl compounds.

1.1

OVERVIEW AND U.S. EXPOSURES

The perfluoroalkyl compounds discussed in this profile primarily consist of perfluorinated aliphatic
carboxylic acids (PFCAs), perfluorinated aliphatic sulfonic acids (PFSAs), and some polyfluorinated
substances that may degrade or be metabolized to some important perfluorinated substances such as
PFOA or PFOS. These substances have been used extensively in surface coating and protectant
formulations due to their unique surfactant properties (Kissa 2001; Schultz et al. 2003). Major
applications have included protectants for paper and cardboard packaging products, carpets, leather
products, and textiles that enhance water, grease, and soil repellency (3M 1999; Hekster et al. 2003; Kissa
2001; Schultz et al. 2003), and in firefighting foams (Schultz et al. 2003). Perfluoroalkyls such as PFOA
have also been used as processing aids in the manufacture of fluoropolymers such as nonstick coatings on
cookware (DuPont 2008; EPA 2008a).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

2
1. RELEVANCE TO PUBLIC HEALTH

Perfluoroalkyls are human-made substances that do not occur naturally in the environment. The
perfluoroalkyl substances discussed in this profile, especially PFOS and PFOA, have been detected in air,
water, and soil in and around fluorochemical facilities; however, these industrial releases have been
declining since companies began phasing out the production and use of several perfluoroalkyls in the
early 2000s (3M 2007b, 2008a, 2008b; Barton et al. 2007; Davis et al. 2007; DuPont 2008; EPA 2007a,
2008a, 2016a). PFOA and PFOS are no longer manufactured or imported into the United States;
however, there could be some imported goods containing trace amounts of these substances as impurities.
Information regarding current releases of shorter-chain perfluoroalkyls (perfluorinated carboxylic acids
with six or fewer carbons and perfluorosulfonic acids with five or fewer carbons) that are now being used
in surface treatment products or perfluoropolyethers that are used as a replacement for PFOA in emulsion
polymerization processes has not been located. In the environment, some of the perfluoroalkyls discussed
in this profile can also be formed from environmental degradation of precursor compounds released
during the manufacture and use of consumer products containing perfluoroalkyls (D’eon and Mabury
2007; D’eon et al. 2009; Martin et al. 2006; Prevedouros et al. 2006). Under the PFOA Stewardship
Program with the U.S. Environmental Protection Agency (EPA), eight major fluoropolymer producers
have phased out PFOA, precursor substances that can degrade to long-chain perfluoroalkyls such as
PFOA, and higher homologues from emissions and products (EPA 2008a, 2016a).

Due to their chemical structure, perfluoroalkyls are very stable in the environment and are resistant to
biodegradation, photoxidation, direct photolysis, and hydrolysis (3M 2000; EPA 2008a; OECD 2002,
2007; Schultz et al. 2003). The perfluoroalkyl carboxylic acids and sulfonic acids have very low
volatility due to their ionic nature (Kissa 2001; Prevedouros et al. 2006; SPARC 2008). As a group,
perfluoroalkyls are persistent in soil and water (3M 2000; Prevedouros et al. 2006). Perfluoroalkyls are
mobile in soil and leach into groundwater (Davis et al. 2007). Volatile fluorotelomer alcohols may be
broken down into substances like PFOA, and atmospheric deposition can lead to contamination of soils
and leaching into groundwater away from point sources. Perfluoroalkyls have been detected in many
parts of the world, including oceans and the Arctic, indicating that long-range transport is possible
(Armitage et al. 2006; Barber et al. 2007; Prevedouros et al. 2006; Wania 2007; Wei et al. 2007a;
Yamashita et al. 2005, 2008).

Perfluoroalkyls have been detected in all environmental media including air, surface water, groundwater
(including drinking water), soil, and food. Human exposure may occur from all of these media.
Contaminated drinking water led to high levels of exposure to PFOA, PFOS, and other perfluoroalkyls for
some populations residing near fluoropolymer manufacturing facilities (ATSDR 2008; Emmett et al.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

3
1. RELEVANCE TO PUBLIC HEALTH

2006a; Steenland et al. 2009b). Median PFOA serum levels of in 45,276 non-occupationally exposed
individuals residing in southeastern Ohio and West Virginia who were exposed to PFOA via
contaminated drinking water were approximately 6 times greater than the median concentration of the
general population when compared to NHANES data (Shin et al. 2011). Serum levels of PFOA and
PFOS in the general population of the United States have decreased dramatically in recent years as U.S.
production of these substances ceased (CDC 2018). For example, the geometric mean concentrations of
PFOA and PFOS in the general population were 5.2 and 30.4 ng/mL (ppb), respectively, in 1999–2000,
but have decreased to 1.94 ng/mL (PFOA) and 4.99 ng/mL (PFOS) in 2013–2014 (CDC 2018).

Based on environmental measurements and theoretical models, one study has proposed that the major
exposure pathways for PFOS for the general population in Europe and North America are food and water
ingestion, dust ingestion, and hand-to-mouth transfer from mill-treated carpets (Trudel et al. 2008). For
PFOA, major exposure pathways were proposed to be oral exposure resulting from migration from paper
packaging and wrapping into food, general food and water ingestion, inhalation from impregnated
clothes, and dust ingestion. This includes exposure to 8:2 fluorotelomer alcohol in food packaging and
air, which can be broken down into PFOA. PFOS and PFOA exposure pathways are proposed to be
similar for children except that exposure from hand-to-mouth transfer from treated carpets is expected to
be much greater in children. Based on these exposure pathways, adult uptake doses estimated for highexposure scenarios were approximately 30 and 47 ng/kg/day for PFOS and PFOA, respectively (Trudel et
al. 2008). PFOS and PFOA doses estimated for children under the age of 12 under high exposure
scenarios were 101–219 and 65.2–128 ng/kg/day, respectively. Since PFOA and PFOS are no longer
produced or used in the United States, current exposure levels may be lower than those predicted by
Trudel et al. (2008). A study by Vestergren and Cousins (2009) evaluated potential exposure to
perfluorocarboxylate homologues for different populations and also concluded that dietary intake was the
primary background exposure pathway for the general population, while inhalation of indoor air was the
main exposure pathway for occupationally exposed individuals with estimated intakes >150 ng/kg/day.

Perfluoroalkyls have been detected in human breast milk and umbilical cord blood. The reported
maximum concentrations of PFOS and PFOA measured in human breast milk samples were 0.360–
0.639 and 0.210–0.490 ng/mL, respectively (Kärrman et al. 2007; So et al. 2006b; Völkel et al. 2008).
Maximum concentrations of other perfluoroalkyl compounds were <0.18 ng/mL. In most umbilical cord
samples, the concentrations of PFOS and PFOA ranged from 4.9 to 11.0 and from 1.6 to 3.7 ng/mL,
respectively (Apelberg et al. 2007a, 2007b; Fei et al. 2007; Inoue et al. 2004; Midasch et al. 2007). Other
perfluoroalkyls have been detected less frequently.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

4
1. RELEVANCE TO PUBLIC HEALTH

1.2

SUMMARY OF HEALTH EFFECTS

Perfluoroalkyls are ubiquitous chemicals in the environment; they are readily absorbed following
inhalation or oral exposure and are not metabolized in humans or laboratory animals. The toxicity of
perfluoroalkyl compounds, particularly PFOA and PFOS, has been extensively evaluated in humans and
laboratory animals. However, comparison of the toxicity of perfluoroalkyls across species is problematic
due to differences in elimination half-lives, lack of adequate mechanistic data, species differences in the
mechanism of toxicity for some endpoints, and differences in measurement of exposure levels between
epidemiology and experimental studies. Substantial differences in the rate of elimination of
perfluoroalkyls exist across species. Table 1-1 lists half-lives for PFOA, PFOS, PFHxS, PFBuS, and
PFBA for human, nonhuman primates, rats, and mice to illustrate some of the species differences. For
example, for PFOA, the estimated elimination half-life ranges from 8 years in humans to 1.9 hours in
female rats.

Table 1-1. Summary of Estimated Elimination Half-lives for Select Perfluoroalkyls
Humans

Nonhuman primates Ratsa

PFOA

8 years
(Olsen et al. 2007a)

PFOS

5.4 years
(Olsen et al. 2007a)

PFHxS

8.5 years
(Olsen et al. 2007a)

PFBuS

665 hours
(Olsen et al. 2009)

PFBA

72 hours
(Chang et al. 2008b)

20.1–32.6 days
(Butenhoff et al.
2004c)
110–170 days
(Chang et al. 2012;
Seacat et al. 2002)
87–141 days
(Sundström et al.
2012)
8.0–95.2 hours
(Chengelis et al. 2009;
Olsen et al. 2009)
40.3–41.0 hours
(Chang et al. 2008b)

aSee

Micea

Males: 44–322 hours
Females: 1.9–16.2 hours
179–1,968 hours

731–1,027 hours

Males: 382–688 hours
597–643 hours
Females: 1.03–41.28 hours
2.1–7.42 hours

1.03–9.22 hours

2.79–13.34 hours

Section 3.1.4 for citations.

The mechanisms of toxicity of perfluoroalkyl compounds have not been fully elucidated. There is strong
evidence that some effects observed in rodents, such as hepatotoxicity, immunotoxicity, and
developmental toxicity, involve the activation of peroxisome proliferator-activated receptor-α (PPARα);
however, humans and nonhuman primates are less responsive to PPARα agonists than rodents.
Additionally, PPARα-independent mechanisms are also involved and it is not known if species

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

5
1. RELEVANCE TO PUBLIC HEALTH

differences exist for these mechanisms. In general, epidemiology studies use serum perfluoroalkyl levels
as a biomarker of exposure, which contrasts with experimental studies that utilize dose, expressed in
mg/kg body weight/day units, or air concentrations as the dose metric. Although physiologically based
pharmacokinetic (PBPK) models have been developed for rodents and humans, these models are not
sufficient to allow for comparisons between administered doses in laboratory animals and serum
concentrations in humans.
Effects in Humans. Perfluoroalkyl compounds have been detected in the serum of workers, residents
living near perfluoroalkyl facilities, and the general population. A large number of epidemiology studies
have evaluated possible associations between perfluoroalkyl exposure and a wide range of adverse health
outcomes. Most of the studies have focused on PFOA and/or PFOS; fewer studies have evaluated a
smaller number of potential health outcomes for the remaining 12 perfluoroalkyls included in this
toxicological profile. Most of the epidemiology studies lack exposure monitoring data, and there is a
potential for multiple routes of exposure (inhalation and oral); however, most of the studies used serum
perfluoroalkyl level as a biomarker of exposure. The three primary sources of this information are
occupational exposure studies, studies of communities living near a PFOA manufacturing facility with
high levels of PFOA in the drinking water, and studies of populations exposed to background levels of
perfluoroalkyl compounds (referred to as general population studies). Of the three categories of subjects,
workers have the highest potential exposure to perfluoroalkyls, followed by the highly-exposed residents
in the Mid-Ohio Valley, and then the general population. In one study of workers at the Washington
Works facility in West Virginia, the average serum PFOA level in 2001–2004 was 1,000 ng/mL (Sakr et
al. 2007a); the mean PFOA level in highly-exposed residents (without occupational exposure) near this
facility was 423 ng/mL in 2004–2005 (Emmett et al. 2006a). By comparison, the geometric mean
concentration of PFOA in the U.S. population was 3.92 ng/mL in 2005–2006 (CDC 2013). Although a
large number of epidemiology studies have examined the potential of perfluoroalkyl compounds to induce
adverse health effects, most of the studies are cross-sectional in design and do not establish causality.
Based on a number of factors (described in Section 2.1) including the consistency of findings across
studies, the available epidemiology studies suggest associations between perfluoroalkyl exposure and
several health outcomes:
•

Pregnancy-induced hypertension/pre-eclampsia (PFOA, PFOS)

•

Liver damage, as evidenced by increases in serum enzymes and decreases in serum bilirubin
levels (PFOA, PFOS, PFHxS)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

6
1. RELEVANCE TO PUBLIC HEALTH

•

Increases in serum lipids, particularly total cholesterol and low-density lipoprotein (LDL)
cholesterol (PFOA, PFOS, PFNA, PFDeA)

•

Increased risk of thyroid disease (PFOA, PFOS)

•

Decreased antibody response to vaccines (PFOA, PFOS, PFHxS, PFDeA)

•

Increased risk of asthma diagnosis (PFOA)

•

Increased risk of decreased fertility (PFOA, PFOS)

•

Small (<20 g or 0.7 ounces per 1 ng/mL increase in blood perfluoroalkyl level) decreases in birth
weight (PFOA, PFOS)

The International Agency for Research on Cancer (IARC 2017) concluded that PFOA is possibly
carcinogenic to humans (Group 2B) and EPA (2016e, 2016f) concluded that there was suggestive
evidence of the carcinogenic potential of PFOA and PFOS in humans. Increases in testicular and kidney
cancer have been observed in highly exposed humans.

There is also some suggestive evidence for associations between perfluoroalkyls and additional health
outcomes; there is less certainty in these associations due to the higher degree of inconsistencies across
studies and/or a smaller number of studies examining a specific outcome. These health outcomes include
osteoarthritis in women under 50 years of age (PFOA, PFOS) and decreased antibody response to
vaccines (PFNA, PFUA, PFDoA). Additionally, associations between serum PFOA and PFOS and
decreases in glomerular filtration rate and increases in serum uric acid levels and between serum PFOA,
PFOS, PFHxS, and PFNA and increased risk of early menopause have been observed; these effects may
be due to reverse causation and not perfluoroalkyl toxicity.
Effects in Laboratory Animals. Most of the information regarding the effects of perfluoroalkyl
compounds in animals is derived from oral studies; considerably less information is available from
inhalation and dermal exposure studies. PFOA and PFOS are the most studied perfluoroalkyl
compounds, with considerably less data for the other compounds. Of the 187 animal studies reviewed in
this toxicological profile, 48% examined PFOA, 34% examined PFOS, and 18% examined other
perfluoroalkyls (4 studies on PFHxS, 16 studies on PFNA, 1 study on PFUA, 4 studies on PFBuS,
6 studies on PFBA, 6 studies on PFDeA, 2 studies on PFDoA, 1 study on PFOSA, and 3 studies on
PFHxA). The primary effects observed in laboratory animals exposed to perfluoroalkyl compounds are
liver toxicity, developmental toxicity, and immune toxicity (see Figures 1-1, 1-2, and 1-3; not all of these
effects have been observed or examined for all perfluoroalkyl compounds. Based on limited data, the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

7
1. RELEVANCE TO PUBLIC HEALTH

Figure 1-1. Health Effects Found in Animals Following Oral Exposure to PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

8
1. RELEVANCE TO PUBLIC HEALTH

Figure 1-2. Health Effects Found in Animals Following Oral Exposure to PFOS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

9
1. RELEVANCE TO PUBLIC HEALTH

Figure 1-3. Health Effects Found in Animals Following Oral Exposure to Other
Perfluoroalkyls

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

10
1. RELEVANCE TO PUBLIC HEALTH

toxicity of perfluoroalkyl compounds does not appear to be specific to the route of administration. It
should be noted that, for the most part, adverse health effects in studies in animals have been associated
with exposure concentrations or doses that resulted in blood levels of perfluoroalkyl compounds that were
significantly higher than those reported in perfluoroalkyl workers or in the general population.
Furthermore, there are profound differences in the toxicokinetics of perfluoroalkyls between humans and
experimental animals. The elimination t1/2 of PFOA is approximately 4 years in humans compared with
days or hours in rodents. These factors, plus issues related to the mode of action of perfluoroalkyls (see
below), make it somewhat difficult at this time to determine the true relevance of some effects reported in
animal studies to human health.

Many of the adverse health effects observed in laboratory animals result from the ability of these
compounds (with some structural restrictions) to activate the PPARα, which can mediate a broad range of
biological responses (Issemann and Green 1990). Species differences in the response to PPARα agonists
have been found; rats and mice are the most sensitive species and guinea pigs, nonhuman primates, and
humans are less responsive. Although humans are less responsive to PPARα agonists, they do have a
functional PPARα. Several explanations for these species differences have been suggested (e.g.,
differences in the ability of PPARα to be induced after exposure to a peroxisome proliferator and
differences in the pattern and level of tissue-specific expression of PPARα). Activation of this receptor in
rodents initiates a characteristic sequence of morphological and biochemical events, principally, but not
exclusively, in the liver (Cattley et al. 1998; Kennedy et al. 2004; Klaunig et al. 2003). The proliferation
of peroxisomes has been associated with a variety of effects, including hepatocellular hypertrophy,
alterations in lipid metabolism, and decreased pup survival and immune effects. Studies in PPARα-null
mice provide evidence that PPARα-independent mechanisms are also involved in PFOA and PFOS
toxicity, including liver and immune toxicity. A more complete discussion of the mechanisms of PFOA
and PFOS toxicity is presented in Section 2.20.

Liver Effects. Many studies have described morphological and biochemical alterations in the liver from
rodents following acute and longer-term oral exposure to PFOA. Some of the effects observed in rats
include increases in liver weight, hepatocellular hypertrophy, and decreases in serum cholesterol and
triglyceride levels (e.g., Butenhoff et al. 2004b; Liu et al. 1996; Pastoor et al. 1987; Yang et al. 2001).
The observed hepatomegaly and hypertrophy are likely due to expansion of the smooth endoplasmic
reticulum and proliferation of peroxisomes, as confirmed by increased activity of biochemical markers
and light and electron microscopy (Pastoor et al. 1987). It is important to note also that there appear to be
different sensitivities for different endpoints. For example, in male rats dosed with PFOA for 14 days,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

11
1. RELEVANCE TO PUBLIC HEALTH

absolute liver weight and fatty acid β-oxidation activity were significantly increased at 2 mg/kg/day,
whereas hepatic microsomal concentration of total cytochrome P450 was significantly increased at
20 mg/kg/day (Liu et al. 1996). In general, longer-term studies with PFOA have shown that the hepatic
effects are reversible once dosing ceases and that recovery tends to parallel the decline in blood levels of
PFOA (Perkins et al. 2004). Studies in mice have provided similar results. However, studies in
PPARα-null mice suggest that hepatomegaly may also be due to a PPARα-independent process in mice
(Yang et al. 2002b), since PFOA induced hepatomegaly to the same extent in wild-type mice and
PPARα-null mice, but failed to increase acyl-CoA oxidase activity in PPARα-null mice. PFOA exposure
also resulted in increases in absolute liver weight in monkeys treated with ≥3 mg/kg/day for 26 weeks, an
effect that was partly associated with significant mitochondrial proliferation, but not peroxisome
proliferation (Butenhoff et al. 2002).

Similar to PFOA, PFOS exposure results in increases in liver weight, hepatocellular hypertrophy, and
decreases in serum cholesterol and triglyceride levels in rodents (e.g., Elcombe et al. 2012a, 2012b; Era et
al. 2009; Seacat et al. 2003; Thibodeaux et al. 2003). PFOS induced an increase in absolute liver weight,
a decrease in serum cholesterol, and hepatocellular hypertrophy and lipid vacuolation in monkeys in a
26-week study (Seacat et al. 2002). Not unexpectedly, there was no evidence of peroxisome proliferation
and no increase in hepatic palmitoyl-CoA oxidase, consistent with the fact that monkeys (and humans)
seem to be refractory to peroxisome proliferative responses (Cattley et al. 1998; Klaunig et al. 2003).

Studies with other perfluoroalkyl compounds have shown that, in general, liver weight and parameters of
fatty acid β-oxidation are more severely affected as the carbon length increases up to about a 10-carbon
chain length (Butenhoff et al. 2009a, 2012a; Goecke-Flora and Reo 1996; Goecke et al. 1992; Hoberman
and York 2003; Kudo et al. 2000, 2006; Permadi et al. 1992, 1993; van Otterdijk 2007a, 2007b).
Significant peroxisome activity seems to require a carbon length >7 (Goecke-Flora and Reo 1996;
Goecke et al. 1992), but increases over control levels have been reported with a four-carbon chain length
(Permadi et al. 1993; Wolf et al. 2008a). In an in vitro study in mouse COS-1 cells, PFOA had the lowest
effective concentration needed for PPARα activation followed by PFNA and PFDeA, PFHxA, and PFBA
(Wolf et al. 2008a). This pattern was not found for the sulfonates; the lowest effective concentration was
for PFHxS followed by PFOS and PFBuS. Wolf et al. (2008a) also found that carboxylate
perfluoroalkyls activated PPARα at lower concentrations than the sulfonate perfluoroalkyls. Studies have
shown that the differential activity is also related to differential accumulation of the perfluoroalkyl
compound in the liver (Kudo and Kawashima 2003; Kudo et al. 2000, 2006). Hydrophobicity, which

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

12
1. RELEVANCE TO PUBLIC HEALTH

increases as carbon length increases, seems to favor biliary enterohepatic recirculation, resulting in a more
protracted toxicity (Goecke-Flora and Reo 1996).

Developmental Effects. PFOA and PFOS have induced developmental effects in rodents. Most studies
with PFOA have been conducted in mice, probably because of the relatively short half-life for PFOA in
female rats, which would prevent accumulation of PFOA during the dosing period. Specific effects
reported include prenatal loss, reduced neonate weight and viability, neurodevelopment toxicity, and
delays in mammary gland differentiation, eye opening, vaginal opening, and first estrus (Abbott et al.
2007; Albrecht et al. 2013; Cheng et al. 2013; Johansson et al. 2008; Koskela et al. 2016; Lau et al. 2006;
Macon et al. 2011; Ngo et al. 2014; Onishchenko et al. 2011; Sobolewski et al. 2014; White et al. 2007,
2009, 2011; Wolf et al. 2007; Yahia et al. 2010). These effects occurred generally in the absence of overt
maternal toxicity. Some of these effects, such as reduced pup survival from birth to weaning, have been
observed in mice treated with as low as 0.6 mg/kg/day PFOA on gestation days (GDs) 1–17 (Abbott et al.
2007). This dose level resulted in mean serum PFOA concentrations of 5,200 and 3,800 ng/mL in dams
and pups, respectively, on postnatal day (PND) 22. A cross-fostering study in mice showed that in utero,
lactation only, and in utero and lactation exposure resulted in significant decreases in postnatal growth
(Wolf et al. 2007). Alterations in spontaneous behavior were reported in 2- or 4-month-old male mice
that were administered a single gavage dose of PFOA at the age of 10 days (Johansson et al. 2008).
Increases in motor activity were also observed following in utero exposure to PFOA (Cheng et al. 2013;
Onishchenko et al. 2011). A cross-fostering study showed that the delays in mammary gland
development were observed following in utero exposure and following lactation-only exposure (White et
al. 2009); however, the results of a 2-generation study showed that the delayed development did not
appear to affect lactational support (White et al. 2011). No fetal toxicity or teratogenicity was reported in
offspring of rabbits exposed to up to 50 mg/kg/day PFOA on GDs 6–18 (Gortner et al. 1982), suggesting
that rabbits are less susceptible than mice to the developmental effects of PFOA, although comparing
administered doses is probably not very informative. There were significant increases in body weight
gain in mice aged 10–40 weeks that were exposed to low levels of PFOA (0.01–0.3 mg/kg/day) on
GDs 1–17 (Hines et al. 2009). Increases in serum insulin and leptin levels were also observed, but there
was no change in serum glucose or the response to a glucose challenge. A comparison of the effects of in
utero exposure (GDs 1–17) to adult exposure (17 days at age 8 weeks) demonstrated that in utero
exposure resulted in higher body weights, white fat weight, and brown fat weight at age 18 months (Hines
et al. 2009).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

13
1. RELEVANCE TO PUBLIC HEALTH

Studies conducted with wild-type and PPARα knockout mice showed that PPARα was required for
PFOA-induced postnatal lethality and that the expression of one copy of the gene was sufficient to
mediate this effect (Abbott et al. 2007). Strain or PPARα expression did not affect serum PFOA levels.
The mechanism of reduced postnatal viability has not been elucidated. Alterations in gene expression in
both fetal liver and lung have been reported following exposure of mice to PFOA during pregnancy
(Rosen et al. 2007). In the liver, PFOA altered the expression of genes linked to fatty acid catabolism,
lipid transport, ketogenesis, glucose metabolism, lipoprotein metabolism, cholesterol biosynthesis, steroid
metabolism, bile acid biosynthesis, phospholipid metabolism, retinol metabolism, proteosome activation,
and inflammation. In the lung, transcriptional-related changes were predominantly associated with fatty
acid catabolism. Although decreased pup survival appears to be linked to PPARα expression, there are
insufficient data to determine whether other developmental effects observed in rats and mice are PPARαindependent.

PFOS significantly decreased birth weight and survival in neonatal rats exposed in utero (Chen et al.
2012b; Lau et al. 2003; Xia et al. 2011), and cross-fostering exposed pups with unexposed dams failed to
improve survival rates (Lau et al. 2003). PFOS serum levels of pups at birth associated with significant
decreased survival were approximately ≥70,000 ng/mL. Dosing rats late during gestation (GDs 17–20)
caused significantly more lethality than dosing early (GDs 2–5) (Grasty et al. 2003). Since pups had
difficulty breathing within minutes of birth and their lungs showed evidence of delayed lung maturation
and other histological alterations (Grasty et al. 2003, 2005; Yahia et al. 2008), the possibility that this
caused the early death has been suggested. Other effects included decreases in birth weight or pup body
weight, delays in eye opening, cleft palate, and neurodevelopmental alterations (Butenhoff et al. 2009b;
Case et al. 2001; Chen et al. 2012b; Era et al. 2009; Fuentes et al. 2006, 2007a, 2007b; Lau et al. 2003;
Luebker et al. 2005a, 2005b; Onishchenko et al. 2011; Thibodeaux et al. 2003; Wang et al. 2015c; Yahia
et al. 2008). Alterations in spontaneous motor activity were observed in mice. A decrease in activity was
observed when mice were placed in a novel environment (Fuentes et al. 2007a; Onishchenko et al. 2011);
another study found a decrease in motor activity followed by increased activity (Johansson et al. 2009).
Evaluation of immunological parameters in 8-week-old pups from mice exposed to PFOS during
gestation showed reduced natural killer (NK) cell activity, suppressed IgM response to immunization, and
alterations in splenic and thymic lymphocyte subpopulations (Keil et al. 2008).

Similar to PFOA and PFOS, increases in fetal mortality were observed in mice exposed to PFDeA on
GDs 6–15 (Harris and Birnbaum 1989) and decreases in litter size and pup survival were observed in
mice exposed to PFNA (Wolf et al. 2010). In contrast, gestational exposure to PFBA, PFBuS, or PFHxS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

14
1. RELEVANCE TO PUBLIC HEALTH

did not result in alterations in pup survival or pup body weight (Das et al. 2008; Hoberman and York
2003; Lieder et al. 2009b). Decreases in spontaneous activity followed by an increase in activity were
observed in mice exposed to PFHxS on PND 10 (Viberg et al. 2013); no alterations were observed in
mice similarly exposed to PFDeA (Johansson et al. 2008).

Immunological Effects. A number of studies have examined the immunotoxicity of perfluoroalkyls in
rats and mice; these data suggest that mice are considerably more sensitive than rats. PFOA- and
PFOS-induced immunological alterations in adult mice are characterized by thymus and spleen atrophy,
alterations in thymic and splenic lymphocyte phenotypes, and impaired response to T-dependent antigens
(DeWitt et al. 2008, 2009; Dong et al. 2009; Guruge et al. 2009; Lefebvre et al. 2008; Loveless et al.
2008; Qazi et al. 2012; Yang et al. 2000, 2002a; Zheng et al. 2009). The lowest lowest-observed-adverseeffect level (LOAEL) for immune effects in mice exposed to PFOA was 3.75 mg/kg/day administered for
15 days; this dosing level resulted in a mean PFOA serum level of 75,000 ng/mL (DeWitt et al. 2008).
For PFOS, several studies identified LOAELs of 0.02–0.8 mg/kg/day (Dong et al. 2009, 2011; Zheng et
al. 2009) and one study identified a LOAEL of 0.00166 mg/kg/day for suppressed response to a
T-dependent antigen (Peden-Adams et al. 2008). PFOA applied to the skin of mice increased serum IgE
levels following a challenge with ovalbumin relative to mice treated with ovalbumin alone, which led the
investigators to suggest that PFOA may increase the IgE response to environmental allergens (Fairley et
al. 2007). More limited data are available for other perfluoroalkyls. Thymic and/or splenic alterations
were observed in rats and mice administered ≥1 mg/kg/day PFNA (Fang et al. 2008, 2009, 2010). No
histological alterations were observed in rodents exposed to PFHxS (Butenhoff et al. 2009a), PFDeA
(Harris et al. 1989), PFBuS (3M 2001), or PFBA (3M 2007a; Butenhoff et al. 2012a; van Otterdijk 2007a,
2007b).
Cancer Effects. PFOA, as many other PPARα agonists, induced hepatocellular adenomas, Leydig cell
adenomas, and pancreatic acinar cell adenomas in rats (Biegel et al. 2001). Liver tumors induced by
PFOA are believed to be mediated largely through PPARα activation, and considered to be of limited or
no relevance to humans (EPA 2016h), based on species differences in response to PPARα activation.
Although Leydig cell tumors are also commonly induced by peroxisome proliferating agents, the mode of
action by which these tumors are induced by PFOA, and thus their relevance to humans, is much less
clear (Corton et al. 2014; EPA 2016h; Klaunig et al. 2003). One mode of action proposed for the
induction of Leydig cell tumors involves PFOA-induced decreases in circulating testosterone levels,
leading to increased production of gonadotropin releasing hormone and circulating luteinizing hormone
(LH), which promotes Leydig cell proliferation. Reduced testosterone levels may occur through

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

15
1. RELEVANCE TO PUBLIC HEALTH

decreased biosynthesis, or via the conversion of testosterone to estradiol via the enzyme aromatase, both
of which may be related to PPARα activation (EPA 2016h). However, the data supporting a PPARαdependent mode of action for Leydig cell tumors is not sufficiently established to rule out human
relevance (EPA 2016h). Likewise, the mechanism of PFOA-induced pancreatic acinar cell tumors may
include a PPARα-dependent component, but the mechanism has not been fully elucidated, and relevant
data are limited. A proposed mode of action involves stimulation of PPARα leading to reduced bile flow
and/or changes in bile acid composition with subsequent increase in cholecystokinin (CCK), which
stimulates pancreatic cell proliferation and tumor formation (EPA 2016h). Support for this mode of
action is limited to information demonstrating increased biliary excretion of PFOA in wild-type and
PPARα null mice (Minata et al. 2010) and data showing altered expression of bile acid transporters
(OATPs and MRPs) in exposed laboratory animals (Cheng and Klassen 2008a; Maher et al. 2008). The
limitations in available data on the mode of action for pancreatic tumor preclude a conclusion regarding
the human relevance of PFOA-induced pancreatic tumors (EPA 2016h).

1.3

MINIMAL RISK LEVELS (MRLs)

A summary of the provisional MRLs derived for perfluoroalkyl compounds is presented in Table 1-2.
The database was not considered adequate for derivation of inhalation MRLs. Though inhalation data are
available for PFOA and PFNA, these studies examined a limited number of endpoints and the data are not
adequate for identifying the most sensitive targets of toxicity or establishing dose-response relationships.
No inhalation data are available for other perfluoroalkyl compounds.

Table 1-2. Overview of Provisional Minimal Risk Levels Derived for
Perfluoroalkyl Compounds

Compound
PFOA

Acute
Xa

Inhalation MRLs
Intermediate Chronic
X

X

Acute
X

PFOS

X

X

X

X

PFHxS

X

X

X

X

PFNA

X

X

X

X

PFDeA
PFUA
PFHpA

X
X
X

X
X
X

X
X
X

X
X
X

***DRAFT FOR PUBLIC COMMENT***

Oral MRLs
Intermediate
3x10-6 mg/kg/day
(Table 1-3)
2x10-6 mg/kg/day
(Table 1-4)
2x10-5 mg/kg/day
(Table 1-5)
3x10-6 mg/kg/day
(Table 1-6)
X
X
X

Chronic
X
X
X
X
X
X
X

 PERFLUOROALKYLS

16
1. RELEVANCE TO PUBLIC HEALTH

Table 1-2. Overview of Provisional Minimal Risk Levels Derived for
Perfluoroalkyl Compounds

Compound
PFBuS
PFBA
PFDoA
PFHxA
PFOSA
Me-PFOSA-AcOH
Et-PFOSA-AcOH
aX

Acute

Inhalation MRLs
Intermediate Chronic

X
X
X
X
X
X
X

X
X
X
X
X
X
X

X
X
X
X
X
X
X

Acute
X
X
X
X
X
X
X

Oral MRLs
Intermediate

Chronic

X
X
X
X
X
X
X

X
X
X
X
X
X
X

indicates that no MRL was derived.

Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; Me-PFOSA-AcOH = 2-(N-methylperfluorooctane sulfonamide); acetic acid; PFBA = perfluorobutyric acid; PFBuS = perfluorobutane sulfonic acid;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxA = perfluorohexanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide;
PFUA = perfluoroundecanoic acid

The oral databases were considered adequate for derivation of provisional intermediate-duration oral
MRLs for PFOA, PFOS, PFHxS, and PFNA based on laboratory animal data. The databases were not
considered adequate for derivation of MRLs for the other perfluoroalkyl compounds. Hepatic, immune,
and developmental endpoints were the most sensitive targets in laboratory animals exposed to PFOA (see
Figure 1-4) and PFOS (see Figure 1-5), respectively. The most sensitive targets were hepatic and thyroid
endpoints for PFHxS and body weight and developmental endpoints for PFNA. The provisional MRL
values for PFOA, PFOS, PFHxS, and PFNA are summarized in Tables 1-3, 1-4, 1-5, and 1-6 and
discussed in greater detail in Appendix A.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

17
1. RELEVANCE TO PUBLIC HEALTH

Figure 1-4. Summary of Sensitive Targets of PFOA – Oral
Developmental endpoints are the most sensitive target of PFOA.
Numbers in circles are the lowest LOAELs for all health effects in animals.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

18
1. RELEVANCE TO PUBLIC HEALTH

Figure 1-5. Summary of Sensitive Targets of PFOS – Oral
The immune system and developing organism are the most sensitive targets of PFOS.
Numbers in circles are the lowest LOAELs for all health effects in animals.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

19
1. RELEVANCE TO PUBLIC HEALTH

Table 1-3. Provisional Minimal Risk Levels (MRLs) for PFOAa
Exposure
duration

MRL

Critical effect

Point of
departure

Inhalation exposure
Acute
Insufficient data for MRL derivation
Intermediate Insufficient data for MRL derivation
Chronic
Insufficient data for MRL derivation
Oral exposure (mg/kg/day)
Acute
Insufficient data for MRL derivation
Intermediate 3x10-6 Neurodevelopmental
0.000821
and skeletal effects in
(LOAELHED)
mice
Chronic
aSee

Uncertainty factor Reference

300

Koskela et al.
2016;
Onishchenko
et al. 2011

Insufficient data for MRL derivation

Appendix A for additional information.

HED = human equivalent dose; LOAEL = lowest-observed-adverse-effect level; PFOA = perfluorooctanoic acid

Table 1-4. Provisional Minimal Risk Levels (MRLs) for PFOSa
Exposure
duration

MRL

Critical effect

Point of
departure

Inhalation exposure
Acute
Insufficient data for MRL derivation
Intermediate Insufficient data for MRL derivation
Chronic
Insufficient data for MRL derivation
Oral exposure (mg/kg/day)
Acute
Insufficient data for MRL derivation
Intermediate 2x10-6 Delayed eye opening
0.000515
and decreased pup
(NOAELHED)b
weight in rats
Chronic
Insufficient data for MRL derivation
aSee

Uncertainty and
modifying factors

30
10

Reference

Luebker et
al. 2005a

Appendix A for additional information.

HED = human equivalent dose; NOAEL = no-observed-adverse-effect level; PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

20
1. RELEVANCE TO PUBLIC HEALTH

Table 1-5. Provisional Minimal Risk Levels (MRLs) for PFHxSa
Exposure
duration

MRL

Critical effect

Point of
departure

Inhalation exposure
Acute
Insufficient data for MRL derivation
Intermediate Insufficient data for MRL derivation
Chronic
Insufficient data for MRL derivation
Oral exposure (mg/kg/day)
Acute
Insufficient data for MRL derivation
Intermediate 2x10-5 Thyroid follicular cell
0.0047
damage in rats
(NOAELHED)
Chronic
Insufficient data for MRL derivation
aSee

Uncertainty and
modifying factors

30
10

Reference

Butenhoff et
al. 2009a

Appendix A for additional information.

HED = human equivalent dose; NOAEL = no-observed-adverse-effect level; PFHxS = perfluorohexane sulfonic acid

Table 1-6. Provisional Minimal Risk Levels (MRLs) for PFNAa
Exposure
duration

MRL

Critical effect

Point of
departure

Uncertainty and
modifying factors

Reference

Inhalation exposure
Acute
Insufficient data for MRL derivation
Intermediate Insufficient data for MRL derivation
Chronic
Insufficient data for MRL derivation
Oral exposure (mg/kg/day)
Acute
Insufficient data for MRL derivation
Intermediate 3x10-6 Decreased body weight 0.001
and developmental
(NOAELHED)
delays in mice
Chronic
Insufficient data for MRL derivation
aSee

30
10

Das et al.
2015

Appendix A for additional information.

HED = human equivalent dose; NOAEL = no-observed-adverse-effect level; PFNOA = perfluorononanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

21

CHAPTER 2. HEALTH EFFECTS
2.1

INTRODUCTION

The primary purpose of this chapter is to provide public health officials, physicians, toxicologists, and
other interested individuals and groups with an overall perspective on the toxicology of perfluoroalkyls.
It contains descriptions and evaluations of toxicological studies and epidemiological investigations and
provides conclusions, where possible, on the relevance of toxicity and toxicokinetic data to public health.

This document discusses information on perfluoroalkyl compounds that have been measured in the serum
collected from a representative U.S. population ≥12 years of age in the 2003–2004 NHANES (Calafat et
al. 2007b), as well as 2 compounds (PFBA and PFHxA) that have been identified in other monitoring
studies. These compounds include:
Perfluorooctanoic acid (PFOA)
Perfluorooctane sulfonic acid (PFOS)
Perfluorohexane sulfonic acid (PFHxS)
Perfluorononanoic acid (PFNA)
Perfluorodecanoic acid (PFDeA)
Perfluoroundecanoic acid (PFUA)
Perfluoroheptanoic acid (PFHpA)
Perfluorobutane sulfonic acid (PFBuS)
Perfluorobutyric acid (PFBA)
Perfluorododecanoic acid (PFDoA)
Perfluorohexanoic acid (PFHxA)
Perfluorooctane sulfonamide (PFOSA)
2-(N-Methyl-perfluorooctane sulfonamide) acetic acid (Me-PFOSA-AcOH)
2-(N-Ethyl-perfluorooctane sulfonamide) acetic acid (Et-PFOSA-AcOH)
The term perfluoroalkyls used throughout the profile is referring to these 14 compounds and may
not be applicable to other perfluoroalkyl compounds.

A glossary and list of acronyms, abbreviations, and symbols can be found at the end of this profile.

To help public health professionals and others address the needs of persons living or working near hazardous
waste sites, the information in this section is organized by health effect. These data are discussed in terms of
route of exposure (inhalation, oral, and dermal) and three exposure periods: acute (≤14 days), intermediate
(15–364 days), and chronic (≥365 days).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

22
2. HEALTH EFFECTS

As discussed in Appendix B, a literature search was conducted to identify relevant studies examining health
effect endpoints. Figures 2-1, 2-2, and 2-3 provide an overview of the database of studies in humans or
experimental animals included in this chapter of the profile. These studies evaluate the potential health
effects associated with inhalation, oral, or dermal exposure to perfluoroalkyls, but may not be inclusive of
the entire body of literature.

Summaries of the epidemiology studies, including details on the study design and results, are presented in
tables in the Supporting Document for Epidemiological Studies for Perfluoroalkyls; briefer summaries of
the studies are presented in summary tables for each endpoint. For studies in which the population was
divided into perfluoroalkyl exposure categories, such as quartiles, the risk ratio reported in the summary
table is for the lowest exposure category with a statistically significant association; risk ratios for higher
exposure categories are presented in the Supporting Document for Epidemiological Studies for
Perfluoroalkyls tables. Summaries of experimental studies are separated by exposure route and are
presented in Tables 2-1, 2-2, 2-3, 2-4, 2-5, and 2-6. The inhalation data for PFOA and other
perfluoroalkyls are presented in Tables 2-1 and 2-2, respectively. The oral data for PFOA, PFOS, and
other perfluoroalkyls are presented in Tables 2-3, 2-4, and 2-5, respectively. The dermal data for PFOA
is presented in Table 2-6. In addition, the NOAEL and LOAEL values from inhalation and oral studies
are graphically presented in Figures 2-4, 2-5, 2-6, 2-7, and 2-8.

Levels of significant exposure (LSEs) for each route and duration are presented in tables and illustrated in
figures. The points in the figures showing no-observed-adverse-effect levels (NOAELs) or lowestobserved-adverse-effect levels (LOAELs) reflect the actual doses (levels of exposure) used in the studies.
LOAELs have been classified into "less serious" or "serious" effects. "Serious" effects are those that
evoke failure in a biological system and can lead to morbidity or mortality (e.g., acute respiratory distress
or death). "Less serious" effects are those that are not expected to cause significant dysfunction or death,
or those whose significance to the organism is not entirely clear. ATSDR acknowledges that a
considerable amount of judgment may be required in establishing whether an endpoint should be
classified as a NOAEL, "less serious" LOAEL, or "serious" LOAEL, and that in some cases, there will be
insufficient data to decide whether the effect is indicative of significant dysfunction. However, the
Agency has established guidelines and policies that are used to classify these endpoints (ATSDR 2003).
ATSDR believes that there is sufficient merit in this approach to warrant an attempt at distinguishing
between "less serious" and "serious" effects. The distinction between "less serious" effects and "serious"
effects is considered to be important because it helps the users of the profiles to identify levels of
exposure at which major health effects start to appear. LOAELs or NOAELs should also help in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

23
2. HEALTH EFFECTS

determining whether or not the effects vary with dose and/or duration, and place into perspective the
possible significance of these effects to human health.

A User's Guide has been provided at the end of this profile (see Appendix C). This guide should aid in
the interpretation of the tables and figures for LSEs and MRLs.

The discussion of the available data for each health effect is divided into several subsections. Each health
effect section begins with an overview, which contains a brief discussion of the available data and
conclusions that can be drawn from the data. Compound-specific discussions follow the overview; the
perfluoroalkyls are discussed in the following order: PFOA, PFOS, PFHxS, PFNA, PFDeA, PFUA,
PFHpA, PFBuS, PFBA, PFDoA, PFHpA, PFOSA, Me-PFOSA-AcOH, and Et-PFOSA-AcOH. It is
noted that for most health effects, there are no data for a number of the perfluoroalkyl compounds. The
compound-specific discussions are further divided into Epidemiology Studies and Laboratory Animal
Studies; for data-rich endpoints, a compound-specific summary is also included. Each perfluoroalkyl is
treated separately in this chapter. Although there is some evidence of similar health outcomes for some
compounds, there is evidence of qualitative and mechanistic differences (Peters and Gonzalez 2011).

The health effects of perfluoroalkyls have been evaluated in a large number of epidemiology and animal
studies; the literature search framework for identifying these studies is discussed in Appendix B. As
illustrated in Figures 2-1, 2-2, and 2-3, most of the health effects data come from epidemiology studies.
For PFOA, PFOS, and other perfluoroalkyls, 67, 70, and 73%, respectively, of the health effect studies
were in humans; it is noted that a number of epidemiology studies examined more than one perfluoroalkyl
compound. Three population categories were examined in epidemiology studies: workers at facilities
involved in the production or use of perfluoroalkyls (most of the studies involved workers at two U.S.
facilities), communities living near a PFOA manufacturing facility with high levels of PFOA in the
drinking water (almost all of the studies involved residents living near a PFOA production facility in
West Virginia), and populations exposed to background levels of perfluoroalkyl compounds (referred to
as general population studies). Most of the studies of communities living near perfluoroalkyl
manufacturing facilities are part of the C8 Health Project and C8 Health Study (C8 is a synonym for
PFOA. The C8 Health Project was a population study of Ohio and West Virginia residents living near the
DuPont Washington Works facility in West Virginia and was funded by DuPont as part of a class action
settlement agreement. The Washington Works facility began using PFOA in 1951 and peak use was in
the late 1990s. At the time of enrollment (2005–2006), blood samples were collected from over
69,000 participants who lived, worked, or attended school in six contaminated water districts surrounding

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

24
2. HEALTH EFFECTS

the facility for at least 12 months between 1950 and December 2004 (Frisbee et al. 2009); the six water
districts were Little Hocking Water Association, Tuppers Plains Chester Water District, Village of
Pomeroy, Lubeck Public Service District, Mason County Public Service District, or private water sources
within these areas. The participants ranged in age from 1.5 to >100 years, with an average age of
39.1 years.

Serum perfluoroalkyl levels were used as the biomarker of exposure in almost all of the epidemiology
studies. The highest levels of serum PFOA were found among workers, followed by the community
members, and then the general population. One study of PFOA workers in 2004–2005 reported an
average serum PFOA level of 1,000 ng/mL (Sakr et al. 2007a). A study of community members living
near this same facility reported a mean serum PFOA level of 423 ng/mL in 2004–2005. In the United
States, the mean geometric mean serum PFOA level in 2005–2006 was 3.92 ng/mL (CDC 2018). In a
study of two PFOS facilities, mean serum PFOS levels in workers were 960–1,400 ng/mL in 2000 (Olsen
et al. 2003a); the geometric mean serum PFOS levels in the U.S. general population in 1999–2000 was
30.4 ng/mL (CDC 2018). Most of the epidemiology studies provided a single serum perfluoroalkyl
concentration, which has been shown to be a reliable biomarker of recent exposure; however, it does not
provide information on historical exposure. The lack of historical exposure data is a particular limitation
of the occupational and community population studies where past exposures were typically higher than
current exposures. Another limitation of the epidemiology studies involves co-exposure to multiple
perfluoroalkyl compounds. A number of the epidemiology studies have found strong correlations
between serum levels of different perfluoroalkyl compounds. In vitro studies (Carr et al. 2013; Wolf et al.
2014) have shown that at lower concentrations, binary pairs of perfluoroalkyl compounds demonstrate
concentration and response additivity, but deviate from additivity at higher concentrations (Wolf et al.
2014). These possible interactions (or dose additivity) complicate the interpretation of the epidemiology
data.

Although a large number of epidemiology studies have examined the potential of perfluoroalkyl
compounds to induce adverse health effects, most of the studies are cross-sectional in design and do not
establish causality. ATSDR used a weight-of-evidence approach to evaluate whether the available data
supported a link between perfluoroalkyl exposure and a particular health effect. This weight-of-evidence
approach takes into consideration the consistency of the findings across studies, the quality of the studies,
dose-response, and plausibility. It should be noted that although the data may provide strong evidence for
an association, it does not imply that the observed effect is biologically relevant because the magnitude of
the change may be within the normal limits or not indicative of an adverse health outcome. Plausibility

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

25
2. HEALTH EFFECTS

depends primarily on experimental toxicology studies that establish a plausible biological mechanism for
the observed effects.

The available epidemiology studies suggest links between perfluoroalkyl exposure and several health
outcomes:
•

Hepatic effects. Increases in serum enzymes and decreases in serum bilirubin, observed in
studies of PFOA, PFOS, and PFHxS, are suggestive of liver damage. In addition, the results of
epidemiology studies of PFOA, PFOS, PFNA, and PFDeA suggest a link between perfluoroalkyl
exposure and increases in serum lipid levels, particularly total cholesterol and LDL cholesterol.

•

Cardiovascular effects. There is suggestive epidemiological evidence for an association
between serum PFOA and PFOS and pregnancy-induced hypertension and/or pre-eclampsia.

•

Endocrine effects. Epidemiology studies provide suggestive evidence of a link between serum
PFOA and PFOS and an increased risk of thyroid disease.

•

Immune effects. Evidence is suggestive of a link between serum PFOA, PFOS, PFHxS, and
PFDeA levels and decreased antibody responses to vaccines. A possible link between serum
PFOA levels and increased risk of asthma diagnosis has also been found.

•

Reproductive effects. A suggestive link between serum PFOA and PFOS levels and an
increased risk of decreased fertility has been found.

•

Developmental effects. Evidence is suggestive of a link between serum PFOA and PFOS and
small decreases in birth weight; the decrease in birth weight is <20 g (0.7 ounces) per 1 ng/mL
increase in blood PFOA or PFOS level.

As presented in Figures 2-1, 2-2, and 2-3, most of the available literature on the health effects of
perfluoroalkyl compounds in laboratory animals was conducted in oral studies, with a few inhalation and
dermal exposure studies identified. The most commonly examined endpoints were liver, body weight,
developmental, reproductive, and immunological.

The results of the animal studies suggest the following:
•

Hepatic effects. Evidence from acute, intermediate, and/or chronic oral studies in rats, mice, and
monkeys indicates that the liver is a sensitive target of PFOA, PFOS, PFHxS, PFNA, PFDeA,
PFUA, PFBA, PFBuS, PFDoA, and PFHpA toxicity. The effects include increases in liver
weight, hepatocellular hypertrophy, and decreases in serum lipid levels. The effects were
considered specific to rodents and were not considered relevant to humans. Some degenerative
and necrotic effects that are likely relevant to humans have also been observed for PFOA, PFOS,
and PFHpA.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

26
2. HEALTH EFFECTS

•

Immune effects. Evidence from acute and intermediate oral studies in mice indicates that
immune endpoints are sensitive targets of PFOA and PFOS toxicity. The most commonly
reported effect was an impaired response to antigens. No alteration in antigen response was
observed in the one study of PFNA. Immune function has not been tested for the other
perfluoroalkyl compounds examined in this profile.

•

Reproductive effects. Decreases in mammary gland development have been observed in mice
orally exposed to PFOA. In general, studies of PFOA and PFOS have not found alterations in
fertility.

•

Developmental effects. Evidence from acute and intermediate oral studies in rats and/or mice
indicates that developmental endpoints are targets of PFOA, PFOS, PFHxS, PFNA, PFDeA,
PFUA, and PFBA toxicity. The developmental effects include decreases in pup body weight,
decreases in pup survival, and alterations in locomotor activity.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

27
2. HEALTH EFFECTS

Figure 2-1. Overview of the Number of Studies Examining PFOA Health Effects*
Developmental, hepatic, and body weight effects of PFOA were the most widely examined potential toxicity outcomes
More studies evaluated health effects in humans than animals (counts represent studies examining endpoint)

*Includes studies discussed in Chapter 2. A total of 271 studies (including those finding no effect) have examined toxicity; most animal studies examined multiple
endpoints. In this figure, the number of human studies is referring to the number of publications.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

28
2. HEALTH EFFECTS

Figure 2-2. Overview of the Number of Studies Examining PFOS Health Effects*
Developmental, hepatic, and reproductive effects of PFOS were the most widely examined potential toxicity outcomes
More studies evaluated health effects in humans than animals (counts represent studies examining endpoint)

*Includes studies discussed in Chapter 2. A total of 218 studies (including those finding no effect) have examined toxicity; most animal studies examined multiple
endpoints. In this figure, the number of human studies is referring to the number of publications.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

29
2. HEALTH EFFECTS

Figure 2-3. Overview of the Number of Studies Examining Other Perfluoroalkyls Health Effects*
Developmental, hepatic, and body weight effects of other perfluoroalkyls were the most widely examined potential toxicity outcomes
More studies evaluated health effects in humans than animals (counts represent studies examining endpoint)

*Includes studies discussed in Chapter 2. A total of 127 studies (including those finding no effect) have examined toxicity; most animal studies examined multiple
endpoints. In this figure, the number of human studies is referring to the number of publications.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

30
2. HEALTH EFFECTS

Table 2-1. Levels of Significant Exposure to PFOA – Inhalation

Species
Figure (strain)
Exposure
Doses
keya No./group parameters (mg/m3)
ACUTE EXPOSURE
1

Rat
(albino)
5 M,F

1 hour
(NS)

18,600

Parameters
monitored Endpoint

NOAEL
(ppm)

CS, BW, GN, Resp
HP
Ocular
Neuro

Less
serious
LOAEL
(ppm)

Serious
LOAEL
(ppm)

Effect

18,600

Red nasal discharge; dry rales

18,600
18,600

Red material around the eyes; lacrimation
Excessive salivation

Griffith and Long 1980
APFO
2

Rat
(CD)
36 M

4 hours

380, 810, CS, HE, BI,
830, 2,200, GN, HP
4,800,
5,700

Death
Bd Wt

980 M
380 M

Resp
Gastro
Hepatic

380 M
810

Ocular

380 M

Bd Wt

7.6

Resp

84

Cardio
Gastro
Hemato
Musc/skel
Hepatic

84
84
84
84
84

Renal
Dermal
Ocular
Endocr
Neuro
Repro

84
84
84
84
84
84

380 M

810 M

LC50
Weight loss for 1–2 days after exposure
(data not shown)
Pulmonary edema
Stomach irritation
Liver enlargement at 810 mg/m3; no
histological alterations
Corneal opacity and corrosion

Kennedy et al. 1986
APFO
3

Rat
(CD)
24 M

2 weeks
0, 1, 7.6,
6 hours/day 84
5 days/week

CS, HE, BI,
GN, HP

84

Kennedy et al. 1986
***DRAFT FOR PUBLIC COMMENT***

7% lower body weight on exposure day 5

Increased absolute and relative liver weight;
hepatocellular hypertrophy at 7.6 mg/m3

No histological alterations
No histological alterations

 PERFLUOROALKYLS

31
2. HEALTH EFFECTS

Table 2-1. Levels of Significant Exposure to PFOA – Inhalation

Species
Figure (strain)
Exposure
Doses
keya No./group parameters (mg/m3)

Parameters
monitored Endpoint

NOAEL
(ppm)

Less
serious
LOAEL
(ppm)

Serious
LOAEL
(ppm)

Effect

25

3/12 deaths on GDs 12, 13, and 17

APFO
4

Rat
(SpragueDawley)
12 F

GDs 6–15
6 hours/day

0, 0.1, 1,
10, 25

MX, DX, OW, Death
CS, HP
Bd Wt

1

Hepatic

25

Develop

10

10

25

12% decrease weight gain on GDs 6–15
18% increase absolute liver weight at
25 mg/m3
10% decreased neonatal body weight on
PND 1

Staples et al. 1984
APFO
aThe

number corresponds to entries in Figure 2-4.

APFO = ammonium perfluorooctanoate (ammonia salt of PFOA); BI = biochemical changes; BW or Bd wt = body weight; F = female(s); Cardio = cardiovascular;
CS = clinical signs; Develop = developmental; DX = developmental toxicity; Endocr = endocrine; Gastro = gastrointestinal; GD = gestation day; GN = gross
necropsy; HE or Hemato = hematology; HP = histopathology; LC50 = lethal concentration, 50% kill; LE = lethality; LOAEL = lowest-observed-adverse-effect level;
M = male(s); Musc/skel = musculoskeletal; MX = maternal toxicity; Neuro = neurological; NOAEL = no-observed-adverse-effect level; NS = not specified;
OW = organ weight; PFOA = perfluorooctanoic acid; PND = postnatal day; Repro = reproductive; Resp = respiratory

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

32
2. HEALTH EFFECTS

Figure 2-4. Levels of Significant Exposure to PFOA – Inhalation
Acute (≤14 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

33
2. HEALTH EFFECTS

Table 2-2. Levels of Significant Exposure to Other Perfluoroalkyls – Inhalation
Species
Figure (strain)
keya
No./group

Exposure
parameters

Doses
(mg/m3)

Parameters
monitored Endpoint

NOAEL
(ppm)

Rat
4 hours
(CD)
6M
Kinney et al. 1989
Exposure was nose-only.

67–4,600

LE

Death

2

0, 67, 590

BW, OW

Bd Wt
Resp

67
67

Hepatic

67

Less
serious
LOAEL
(ppm)

Serious
LOAEL
(ppm)

Effect

820

14-day LC50

ACUTE EXPOSURE
PFNA
1

Rat
(CD)
10 M

4 hours

590
590

Reduced 18% 5 days after exposure
Lung noise; labored breathing during and
after exposure
28% increase in absolute liver weight 5 days
after exposure to ≥67 mg/m3

Kinney et al. 1989
Exposure was nose-only.
aThe

number corresponds to entries in Figure 2-5.

BW or Bd wt = body weight; LC50 = lethal concentration, 50% kill; LE = lethality; LOAEL = lowest-observed-adverse-effect level; M = male(s);
NOAEL = no-observed-adverse-effect level; OW = organ weight; PFNA = perfluorononanoic acid; Resp = respiratory

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

34
2. HEALTH EFFECTS

Figure 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Inhalation
Acute (≤14 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

35
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
ACUTE EXPOSURE
1

Monkey
(Rhesus)
10 M,F
Griffith and Long 1980
APFO

2 weeks
(G)

0, 3, 10, 30,
100

LE, CS, HE,
BI, GN, HP

Death

2

Rat
(CD)
10 M
Biegel et al. 1995
APFO

14 days
(GW)

0, 25

BI, OW

Repro

3

14 days
(GW)

0, 1, 10, 25,
50

BW, OW, BI, Bd Wt
HP
Hepatic

Rat
(CD)
75 M

100

10

25

184% increase in serum estradiol

25

14% reduction in final body weight

10

46% increase in relative liver
weight at ≥10 mg/kg/day
63% increase in serum estradiol

50

Repro

1

0, 18, 23

Hepatic

23

100, 215,
464, 1,000,
2,150

Death

Unspecified number out of 4 died
on week 2

Cook et al. 1992
APFO
4

Rat
1 or 7 days
(Sprague-Dawley) (F)
18 M

Increased liver weight, decreased
serum cholesterol, triglyceride,
hepatocellular hypertrophy at
≥18 mg/kg/day

Elcombe et al. 2010
APFO
5

Rat
(albino)
25 M,F
Griffith and Long 1980
APFO

Once
(GO)

***DRAFT FOR PUBLIC COMMENT***

680 M
430 F

LD50
LD50

 PERFLUOROALKYLS

36
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
6

Rat
(albino)
40 M,F

28 days
(F)

M: 0, 3, 10,
30, 10, 300,
1,000, 3,000;
F: 0, 3.4,
11.3, 34, 113,
340, 1,130,
3,400

7 days
ad lib
(F)

0, 16

Death

1,000 M
1,130 F

5/5 males and 5/5 females died
before end of 1st week of study

Griffith and Long 1980
APFO
7

Rat
(Wistar)
8M

BW, OW, BI, Bd Wt
EA

16

Hepatic

16

66% increase in absolute liver
weight

45% increase in relative liver
weight

Haughom and Spydevold 1992
APFO
8

Rat (SpragueDawley)
3M

14 days
(F)

0, 20

OW, EA

Hepatic

20

14 days
(GW)

0, 0.5, 5, 50

BW, OW, CS, Bd Wt
HE, BI
Hepatic

50
50

2-fold increased mean relative liver
weight at 50 mg/kg/day

Immuno

50

No alterations in spleen weight or
splenocyte phenotype

Ikeda et al. 1985
PFOA
9

Rat (SpragueDawley)
16 M

Iwai and Yamashita 2006
APFO

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

37
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
10

Rat
(Wistar)
5–12 M

Once
(GO)

0, 50

BW, FX

BW

50

Neuro

50

No alteration in performance on
novel object recognition test

Kawabata et al. 2017
PFOA
11

Rat
(Wistar)
30 M

1 week
(F)

0, 1.2, 2.4,
4.7, 9.5

BW, OW, EA, Bd Wt
HP
Hepatic

38

BW, OW, BC Bd Wt

1

9.5

Significant increase in absolute
and relative liver weight at
≥4.7 mg/kg/day

Kawashima et al. 1995
PFOA
12

Rat
(SD-IGS BR)
10 M

14 days
(GW)

0, 0.3, 1, 3,
10, 30,

Hepatic

3

30

24% decrease in overall body
weight gain
Decreased serum cholesterol
levels at ≥0.3 mg/kg/day

Loveless et al. 2006
APFO
13

Rat
(CD)
15 M

14 days
(G)

0, 0.2, 2, 20, BW, OW, EA Bd Wt
40
Hepatic

Repro

2

20

14% lower final body weight

2

34% increase in absolute and
relative liver weight at
≥2 mg/kg/day
2-fold increase in serum estradiol

50

17% weight loss

40

0.2

Liu et al. 1996
APFO
14

Rat
1, 3, 7 days 0, 50
(Sprague-Dawley) (GW)
24 M

BW, BI, EA,
HP

Bd Wt
Hepatic

50

Pastoor et al. 1987
APFO

***DRAFT FOR PUBLIC COMMENT***

2-fold increase in relative and
absolute liver weight

 PERFLUOROALKYLS

38
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
15

Rat
GDs 6–15
(Sprague-Dawley) (GO)
12 or 25 F

0, 100

DX, MX, BW Bd Wt

100

Develop

100

Bd Wt

10

Hepatic

10

Immuno

7.5

33% reduced maternal body weight
gain
No alterations in fetal body weight
or teratology

Staples et al. 1984
APFO
16

Mouse
7 days
(SV129 wild type) (G)
4M

0, 10

BW, HP

Hepatocellular hypertrophy,
steatosis, and increased hepatic
triglyceride levels

Das et al. 2017
PFOA
17

Mouse
(C57BL/6N)
6F
DeWitt et al. 2009
PFOA

10 days
(W)

0, 3.75, 7.5,
15

18

28 days
(F)

M: 0, 5.4,
18.0, 54, 180,
540, 1,800,
5,400; F: 0,
5.8, 19.5, 58,
195, 580,
1,950, 5,800

Death

GDs 6–17
(DW)

0, 0.5, 1

Bd Wt

Mouse
(CD)
40 M,F

FX

15

Altered response to sRBC

180 M
195 F

5/5 died before 2nd week of study
5/5 died before 2nd week of study

Griffith and Long 1980
APFO
19

Mouse
(C57BL/6N)
16 F

Develop

1
0.5

Hu et al. 2010
APFO

***DRAFT FOR PUBLIC COMMENT***

7–10% decrease in litter weight on
PND 2

 PERFLUOROALKYLS

39
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
20

Mouse
(CD-1)
10 M
Johansson et al. 2008
APFO

Once
(G)

0, 0.58, 8.70 CS, OF, DX

Develop

0.58

Decreased spontaneous activity
and altered response to cholinergic
stimulant

10-day-old mice were administered a single dose of PFOA; neurodevelopmental testing was conducted when the pups were 2 or 4 months of age
21

Mouse
(CD-1)
5 M,F
Kennedy 1987
APFO

14 days
(F)

0, 5.3, 54,
537

OW

22

14 days
(GW)

0, 0.3, 1, 3,
10, 30,

BW, OW, BC Bd Wt

Mouse
(CD-1)
10 M

Hepatic

Hepatic

537

3

123–155% increase in absolute
liver weight in 14 days at
≥5.3 mg/kg/day

10

30

6–12% decreased in body weight
gain
Decreased serum cholesterol
levels at ≥0.3 mg/kg/day

Loveless et al. 2006
APFO
23

Mouse
(C57BL/6N)
3M

2–10 days
(F)

0, 78, 390

BW, OW, EA Bd Wt
Hepatic

78

25% body weight loss after
10 days of treatment
74% increase in absolute liver
weight at ≥78 mg/kg/day

78

25% body weight loss after 5 days
of treatment
74% increase in absolute liver
weight in 5 days at ≥78 mg/kg/day

390

Permadi et al. 1992
PFOA
24

Mouse
(C57BL/6N)
4M

2–10 days
(F)

0, 78, 390

BW, OW, EA Bd Wt
Hepatic

390

Permadi et al. 1993
PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

40
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
25

Mouse
(BALB/c)
5F

7 days
(GW)

20

BW, OW, BC Bd wt
Hepatic

20
20

36% decrease in body weight gain
32% increase in relative liver
weight

Immuno

20

Inhibition of T-and B-lymphocyte
proliferation in response to sRBC;
decreased phagocytosis by
peripheral blood cells and NK cell
activity; decreased IgM antibody
formation in response to OVA

Develop

5

Altered mammary gland
development in female pups;
reduced pup weight on PND 20

Vetvicka and Vetvickova 2013
PFOA
26

Mouse
(CD-1)
14 F
White et al. 2007
APFO

GDs 8–17
GDs 1217
(GW)

0, 5

DX, GN

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

41
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
27

Mouse
(CD-1)
56 F

GDs 8–17
(GW)

0, 5

Bd Wt

5

Hepatic

5

40–120% increased relative liver
weight in lactating dams on
PNDs 1–10

Repro

5

Immature mammary gland
morphology in lactating dams on
PNDs 1–10

Develop

5

Delayed mammary gland
development (30–60%) in female
pups on PNDs 1–10

Develop

5

Delayed mammary gland
development (31–47%) in female
pups on PNDs 22–32 and at
18 months

White et al. 2009
APFO
28

Mouse
(CD-1)
12–14 F

GDs 7–17 0, 5
GDs 10–17
GDs 13–17
GDs 15–17

White et al. 2009
APFO
29

Mouse
(CD-1)
6–14 F

GDs 7–17, 0, 5, 20
GDs 10–17
GDs 13–17
GDs 15–17
(GW)

DX, MX, BW, Bd Wt
OW
Hepatic

20

No alterations in dams dosed on
GDs 15–17

20

Increase in relative liver weight in
dams dosed on GDs 13–17, 10–
17, or 7–17 at ≥5 mg/kg/day

Develop

5

Reduced pup body weight at
weaning, 43% in males and 35% in
females

24

>10% reduced final body weight

Wolf et al. 2007
APFO
30

Mouse
(C57BL/6N)
8M

7 days
(F)

0, 24

CS, BW, OW, Bd Wt
BI
Hepatic

24

Xie et al. 2003
PFOA

***DRAFT FOR PUBLIC COMMENT***

2-fold increase in absolute liver
weight

 PERFLUOROALKYLS

42
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
31

Mouse
(C57BL/6N)
8M

10 days
(F)

0, 30

BW, OW, BI, Bd Wt
CS
Hepatic

30

17% decrease in final body weight

30

>90% increase in absolute and
relative liver weight

Immuno

30

86% reduction in absolute thymus
weight; 30% reduction in absolute
spleen weight

Yang et al. 2000
PFOA
32

Mouse
(C57BL/6N)
8M

10 days
(F)

0, 1, 3.5,
11.5, 23, 58

CS, BW, OW, Hepatic
OF, BI

1

35% increase in absolute liver
weight at ≥1 mg/kg/day

Immuno

11.5

40–50% decrease in spleen and
thymus weights

Yang et al. 2001
PFOA
33

Mouse
(C57BL/6N)
8–12 M
Yang et al. 2002a
PFOA

7 days
(F)

0, 24

CS, BW, BI,
OF

Immuno

24

Decreased humoral response to
immunization with horse red blood
cells

34

7 days
ad lib
(F)

0, 33

BW, OW, BI, Bd Wt
OF
Hepatic

33

14% decreased mean body weight

Mouse
(C57BL/6N)
16 M

33

Immuno

86% increase in absolute liver
weight
33

Yang et al. 2002b
PFOA
Experiments with PPARα-null mice suggested PPARα-dependent and -independent immune effects
35

Rabbit
(New Zealand)
18 F
Gortner et al. 1982
PFOA

GDs 6–18
1 time/day
(GW)

0, 1.5, 5, 50

CS, MX, DX, Develop
BW

50

***DRAFT FOR PUBLIC COMMENT***

40% reduction in spleen weight
and 79% reduction in thymus
weight

 PERFLUOROALKYLS

43
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
INTERMEDIATE EXPOSURE
36

Monkey
(Cynomolgus)
4–6 M

26 weeks
1 time/day
(C)

0, 3, 10,
30/20

BC, BW, CS, Bd Wt
EA, FI, GN, Resp
HE, HP, LE,
OP, OW, UR Cardio
Gastro
Hemato

10

20

12% decrease by week 10

3

36% increase in absolute liver
weight at ≥3 mg/kg/day; increased
serum triglyceride levels at
30/20 mg/kg/day

10

Significant decrease in serum TT4
(27–35%) and FT4 (30–38%)
No histological alterations
No histological alterations
No histological alterations

20
20
20
20

Musc/skel 20
Hepatic

Renal

20

Dermal

20

Ocular
Endocr

20
3

Immuno
Neuro
Repro

20
20
20

Butenhoff et al. 2002
APFO

37

Monkey
(Rhesus)
10 M,F

90 days
1 time/day
(G)

0, 3, 10, 30,
100

LE, CS, HE,
BI, GN, HP

Death
Bd Wt

10

Cardio

10

Gastro

10

Hemato

30

Hepatic

10

Renal

10

Immuno
Neuro

30

One male and two females died
during weeks 7–12

30

33% body weight loss by week 6

30

Emesis

10

30

10

30

Atrophy of lymphoid follicles in
spleen and lymph nodes
Hypoactivity and prostration

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

44
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
Repro

100

Bd Wt

20

Hemato

20

Hepatic

20

Endocr

20

Repro

20

No histological alterations in testes
or ovaries

Griffith and Long 1980
APFO
38

Monkey
(Cynomolgus)
8M

30 day
1 time/day
(C)

0, 2, 20

CS, BW, FI,
BI, HE, EA,
GN, HP

No alterations in serum levels of
thyroid hormones and TSH and
histopathology of adrenals
No alterations in serum estradiol,
estriol, or histopathology of the
testes

Thomford 2001
APFO
39

Rat
(CD)
156 M

1 year
ad lib
(F)

0, 13.6

CS, BW, FI, Bd Wt
OW, GN, HP Hepatic

13.6
13.6

Repro

>10% reduced weight gain
Increased relative liver weight

13.6

Significant increase in serum
estradiol at 1, 3, 6, 9, and
12 months (~100–180%); prolactin
was decreased at all time points,
but not always significantly

3

10

>11% reduced body weight

1

3

Increased absolute and relative
liver weight in P-generation (36%)
and F1-generation (30%) males;
hepatocellular hypertrophy and
less commonly necrosis in F1
males (incidence not reported)

3

Increased absolute and relative
kidney weight in P-generation
(14%) and F1-generation (11%)
males

Biegel et al. 2001
APFO
40

Rat
70–90 days 0, 1, 3, 10, 30 MX, DX, OF, Bd Wt
(Sprague-Dawley) 1 time/day
BW, GN, HP Hepatic
30 M,F
(GW)

Renal

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

45
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
Endocr

10

Repro

30

Develop

10

30

30

Vacuolation of zona glomerulosa of
adrenal gland
No alteration in reproductive
performance in P0 or F1
generation
Increased number of dead pups on
PNDs 6–8

Butenhoff et al. 2004b
APFO
41

Rat
(Wistar)
5F

GD 1 to
PND 21
ad lib
(W)

0, 1.6

Develop

1.6

17–18% reduced motor
coordination and increased
locomotor activity in pups on
PNDs 34–35

Resp

5

Hepatic

5

Cytoplasmic vacuolization,
necrosis, hypertrophy, increased
liver weight at ≥5 mg/kg/day; fatty
degeneration, angiectasis and
congestion in the hepatic sinusoid
or central vein at 20 mg/kg/day

Neuro

5

Cachexia and lethargy

Hepatic

18

Increased liver weight (43% on
day 29), decreased serum
cholesterol (39% on day 29) and
triglyceride (73% on day 29),
hepatocellular hypertrophy and
hyperplasia

Cheng et al. 2013
PFOA
42

Rat
Daily
(Sprague-Dawley) 28 days
10 M
(G)

0, 5, 20

OW, HP

Cui et al. 2009
PFOA
43

Rat
28 days
(Sprague-Dawley) (F)
10 M

0, 18

Elcombe et al. 2010
APFO

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

46
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
44

Rat
(CD)
40 M,F

28 days
ad lib
(F)

M: 0, 3, 10, BW, FI, HP
30, 100, 300,
1,000, 3,000;
F: 0, 3.4,
11.3, 34, 113,
340, 1,130,
3,400

Bd Wt

10 M

Hepatic

3M

M: 0, 1, 3, 10, CS, BW, FI,
30, 100; F: 0, HE BI, GN,
1.1, 3.4, 11, HP, OW
34, 110

Bd Wt

30 M
110 F

Resp

100 F

Cardio

110 F

Gastro

110 F

30 M

100 M

30 mg/kg/day: 11% reduction in
final body weight; 100 mg/kg/day:
33% reduction in final body weight
Hepatocyte hypertrophy

Griffith and Long 1980
APFO
45

Rat
(CD)
30 M,F

90 days
ad lib
(F)

100 M

33% reduction in final mean body
weight

Musc/skel 110 F
Hepatic

100 M

Renal

110 F

Dermal

110 F

Ocular
Endocr
Immuno
Neuro
Repro

110 F
110 F
110 F
110 F
100 M
110 F

Hepatocyte hypertrophy; 50%
increase in absolute liver weight

No histological alterations
No histological alterations

Griffith and Long 1980
APFO
46

Rat
GDs 12–18 0, 5, and 20
(Sprague-Dawley) (GO)
10 dams; 12–
13 pups

BC, BW, DX, Bd Wt
OW
Develop

5

20

30% reduction in body weight of
dams

5

20

13% reduction in body weight of
male pups

Li et al. 2016
PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

47
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
47

Rat
(CD)
10 M

28 days
(G)

0, 0.29, 0.96,
9.6, 29

Bd Wt

0.96

Hemato

29

Hepatic

0.29

Immuno

29

9.6

10% decrease in final body weight

29

34% decrease in serum triglyceride
levels, minimal hepatocellular
hypertrophy at ≥0.29 mg/kg/day;
minimal focal necrosis at
29 mg/kg/day

Loveless et al. 2008
APFO
48

Rat
(CD)
55 M

13 weeks
ad lib
(F)

0, 0.06, 0.64, CS, BW, FI, Bd Wt
1.94, 6.5
OW, GN, HP Resp

6.5
6.5

Hepatic

6.5

Neuro
Repro

6.5
6.5

Minimal to moderate hepatocellular
hypertrophy at ≥0.64 mg/kg/day
No histological alterations
No histological alterations

Perkins et al. 2004
APFO
49

Mouse
(129S1/SvlmJ
WT)
17 F

GDs 1–17
(GW)

0, 0.1, 0.3,
0.6, 1, 3, 5,
10, 20

MX, DX, BW Bd Wt

10 F

Hepatic

20

Develop

0.3

Hepatic

3

Increased absolute and relative
liver weight of dams on PND 22 at
≥1 mg/kg/day
0.6

Significantly reduced pup survival
(46%) from birth to weaning

Abbott et al. 2007
APFO
Body weight NOAEL is for changes during pregnancy.
50

Mouse
GDs 1–17
(wild-type Sv/129) 1 time/day
5–6 F
(GW)

0, 3

Develop
Albrecht et al. 2013
PFOA
***DRAFT FOR PUBLIC COMMENT***

28% increased liver weight,
hepatocellular hypertrophy with
increased peroxisomes
3

31.5% reduced pups per litter on
PND 20

 PERFLUOROALKYLS

48
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
51

Mouse
(C57BL/6N)
8F

15 days
ad lib
(W)

0, 0.94, 1.88, BW, OW, OF Bd Wt
3.75, 7.5, 15,
Immuno
30

7.5
1.88

15

Weight loss (~5%)

3.75

Reduced sRBC-specific response
to IgM antibody titers

30

14–20% decrease in body weight
in wild-type mice

30

16 and 14% reduction in sRBCspecific antibody response in wildtype and PPARα knockout mice,
respectively; 29.8% decrease in
relative spleen weight in wild-type
mice

1.88

Decrease in DNP-specific IgM
antibody responses; decreases in
relative spleen (17%) and thymus
weights (14%) at 7.5 mg/kg/day

DeWitt et al. 2008
APFO
52

Mouse
15 days
(wild-type
(W)
C57BL/6-Tac and
PPARα knockout)
8F

0, 7.5, 30

BW, OW, BC Bd Wt
Immuno

7.5

15 days
(W)

0, 0.94, 1.88, BW, OW, BC Bd Wt
3.75, 7.5
Immuno

7.5

10, 13,
15 days
(W)

0, 3.75, 7.5

7.5

DeWitt et al. 2016
PFOA
53

Mouse
(C57BL/6N)
8F

0.94

DeWitt et al. 2016
PFOA
54

Mouse
(C57BL/6N)
8F

BC

Bd Wt

DeWitt et al. 2016
PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

49
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
55

Mouse
(CD-1)
6–14 dams; 21–
37 F offspring

GDs 1–17
(GW)

0, 0.01, 0.1,
0.3, 1, 5

HP

Hepatic

0.3

1

Increase in severity of chronic
inflammation at ≥1 mg/kg/day; Ito
cell hypertrophy and centrilobular
hepatocellular hypertrophy at
5 mg/kg/day

Filgo et al. 2015a, 2015b
PFOA
Animals exposed in utero on GDs 1–17 and examined at 18 months of age; the offspring were also examined by Hines et al. 2009
56

Mouse
GDs 1–17 0, 0.1, 0.3,
HP
Hepatic
1
Increased severity of centrilobular
(129/Sv WT)
(GW)
0.6, 1
hepatocyte hypertrophy at
3–7 dams; 6–10 F
≥0.3 mg/kg/day
offspring
Filgo et al. 2015a, 2015b
PFOA
Animals exposed in utero on GDs 1–17 and examined at 18 months of age; the offspring were also examined by Abbott et al. 2007
57

Mouse
GDs 1–17 0, 0.1, 0.3, 1, HP
Hepatic
1
3
Centrilobular hepatocyte
(129/Sv PPARα- (GW)
3
hypertrophy and bile duct
knockout)
hyperplasia at 3 mg/kg/day
5–9 dams; 6–10 F
offspring
Filgo et al. 2015a, 2015b
PFOA
Animals exposed in utero on GDs 1–17 and examined at 18 months of age; the offspring were also examined by Abbott et al. 2007
58

Mouse
(CD)
40 M,F

28 days
ad lib
(F)

M: 0, 5.4,
BW, FI, HP
18.0, 54, 180,
540, 1,800,
5,400; F: 0,
5.8, 19.5, 58,
195, 580,
1,950, 5,800

Death
Bd Wt

Hepatic

5.4 M

18

Griffith and Long 1980
APFO

***DRAFT FOR PUBLIC COMMENT***

54 M
58 F

4/5 died before end of 4th week
5/5 died before 4th week of study

5.8 F

Males: final body weight 20% lower
than controls; females: final body
weight 25% lower than controls
3-fold or greater increased
absolute and relative liver weight
and hepatocellular hypertrophy at
≥5.4 mg/kg/day

 PERFLUOROALKYLS

50
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
59

Mouse
(CD-1)
5–14F
Hines et al. 2009
APFO

GDs 1–17
(GW)

0, 0.01, 0.1,
0.3, 1, 3, 5

60

21 days
ad lib
(F)

0, 0.0018,
OW
0.0054,
0.018, 0.054,
0.18, 0.54,
1.8, 5.4

Mouse
(CD-1)
5 M,F

BW FX

Develop

Hepatic

5F

Decreased birth weight
(approximately 8%) and body
weight at weaning (24%)

5.4

39–41% increase in absolute liver
weight at ≥5.4 mg/kg/day

Kennedy 1987
APFO
61

Mouse
GDs 1–21 0.3
BW, DX
(C57BL/6)
(F)
6F
Koskela et al. 2016
PFOA
Offspring were examined at 13 and 17 months of age
62

Mouse
(CD-1)
9–45 M

GDs 1–17
1 time/day
(GW)

0.3b

Develop

0, 1, 3, 5, 10, MX, DX, BW, Bd Wt
20, 40
OW
Hepatic
Develop

5

Altered femur and tibial bone
morphology, decreased tibial
mineral density

10

1

32% reduced weight gain during
pregnancy
38% increase in absolute liver
weight at ≥1 mg/kg/day

1

Lau et al. 2006
APFO

***DRAFT FOR PUBLIC COMMENT***

5

Reduced ossification of proximal
phalanges and advanced preputial
separation at ≥1 mg/kg/day; 20%
decrease in pup body weight on
PND 23 at ≥3 mg/kg/day;
increased number of dams with full
litter resorptions, decrease in
neonatal survival, tail and limb
defects, delay in eye opening, and
delay in first estrus at
≥5 mg/kg/day

 PERFLUOROALKYLS

51
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
63

Mouse
(CD)
20 M

28 days
(G)

0, 0.29, 0.96,
9.6, 29

Bd Wt

0.96

Hepatic

0.29

0.96

9.6

Weight loss (86% of controls)
Mild hepatocellular hypertrophy at
≥0.29, moderate to severe
hypertrophy and single cell
necrosis at ≥0.96 mg/kg/day;
decreased serum cholesterol
(31%) and triglyceride (53%) at
≥9.6 mg/kg/day

Immuno

0.96 M

9.6 M

Decreased response to sRBC,
decreased number of splenic and
thymic lymphocytes

Loveless et al. 2008
APFO
64

Mouse
(CD-1)
13 F
Macon et al. 2011
PFOA

GDs 1–17
(G)

65

Mouse
(CD-1)
13 F
Macon et al. 2011
PFOA
66

DX

Develop

0.3

Impaired development of
mammary glands in offspring

GDs 10–17 0, 0.01, 0.1,
(G)
1.0

DX

Develop

0.01

Developmental delays in mammary
gland development

Mouse
(C57BL/6JApc+/+)
20–21 F
Ngo et al. 2014
PFOA

GDs 1–17
(GW)

0, 0.01, 0.1,
3.0

BC, BW, FI,
HE, OW

Develop

3

Decrease in number of successful
births

67

GDs 1–21
ad lib
(F)

0, 0.3

0.3b

Increased locomotor activity in
adult offspring

Mouse
(C57BL/6/
Bk1)
6F

0, 0.3, 1.0,
3.0

Develop

0.1

Onishchenko et al. 2011
PFOA
***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

52
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
68

Mouse
GDs 1–17
(CD-1)
(GW)
17–21 F dams; 4–
6 M,F pups

0, 0.01, 0.1,
0.3, 1

BI, BW, HP,
OF, OW

Hepatic

0.01

Increased severity of
hepatocellular hypertrophy at
PND 91 and periportal
inflammation on PND 21 at
≥0.01 mg/kg/day (incidence was
not reported); decreased serum
total cholesterol, LDL, and HDL
levels in high-fat fasted animals on
PND 91 at ≥0.3 mg/kg/day

Quist et al. 2015a, 2015b
PFOA
Subgroup of female offspring were fed a high-fat diet (50% calories from diet) for 6 weeks
69

Mouse
(C57BL/6)
6 M,F

6 weeks
(F)

0.55

BC, BW, FI,
OW

Hepatic

0.55

66–67% increase in relative liver
weight in males; increased plasma
cholesterol levels in males (35%)
and females (70%)

Mouse
(BALB/c)
6 M,F
Rebholz et al. 2016
PFOA

6 weeks
(F)

0.55

BC, BW, FI,
OW

Hepatic

0.55

54–65% increase in relative liver
weight; 20% increase in plasma
cholesterol in males

71

GD 7 to
PND 21
(F)

0, 0.1

BH, BW

Develop

Rebholz et al. 2016
PFOA
70

Mouse
(C57BL/6)
6F

0.1

Sobolewski et al. 2014
PFOA

***DRAFT FOR PUBLIC COMMENT***

Increased horizontal and
ambulatory locomotor activity and
decreased resting time in males;
decrease in novel object
recognition in males and females

 PERFLUOROALKYLS

53
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
72

Mouse
(ICR)
10 M

21 days
ad lib
(W)

0, 0.5, 2.6,
18, 47

BW, OW,
GN, HP

Bd Wt

2.6

Hepatic

18

Renal

47

18

17% decrease in weight gain

47

27% increase in relative liver
weight at ≥0.5 mg/kg/day;
increases in ALT at
≥2.6 mg/kg/day; hepatocytomegaly
at 18 mg/kg/day; necrosis at
47 mg/kg/day

Son et al. 2008
APFO
73

Mouse
(ICR)
10 M
Son et al. 2009
PFOA

21 days
(W)

0, 0.49, 2.64, FX HP
17.63, 47.21

Immuno

47.21

Marked hyperplasia in spleen white
pulp and thymic atrophy

74

Mouse
(C57BL/6N)
7–8 M
Tan et al. 2013
PFOA

3 weeks
(F)

0, 5

HP

Hepatic

5

Hepatocellular hypertrophy and
degeneration

75

Mouse
(CD-1)
4–12 F
Tucker et al. 2015
PFOA

GDs 1–17
(GW)

0, 0.01, 0.1,
0.3, 1.0

BW, DX, OW Develop

0.01

Developmental delays in the
mammary glands on PNDs 35
(26%) and 56 (30%)

76

GDs 1–17
(GW)

0, 0.01, 0.1,
0.3, 1.0

BW, DX, OW Develop

0.3

Developmental delays in the
mammary glands on PNDs 21
(38%) and 61 (25%)

Mouse
(C57BL/6)
2–6 F
Tucker et al. 2015
PFOA

0.1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

54
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
77

Mouse
(CD-1)
14–16 F

GDs 1–17 0, 5
GDs 8–17
1 time/day
GDs 12–17
(GW)

MX, DX, GN, Repro
HP

5

Delayed mammary gland
differentiation

78

Mouse
(CD-1)
14 F
White et al. 2007
APFO

GDs 1–17
1 time/day
(GW)

0, 5

DX, GN

79

Mouse
(CD-1)
28–48 F
White et al. 2009
APFO

GDs 1–17
1 time/day
(GW)

0, 3, 5

Develop

3

Delayed mammary gland
development in female pups on
PNDs 22–63 and at 18 months

80

GDs 1–17
1 time/day
(GW)

0, 1, 5

Repro

1

Delayed mammary gland
lactational differentiation in dams
on PND 22

White et al. 2007
APFO

Mouse
(CD-1)
10–12 F

Develop

Develop

5

1

5

Increased prenatal loss; 40%
reduced neonatal body weight on
PNDs 5 and 10

323% increased prenatal loss,
16.7% decreased live fetuses,
24.3% decreased neonatal survival

White et al. 2011
APFO
81

Mouse
(CD-1)
10–12 F

GDs 1–17 0, 0.0024,
1 time/day 1.0024
(GW)
GD 7 to
PND 22
(W);
3-generation
study

Repro

0.0024

Delayed mammary gland
lactational differentiation in dams
on PND 22

Develop

0.0024

Delayed mammary gland
development in female pups on
PNDs 22–63

White et al. 2011
APFO

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

55
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
82

Mouse
(CD-1)
28–48 F

GDs 1–17
1 time/day
(GW)

0, 3, 5

CS, BW, MX, Bd Wt
DX, OW
Hepatic

5
5

Develop

Significant increase in relative and
absolute maternal liver weight on
PND 22 at ≥3 mg/kg/day
3

5

3 mg/kg/day: reduced weight gain
through lactation (14.8% in males
and 20.6% in females); delayed
eye opening and hair growth at
≥3 mg/kg/day; decreased pup
survival from birth to weaning at
5 mg/kg/day

Wolf et al. 2007
APFO
83

Mouse
(ICR)
5–10 F

GDs 0–17 or 0, 1, 5, 10
GDs 0–18
(GW)

BW, FI, WI, Hepatic
OW, HP, DX

1

Renal

1

Develop

1

10

Yahia et al. 2010
PFOA

***DRAFT FOR PUBLIC COMMENT***

35% increased maternal relative
liver weight with hepatocellular
hypertrophy at ≥1 mg/kg/day;
single cell necrosis and mild
calcification at 10 mg/kg/day
35% increased maternal relative
kidney weight with renal
hypertrophy
5

14% increased neonatal mortality,
9.5% reduced fetal body weight

 PERFLUOROALKYLS

56
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
CHRONIC EXPOSURE
84

Rat
2 years
(Sprague-Dawley) ad lib
65 M,F
(F)

0, 1.5, 15

CS, FI, BW,
OW, HE, BI,
GN, HP

Bd Wt
Resp

1.5 F
1.5 M

Cardio

15

Gastro

15

Hemato

15

Hepatic

1.5

Renal

15

Ocular

15

Endocr
Immuno
Neuro
Repro

15
15
15
1.5 M

Other
noncancer
Cancer

15 F
15 M

10.3% lower terminal body weight
Lung hemorrhage

15

Increased serum ALT and AST and
hepatocellular hypertrophy at
≥1.5 mg/kg/day; hepatocellular
necrosis at 15 mg/kg/day after
1 year of exposure

15 M

Vascular mineralization in the
testes
Tubular hyperplasia in the ovaries
Inflammation of the salivary gland

1.5 F
1.5 M

3M 1983
PFOA

***DRAFT FOR PUBLIC COMMENT***

No significant increases in the
incidence of neoplastic lesions

 PERFLUOROALKYLS

57
2. HEALTH EFFECTS

Table 2-3. Levels of Significant Exposure to PFOA – Oral
Less
serious
Serious
Figure Species (strain) Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
85

Rat
(CD)
156 M

2 years
ad lib
(F)

0, 13.6

CS, BW, FI, Bd Wt
OW, GN, HP
Hepatic
Repro

Other
noncancer
Cancer

13.6
13.6
13.6

13.6

>10% reduction in weight gain
most of the study
Increased relative liver weight
Increased incidence of Leydig cell
hyperplasia; elevated serum LH at
18 months
Increased incidence of acinar cell
hyperplasia in pancreas
No significant increases in the
incidence of neoplastic lesions

Biegel et al. 2001
APFO
aThe

number corresponds to entries in Figure 2-6.
to derive an intermediate-duration oral MRL of 3x10-6 mg/kg/day based on the predicted TWA serum PFOA level of 8.29 µg/mL at the LOAEL dose and an
empirical clearance model to estimate a HED. The LOAELHED of 0.000821 mg/kg/day was divided by an uncertainty factor of 300 (10 for the use of a LOAEL, 3
for extrapolation from animals to humans with dosimetric adjustment, and 10 for human variability).
bUsed

ALT = alanine aminotransferase; APFO = ammonium perfluorooctanoate (ammonium salt of PFOA); AST = aspartate aminotransferase; BC = biochemistry; BI =
biochemical changes; BW or Bd wt = body weight; C = capsule; Cardio = cardiovascular; CS = clinical signs; Develop = developmental; DW = drinking water;
DX = developmental toxicity; EA = enzyme activity; Endocr = endocrine; (F) = feed; F = female(s); FI = food intake; FT4 = free thyroxine; FX = fetal toxicity;
G = gavage; Gastro = gastrointestinal; GD = gestation day; GN = gross necropsy; GO = gavage in oil vehicle; GW = gavage in water vehicle; HDL = high-density
lipoprotein; HE or Hemato = hematology; HED = human equivalent dose; HP = histopathology; Immuno = immunotoxicological; LD50 = lethal dose, 50% kill;
LE = lethality; LDL = low-density lipoprotein; LH = luteinizing hormone; LOAEL = lowest-observed-adverse-effect level; M = male(s); MRL = Minimal Risk Level;
Musc/skel = musculoskeletal; MX = maternal toxicity; Neuro = neurological; NK = natural killer; NOAEL = no observed-adverse-effect level; OF = organ function;
OP = ophthalmology; OW = organ weight; PFOA = perfluorooctanoic acid; PND = postnatal day; PPARα = peroxisome proliferator-activated receptor-α;
Repro = reproductive; Resp = respiratory; sRBC = sheep red blood cell; TSH = thyroid-stimulating hormone; TT4 = total thyroxine; TWA = time-weighted
average; UR = urinalysis; W = water; WI = water intake

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

58
2. HEALTH EFFECTS

Figure 2-6. Levels of Significant Exposure to PFOA – Oral
Acute (≤14 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

59
2. HEALTH EFFECTS

Figure 2-6. Levels of Significant Exposure to PFOA – Oral
Intermediate (15–364 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

60
2. HEALTH EFFECTS

Figure 2-6. Levels of Significant Exposure to PFOA – Oral
Intermediate (15–364 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

61
2. HEALTH EFFECTS

Figure 2-6. Levels of Significant Exposure to PFOA – Oral
Chronic (≥365 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

62
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral
Less
Species
serious
Serious
Figure (strain)
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
ACUTE EXPOSURE
1

Rat
Once
(Sprague(GW)
Dawley)
5–15 F
Chang et al. 2008b
PFOS potassium salt

0, 15

2

Rat
1 day
(Sprague(F)
Dawley)
30 M
Elcombe et al. 2012a
PFOS potassium salt

0, 1.97, 10.3 HP

3

0, 0, 1.72,
8.17

Rat
(SpragueDawley)
30 M

7 day
(F)

BI, OF

HP

Endocr

15

Hepatic

10.3

Endocr

10.3

Hepatic

8.17

Endocr

8.17

Hepatic

8.96

Endocr

8.96

Transient decrease in serum TT4 (24,
38, and 53% after 2, 6, and 24 hours,
respectively)

Decreased serum cholesterol (38%)
and triglyceride (55%) levels at
8.17 mg/kg/day

Elcombe et al. 2012a
PFOS potassium salt
4

Rat
(SpragueDawley)
10 M

7 day
(F)

0, 1.79, 8.96 HP

Hepatocellular hypertrophy, incr liver
weight, decreased serum cholesterol at
≥1.79 mg/kg/day

Elcombe et al. 2012b
PFOS potassium salt
5

Rat
(SpragueDawley)
NS F
Grasty et al. 2003
PFOS potassium salt

2 day
0, 25, 50
GDs 19–20
1 x/d
(GW)

BW, MX, DX Develop

***DRAFT FOR PUBLIC COMMENT***

25

Decreased neonatal survival (82% of
controls on PND 1)

 PERFLUOROALKYLS

63
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
6

Rat
(SpragueDawley)
NS F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
4 days
0, 25
GDs 2–5, 6–
9, 10–13,
14–17, or
17–20
(GW)

MX, DX

Bd Wt

25

Weight loss during treatment when
treated on GDs 2–5 (22%) or 6–9
(17%)

Develop

25

Decreased neonatal survival (90%
survival on GDs 2–5; 30% survival on
GDs 17–20)

Develop

25

Increased neonatal mortality

Grasty et al. 2003
PFOS potassium salt
7

Rat
(SpragueDawley)
NS F

2 days
0, 25, 50
GDs 19–20
1 x/d
(G)

Grasty et al. 2005
PFOS potassium salt
8

Rat
7 days
(Wistar)
ad lib
8M
(F)
Haughom and Spydevold 1992
PFOS potassium salt

0, 15

9

BW, OW, BI, Hepatic
EA

Mouse
GDs 15–18 0, 4.5, 6.5,
(wild-type
1 time/day 8.5, 10.5
129S1/Svlm) (GW)
8–20 F
Abbott et al. 2009
PFOS potassium salt

DX

10

BW, OW, DX Hepatic

Mouse
(ICR)
5–7 F

GDs 11–15 0, 50
1 time/day
(GW)

15

40% increase in absolute liver weight

Develop

Develop

4.5

50

31% reduced percentage of live pups
per litter on PND 15

103% increased maternal relative liver
weight
50

Era et al. 2009
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

6.1% increased cleft palate and 12.7%
reduced body weight in fetuses

 PERFLUOROALKYLS

64
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
11

Mouse
(CD-1)
10–11 F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
GDs 6–18
1 time/day
(GW)

0, 1.5, 3, 6

MX, DX, BW, Bd Wt
CS, OW
Hepatic

6
6

21% increase in absolute liver weight at
≥3 mg/kg/day

Endocr

6

No alterations in serum T3 or T4 levels

Develop

6

CS, BW, BH, Bd Wt
MX, DX
Develop

6

Fuentes et al. 2006
PFOS potassium salt
12

Mouse
(CD-1)
8–10 F

GDs 12–18 0, 6
(GW)

6

Reduced body weight of pups on
PNDs 4 and 8

6

Decreased distanced traveled in open
field test at 3 months of age

11.3

Altered spontaneous behavior (≤60,
87.5, or 60% changes in total activity,
rearing, and locomotion)

0.75 M

24% decreased total spontaneous
activity at 2 months of age; no
significant alteration at 4 months

Fuentes et al. 2007b
PFOS potassium salt
13

Mouse
GDs 12–18 0, 6
(CD-1)
(GW)
8–10 F
Fuentes et al. 2007a
PFOS potassium salt

DX

14

BH, BW, OF Develop

Mouse
(NMRI)
12 M pups
Hallgren et al. 2015
PFOS
15

Single dose 0, 11.3
(GO)

Mouse
Once
(CD-1)
(G)
10 M pups
Johansson et al. 2008
PFOS potassium salt

Develop

0, 0.75, 11.3 CS, OF, DX

Develop

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

65
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
16

Mouse
(CD-1)
10 F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
GDs 11–16 0, 0.5, 2.0,
(G)
8.0

BC, BW, DX, Bd Wt
HP

8.0

21% reduction in maternal body weight
gain on GDs 14–17

Repro

0.5

Develop

0.5

Decreases in mean fetal placental
weight and placental capacity
Post-implantation losses at
≥0.5 mg/kg/day; 24 and 35% reduction
fetal body weight and 31 and 52%
reduction in the number of live fetuses
at 2.0 and 8.0 mg/kg/day

2.0

Lee et al. 2015a
PFOS
17

Mouse
(BALB/c)
5F

7 days
(GW)

20

BW, OW, BC Bd Wt
Hepatic

20

41% decrease body weight gain

20

Immuno

59% increase in relative liver weight
20

Inhibition of T lymphocyte proliferation
in response to sRBC; decreased
phagocytosis by peripheral blood cells
and NK cell activity; decreased IgM
antibody formation in response to OVA

Vetvicka and Vetvickova 2013
PFOS
18

Mouse
(CD-1)
4M

14 days
1 time/day
(GO)

0, 1, 5, 10

Bd Wt

10

Hepatic

10

Repro

10

~70% increased absolute liver weight
at ≥5 mg/kg/day

Wan et al. 2011
PFOS potassium salt
19

Mouse
(C57BL/6)
10 M
Xing et al. 2016
PFOS

1 day
(GO)

0, 300, 400, BW, CS, HP, Death
500, 600, 700 LE

***DRAFT FOR PUBLIC COMMENT***

579

LD50

 PERFLUOROALKYLS

66
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
20

Mouse
(C57BL/6N)
12 M
Zheng et al. 2009
PFOS
21

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
7 days
(G)

Rabbit
GDs 6–20
(New Zealand) 1 time/day
22 F
(GW)

0, 5, 20, 40

FX

Immuno

0, 0.1, 1.0,
2.5, 3.75

MX, DX, BW, Bd wt
CS
Develop

5

0.1

Impaired response to T-cell mitogens;
suppressed response to sRBC

1F

21% decreased mean maternal body
weight gain on GDs 7–21; no effect on
food consumption

3.75

Decreased fetal body weight; 10% at
2.5 mg/kg/day and 24% at
3.75 mg/kg/day; 10/22 does aborted
between GD 22 and 28 at
3.75 mg/kg/day

1

2.5

0.15 M
0.75 F

0.75 M

13.5% reduction in final body weight

0.75

47–55% increased absolute liver
weight; 50–60% decreased serum
cholesterol; hepatocellular hypertrophy,
mild bile stasis, and lipid vacuolation at
0.75 mg/kg/day

Case et al. 2001
PFOS potassium salt

INTERMEDIATE
22

Monkey
26 weeks
(Cynomolgus) 1 time/day
4–6 M, 4–6 F (C)

0, 0.03, 0.15, CS, BW, OW, Bd Wt
0.75
HE, BI, GN,
HP
Resp

0.75

Cardio

0.75

Gastro
Hemato
Musc/skel
Hepatic

0.75
0.75
0.75
0.15

Renal
Dermal

0.75
0.75

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

67
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
Ocular
Endocr
Immuno
Neuro
Repro

0.75
0.15
0.75
0.75
0.15

0.75

Increased TSH and decreased total T3

0.75 M

No histological alterations
Significant decrease in serum estradiol
on days 62 (48%), 91 (42%), and 182
(96%); no histological alterations

Seacat et al. 2002
PFOS potassium salt

23

Monkey
4 weeks
(Cynomolgus) 1 time/day
6 M,F
(C)

0, 0.02, 2

CS, BW, FC, Bd Wt
HE, BI, GN,
Resp
HP, EA
Hemato

2

Hepatic

2

Renal

2

Ocular

2

Endocr

2

Immuno

2

No histological alteration

Repro

2

No histological alterations

Repro

1

Develop

0.3

2
2

Thomford 2002a
PFOS potassium salt

24

Rat
(SpragueDawley)
25 F

GD 0 to
PND 20
1 time/day
(GW)

0, 0.1, 0.3, 1

GD 0 to
PND 20
1 time/day
(GW)

0, 0.1, 0.3, 1

1

~30% increased locomotor activity and
concurrent failure to habituate to test
environment in male pups on PND 17

1

2.1-fold increased fetal thyroid cell
proliferation on GD 20

Butenhoff et al. 2009b
PFOS potassium salt

25

Rat
(SpragueDawley)
25 F

Develop

Chang et al. 2009
PFOS potassium salt
***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

68
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
Rat
26
(SpragueDawley)
10 F
Chen et al. 2012b
PFOS

27

Rat
(SpragueDawley)
10 M

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
GDs 1–21
1 time/day
(GW)

0, 0.1, 2

Daily
28 days
(G)

0, 5, 20

Develop

HP

0.1

Death

2

~5-fold increased postnatal mortality
and severe lung histopathology in pups

20

100% by day 26

Resp

5

Pulmonary congestion

Hepatic

5

Hepatocellular hypertrophy and focal
necrosis at ≥5 mg/kg/day; fatty
degeneration at 20 mg/kg/day

Neuro

5

Cachexia and lethargy

7.01 F

Decreased red blood cells (8.9%),
hemoglobin (10%), hematocrit (8.8%)

Cui et al. 2009
PFOS
28

Rat
(SpragueDawley)
15M, 15F

Daily
28 days
(F)

M: 0, 0.13,
1.23, 2.98,
5.89
F: 0, 0.14,
1.33, 3.47,
7.01

Cardio

5.89 M

Hemato

3.47 F

Hepatic

5.89 M

Renal

5.89 M

Endocr

0.14

Hepatic

7.34

Endocr

7.34

Increased relative liver weight;
hepatocellular hypertrophy in males at
5.89 mg/kg/day
1.23

Decreased T4 level (82% in males;
48% in females)

Curran et al. 2008
PFOS potassium salt
29

Rat
(SpragueDawley)
30 M

28 days
(F)

0, 1.54, 7.34 HP

Elcombe et al. 2012a
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

Hepatocellular hypertrophy and
decreased serum cholesterol levels at
≥1.54 mg/kg/day

 PERFLUOROALKYLS

69
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

30

Rat
Daily
(Wistar)
28 days
5–6 M
(G)
Kawamoto et al. 2011
PFOS

0, 0.12, 0.5,
2.0, 8.5

31

Rat
GDs 2–21
(Sprague(GW)
Dawley)
NS
Lau et al. 2003
PFOS potassium salt

0, 1, 2, 3, 5,
10

32

M: 0, 0.14,
1.33, 3.21,
6.34; F: 0,
0.15, 1.43,
3.73, 7.58

OW, BW

0, 0.5, 1.0,
3.0, 6.0

BC, BW, HP, Neuro
OF, OW

0.5

1

Degeneration of gonadotropic cells of
the pituitary gland at ≥1.0 mg/kg/day;
dense chromatin, condensed
ribosomes, loss of morphology in the
hypothalamus at ≥3.0 mg/kg/day

Repro

0.5

1

Loss/degeneration of spermatozoids,
marked edema in the testes

Rat
(SpragueDawley)
15M, 15F

28 days
(F)

FX

Neuro

2

Develop

1

Bd Wt

1.33 M

Hepatic

6.34 M

Immuno

6.34 M

8.5

Tonic convulsions in response to
stimuli

2

3.21 M

Reduced serum T4 in pups at
1 mg/kg/day; approximately 60%
survival at weaning versus 80% in
controls at 2 mg/kg/day

12% decrease in terminal body weight
Increased relative liver weight at
≥0.14 mg/kg/day

Lefebvre et al. 2008
PFOS
33

Rat
(SpragueDawley)
19 M

28 days
(GO)

Lopez-Doval et al. 2014
PFOS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

70
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
34

Rat
(SpragueDawley)
35 M,F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
84 days
0, 0.1, 0.4,
(6 weeks
1.6, 3.2
prior to
mating GD 0
to PND 21)
1 time/day
(GW)

MX, DX, BW, Bd Wt
OW, OF, GN, Repro
HP, FC
Develop

1.6

3.2

>10% reduction in body weight

3.2
0.1b

No alterations in mating and fertility
parameters
0.4

1.6

Delayed eye opening and transient
decrease in F2 pup body weight (13%)
on LDs 7–14 at ≥0.4 mg/kg/day;
decreased pup survival to postpartum
day 21 at ≥1.6 mg/kg/day

1.6

Increased pup mortality during
PNDs 1–4

Luebker et al. 2005a
PFOS potassium salt
35

Rat
90 day
(Sprague1 time/day
Dawley)
50 F
Luebker et al. 2005a
PFOS potassium salt
Cross-foster study
36

Rat
(SpragueDawley)
20 F

0, 1.6

MX, DX, BW,
OW, OF, GN

Develop

62–67 days 0, 0.4, 0.8, 1, MX, DX, BW, Bd Wt
42 days
1.2, 1.6, 2
CS, BI
prior to
mating
Hepatic
through
GD 20 or
PND 4
(G)
Endocr
Repro
Develop

1.6

2

22% reduction in body weight gain
during premating; food consumption
reduced 5.8%

2

16% reduction in serum total
cholesterol on PND 5 at
≥0.4 mg/kg/day Increased liver weight
in dams at ≥0.8 mg/kg/day;
0.4

46% reduction in total T4 on PND 5

2

No alteration in fertility
0.4

Luebker et al. 2005b
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

1.6

>10% decrease in mean pup weight
per litter on PND 5 at ≥0.4 mg/kg/day;
~50% decrease mean pup survival per
litter on PND 5 at ≥1.6 mg/kg/day

 PERFLUOROALKYLS

71
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
37

Rat
(SpragueDawley)
10 M

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
25 days
(GW)

0, 0.5, 1.0,
3.0 and 6.0

OF, OW

Endocr

GDs 2–6
(G)

0, 18.75

BW, OF, OW Bd Wt

0.5

Decreases in serum corticosterone
(~58%) and ACTH levels (~11%),
decrease in corticotrophin releasing
hormone levels in hypothalamus (~8%);
decrease in relative adrenal weight
(~43%),

18.75

Reduced body weight in dams;
approximately 98% on GD 8 and 33%
on GD 20

18.75

Decreased birth weight in females only
(approximately 11%); increased
systolic blood pressure in male
offspring at 7 and 52 weeks and in
female offspring at 37 and 65 weeks;
reduced nephron endowment

3

Decreased serum prolactin (~78%) and
estradiol concentrations (~18%)

Pereiro et al. 2014
PFOS
38

Rat
(SpragueDawley)
21 dams; 10–
12 M,F
pups/litter

Develop

Rogers et al. 2014
PFOS
39

Rat
(SpragueDawley)
7M
Salgado et al. 2015
PFOS

28 days
(GW)

0, 3.0, and
6.0

BI, BW, OW

Bd Wt
Endocr

6

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

72
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
40

Rat
(SpragueDawley)
25 M,F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
4 weeks
ad lib
(F)

M: 0, 0.05,
0.18, 0.37,
1.51; F: 0,
0.05, 0.22,
0.47, 1.77

CS, CO, OW, Bd Wt
HE, BI, GN, Hemato
HP
Hepatic

1.77 F
1.77 F
1.77 F

Renal

1.77 F

Ocular

1.77 F

Immuno

1.77 F

No histological alterations

Neuro

1.77 F

No histological alterations

Repro

1.51 M
1.77 F

No histological alterations

Seacat et al. 2003
PFOS potassium salt
41

Rat
(SpragueDawley)
25 M,F

14 weeks
ad lib
(F)

M:0, 0.03,
0.13, 0.34,
1.33; F:0,
0.04, 0.15,
0.4, 1.56

CS, CO, OW, Bd Wt
HE, BI, GN, Hemato
HP

1.56 F

Hepatic

1.33 M

Renal

1.56 F

Ocular

1.56 F

Endocr

1.56 F

Immuno
Neuro
Repro

1.56 F
1.56 F
1.33 M
1.56 F

0.34 M

1.33 M

Seacat et al. 2003
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

45% increase in non-segmented
neutrophils
Increased absolute and relative liver
weight; increased serum ALT;
hepatocyte hypertrophy and
vacuolation at 1.33/1.55 mg/kg/day

No histological alterations
No histological alterations
No histological alterations

 PERFLUOROALKYLS

73
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
42

Rat
(SpragueDawley)
25–50 F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
GDs 2–20
(GW)

0, 1, 2, 3, 5,
10

MX, DX, BW, Bd Wt
FI, WI, OW,
BI

2

Decreased mean body weight gain,
10% at 2 mg/kg/day and 33% at
5 mg/kg/day

Endocr

1

Reduced total and free T4 and T3

Develop

10

Increased incidences of cleft palate

Develop

0.8

Decreased spatial learning ability in
prenatally or postnatally exposed
offspring at ≥0.8 mg/kg/day and in
offspring exposure pre- and postnatally
at 2.4 mg/kg/day; decreased memory
ability in offspring exposure pre- and
postnatally at 2.4 mg/kg/day

Hepatic

1

10

Thibodeaux et al. 2003
PFOS potassium salt
43

Rat
(Wistar)
10 or 15 dams;
6–10 M,F pups

GD 1 to
0, 0.8, 2.4
PND 1,
PNDs 1–7
or 35,
or GD 1 to
PND 7 or 35

DX

Wang et al. 2015c
PFOS
44

Rat
(SpragueDawley)
10 F
Xia et al. 2011
PFOS

GDs 2–21
1 time/day
(GW)

0, 0.1, 0.6, 2

Develop

45

91 days
(W)

0, 0.27, 0.79, BC
2.37

Endocr

0.27 M

42% decrease in total T4 levels

Develop

3.2 F

19–36% reduced serum T4 levels in
pups on PNDs 21–35 after gestationand/or postnatal-only exposure

Rat
(SpragueDawley)
8–10 M

0.6

2

Yu et al. 2009a
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

5-fold increased neonatal mortality on
PNDs 1–3

 PERFLUOROALKYLS

74
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
46

Mouse
(E3L CETP)
6–8 M

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
4–6 weeks
(F)

0, 3

BW, FI, BC,
OW

Bd wt

3

Hepatic

3

60 days
(G)

0, 0.00833,
0.08333,
0.41667,
0.83333,
2.0833

FX

Immuno

0.0083

0.0833

Impaired response to sRBC

60 days
(G)

0, 0.00833,
0.0167,
0.0833,
0.4167,
0.8333

FX

Immuno

0.0167

0.0833

Impaired response to sRBC

GDs 1–17
1 time/day
(GW)

0, 20

Bd Wt

20

35% reduced maternal body weight

Develop

20

89% increased cleft palate, 25%
reduced body weight in fetuses

3

Impaired retention of the task in the
water maze test

Decreased plasma triglyceride, total
cholesterol, non-HDL cholesterol, and
HDL cholesterol levels; increased
hepatic triglyceride levels, increased
liver weight

Bijland et al. 2011
PFOS
47

Mouse
(C57BL/6N)
10 M

Dong et al. 2009
PFOS potassium salt
48

Mouse
(C57BL/6N)
12 M

Dong et al. 2011
PFOS potassium salt
49

Mouse
(ICR)
6–8 F

Era et al. 2009
PFOS potassium salt
50

Mouse (CD-1) 4 weeks
10 M
(GW)
PFOS potassium salt
Fuentes et al. 2007c

0, 3, 6

CS, BW, BH, Develop
MX, DX

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

75
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
51

Mouse
(B6C3F1)
30 F
Guruge et al. 2009
PFOS

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
21 days
(G)

52

0, 0.005,
0.025

Immuno

Mouse
GDs 1–17
(B6C3F1)
1 time/day
10–12 F
(GW)
Keil et al. 2008
PFOS potassium salt

0, 0.1, 1, 5

MX, DX, BW, Develop
CS, OF

53

GDs 1–17
1 time/day
(GW)

0, 1, 5, 10,
15, 20

DX, OW, BI

Develop

GDs 1–17
(GO)

3

BC, EA, OF

Hepatic

Mouse
(CD-1)
21–22 F

0.005

0.025

Decreased host resistance to influenza
virus

0.1

1

42.5% reduced NK cell activity in male
pups at 8 weeks of age

1

10

1 mg/kg/day: delayed eye opening at
≥1 mg/kg/day, ~ 0% postnatal survival
at weaning versus 90% in controls at
10 mg/kg/day

Lau et al. 2003
PFOS potassium salt
54

Mouse
(CD-1)
4 dams;
8 fetuses

3

Develop

No alteration in maternal hepatic lipid
levels
3

Significant increase in cholesterol
levels in fetal livers

2.15

Impaired spatial learning and memory

Lee et al. 2015b
PFOS
55

Mouse
(C57BL/6)
15 M,F
Long et al. 2013
PFOS

Daily
3 months
(G)

0, 0.43, 2.15, FX
10.75

Neuro

0.43

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

76
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

56

3 weeks

Mouse
(BALB/c)
28 M

0, 2.5, 5 and BW, OF, OW Bd Wt
10
Immuno

10 M

~15% reduction in body weight during
the recovery period
~36% decrease in spleen index during
recovery; ~15% inhibition in Con Ainduced T-cell proliferation during
treatment; 32% increase in CD3+ cells
after recovery; ~60% increase in
CD3+CD8+ cells and ~56% increase in
CD3+CD4+ cells on week 2; 15%
inhibition in Con A-induced T-cell
proliferation during recovery

2.5 M

5M

0.1

3

Decrease in number of successful
births

0.3

Decreased locomotion, muscle strength
and motor coordination in adult
offspring

0.00166 M

Suppressed response to sRBC (~60%)

Lv et al. 2015
PFOS
57

Mouse
(C57BL/6JApc+/+)
20–21 F

GDs 1–7
(GW)

0, 0.01, 0.1,
3.0

BC, BW, FI,
HE, OW

Develop

Ngo et al. 2014
PFOS
58

Mouse
GDs 1–21
(C57BL/6/
ad lib
Bk1)
6F
Onishchenko et al. 2011
PFOS, potassium salt

0, 0.3

Develop

59

0, 0.000166, OW, FX
0.00166,
0.00331,
0.0166,
0.0331, 0.166

Immuno

Mouse
(B6C3F1)
5M, 5F

28 days
(G)

0.000166 M

Peden-Adams et al. 2008
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

77
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
60

Mouse
(B6C3F1)
5M

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
28 days
(F)

0.20

NS

Bd Wt

0.2

21% reduction in body weight

Immuno

0.2

No alterations in thymic lymphocyte
phenotypes, response to sRBC, or IgM
antibodies to LPS

Develop

5

Peroxisome proliferation in fetal liver at
≥5 mg/kg/day

Qazi et al. 2010b
PFOS
61

Mouse
GDs 1–17
(CD-1)
1 time/day
5F
(GW)
Rosen et al. 2009
PFOS, potassium salt

0, 5, 10

62

0, 1, 5, 10,
15, 20

Mouse
(CD-1)
60–80 F

GDs 1–17
(GW)

BW, OW, BI, Bd Wt
DX
Hepatic

20

Endocr

15

20

Develop

1

5

BW, BI, OW, Bd Wt
HP
Hepatic

5

10

20

Increase in absolute and relative liver
weight and decreased serum
triglycerides at ≥5 mg/kg/day
Decreased total T4 on GD 6
20

Increased incidences of sternal defects
at ≥5 mg/kg/day; reduced percentage
of live fetuses (9%) at 20 mg/kg/day

Thibodeaux et al. 2003
PFOS potassium salt
63

Mouse
(CD-1)
4M

21 days
1 time/day
(GO)

0, 1, 5, 10

Repro

10
5

~15% reduced body weight
Increased absolute liver weight at
≥5 mg/kg/day

10

Wan et al. 2011
PFOS potassium salt

***DRAFT FOR PUBLIC COMMENT***

~17% reduced serum testosterone,
~38% reduced epididymal sperm count

 PERFLUOROALKYLS

78
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
64

Mouse
(CD-1)
6 F (dams)

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
GD 3 to
PND 21 or
GD 3 to
PND 63
(GO)

0, 0.3, 3

BW, OF, OW Hepatic

3

~24% increase in relative liver weight in
dams at 3 mg/kg/day

Develop

3

Increase in relative liver weight at
3 mg/kg/day in male and female pups
on PND 21 (~20–32%), in male STDfed adults (~11%), and in male HDFfed adults (~33%)

Develop

0.3

In PND 63 offspring fed a high fat diet,
increased serum glucose levels at
≥0.3 mg/kg/day in females (~40%) and
3 mg/kg/day in males (~8%); increased
serum insulin in males (~109%) and
females (~85%)increased response to
oral glucose tolerance test, increased
HOMA-IR index, and 33% increased
relative liver weight at 3 mg/kg/day
No significant alteration in fasting
serum insulin or glucose levels;
significant increase in HOMA-IR index
at 0.3 and 3 mg/kg/day

10

31% reduction in body weight
(correlated with 68% reduction in feed
consumption)

2.5

Increased liver weight (35%) and
serum AST(~12%) and GGT levels
(~98%) at ≥2.5 mg/kg/day; increases in
ALT (~45%) and ALP (~36%) at
≥5 mg/kg/day; cytoplasmic vacuolation,
focal or flake-like necrosis, and
hepatocellular hypertrophy observed,
but no incidence data provided

Other
3F
noncancer

Wan et al. 2014b
PFOS
65

Mouse
(C57BL/6)
10 M

30 days
(GO)

0, 2.5, 5, 10

BW, HP, OF, Bd Wt
OW
Hepatic

Renal

10

Xing et al. 2016
PFOS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

79
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group
66

Mouse
(ICR)
5F

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
GDs 0–17
GDs 0–18
(GW)

0, 1, 10, 20

Hepatic

20

Develop

60% increased absolute liver weight at
≥10 mg/kg/day
1

Develop

20

GDs 0–17: 15.8% increased sternal
defects in fetuses at ≥1 mg/kg/day;
8.8% decrease in number of live
fetuses at 20 mg/kg/day

10

GDs 0–18: decreased survival (55.2%)
at 10 mg/kg/day on PND 4, decreased
neonatal BW, intracranial blood vessel
dilatation, lung atelectasis

Yahia et al. 2008
PFOS potassium salt
CHRONIC EXPOSURE
67
Rat
104 weeks
(Spraguead lib
Dawley)
(F)
70 M,F

0, 0.025,
0.10, 0.25,
1.04

CS, BW, FC, Bd Wt
GN, HP, BI

0.25 F

Resp

1.04

Cardio

1.04

Gastro

1.04

Hemato

1.04

1.04 F

14% reduction in final body weight

1.04 M

Hepatocellular hypertrophy at
≥0.1 mg/kg/day; single cell necrosis
and cystic degeneration at
1.04 mg/kg/day

Musc/skel 1.04
Hepatic

0.25 M

Renal
Dermal
Ocular
Endocr
Immuno

1.04
1.04
1.04
1.04
1.04

***DRAFT FOR PUBLIC COMMENT***

No histological alterations

 PERFLUOROALKYLS

80
2. HEALTH EFFECTS

Table 2-4. Levels of Significant Exposure to PFOS – Oral

Species
Figure (strain)
keya No./group

Less
serious
Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
Neuro
Repro

1.04
1.04

No histological alterations
No histological alterations

Butenhoff et al. 2012b; Thomford 2002b
PFOS potassium salt
aThe

number corresponds to entries in Figure 2-7.
to derive an intermediate-duration oral MRL of 2x10-6 mg/kg/day based on the predicted TWA serum PFOA level of 29.7 µg/mL at the NOAEL dose and an
empirical clearance model to estimate a HED. The NOAELHED of 0.000515 mg/kg/day was divided by an uncertainty factor of 30 (3 for extrapolation from animals
to humans with dosimetric adjustment and 10 for human variability) and a modifying factor of 10 for concern that immunotoxicity may be a more sensitive endpoint
than developmental toxicity.
bUsed

ALP = alkaline phosphatase; ALT = alanine aminotransferase; AST = aspartate aminotransferase; BC = biochemistry; BH = behavioral; BI = biochemical changes;
BW or Bd wt = body weight; C = capsule; Cardio = cardiovascular; CS = clinical signs; Develop = developmental; DX = developmental toxicity; EA = enzyme
activity; Endocr = endocrine; (F) = feed; F = female(s); FX = fetal toxicity; FI = food intake; FX = fetal toxicity; G = gavage; Gastro = gastrointestinal;
GD = gestation day; GGT = gamma-glutamyl transferase; GN = gross necropsy; GO = gavage in oil vehicle; GW = gavage in water vehicle; HDL = high-density
lipoprotein; HE or Hemato = hematology; HED = human equivalent dose; HOMA IR = Homeostatic Model Assessment of Insulin Resistance; HP = histopathology;
Immuno = immunotoxicological; LD = lactation day; LD50 = lethal dose, 50% kill; LE = lethality; LOAEL = lowest-observed-adverse-effect level;
LPS = lipopolysaccharide; M = male(s); MRL = Minimal Risk Level; Musc/skel = musculoskeletal; MX = maternal toxicity; Neuro = neurological; NK = natural killer;
NOAEL = no observed-adverse-effect level; NS = not specified; OF = organ function; OP = ophthalmology; OW = organ weight; PFOS = perfluorooctane sulfonic
acid; PND = postnatal day; Repro = reproductive; Resp = respiratory; sRBC = sheep red blood cell; T3 = triiodothyronine; T4 = thyroxine; TSH = thyroidstimulating hormone; TT4 = total thyroxine; TWA = time-weighted average; W = water

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

81
2. HEALTH EFFECTS

Figure 2-7. Levels of Significant Exposure to PFOS – Oral
Acute (≤14 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

82
2. HEALTH EFFECTS

Figure 2-7. Levels of Significant Exposure to PFOS – Oral
Intermediate (15–364 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

83
2. HEALTH EFFECTS

Figure 2-7. Levels of Significant Exposure to PFOS – Oral
Intermediate (15–364 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

84
2. HEALTH EFFECTS

Figure 2-7. Levels of Significant Exposure to PFOS – Oral
Chronic (≥365 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

85
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Less serious Serious
Figure (strain)
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
ACUTE EXPOSURE
PFHxS
1

Mouse
7 days
(SV129 WT (G)
and PPARα
null)
4M

0, 10
PFHxS

BW, HP

Bd Wt

10

Hepatic

10

Develop

6.1

9.2

Altered spontaneous behavior and
habituation in adults exposed as
neonates

1

3

Decreases in body weight at 3 and
5 mg/kg/day (18 and 39%)

1

24% increase in relative thymus weight
at 1 mg/kg/day; 20% decrease in thymus
weight at 3 or 5 mg/kg/day, increases in
thymic cortex:medulla ratios, alterations
in cytokine levels at ≥3 mg/kg/day

5

Decreased relative spleen weight and
increases in cytokine levels

Hepatocellular hypertrophy, steatosis,
and increased hepatic triglyceride levels

Das et al. 2017
2

Mouse
(NMRI)
14–18 M,F
Viberg et al. 2013

PND 10
Once
(GO)

0, 0.61, 6.1,
9.2
PFHxS

14 days
(GW)

0, 1, 3, 5
PFNA

PFNA
3

Rat
(SpragueDawley)
10 M

BC, BW, OW Bd wt
Immuno

Fang et al. 2009
4

Rat
(SpragueDawley)
10 M
Fang et al. 2010

14 days
(GW)

0, 1, 3, 5
PFNA

BC, OW

Immuno

3

5

14 days
(GW)

0, 0.2, 1, 5
PFNA

BC

Hepatic

5

14 days
(GW)

0, 0.2, 1, 5
PFNA

HP

Rat
(SpragueDawley)
6M

Other
0.2
noncancer
(glucose)

Decreased HDL levels at ≥1 mg/kg/day
1

Increased serum glucose levels
(1.11-fold at 1 mg/kg/day and 1.16-fold at
5 mg/kg/day)

Fang et al. 2012a
6

Rat
(SpragueDawley)
6M
Fang et al. 2012b

Hepatic

5

***DRAFT FOR PUBLIC COMMENT***

Hepatocellular vacuolation and lipid
accumulation at 5 mg/kg/day

 PERFLUOROALKYLS

86
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

7

14 days
(GW)

0, 1, 3, 5
PFNA

BC, HP

Repro

3

5

85.4% decrease in serum testosterone
and 105% increase in estradiol levels at
5 mg/kg/day; atrophy of seminiferous
tubule epithelium

14 days
(GW)

0, 1, 3, 5
PFNA

HP

Repro

3

5

Large vacuoles between testicular Sertoli
cells and spermatogonia

14 days
(GO)
8 or 10M

0, 0.0125,
0.25, 5
PFNA

BI, BW, OW, Bd Wt
GN, HP

5

Decreased body weight; magnitude of
effect was not reported

5

Decreased androstenedione and
testosterone concentrations (data not
shown)

Mouse
7 days
(SV129 WT (G)
and PPARα
null)
4M
Das et al. 2017

0, 10
PFNA

BW, HP

11

14 days
(G)

0, 1, 3, 5
PFNA

FX

Immuno

14 days
(F)

0, 0.5, 1.8,
5.3, 54, 537
PFNA

LE, OW

Death

Rat
(SpragueDawley)
6M

Feng et al. 2009
8

Rat
(SpragueDawley)
6M

Feng et al. 2010
9

Rat
(Wistar)

Endocr

Hadrup et al. 2016
10

Mouse
(BALB/c)
6M

Bd Wt

10

Hepatic

10

Hepatocellular hypertrophy, steatosis,
and increased hepatic triglyceride levels

1

Decreases in the percentages of F4/80+
and CD49b+ cells in the spleen; no
alteration in the response of splenic
lymphocytes to ConA at ≤5 mg/kg/day

Fang et al. 2008
12

Mouse
(CD-1)
5 M,F

Hepatic

54
5.3

Kennedy 1987

***DRAFT FOR PUBLIC COMMENT***

100% mortality before day 14
50–70% increase in absolute liver weight
at ≥0.5 mg/kg/day

 PERFLUOROALKYLS

87
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

13

14 days
(G)

0, 0.2, 1, 5
PFNA

BW, OW, BC Bd Wt

Once
(GO)

0, 50
PFDeA

BW, FX

Mouse
(BALB/c)
8M

Hepatic

1

5

~25% decrease in body weight

1

5

Increases in liver weight (~160%) and
increases in hepatic triglyceride (~66%)
and cholesterol levels (~26%) at
≥0.2 mg/kg/day; decreases in serum
triglyceride (~67%) and cholesterol levels
(~32%) and increases in serum ALT
(~900%) and AST (~280%) levels at
5 mg/kg/day

Wang et al. 2015a

PFDeA
14

Rat
(Wistar)
5–12 M

BW

50

Neuro

50

BE, OW, EA Bd Wt

4.7

8% weight loss was observed 10-days
post-exposure
No alteration in performance on novel
object recognition test

Kawabata et al. 2017
15

Rat
(Wistar)
25 M

1 week
(F)

0, 1.2, 2.4,
4.7, 9.5
PFDeA

9.5

~32% weight loss

Hepatic

9.5

Increases in liver weight at
≥2.4 mg/kg/day; increases in hepatic
cholesterol at 9.5 mg/kg/day

Hepatic

160

Increase in hepatic lipids and liver weight
2 days post-exposure at 40 mg/kg/day

Kawashima et al. 1995
16

Mouse
Once
(C57BL/6N) (GO)
4F

0, 40, 80,
OW, EA
100, 120, 160
PFDeA

Brewster and Birnbaum 1989
17

Mouse
GDs 6–15
(C57BL/6N) (GO)
12–14 F

0, 0.03, 0.1, MX, DX, BW, Bd Wt
0.3, 1, 3, 6.4, OW
12.8
PFDeA
Develop

3

6.4

12.8

No weight gain at 6.4 mg/kg/day and
weight loss at 12.8 mg/kg/day (net
change of -2.4)

0.3

1

12.8

18–22% decreases in fetal weight per
litter at ≥1 mg/kg/day; decreases in live
fetuses per litter at 12.8 mg/kg/day
(4.6 versus 7.2 in controls)

Harris and Birnbaum 1989

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

88
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group
18

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

Mouse
Once
(C57BL/6N) (GO)
10 F

0, 20, 40, 80, BW, OW,
160, 320
GN, HP
PFDeA

Death

120
80

LD50 in 30-day observation period

Bd Wt

40

12% decreased body weight 30 days
post-exposure

Cardio

80

No histological alterations in the heart
30 days post-exposure; decreased
relative heart weight at 80 mg/kg/day

Hepatic

80

Increases in liver weight and pancellular
hypertrophy at ≥20 mg/kg/day 30 days
post-exposure

Renal

80

No histological alterations 30 days postexposure

Endocr

40

80

Immuno

40

80

Develop

10.8

2-fold increase in T3 and 4-fold increase
in T4 levels 30 days post-exposure
160

28% decrease in relative spleen weight
at 80 mg/kg/day; atrophy and lymphoid
depletion in thymus and spleen at
160 mg/kg/day

Harris et al. 1989
19

Mouse
Once on
(CD-1)
PND 10
10 M
(G)
Johansson et al. 2008

0, 0.72, 10.8 CS, DX
PFDeA

20

0, 78
PFDeA

Mouse
10 days
(C57BL/6N) (F)
4M
Permadi et al. 1992, 1993

BW, OW, EA Bd Wt
Hepatic

78
78

PFBA
21

Rat
(SpragueDawley)
3 M,F

5 days
1 time/day
(GW)

No alteration in spontaneous activity or
habituation at 2–4 months of age

0, 18, 58, 184 CS, BW, OW, Bd Wt
PFBA
HE, BI, GN, Resp
HP
Cardio

184

Gastro

184

Hemato

184

184
184

Musc/skel 184
Hepatic

184

Renal

184

***DRAFT FOR PUBLIC COMMENT***

33% weight loss
36% increase in liver weight

 PERFLUOROALKYLS

89
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
Endocr

184

Immuno

184

No histological alterations

Neuro

184

No histological alterations

Repro

184

No histological alterations

Hepatic

20

Biochemical and ultrastructural evidence
of peroxisome proliferation

3M 2007a
22

Rat
(SpragueDawley)
3M
Ikeda et al. 1985

14 days
ad lib
(F)

0, 20
PFBA

OW, EA

Mouse
10 days
(C57BL/6N) (F)
4M
Permadi et al. 1992, 1993

0, 78
PFBA

BW, OW, EA Bd Wt

0, 5, 20, 50
PFDoA

BW, FX

23

Hepatic

78
78

63% increase in absolute liver weight

PFDoA
24

Rat
(Wistar)
5–12 M

Once
(GO)

BW

50

Neuro

50

44% decrease in body weight gain
(measured 10-days post-exposure)
Impaired performance on novel object
recognition test

Kawabata et al. 2017
25

Rat
(Wistar)
10 M

Once
(GO)

0, 50
PFDoA

FX

Neuro

50

No alterations in open field activity

0, 50
PFDoA

FX

Neuro

50

No alterations in tests of working memory
or depressive behavior.

Kawabata et al. 2017
26

Rat
(Wistar)
8M

Once
(GO)

Kawabata et al. 2017

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

90
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

27

14 days
1 time/day
(GW)

Rat
(SpragueDawley)
10 M

Shi et al. 2007
28
Rat
(SpragueDawley)
10 M

14 days
1 time/day
(GW)

0, 1, 5, 10
PFDoA

0, 1, 5, 10
PFDoA

BW, OW, BI

OW, BI, HP

Bd Wt

1

Hepatic

10

Repro

1

Hepatic

5

25% reduction in final body weight
35% increase in total serum cholesterol
at 10 mg/kg/day

5

10

Decreased serum testosterone (38%)
and estradiol (~38%), and ultrastructural
alterations in testes at ≥5 mg/kg/day;
decreased testicular weight at
10 mg/kg/day (22%)
Increased liver weight, increase in
increased hepatic triglyceride and
cholesterol levels at ≥5 mg/kg/day;
increased serum triglyceride levels at
10 mg/kg/day

Zhang et al. 2008

PFOSA
29

Rat
(SpragueDawley)
15 M

Once
(GO)

0, 5
PFOSA

BW, OW

Bd Wt

5

Hepatic

5

0, 100, 350,
500
PFHxA

CS, DX

Develop

100

350

12.5% decrease in birth weight and
delayed eye opening at ≥350 mg/kg/day;
increased pup mortality (PND 0–3) and
decreased pup survival at 500 mg/kg/day

0, 7, 35, 175 CS, DX
PFHxA

Develop

35

175

Increase in stillborn pups and 12.5%
decrease in birth weight

No alterations in liver weight

Seacat and Luebker 2000

PFHxA
30

Mouse
(CD1)
20 F

GDs 6–18
(GW)

Iwai and Hoberman 2014
31

Mouse
GDs 6–18
(CD1)
(GW)
20 F
Iwai and Hoberman 2014

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

91
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Less serious Serious
Figure (strain)
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
keya No./group parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
INTERMEDIATE EXPOSURE
PFHxS
32

Rat
(SpragueDawley)
15 M,F

42–56 days 0, 0.3, 1, 3,
1 time/day 10
(GW)
PFHxS

CS, BW, MX, Bd Wt
DX, OW, OF, Resp
BI, HE, GN,
Cardio
HP, FI
Gastro

10
10
10
10

Hemato

10 F

0.3 M

6% increase in prothrombin time in males
at ≥0.3 mg/kg/day; decreases in
hemoglobin (4%), hematocrit (3%), and
RBC levels (2.4%) in males at
≥3 mg/kg/day

Hepatic

10

Renal

3M

10 M

31% increase in BUN in males

Endocr

1b M
10 F

3M

Hypertrophy and hyperplasia of thyroid
follicular cells in males

Immuno

10

No histological alterations

Neuro

10

No histological alterations

Repro

10

No histological alterations or effects on
fertility

Develop

10 F

Bd wt

6

Hepatic

6

Increased liver weight; centrilobular
hepatocellular hypertrophy in males at
≥3 mg/kg/day

Butenhoff et al. 2009a; Hoberman and York 2003
33

Mouse
4–6 weeks
(E3L CETP) (F)
6–8 M

0, 6
PFHxS

BW, FI, BC,
OW

Bijland et al. 2011

***DRAFT FOR PUBLIC COMMENT***

Decreased plasma triglyceride,
cholesterol, non-HDL cholesterol, and
HDL cholesterol levels; increased hepatic
triglyceride levels, increased liver weight

 PERFLUOROALKYLS

92
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group
PFNA

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

34

GDs 1–20
(GW)

Rat
(SpragueDawley)
18 dams;
10–12 M,F
pups/litter

5
PFNA

BW, OF, OW Bd Wt

Develop

5

Reduced maternal body weight;
approximately 33% on GD 7 and 10% on
GD 20

5

Decreased birth weight in female pups
only (approximately 11%); increased
systolic blood pressure in 10-week-old
male and female offspring; reduced
nephron endowment

Rogers et al. 2014
35

Mouse
(CD-1)
8–10F

GDs 1–17
(GW)

1, 3, 5, 10
PFNA

BC, DX, FX,
MX, OW

Bd Wt

10

Hepatic

10

Develop

1c

43% maternal weight loss at GD 13
Increases in absolute and relative liver
weights in dams on GD 17 and on postweaning day 28 at ≥1 mg/kg/day

3

5

Delayed postnatal development [eye
opening, preputial separation and vaginal
opening] and decreased body weight
gain persisting in males up to PND 287
at ≥3 mg/kg/day; decreased postnatal
survival between PND 2 and 10 with 80%
mortality at ≥5 mg/kg/day; full litter
resorptions at 10 mg/kg/day

Das et al. 2015
36

Mouse
(129S1/
Svlm)
8F

GDs 1–18
(G)

0, 0.83, 1.1,
1.5, 2.0
PFNA

BW, OW, DX Bd Wt

2.0

No alterations in maternal body weight or
gestational weight gain

Hepatic

2.0

Increase in dam liver weight at
≥0.83 mg/kg/day

Develop

0.83c

1.1

Wolf et al. 2010

***DRAFT FOR PUBLIC COMMENT***

Decreased number of live births (36 and
31%) and pup survival at 1.1 and
2.0 mg/kg/day, but not 1.5 mg/kg/day;
decreased number of live pups per litter
and decreased pup body weight gain in
females at 2 mg/kg/day; increased pup
liver weight at ≥0.83 mg/kg/day

 PERFLUOROALKYLS

93
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

37

GDs 1–18
(G)

Mouse
(PPARα
knockout)
8F

0, 0.83, 1.1,
1.5, 2.0
PFNA

BW, OW, DX Bd Wt

2.0

No alterations in maternal body weight or
gestational weight gain

Hepatic

2.0

Increases in liver weight at
≥1.5 mg/kg/day non-pregnant adults, but
not in the dams

Develop

2.0

No alterations in the number of live pups
per litter, birth weight, pup survival, day
of eye opening, or pup body weight gain;
no increases in pup liver weight were
observed

Wolf et al. 2010

PFUA
38

Rat
41–46 days 0, 0.1, 0.3,
(Crl:CD[SD]) (GO)
1.0
12 M,F
PFUA
(main);
5 M,F
(other)

BH, BW, CS, Bd Wt
HP, OF, OW,
UR

0.3

1.0

Decreased body weight (~10%) in males
during exposure and recovery and in
satellite females during dosing (~23% on
day 40) and recovery (~10%)

Hemato

0.3

1.0

Hepatic

1.0

Renal

0.3

1.0

Develop

0.3 F

1.0 F

Main study males: decreased MCV (5%),
MCH (5%), APTT (25%), and fibrinogen
(33%) and increased platelet counts
(7%); satellite males: increased WBC
(52%) and decreased APTT (16%) and
fibrinogen (19%); main study females:
increased MCV (10%) and MCH (10%)
and decreased fibrinogen (32%)
Increased absolute and relative liver
weight and centrilobular hypertrophy at
1.0 mg/kg/day
Increased BUN (61%) and ALP (140%);
decreased total protein (11%) and
albumin (7%) in main group males
Decreased body weight in pups on
PNDs 0 and 4 (13–19% in males and
12–16% in females)

Takahashi et al. 2014

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

94
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group
PFBuS

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

39

28 days
1 time/day
(GO)

Rat
(SpragueDawley)

0, 100, 300,
900
PFBuS

CS, BW, FI,
HE, BI, GN,
HP, OF

Bd Wt

900

Resp
Cardio
Gastro
Hemato
Musc/Skel
Hepatic

900
900
900
900
900
900

Renal
Ocular
Endocr
Immuno
Neuro
Repro

900
900
900
900
900
900

Increased absolute and relative liver
weight at 900 mg/kg/day

No histological alterations
No histological alterations
No histological alterations

3M 2001
40

Rat
(SpragueDawley)
10 NS

90 days
(G)

0, 60, 200,
600
PFBuS

LT, BW, OW, Resp
GN, HP, BC, Cardio
CS, BI, BH,
Gastro
HE
Hemato

600
600
200

600

60 M

200 M

Necrosis of individual squamous cells in
forestomach and hyperplasia and
hyperkeratosis of limiting ridge
Decreased hemoglobin (4.9%) and
hematocrit (5.2%)

Musc/skel 600
Hepatic

600

Renal

200

Endocr

600

Neuro

600

600

Lieder et al. 2009a

***DRAFT FOR PUBLIC COMMENT***

Hyperplasia of medullary and papillary
tubules and medullary ducts and focal
papillary edema

 PERFLUOROALKYLS

95
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

41

P0: starting 0, 30, 100,
70 days
300, 1,000
PFBuS
prior to
mating;
F1: starting
at weaning
(G)

Rat
(SpragueDawley)

BW, OW, FI, Bd Wt
GN, HP, FX, Hepatic
MX, DX, TG

1,000
1,000

Renal

100

Repro

1,000

Develop

1,000

Bd wt

30

Hepatic

30

Increased liver weight in males at
≥300 mg/kg/day
300

Medullary/papillary tubular and ductal
hyperplasia in P0 and F1

Lieder et al. 2009b
42

Mouse
4–6 weeks
(E3L CETP) (F)
6–8 M

0, 30
PFBuS

BW, FI, BC,
OW

Decreased plasma triglyceride levels

Bijland et al. 2011

PFBA
43

Rat
(SpragueDawley)
10 M,F

28 days
1 time/day

0, 6, 30, 150 CS, BW, OW, Bd Wt
PFBA
FI, BI, HE,
GN, HP, OF Resp

150

Cardio

150

Gastro

150

Hemato

150

150

Musc/skel 150
Hepatic

150

Renal

150

Dermal
Ocular
Endocr

150
150
6M

Immuno
Neuro
Repro

150
30 M
150

Increased absolute and relative liver
weight and decreased serum cholesterol
in males at ≥30 mg/kg/day;
hepatocellular hypertrophy in males at
150 mg/kg/day

30 M

150 M

Butenhoff et al. 2012a; van Otterdijk 2007a
***DRAFT FOR PUBLIC COMMENT***

Hyperplasia/hypertrophy of follicular
epithelium of the thyroid
No histological alterations
Delayed pupillary reflex
No histological alterations

 PERFLUOROALKYLS

96
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

44

90 days
1 time/day

Rat
(SpragueDawley)
10 M,F

0, 1.2, 6, 30
PFBA

CS, BW, FI,
OW, BI, HE,
GN, HP

Bd Wt

30

Resp

30

Cardio

30

Gastro

30

Hemato

6

30 M

Reduced erythrocyte counts (3.8%),
hemoglobin (5.7%), and hematocrit
(4.5%)

Musc/skel 30
Hepatic

30

Renal
Dermal
Ocular
Endocr

30
30
30
6M

Immuno
Neuro
Repro

30
30
30

Diffuse panlobular hepatocyte
hypertrophy at 30 mg/kg/day in males

30 M

Hypertrophy/hyperplasia of follicular
epithelium of the thyroid gland
No histological alterations
No histological alterations
No histological alterations

Butenhoff et al. 2012a; van Otterdijk 2007b
45

Mouse
(CD-1)

18 days
GDs 1–17
1 time/day
(GW)

0, 35, 175,
350
PFBA

BW, MX, DX Hepatic

350

Develop

Significant increase in absolute and
relative liver weight at ≥175 mg/kg/day

35

Eye opening delayed approximately
1 day

3

40% reduced serum estradiol and 20%
increased serum cholesterol in pubertal
females

Das et al. 2008

PFDoA
46

Rat
(SpragueDawley)
8F
Shi et al. 2009

28 days
0, 0.5, 1.5, 3
PNDs 24–72 PFDoA
1 time/day
(GW)

Develop

1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

97
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect

CHRONIC EXPOSURE

PFHxA
47

Rat
104 weeks
(Sprague(GW)
Dawley)
60 or 70 M,F

M: 0, 2.5, 15, BC, BW, CS, Death
100 F: 0, 5, GN, HP, LE,
30, 200
OP, OW, UR
PFHxA
Bd Wt
Hemato

Hepatic

200 F

100 M
200 F
100 M

100 M
30F

200 F

200 F

***DRAFT FOR PUBLIC COMMENT***

36, 43, 33, and 22% survival rate in
females at 0, 5, 30, and 200 mg/kg/day,
respectively

8.1% reduction in mean red blood cell
count and 5.2% reduction in hemoglobin
at 51 weeks; 23.6 and 53.6% increase in
reticulocyte counts at weeks 25 and 51,
respectively
Males: 42% decrease in triglycerides,
19% decrease in free fatty acids in males
at 100 mg/kg/day; hepatocellular
necrosis; 66% increase in triglycerides,
44% decrease in non-HDL cholesterol in
females at 200 mg/kg/day

 PERFLUOROALKYLS

98
2. HEALTH EFFECTS

Table 2-5. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Species
Figure (strain)
keya No./group

Less serious Serious
Exposure Doses
Parameters
NOAEL
LOAEL
LOAEL
parameters (mg/kg/day) monitored Endpoint (mg/kg/day) (mg/kg/day) (mg/kg/day) Effect
Renal

100 M

Neuro

100 M
200 F

200 F

Mild renal tubular degeneration and mild
to severe papillary necrosis; increased
mean urine volume (109%) and reduced
specific gravity (0.96%)

Klaunig et al. 2015
aThe

number corresponds to entries in Figure 2-8.
to derive an intermediate-duration oral MRL of 2x10-5 mg/kg/day for PFHxS based on a measured serum PFHxS level of 89.12 µg/mL at the NOAEL dose
and an empirical clearance model to estimate a HED. The NOAELHED of 0.0047 mg/kg/day was divided by an uncertainty factor of 30 (3 for extrapolation from
animals to humans with dosimetric adjustment and 10 for human variability) and a modifying factor of 10 for database deficiencies.
cUsed to derive an intermediate-duration oral MRL of 3x10-6 mg/kg/day for PFNA based on a measured serum PFNA level of 8.91 µg/mL at the NOAEL dose and
an empirical clearance model to estimate a HED. The NOAELHED of 0.001 mg/kg/day was divided by an uncertainty factor of 30 (3 for extrapolation from animals
to humans with dosimetric adjustment and 10 for human variability) and a modifying factor of 10 for database deficiencies.
bUsed

ALT = alanine aminotransferase; APTT = activated partial thromboplastin time; AST = aspartate aminotransferase; BC = biochemistry; BH = behavioral;
BI = biochemical changes; BUN = blood urea nitrogen; BW or Bd wt = body weight; Cardio = cardiovascular; CI = confidence interval; CS = clinical signs;
Develop = developmental; DX = developmental toxicity; EA = enzyme activity; Endocr = endocrine; (F) = feed; F = female(s); FI = food intake; FX = fetal toxicity;
G = gavage; Gastro = gastrointestinal; GD = gestation day; GN = gross necropsy; GO = gavage in oil vehicle; GW = gavage in water vehicle; HDL = high-density
lipoprotein; HE or Hemato = hematology; HP = histopathology; Immuno = immunotoxicological; LD50 = lethal dose, 50% kill; LE = lethality; LOAEL = lowestobserved-adverse-effect level; M = male(s); MCH = mean corpuscular hemoglobin; MCV = mean corpuscular volume; Musc/skel = musculoskeletal;
MX = maternal toxicity; Neuro = neurological; NOAEL = no observed-adverse-effect level; NS = not specified; OF = organ function; OP = ophthalmology;
OW = organ weight; PFBA = perfluorobutyric acid; PFBuS = perfluorobutane sulfonic acid; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid;
PFHxA = perfluorohexanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOSA = perfluorooctane sulfonamide;
PFUA = perfluoroundecanoic acid; PPARα = peroxisome proliferator-activated receptor-α; Repro = reproductive; Resp = respiratory; TG = teratogenicity;
UR = urinalysis

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

99
2. HEALTH EFFECTS

Figure 2-8. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Acute (≤14 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

100
2. HEALTH EFFECTS

Figure 2-8. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Acute (≤14 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

101
2. HEALTH EFFECTS

Figure 2-8. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Intermediate (15–364 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

102
2. HEALTH EFFECTS

Figure 2-8. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Intermediate (15–364 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

103
2. HEALTH EFFECTS

Figure 2-8. Levels of Significant Exposure to Other Perfluoroalkyls – Oral
Chronic (≥365 days)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

104
2. HEALTH EFFECTS

Table 2-6. Levels of Significant Exposure to PFOA – Dermal
Species
(strain)
No./group

Exposure
parameters Doses

Parameters
monitored Endpoint

NOAEL

Less
serious
LOAEL

Serious
LOAEL

Effect

7,000 M

14-day LD50

ACUTE EXPOSURE

Rat
(CD)
15 M,F

Once

3,000, 5,000,
7,500 mg/kg

CS, LE

Death

Bd Wt
Dermal

3,000 M F

Bd Wt

20 M

Resp

2,000 M

Cardio

2,000 M

Gastro
Hemato
Hepatic
Renal
Dermal
Ocular
Endocr
Immuno
Neuro
Repro

2,000 M
2,000 M

5,000 M F

3,000 M F 5–7% transient weight loss
Mild skin irritation

Kennedy 1985
APFO
LD50 in females was >7,500 mg/kg
Rat
(CD)
60 M

2 weeks
0, 20, 200,
CS, BW,
6 hours/day 2,000 mg/kg/day HE, BI,
5 days/week
GN, HP

200 M

20 M
2,000 M
20 M
2,000 M
2,000 M
2,000 M
2,000 M
2,000 M

200 M

14% weight loss

Foci of coagulative necrosis
2,000 M

Skin irritation; acute necrotizing dermatitis

Kennedy 1985
APFO
The immunological NOAEL is for histopathology of the spleen, thymus, and lymph nodes. The neurological NOAEL is for histopathology of the brain. The
reproductive NOAEL is for histopathology of the testes.
Mouse
4 days
0, 0.25, 2.5,
BW, OW, Bd Wt
50 F
(BALB/c)
1 time/day 6.25, 12.5, 25,
OF
Hepatic
2.5 F
6.25 F
52% increase in absolute liver weight
35 F
50 mg/kg/day
Fairley et al. 2007
PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

105
2. HEALTH EFFECTS

Table 2-6. Levels of Significant Exposure to PFOA – Dermal
Species
(strain)
No./group

Exposure
parameters Doses

Parameters
monitored Endpoint

Mouse
4 days
(BALB/c)
1 time/day
35 F
Fairley et al. 2007
PFOA

0, 12.5, 18.8, 25, BW, OW,
50 mg/kg/day
OF

Immuno

Rabbit
Once
(albino)
(NS)
6 NS
Griffith and Long 1980
APFO

100 mg

CS

Ocular

Rabbit
24 hours
(albino)
(NS)
6 NS
Griffith and Long 1980
APFO

500 mg

CS

Dermal

Rabbit
Once
(New Zealand)
17 M

1,500, 3,000,
5,000,
7,500 mg/kg

CS, LE

Death

NOAEL

Less
serious
LOAEL

12.5

18.8

Increased serum IgE following ovalbumin
challenge

100

Moderate eye irritation

4,300

14-day LD50

Serious
LOAEL

Effect

500

Kennedy 1985
APFO
APFO = ammonium perfluorooctanoate; BI = biochemical changes; BW or Bd wt = body weight; Cardio = cardiovascular; CS = clinical signs; Endocr = endocrine;
F = female(s); Gastro = gastrointestinal; GN = gross necropsy; HE or Hemato = hematology; HP = histopathology; Immuno = immunotoxicological; LD50 = lethal
dose, 50% kill; LE = lethality; LOAEL = lowest-observed-adverse-effect level; M = male(s); Neuro = neurological; NOAEL = no-observed-adverse-effect level;
NS = not specified; OF = organ function; OW = organ weight; PFOA = perfluorooctanoic acid; Repro = reproductive; Resp = respiratory

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

106
2. HEALTH EFFECTS

2.2

DEATH

Overview. There are limited data regarding the lethality of perfluoroalkyls in humans; the available data
primarily come from cohort mortality studies in workers. Laboratory animal studies have measured LC50
and LD50 values and reported deaths following inhalation, oral, or dermal exposure to perfluoroalkyls.
These data are presented in Tables 2-1, 2-2, 2-3, 2-4, 2-5, and 2-6 and Figures 2-4, 2-5, 2-6, 2-7, and 2-8.
PFOA

Epidemiology Studies. Five occupational exposure studies at two PFOA manufacturing facilities have
examined the possible associations between PFOA exposure and increases in mortality from all causes
and have not found associations (Gilliland and Mandel 1993; Leonard 2006; Leonard et al. 2008; Lundin
et al. 2009; Raleigh et al. 2014; Steenland and Woskie 2012). Some increases in disease-specific
mortality have been observed; these data are discussed in subsequent sections of this chapter.

Laboratory Animal Studies. Limited data are available regarding death in animals following inhalation
exposure to perfluoroalkyl compounds. Exposure of male and female rats to 18,600 mg/m3 ammonium
perfluorooctanoate (APFO) dusts for 1 hour did not result in deaths during exposure or during a 14-day
observation period (Griffith and Long 1980); APFO is the ammonium salt of PFOA. An LC50 of
980 mg/m3 was reported in male CD rats exposed head-only to APFO dusts for 4 hours (Kennedy et al.
1986). Deaths occurred at all exposure levels (380–5,700 mg/m3) and all deaths occurred within 48 hours
of exposure. Rats dying during exposure had hyperinflated lungs. A similar LC50 value of 820 mg/m3
was calculated for male CD rats exposed nose-only to APFO dusts for 4 hours (Kinney et al. 1989).
Unlike the Kennedy et al. (1986) study, one death was observed at 590 mg/m3 and no deaths occurred at
620 mg/m3. In a developmental study with APFO, whole-body exposure of 12 pregnant rats to 25 mg/m3,
6 hours/day during GDs 6–15 resulted in three deaths on GDs 12, 13, and 17 compared with no deaths in
groups exposed to ≤10 mg/m3 (Staples et al. 1984). The cause of death was not reported.
Oral LD50 values of 680 and 430 mg/kg were reported for male and female albino rats, respectively,
administered single gavage doses of APFO and observed for 14 days (Griffith and Long 1980); all
animals at the highest dose of 2,150 mg/kg died on day 1. Nonlethal signs observed included ptosis,
piloerection, hypoactivity, decreased limb tone, ataxia, and corneal opacity. All signs were intermittent
and there was no apparent dose-response relationship. In a 28-day dietary study with APFO in rats, all
rats (males and females) in groups receiving approximately 1,000–1,130 mg/kg/day APFO died before

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

107
2. HEALTH EFFECTS

the end of the first week (Griffith and Long 1980). In a similar study in mice, all mice receiving doses of
approximately 180–195 mg/kg/day died before the second week of the study (Griffith and Long 1980). In
this study, doses of approximately 54–58 mg/kg/day APFO were lethal to 4/5 male and 5/5 female mice
before the 4th week of the study.

In a 90-day gavage study, treatment of Rhesus monkeys with 100 mg/kg/day APFO by gavage resulted in
the death of an unspecified number of animals (group size was 10/sex) on week 2 (Griffith and Long
1980). Doses of approximately 30 mg/kg/day were lethal to one male and two females during weeks 7–
12. All animals that died in the 30 and 100 mg/kg/day groups had anorexia, emesis, black stool, pale face
and gums, swollen face and eyes, hypoactivity, and prostration. Microscopic examination of tissues
showed marked diffuse lipid depletion in the adrenals, slight to moderate hypocellularity of the bone
marrow, moderate atrophy of the lymphoid follicles of the spleen, and moderate atrophy of the lymphoid
follicles of the lymph nodes. No deaths occurred at 10 mg/kg/day. Deaths were also reported in
intermediate-duration studies in Cynomolgus monkeys (Butenhoff et al. 2002). One monkey exposed to
30/20 mg/kg/day PFOA (12 days of exposure to 30 mg/kg/day, 10 days with no exposure, 23 weeks of
exposure to 20 mg/kg/day) was sacrificed in moribund condition; the animal had a body weight loss of
12.5%, was notably hypoactive, and was cold to the touch (Butenhoff et al. 2002). The investigators
noted that the death was likely due to the high toxicity of the 30 mg/kg/day dose. It is unclear if these
deaths were compound-related; one monkey had pulmonary necrosis with a severe acute recurrence of
pulmonary inflammation and the cause of morbidity for the second monkey was likely hyperkalemia.
Neither effect was observed in the surviving animals.

The dermal LD50 values for APFO were 7,000 mg/kg in male CD rats and >7,500 mg/kg in female rats
(Kennedy 1985). The protocol consisted of application of PFOA (as an aqueous paste) to a clipped area
of the skin, which immediately was covered with gauze pads and wrapped with rubber sheeting around
the trunk. The contact period was 24 hours, at which time the application site was washed with water and
the rats were observed for clinical signs for 14 days. Using the same protocol, the dermal LD50 in male
rabbits was 4,300 mg/kg (Kennedy 1985). Rabbits treated with 1,500 mg/kg showed skin irritation with
formation of a large crusty area at the application site. No deaths occurred at 1,500 mg/kg. Rabbits
treated with 3,000 mg/kg were lethargic and a single death occurred 7 days after treatment. At
5,000 mg/kg, deaths occurred in 3–4 days. These rabbits also showed nasal discharge, pallor, diarrhea,
weakness, severe weight loss, and severe skin irritation along with areas of necrosis.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

108
2. HEALTH EFFECTS

PFOS

Epidemiology Studies. One occupational exposure study evaluated the potential of PFOS to increase
lethality; the study did not find increases in deaths from all causes in workers at a PFOS manufacturing
facility (Alexander et al. 2003). Alterations in disease-specific mortality are discussed in subsequent
sections of this chapter.

Laboratory Animal Studies. Unpublished information summarized by the Organization for Economic
Co-operation and Development (OECD) (2002) indicates that an LC50 of 5,200 mg/m3 was calculated for
PFOS in male and female Sprague-Dawley rats exposed to airborne concentrations of PFOS dusts from
1,890 to 45,970 mg/m3 for 1 hour. All rats exposed to 24,090 mg/m3 died by day 6.

Unpublished information summarized by OECD (2002) indicate that LD50 values of 233 and 271 mg/kg
were calculated for male and female CD rats, respectively, following administration by gavage of single
doses of up to 1,000 mg/kg of powdered PFOS suspended in an acetone/oil mixture and observed for
14 days. All rats (5/sex/dose group) dosed with ≥464 mg/kg PFOS died before the end of the study. The
signs most frequently observed were hypoactivity, decreased limb tone, and ataxia. Gross necropsy
showed stomach distension and signs of irritation of the glandular mucosa, and lung congestion. OECD
(2002) also reported that a different study estimated that the acute oral LD50 for PFOS by gavage in water
in Sherman-Wistar albino rats was >50 and <1,500 mg/kg. An oral LD50 value of 579 mg/kg/day was
reported for male C57/BL/6 mice administered single gavage doses of PFOS and observed for 14 days
(Xing et al. 2016). Mortality occurred within 3 hours of dosing, and moribund mice displayed signs of
neurotoxicity (abdominal breathing, hind limb spasticity, tics, and urinary incontinence).

In a 26-week study, 2/6 male Cynomolgus monkeys administered 0.75 mg/kg/day PFOS via a capsule
died or were sacrificed due to morbidity (Seacat et al. 2002). The cause of death in one monkey was
pulmonary inflammation; the cause of morbidity in the second monkey was not determined, but the
animal did have hyperkalemia.
PFNA
Laboratory Animal Studies. A LC50 of 820 mg/m3 was identified in rats exposed to airborne PFNA for
4 hours (Kinney et al. 1989). In a 14-day study, all mice administered approximately 54 mg/kg/day
PFNA died before the study period ended; no deaths occurred at 5.3 mg/kg/day (Kennedy 1987).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

109
2. HEALTH EFFECTS

PFDeA

Laboratory Animal Studies. An LD50 of 120 mg/kg was estimated for PFDeA in female C57BL/6N mice
administered single doses between 20 and 320 mg/kg/day PFDeA by gavage in corn oil and observed for
30 days (Harris et al. 1989). All mice receiving 160 or 320 mg/kg were dead by 14 days; no mice died at
≤80 mg/kg PFDeA. Early death was associated with mural thrombosis in the left ventricle of the heart.
Without providing any details, George and Andersen (1986) reported that the 30-day oral LD50 for
PFDeA in male Fischer-344 rats was 57 mg/kg.
PFHxA

Laboratory Animal Studies. Decreased survival was observed in female Sprague-Dawley rats
administered 200 mg/kg/day PFHxA via gavage in a 104-week study (Klaunig et al. 2015). There was no
significant effect on survival rates of males. Animals in the 200 mg/kg/day dose group had rales and
increased incidence of struggling behavior.
2.3

BODY WEIGHT

Overview. Epidemiology studies have examined the possible associations between in utero and/or early
life exposure to perfluoroalkyls and body weight, body mass index (BMI), etc. Other studies have
examined possible associations between serum perfluoroalkyl levels in older children or adults and body
weight, adiposity markers, and the risk of being overweight or obese. The results of the epidemiology
studies are summarized in Table 2-7, with more detailed descriptions presented in the Supporting
Document for Epidemiological Studies for Perfluoroalkyls, Table 1. Animal studies have evaluated
changes in body weight, including maternal body weight, in response to inhalation, oral, or dermal
exposure to perfluoroalkyls; these data are summarized in Tables 2-1, 2-2, 2-3, 2-4, 2-5, and 2-6 and
Figures 2-4, 2-5, 2-6, 2-7, and 2-8.

PFOA

Epidemiology Studies. Mixed results were found in studies of monitoring infant growth from 1 to
12 months of age. Andersen et al. (2010) found an inverse association between maternal serum PFOA
and body weight and BMI in male infants at 5 and 12 months of age; no associations were found in girls.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

110
2. HEALTH EFFECTS

Table 2-7. Summary of Childhood Growth in Humansa
Reference and study populationb
PFOA
Barry et al. 2014
Community (C8) (n=8,764 20–40 year olds)
Andersen et al. 2010
General population (n=1,010 infants)

Andersen et al. 2013
General population (n=811 children aged
7 years)
Braun et al. 2016a, 2016b

Serum perfluoroalkyl level Outcome evaluated

Resultc

164.6 and 194.3 ng/mL
(estimated early life median
PFOA)

OR 0.9 (0.7–1.1), males
OR 0.9 (0.7–1.1), females

5.21 ng/mL (maternal median Body weight (age 5 and
Inverse association (p<0.05)*, boys
PFOA)
12 months)
NS (p>0.05), girls
BMI (age 5 and 12 months) Inverse association (p<0.05)*, boys
NS (p>0.05), girls
Height (age 5 and
NS (p>0.05), boys
12 months)
NS (p>0.05), girls
5.25 ng/mL (maternal median BMI
NS (p>0.05)
PFOA)
Waist circumference
NS (p>0.05)

5.3 ng/mL (maternal median
PFOA)

General population (n=204 children)

de Cock et al. 2014
General population (n=89 infants aged 1–
11 months)
Halldorsson et al. 2012
General population (n=665 20 year olds)

Overweight or obesity at
age 20–40 years

Changes in BMI scores
between 2 and 8 years of
age
Overweight/obesity risk

0.9402 ng/mL (cord blood
mean PFOA)

Weight
Height
BMI
Head circumference
3.7 and 5.8 ng/mL (maternal BMI
median PFOA and 4th quartile Waist circumference
median)
Overweight risk
High waist circumference

***DRAFT FOR PUBLIC COMMENT***

Association (p=0.03)*

RR 1.54 (0.77–3.07), 3rd tertile
NS (p=0.350)
NS (p=0.045)
NS (p=0.813)
NS (p=0.774)
Association (p=0.001)*, females
Association (p=0.006)*, females
RR 3.1 (1.4–6.9)*, females
4th quartile
RR 3.0 (1.3–6.8)*, females
4th quartile

 PERFLUOROALKYLS

111
2. HEALTH EFFECTS

Table 2-7. Summary of Childhood Growth in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

2.2–5.1 and 1.1–9.8 ng/mL
Overweight
(maternal 3rd tertile PFOA for
Greenland cohort
Ukraine cohort
General population (n=1,122 children aged 5– Greenland and Ukraine
9 years; n=531 for Greenland cohort and
cohorts)
Waist-to-height ratio >0.5
n=491 for Ukraine cohort)
Greenland cohort
Ukraine cohort
Timmermann et al. 2014
9.3 ng/mL (mean PFOA)
Adiposity markers

Resultc

Høyer et al. 2015b

RR 1.23 (0.87–1.74), 3rd tertile
RR 0.78 (0.47–1.29), 3rd tertile
RR 1.18 (0.80–1.74), 3rd tertile
RR 1.11 (0.48–2.57), 3rd tertile
NS (p>0.05), per 10 ng/mL PFOA
increase

General population (n=499 8–10-year-old
children)
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)
PFOS
Andersen et al. 2010
General population (n=1,010 infants)

2.37 and 2.34 ng/mL (median Growth during childhood
maternal PFOA for boys and
girls)

NS (p>0.05)

33.8 ng/mL (maternal median Body weight (age 5 months)
PFOS)
Body weight (age
12 months)
BMI (age 5 months)

NS (p>0.05), boys
NS (p>0.05), girls
Inverse association (p<0.05)*, boys
NS (p>0.05), girls
NS, boys
NS (p>0.05), girls
Inverse association (p<0.05)*, boys
NS (p>0.05), girls
NS (p>0.05), boys
NS (p>0.05), girls
NS (p>0.05)
NS (p>0.05)

BMI (age 12 months)

Andersen et al. 2013
General population (n=811 children aged
7 years)

Height (age 5 and
12 months)
33.8 ng/mL (maternal median BMI
PFOS)
Waist circumference

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

112
2. HEALTH EFFECTS

Table 2-7. Summary of Childhood Growth in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Braun et al. 2016a, 2016b

13 ng/mL (maternal median
PFOS)

NS (p>0.23)

General population (n=204 children)
Halldorsson et al. 2012
General population (n=665 20 year olds)
Høyer et al. 2015b

Changes in BMI scores
between 2 and 8 years of
age
Overweight/obesity risk
21.5 and 5.8 ng/mL (maternal BMI
median PFOS)
Waist circumference

23.9–87.3 and 5.9–
18.1 ng/mL (maternal 3rd
General population (n=1,122 children aged 5– tertile PFOS for Greenland
9 years; n=531 for Greenland cohort and
and Ukraine cohorts)
n=491 for Ukraine cohort)
Maisonet et al. 2012
General population (n=447 girls)
Timmermann et al. 2014
General population (n=499 8–10-year-old
children)
PFHxS
Braun et al. 2016a, 2016b

Overweight
Greenland cohort
Ukraine cohort
Waist-to-height ratio >0.5
Greenland cohort
Ukraine cohort
>23.0 ng/mL (maternal PFOS Body weight at 20 months
3rd tertile)
(adjusted for birth weight)

RR 0.84 (0.61–1.14), 3rd tertile
RR 0.89 (0.57–1.37), 3rd tertile
RR 1.22 (0.86–1.74), 3rd tertile
RR 1.44 (0.62–3.31), 3rd tertile
β 438.4 g (71.09–805.65 g,
p=0.0195)*, 3rd tertile

41.5 ng/mL (mean PFOS)

Adiposity markers

NS (p>0.05), per 10 ng/mL PFOS
increase

1.4 ng/mL (maternal median
PFHxS)

Changes in BMI scores
between 2 and 8 years of
age
Overweight/obesity risk
Body weight at 20 months

NS (p>0.23)

General population (n=204 children)
Maisonet et al. 2012

RR 1.08 (0.59–1.95), 3rd tertile
NS (p>0.56)
NS (p>0.56)

1.6 ng/mL (maternal median
PFHxS)

General population (n=447 girls)

***DRAFT FOR PUBLIC COMMENT***

RR 1.48 (0.75–2.96), 3rd tertile
NS (p=0.4375 for trend)

 PERFLUOROALKYLS

113
2. HEALTH EFFECTS

Table 2-7. Summary of Childhood Growth in Humansa
Reference and study populationb
PFNA
Braun et al. 2016a, 2016b

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.9 ng/mL (maternal median
PFNA)

NS (p>0.23)

General population (n=204 children)
Halldorsson et al. 2012
General population (n=665 20 year olds)
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)
PFDeA
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)
PFUA
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)
PFDoA
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)

0.3 ng/mL (maternal median
PFNA)

Changes in BMI scores
between 2 and 8 years of
age
Overweight/obesity risk
BMI
Waist circumference

1.55 and 1.58 ng/mL (median Growth during childhood
maternal PFNA for boys and
girls)

RR 1.26 (0.64–2.48), 3rd tertile
NS (p>0.56)
NS (p>0.56)
NS (p>0.05)

0.46 and 0.43 ng/mL (median Growth during childhood
maternal PFDeA for boys and Weight
girls)
Height

Inverse association (p<0.05)*, girls
Inverse association (p<0.05)*, girls

3.52 and 3.31 ng/mL (median Growth during childhood
maternal PFUA for boys and
Weight
girls)
Height

Inverse association (p<0.05)*, girls
Inverse association (p<0.05)*, girls

0.37 and 0.37 ng/mL (median Growth during childhood
maternal PFDoA for boys and Weight
girls)
Height

Inverse association (p<0.05)*, girls
Inverse association (p<0.05)*, girls

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

114
2. HEALTH EFFECTS

Table 2-7. Summary of Childhood Growth in Humansa
Reference and study populationb
PFOSA
Halldorsson et al. 2012
General population (n=665 20 year olds)

Serum perfluoroalkyl level Outcome evaluated

Resultc

1.1 ng/mL (maternal median
PFOSA)

BMI

NS (p>0.56)

Waist circumference

NS (p>0.56)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 1 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

BMI = body mass index; HR = hazard ratio; NR = not reported; NS = not significant; OR = odds ratio; PFDeA = perfluorodecanoic acid;
PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid; RR = relative risk

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

115
2. HEALTH EFFECTS

A study by de Cock et al. (2014) of infants 1–11 months of age did not find associations between cord
blood PFOA levels and weight, height, or BMI. One study of children (Braun et al. 2016a) found an
association between changes in BMI scores between ages 2 and 8 years and maternal PFOA levels;
however, there was no increase in the risk of being overweight or obese. Other studies in children (2–
11 years of age) found no associations between maternal PFOA or cord blood PFOA and growth during
childhood (Wang et al. 2016), risk of being overweight (Andersen et al. 2013; Braun et al. 2016a; Høyer
et al. 2015b), or risk of having a waist-to-height ratio of >0.5 (Høyer et al. 2015b). In a study of children
aged 8–10 years, no associations were found between plasma PFOA levels and markers of adiposity
(BMI, skinfold thickness, waist circumference, adiponectin levels, and leptin levels) (Timmermann et al.
2014).

Two studies in adults have not found associations between PFOA and body weight gain. A general
population study of 20-year-old females found associations between maternal PFOA levels and BMI and
waist circumferences, and increases in the risk of being overweight and having a high waist
circumference (Halldorsson et al. 2012); these associations were not observed in males. No increases in
the risk of being overweight or obese were observed in male or female C8 participants (20–40 years of
age) when estimated early life PFOA exposure was used as the exposure metric (Barry et al. 2014).

Overall, the available epidemiology data do not suggest a connection between serum PFOA levels and
body weight or risk of being overweight/obese.
Laboratory Animal Studies. Male rats that survived a 4-hour inhalation exposure to 380 mg/m3 APFO
dusts lost weight for 1–2 days after exposure, but resumed normal weight gain thereafter (Kennedy et al.
1986). Male rats exposed via inhalation intermittently to 84 mg/m3 APFO dusts for 2 weeks lost
approximately 7% of their body weight by day 5 of exposure (250 g at start of study, 237 g on day 5)
(Kennedy et al. 1986), but recovered by day 16 after exposure ceased. Nose-only exposure of male CD
rats to 590 mg/m3 ammonium perfluorononanoate dusts for 4 hours resulted in 18 and 36% reductions in
body weight 5 and 12 days after exposure, respectively (Kinney et al. 1989). Inhalation exposure to
67 mg/m3 had no significant effect on body weight. In a developmental study, inhalation exposure of
pregnant rats to 25 mg/m3 APFO dusts during GDs 6–15 induced a 37% reduction in body weight gain
relative to controls during the exposure period (Staples et al. 1984); in a pair-fed group, the reduction of
weight gain during the same period was 61% relative to ad libitum controls.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

116
2. HEALTH EFFECTS

Reductions in body weight or body weight gain are typical, although not particularly sensitive, responses
of rodents to oral exposure to perfluoroalkyl compounds. In many cases, this effect is not associated with
reduced food intake, and in some cases, exposed animals have shown an increase in relative food
consumption (grams of food/grams of body weight) relative to controls. For example, in acute-duration
studies, rats administered 25 mg/kg/day APFO for 14 days had a mean terminal body weight 14% lower
than controls (Cook et al. 1992). Administration of 50 mg/kg/day APFO for 7 days resulted in 17%
weight loss; a similar decrease was observed in a pair-fed group (Pastoor et al. 1987). In mice, doses of
approximately 25–30 mg/kg/day PFOA in the food for 7 days reduced terminal body weight by >10%
relative to controls without a significant reduction in food intake (Xie et al. 2003; Yang et al. 2000,
2002a, 2002b). However, administration of the same dose to PPARα-null mice did not cause a reduction
in weight gain, suggesting that the effect on body weight is a specific effect of peroxisome proliferators
possibly due to increased fat utilization (Yang et al. 2002b). In general, body weight recovered once
treatment ceased.

Intermediate-duration oral studies in rats have also reported reduced body weight gain with doses
≥10 mg/kg/day APFO (Butenhoff et al. 2004b; Griffith and Long 1980). In the former study, mean
absolute food consumption was decreased, but mean relative food consumption was increased. In a
2-year bioassay, body weight gain in rats dosed with 15 mg/kg/day PFOA was reduced >10% relative to
controls at the 1 year mark and at termination (Biegel et al. 2001). Similar observations have been made
in mice dosed with approximately ≥18 mg/kg/day APFO for 28 days (Griffith and Long 1980) and in
pregnant mice dosed with ≥10 mg/kg/day APFO during GDs 1–17 (Lau et al. 2006). A study comparing
wild-type mice and PPARα knockout mice (DeWitt et al. 2016) found a decrease in body weight gain in
the wild-type mice, but not in the knockout mice. A 90-day and a 26-week study in monkeys also
reported significant reductions in body weight gain or weight loss associated with decreased food
consumption at dose levels in the range of 20–30 mg/kg/day APFO (Butenhoff et al. 2002; Griffith and
Long 1980), but a 4-week study in monkeys dosed with 20 mg/kg/day PFOA did not (Thomford 2001).

Transient weight loss was reported in rats applied 3,000 mg/kg APFO to the shaven skin for 24 hours
(Kennedy 1985). In the 2-week study, rats in the 200 and 2,000 mg/kg/day groups lost weight during the
treatment period (14 and 24%, respectively, on test day 10), but body weights were comparable to control
after 42 days of recovery. No changes in body weight were reported in mice applied up to 50 mg/kg/day
PFOA daily for 4 days on the dorsal surface of the ears (Fairley et al. 2007).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

117
2. HEALTH EFFECTS

PFOS

Epidemiology Studies. General population studies have evaluated body weight, height, and BMI in
infants, children, and adults to assess whether there were links between growth and maternal serum PFOS
levels. Andersen et al. (2010) found that maternal PFOS levels were inversely related to body weight and
BMI in 12-month-old male infants; no associations were found in females at 12 months of age or in males
and females at 5 months of age. The magnitude of the effect on body weight in the boys was small, 9 g
per 1 ng/mL increase in maternal serum PFOS level. Maisonet et al. (2012) found that at 20 months of
age, girls whose mothers had serum PFOS levels in the 3rd tertile weighed 438 g more than those in the
first tertile. Studies in children (Andersen et al. 2013; Braun et al. 2016a; Høyer et al. 2015b) or young
adults (Halldorsson et al. 2012) did not find associations between maternal PFOS levels and BMI, waist
circumference, and/or risk of being overweight. No associations between plasma PFOS and markers of
adiposity (BMI, skinfold thickness, waist circumference, adiponectin levels, and leptin levels) were found
in a study of children aged 8–10 years (Timmermann et al. 2014). Overall, the epidemiology studies do
not suggest a connection between serum PFOS and body weight or the risk of being overweight/obese.

Laboratory Animal Studies. Dietary treatment of rats with 15 mg/kg/day PFOS (only dose level tested)
for 7 days did not significantly alter body weight (Haughom and Spydevold 1992). Oral treatment of
pregnant rats with 25 mg/kg/day PFOS on GDs 2–5 or 6–9 resulted in weight loss during treatment,
whereas treatment on GDs 10–13, 14–17, or 17–20 resulted in significant reductions in weight gain
(Grasty et al. 2003). In pregnant mice, oral dosing with up to 6 mg/kg/day PFOS on GDs 6–18 or 12–18
did not significantly affect body weight (Fuentes et al. 2006, 2007b). Pregnant rabbits appeared to be
more sensitive as oral doses of 1 mg/kg/day on GDs 6–20 caused a 21% reduction in weight gain during
treatment without altering food consumption (Case et al. 2001).

Reductions in body weight gain of >10% have been reported in intermediate-duration studies in rats
dosed with ≥2 mg/kg/day PFOS associated with reductions in mean absolute and relative food
consumption (Luebker et al. 2005a, 2005b). In a developmental toxicity study, treatment of pregnant rats
with ≥2 mg/kg/day PFOS on GDs 2–20 resulted in significant reductions in body weight gain, which were
associated with significant reductions in mean absolute food and water consumption (Thibodeaux et al.
2003). In a 4-week study, treatment of Cynomolgus monkeys with up to 2 mg/kg/day, administered via a
capsule, did not affect body weight gain (Thomford 2002a). In a 26-week study in Cynomolgus
monkeys, the highest dose of PFOS tested, 0.75 mg/kg/day, produced a 13.5% reduction in final body
weight, at which time the mean concentration of PFOS in serum was 172 µg/mL (Seacat et al. 2002). In a

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

118
2. HEALTH EFFECTS

2-year dietary study in rats, final mean body weight of females that received doses of approximately
1.04 mg/kg/day PFOS was 14% lower than controls; this could have been due, in part, to a tendency of
decreased food consumption during weeks 28 through 104 of the study (Butenhoff et al. 2012b;
Thomford 2002b). No significant effect (<10% difference with controls) was seen in females dosed with
≤0.25 mg/kg/day PFOS.
PFHxS

Epidemiology Studies. Two studies have evaluated the influence of in utero PFHxS exposure on
childhood growth and found no associations between maternal PFHxS levels and body weight at
20 months of age (Maisonet et al. 2012), changes in BMI scores between 2 and 8 years of age (Braun et
al. 2016a), or the risk of childhood overweight/obesity (Braun et al. 2016a).

Laboratory Animal Studies. Treatment of rats with up to 10 mg/kg/day PFHxS by gavage for 40–
60 days did not significantly affect body weight (Butenhoff et al. 2009a; Hoberman and York 2003); the
mean terminal body weights were within 10% of the body weight of the control group. Food
consumption was not affected by treatment with PFHxS.
PFNA

Epidemiology Studies. Several studies have examined the influence of maternal PFNA levels on
childhood growth. These studies did not find associations between maternal PFNA levels and growth
during childhood (Wang et al. 2016), BMI (Braun et al. 2016a; Halldorsson et al. 2012), or overweight/
obesity risk (Braun et al. 2016a).

Laboratory Animal Studies. Decreases in body weight gain have been observed in rats administered
≥3 mg/kg/day for 14 days (Fang et al. 2009, 2010; Hadrup et al. 2016) and in mice administered
5 mg/kg/day for 14 days (Wang et al. 2015a). The NOAEL for body weight effects was 1 mg/kg/day for
both species. In intermediate-duration developmental toxicity studies, decreases in body weight were
observed at 5 mg/kg/day in rats (Rogers et al. 2014) and weight loss was observed in mice at
10 mg/kg/day (Das et al. 2015). No alterations in maternal weight gain were observed in mice at
2.0 mg/kg/day (Wolf et al. 2010).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

119
2. HEALTH EFFECTS

PFDeA

Epidemiology Studies. One study examined the effect of PFDeA levels on childhood growth; Wang et al.
(2016) reported decreases in weight and height in girls associated with increasing maternal serum PFDeA
levels.

Laboratory Animal Studies. Ten days following administration of a single gavage dose of 50 mg/kg,
weight loss was observed in rats (Kawabata et al. 2017). In a 1-week study, exposure to 9.5 mg/kg/day
PFDeA in the diet resulted in a 32% weight loss in rats (Kawashima et al. 1995); the NOAEL was
4.7 mg/kg/day.

Body weight of female C57BL/6N mice administered a single gavage dose of 80 mg/kg PFDeA was
reduced 12% relative to controls 30 days after dosing (Harris et al. 1989); no significant effect was seen at
40 mg/kg PFDeA. In a developmental study, pregnant mice dosed with 6.4 mg/kg/day PFDeA on
GDs 6–15 gained 92% less weight (adjusted weight) on GDs 6–18 than controls; mice dosed with
12.8 mg/kg/day lost weight (Harris and Birnbaum 1989). Food consumption data were not provided in
any of these studies. Weight loss was also observed in C57BL/6N mice exposed to 78 mg/kg/day PFDeA
in the diet for 10 days (Permadi et al. 1992, 1993).
PFUA

Epidemiology Studies. Wang et al. (2016) found an inverse association between maternal serum PFUA
levels and weight and height in girls.

Laboratory Animal Studies. Decreases in body weight gain (10% in males and 23% in females) were
observed in rats exposed to 1.0 mg/kg/day in a 41–46-day developmental toxicity study (Takahashi et al.
2014).
PFBuS

Laboratory Animal Studies. No significant alterations in body weight gain were observed in SpragueDawley rats administered ≤900 mg/kg/day PFBuS via gavage for 28 days (3M 2001) or in SpragueDawley rats administered ≤1,000 mg/kg/day PFBuS via gavage for at least 70 days (Lieder et al. 2009b).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

120
2. HEALTH EFFECTS

PFBA

Laboratory Animal Studies. Alterations in body weight do not appear to be a sensitive outcome of PFBA
exposure in rats or mice. No alterations in body weight gain were observed in Sprague-Dawley rats
administered 184 mg/kg/day PFBA via gavage for 5 days (3M 2007a), C57BL/6 mice exposed to
78 mg/kg/day PFBA in the diet for 10 days (Permadi et al. 1992, 1993), Sprague-Dawley rats
administered 150 mg/kg/day PFBA via gavage for 28 days (Butenhoff et al. 2012a; van Otterdijk 2007a),
or Sprague-Dawley rats administered 30 mg/kg/day PFBA via gavage for 90 days (Butenhoff et al. 2012a;
van Otterdijk 2007b).
PFDoA

Epidemiology Studies. Wang et al. (2016) found an inverse association between maternal serum PFDoA
levels and growth (weight and height) in girls.

Laboratory Animal Studies. Dosing of Sprague-Dawley rats with 5 mg/kg/day PFDoA by gavage for
14 days resulted in a 25% reduction in final body weight relative to a control group or 7% loss of body
weight compared with the starting body weight (Shi et al. 2007). Decreases in body weight gain
(measured 10 days postexposure) were also observed in rats administered a single gavage dose of
50 mg/kg PFDoA (Kawabata et al. 2017).
PFHxA

Laboratory Animal Studies. Gavage administration of 100 or 200 mg/kg/day for 2 years did not result in
alterations in body weight gain in male or female rats, respectively (Klaunig et al. 2015).
PFOSA

Epidemiology Studies. Halldorsson et al. (2012) did not find associations between maternal serum
PFOSA levels and BMI or waist circumference in 20 year olds.

Laboratory Animal Studies. No alterations in body weight were observed in Sprague-Dawley rats
following a single gavage dose of 5 mg/kg PFOSA (Seacat and Luebker 2000).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

121
2. HEALTH EFFECTS

2.4

RESPIRATORY

Overview. A small number of epidemiology studies have examined the potential of perfluoroalkyl
compounds to damage the respiratory tract; detailed descriptions of these studies are presented in the
Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 2. These studies were
primarily conducted in PFOA workers or in residents of nearby communities. The possible associations
between perfluoroalkyl exposure and asthma are discussed along with other immune effects in
Section 2.14. Studies in laboratory animals have examined the potential for perfluoroalkyls to induce
histological lesions in the lungs following inhalation (see Tables 2-1 and 2-2) or oral exposure (see
Tables 2-3, 2-4, and 2-5).

PFOA

Epidemiology Studies. There are limited data on the potential of PFOA to damage the respiratory tract.
Pulmonary function tests and chest roentgenograms conducted on workers potentially exposed to PFOA
at the Washington Works fluoropolymers production facility were within normal limits (Sakr et al.
2007b); the serum PFOA levels ranged from 5 to 9,550 ng/mL. Another study of workers at this facility
did not find an association between estimated cumulative serum PFOA levels and the risk of chronic
obstructive pulmonary disease (Steenland et al. 2015). In contrast, a study of residents living near this
facility found an increase in the risk of chronic bronchitis (standard prevalence ratio [SPR] of 3.60, 95%
confidence interval [CI] 2.92–4.44) and shortness of breath (SPR 2.05, 95% CI 1.70–2.46) (AndersonMahoney et al. 2008); it is noted that results were based on health surveys, and some of the subjects also
worked at the facility. Summaries of these studies are presented in the Supporting Document for
Epidemiological Studies for Perfluoroalkyls, Table 2.
Laboratory Animal Studies. Inhalation exposure of male and female rats to 18,600 mg/m3 APFO dusts
for 1 hour induced a red nasal discharge and dry rales (Griffith and Long 1980). Necropsy conducted
14 days after exposure showed bilateral mottling of the lungs in 8 out of 10 rats. Head-only exposure for
4 hours to 380 mg/m3 APFO dusts, a concentration that was lethal to some rats, produced pulmonary
edema, which disappeared within 1 week of exposure (Kennedy et al. 1986). Examination of the lungs
and trachea from rats exposed head-only to up to 84 mg/m3 APFO dusts 6 hours/day, 5 days/week for
2 weeks showed no significant gross or microscopic alterations (Kennedy et al. 1986). Male CD rats
exposed nose-only to ≥590 mg/m3 ammonium perfluorononanoate dusts for 4 hours exhibited lung noise
and labored breathing during exposure and throughout a 12-day recovery period (Kinney et al. 1989).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

122
2. HEALTH EFFECTS

Oral dosing of male and female CD rats with ≤110 mg/kg/day APFO did not induce gross or microscopic
changes in the lungs (Griffith and Long 1980; Perkins et al. 2004). Dosing for 2 years with 15 mg/kg/day
APFO increased the incidence of lung hemorrhage in males (3M 1983). The incidences were 10/50,
14/50, and 22/50 for groups receiving doses of 0, 1.5, and 15 mg/kg/day, respectively. Pair-wise
comparison between controls and high-dose groups revealed a statistically significant difference (p<0.05).
In a study in monkeys administered up to 20 mg/kg/day APFO, administered via a capsule, for 26 weeks,
no signs of respiratory problems were observed during the study and no gross or microscopic alterations
in the lungs and trachea were observed at termination (Butenhoff et al. 2002).

No gross or microscopic alterations were found in the lung and trachea from male CD rats following
application of up to 2,000 mg/kg/day APFO as an aqueous paste to an area of the shaven back
(approximately 15% of the total body surface) 6 hours/day, 5 days/week for 2 weeks (Kennedy 1985).

PFOS

Laboratory Animal Studies. Unpublished data summarized by OECD (2002) indicate that inhalation
exposure of rats to concentrations of PFOS between 1,890 and 45,970 mg/m3 for 1 hour induced dry rales
and other breathing disturbances.

Dosing of Cynomolgus monkeys with up to 2 mg/kg/day PFOS for 4 weeks had no effect of the gross or
microscopic morphology of the lungs (Thomford 2002a). Administration of doses of up to
0.75 mg/kg/day of PFOS (potassium salt) administered via a capsule to Cynomolgus monkeys for
26 weeks did not produce any gross or microscopic alterations in the lungs or the trachea (Seacat et al.
2002). Dosing rats with up to 1.04 mg PFOS/kg/day in the diet for 104 weeks did not induce significant
gross or microscopic alterations in the lungs or trachea (Butenhoff et al. 2012b; Thomford 2002b).
PFHxS
Laboratory Animal Studies. Examination of the respiratory tract of rats administered ≤10 mg/kg/day
PFHxS by gavage in a reproductive study (40–60 days of dosing) showed no treatment-related effects
(Butenhoff et al. 2009a; Hoberman and York 2003).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

123
2. HEALTH EFFECTS

PFNA

Laboratory Animal Studies. Labored breathing during and after a 4-hour nose-only exposure to
590 mg/m3 PFNA was reported in rats (Kinney et al. 1989).

PFBuS
Laboratory Animal Studies. Administration of PFBuS at gavage doses of ≤900 mg/kg/day for 28 days or
600 mg/kg/day for 90 days had no significant effect on the gross or microscopic morphology of the lungs
or trachea in rats (3M 2001; Lieder et al. 2009a); no increases in nasal lesions were observed in the
90-day study (Lieder et al. 2009a).
PFBA
Laboratory Animal Studies. Administration of PFBA to rats by gavage in doses ≤184 mg/kg/day for
5 days (3M 2007a), ≤150 mg/kg/day for 28 days (Butenhoff et al. 2012a; van Otterdijk 2007a), or
≤30 mg/kg/day for 90 days (Butenhoff et al. 2012a; van Otterdijk 2007b) did not cause morphological
alterations in the respiratory tract.
2.5

CARDIOVASCULAR

Overview. Epidemiology and laboratory animal studies have evaluated the toxicity of perfluoroalkyls to
the cardiovascular system. The epidemiology studies evaluated several cardiovascular outcomes
including ischemic heart disease, cerebrovascular disease, stroke, cardiovascular disease, myocardial
infarction, hypertension, and pregnancy-induced hypertension. The results of these studies are
summarized in Table 2-8, with more detailed descriptions presented in the Supporting Document for
Epidemiological Studies for Perfluoroalkyls, Table 3. The available occupational, community, and
general population studies have not consistently found increases in the risk of heart disease or stroke that
were associated with serum PFOA levels. Considerably less epidemiology data are available for other
perfluoroalkyls; general population studies for PFOS, PFHxS, PFNA, PFDeA, PFUA, and PFHpA have
not found increases in the risk of cardiovascular disease. Most of the available epidemiology studies did
not find a link between serum PFOA and hypertension, and the one study for PFOS did not find an
association. Although mixed results were found in studies of highly exposed community residents, the
strongest methodological study (Darrow et al. 2013) found an increased risk of pregnancy-induced

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

124
2. HEALTH EFFECTS

hypertension that was associated with serum PFOA levels. Increases in the risk of pregnancy-induced
hypertension associated with serum PFOS levels were also found in two community studies. General
population studies have not found associations between serum PFHxS or PFDeA and pre-eclampsia; one
study on PFUA found an inverse association.

Examination of the cardiovascular system in laboratory animals primarily consists of inhalation, oral, and
dermal studies examining the heart for morphological alterations (see Tables 2-1, 2-3, 2-4, 2-5, and 2-6).
The laboratory animal studies did not find increases in the incidence of histological alterations in the heart
following exposure to PFOA, PFOS, PFHxS, PFDeA, PFBuS, or PFBA.
PFOA

Epidemiology Studies—Heart Disease. Possible associations between PFOA exposure and increased
risk of heart disease have been examined in cohort mortality studies of workers, community members
living near a PFOA facility, and the general population. Occupational exposure studies have not found
increases in deaths from all heart disease, cerebrovascular disease, or ischemic heart disease when
compared to U.S. general populations, state populations, and/or a population of workers at other company
facilities (Leonard 2006; Lundin et al. 2009; Raleigh et al. 2014; Steenland and Woskie 2012). One
occupational exposure study found an increase in the risk of cerebrovascular disease in workers with
definite exposure for at least 6 months compared to an internal referent group (Lundin et al. 2009).
However, other studies have not found increased risks of ischemic heart disease (Raleigh et al. 2014; Sakr
et al. 2009), cerebrovascular disease (Raleigh et al. 2014), or coronary artery disease (Steenland et al.
2015). In another occupational exposure study, the investigators noted that electrocardiograms (EKGs)
were within normal limits (Sakr et al. 2007b).

Studies of residents living near the Washington Works facility in West Virginia reported increased risks
of self-reported cardiovascular disease (Anderson-Mahoney et al. 2008), angina (Anderson-Mahoney et
al. 2008), myocardial infarction (Anderson-Mahoney et al. 2008), and stroke (Anderson-Mahoney et al.
2008; Simpson et al. 2013). It is noted that the Anderson-Mahoney et al. (2008) study did not measure
serum PFOA levels; the incidences of self-reported diseases were compared to NHANES rates. Another
community study of residents in this area did not find an increased risk of coronary artery disease
(Winquist and Steenland 2014a). Two general population studies have examined possible associations
between serum PFOA and heart disease risks. A case-control study did not find increases in the risk of
coronary artery disease in subjects with median serum PFOA levels of 4.2 ng/mL (cases) or 4.0 ng/mL

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

125
2. HEALTH EFFECTS

Table 2-8. Summary of Cardiovascular Outcomes in Humansa
Reference and study populationb
PFOA
Leonard 2006
Occupational (n=6,027)

Serum perfluoroalkyl level Outcome evaluated

Resultc

5–9,550 ng/mL (PFOA
range)

SMR 110 (98–123)

Heart disease deaths

Cerebrovascular disease SMR 86 (60–120)
deaths
Ischemic heart disease
deaths

Lundin et al. 2009
Occupational (n=3,993)

Raleigh et al. 2014
Occupational (n=9,027)

Sakr et al. 2009

2,600–5,200 ng/mL (range of Heart disease deaths
definite exposure group)
Cerebrovascular disease
deaths
Ischemic heart disease
deaths
Cerebrovascular disease
risk

Cumulative PFOA exposure Ischemic heart disease
deaths
Cerebrovascular disease
Ischemic heart disease
risk
Cerebrovascular disease
risk
NR
Ischemic heart disease
risk

Occupational (n=4,747)

***DRAFT FOR PUBLIC COMMENT***

SMR 109 (96–124)
SMR 0.7 (0.5–1.3)
SMR 1.6 (0.5–3.7)
SMR 0.8 (0.5–1.4)
HR 4.6 (1.3–17.0)* workers with
definite exposure of ≥6 months
HR 2.1 (1.0–4.6)* workers with
definite exposure ≥5 years
SMR 0.84 (0.74–0.95)*
SMR 0.81 (0.61–1.05)
HR 0.89 (0.66–1.21), 4th quartile
HR 0.98 (0.53–1.81), 4th quartile
NS (p=0.16 for trend)

 PERFLUOROALKYLS

126
2. HEALTH EFFECTS

Table 2-8. Summary of Cardiovascular Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Steenland et al. 2015

Cumulative PFOA exposure Coronary artery disease
risk
Hypertension

NS (p=0.78 for trend), no lag
NS (p=0.75 for trend), 10-year lag
NS (p=0.95 for trend), no lag
NS (p=0.54 for trend), 10-year lag
NS (p=0.35 for trend), no lag
NS (p=0.64 for trend), 10-year lag
SMR 0.93 (0.72–1.19), no lag

Occupational (n=3,713)

Stroke
Steenland and Woskie 2012
Occupational (n=1,084)
Anderson-Mahoney et al. 2008

7,800 ng/mL-years (mean
Ischemic heart disease
cumulative PFOA exposure) deaths
NR

Community (n=566)

Darrow et al. 2013
Community (C8) (n=1,330)
Nolan et al. 2009
Community (C8) (n=1,555 women)
Savitz et al. 2012a
Community (C8) (n=11,737 pregnant women)
Savitz et al. 2012b

Cardiovascular disease
(self-reported)
Angina (self-reported)
Myocardial infarction
Stroke

Hypertension
6.9–<11.1 ng/mL (2nd PFOA Pregnancy-induced
quintile)
hypertension

SPR 4.29 (3.47–5.29)*
SPR 8.07 (6.54–9.95)*
SPR (1.91 (1.40–2.62)*
SPR 2.17 (1.47–3.21)*
SPR 1.18 (0.97–1.43)
OR 2.39 (1.05–5.46)* (2nd quintile)

NR

Pregnancy-induced
hypertension
LHWA residents
OR 1.2 (0.7–2.0), unadjusted
Partial LHWA residents OR 0.8 (0.5–1.4), unadjusted
6.8–<16.6 ng/mL (2nd PFOA Pre-eclampsia
OR 1.2 (1.0–1.5)*
quartile, estimated)
21.0–717.6 ng/mL (5th PFOA Pregnancy induced
quintile, estimated)
hypertension

Community (C8) (n=224 cases of pregnancyinduced hypertension)

***DRAFT FOR PUBLIC COMMENT***

OR 1.0 (0.7–1.3), 5th quintile

 PERFLUOROALKYLS

127
2. HEALTH EFFECTS

Table 2-8. Summary of Cardiovascular Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Savitz et al. 2012b

21.0–717.6 ng/mL (5th PFOA Pregnancy induced
quintile, estimated)
hypertension

OR 1.1 (0.8–1.5)

>178–319 ng/mL
(cumulative, estimated 2nd
PFOA quintile)

Stroke

OR 1.39 (1.11–1.76)*, 2nd quintile

120.6–894.4 ng/mL (4th
PFOA quartile)

Pre-eclampsia

OR 0.9 (0.5–1.8)

Community (C8) (n=4,547 pregnant women)
Simpson et al. 2013
Community (C8) (n=28,541; 11% also had
occupational exposure)
Stein et al. 2009
Community (C8) (n=5.262 pregnant women)
Winquist and Steenland 2014a
Community (C8) (n=28,541; 11% also had
occupational exposure)
Geiger et al. 2014a
General population (NHANES) (n=1,655
adolescents)
Lin et al. 2013a, 2013b

≥3,579 ng/mL (cumulative,
Hypertension
estimated 5th PFOA quintile) Coronary artery disease

HR 0.98 (0.91–1.06), 5th quintile
HR 1.07 (0.93–1.23), 5th quintile

>5.4 ng/mL (4th quartile
PFOA)

Hypertension

OR 0.69 (0.41–1.17), 4th quartile

3.49 ng/mL (median PFOA)

Carotid intima media
thickness

NS (p=0.285 for trend)

General population (n=644)
Mattsson et al. 2015

4.2 and 4.0 ng/mL (median Coronary artery disease
PFOA in cases and controls)

General population (n=231 cases with CHD,
231 controls)
Melzer et al. 2010

10.39 and 9.47 ng/mL (mean Coronary artery disease,
4th quartile PFOA)
angina, and/or heart
General population (NHANES) (n=3,966 adults)
attack
Min et al. 2012
4.00 ng/mL (geometric mean Systolic blood pressure
PFOA)
Hypertension risk
General population (NHANES) (n=2,208)

***DRAFT FOR PUBLIC COMMENT***

OR 0.88 (0.50–1.55), 4th quartile

OR 1.08 (0.70–1.69, p=0.715), 4th
quartile
Association (p=0.0004)*
OR 1.71 (1.23–2.36)*, 4th quartile

 PERFLUOROALKYLS

128
2. HEALTH EFFECTS

Table 2-8. Summary of Cardiovascular Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Shankar et al. 2012

4.0–5.6 and 4.4–6.1 ng/mL
(females and males, 3rd
PFOA quartile)
>5.6 and >6.1 ng/mL
(females and males, 4th
PFOA quartile)

Cardiovascular disease
Peripheral arterial
disease
Coronary heart disease

OR 1.77 (1.04–3.02)*, 3rd quartile
OR 1.78 (1.03–3.08)*, 4th quartile

Stroke

OR 4.26 (1.84–9.89)*, 4th quartile

2.78 ng/mL (median PFOA)

Pre-eclampsia

HR 0.89 (0.65–1.22), per ln unit

General population (NHANES) (n=1,216)

Starling et al. 2014b

OR 2.24 (1.02–4.94)*, 4th quartile

General population (n=976 pregnant women)
PFOS
Darrow et al. 2013
Community (C8) (n=1,330)
Stein et al. 2009
Community (C8) (n=5,262 pregnant women)
Geiger et al. 2014a
General population (NHANES)
(n=1,655 adolescents)
Lin et al. 2013a, 2013b

12.1–<15.9 ng/mL (3rd PFOS Pregnancy-induced
quintile)
hypertension

OR 2.71 (1.33–5.52)* (3rd quintile)

23.2–83.4 ng/mL (4th PFOS
quartile)

Pre-eclampsia

OR 1.6 (1.2–2.3)*

>25.5 ng/mL (4th PFOS
quartile)

Hypertension

OR 0.77 (0.37–1.61), 4th quartile

8.65 ng/mL (median PFOS)

Carotid intima media
thickness

Association (p<0.001 for trend)*

General population (n=644)
Mattsson et al. 2015
General population (n=231 cases with CHD,
231 controls)
Melzer et al. 2010

22.8 and 22.0 ng/mL (median Coronary artery disease
PFOS in cases and controls)

OR 1.07 (0.60–1.92), 4th quartile

Coronary artery disease, OR 0.91 (0.570–1.64, p=0.745), 4th
angina, and/or heart
quartile
General population (NHANES) (n=3,966 adults)
attack
Starling et al. 2014b
12.87 ng/mL (median PFOS) Pre-eclampsia
HR 1.13 (0.84–1.52), per ln unit
57.73 and 50.96 ng/mL
(mean 4th quartile PFOS)

General population (n=976 pregnant women)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

129
2. HEALTH EFFECTS

Table 2-8. Summary of Cardiovascular Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

1.6 ng/mL (median PFHxS in Coronary artery disease
cases and controls)

OR 0.95 (0.54–1.67), 4th quartile

General population (n=231 cases with CHD,
231 controls)
Starling et al. 2014b

0.69 ng/mL (median PFHxS) Pre-eclampsia

HR 0.91 (0.72–1.14), per ln unit

General population (n=976 pregnant women)
PFNA
Lin et al. 2013a, 2013b

0.38 ng/mL (median PFNA)

Carotid intima media
thickness

Inverse association (p=0.014 for
trend)*

0.5 ng/mL (median PFNA in
cases and controls)

Coronary artery disease

OR 0.68 (0.39–1.20), 4th quartile

0.54 ng/mL (median PFNA)

Pre-eclampsia

HR 0.90 (0.70–1.16), per ln unit

PFHxS
Mattsson et al. 2015

General population (n=644)
Mattsson et al. 2015
General population (n=231 cases with CHD,
231 controls)
Starling et al. 2014b
General population (n=976 pregnant women)
PFDeA
Mattsson et al. 2015

0.2 ng/mL (median PFDeA in Coronary artery disease
cases and controls)

OR 0.92 (0.53–1.60), 4th quartile

General population (n=231 cases with CHD,
231 controls)
Starling et al. 2014b

0.10 ng/mL (median PFDeA) Pre-eclampsia

HR 0.88 (0.75–1.04), per ln unit

General population (n=976 pregnant women)
PFUA
Lin et al. 2013a, 2013b

6.59 ng/mL (median PFUA)

NS (p=0.953 for trend)

Carotid intima media
thickness

General population (n=644)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

130
2. HEALTH EFFECTS

Table 2-8. Summary of Cardiovascular Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Mattsson et al. 2015

0.2 ng/mL (median PFUA in
cases and controls)

Coronary artery disease

OR 0.88 (0.51–1.51), 4th quartile

0.17 ng/mL (median PFUA)

Pre-eclampsia

HR 0.78 (0.66–0.92)*, per ln unit

General population (n=231 cases with CHD,
231 controls)
Starling et al. 2014b
General population (n=976 pregnant women)
PFHpA
Mattsson et al. 2015
General population (n=231 cases with CHD,
231 controls)
PFDoA
Mattsson et al. 2015

0.06 and 0.04 ng/mL
Coronary artery disease
(median PFHpA in cases and
controls)

OR 2.58 (1.39–4.78)*, 3rd quartile
OR 1.73 (0.94–3.16), 4th quartile

0.02 ng/mL (median PFDoA Coronary artery disease
in cases and controls)

OR 0.63 (0.35–1.11), 4th quartile

General population (n=231 cases with CHD,
231 controls)
a

See the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 3 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

CHD = coronary heart disease; HR = hazard ratio; LHWA = Little Hocking Water Authority; NHANES = National Health and Nutrition Examination Survey;
NR = not reported; NS = not significant; OR = odds ratio; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid;
PFUA = perfluoroundecanoic acid; SMR = standardized mortality ratio; SPR = standard prevalence ratio

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

131
2. HEALTH EFFECTS

(controls) (Mattsson et al. 2015). Utilizing the NHANES data set, Shankar et al. (2012) found increases
in the risk of peripheral arterial disease, coronary heart disease, or stroke in participants with serum PFOA
levels in the 4th quartile (>5.6 and >6.1 ng/mL in females and males, respectively) and for cardiovascular
disease in participants with serum PFOA levels in the 3rd and 4th quartiles (>4.0 and >4.4 ng/L for females
and males, respectively). In contrast, another NHANES study did not find an association between serum
PFOA and physician-diagnosed coronary artery disease, angina, and/or heart attack (Melzer et al. 2010).
A general population study did not find an association between serum PFOA levels and carotid intima
media thickness (Lin et al. 2013a).

Epidemiology Studies—Hypertension. Occupational, community, and general population exposure
studies have investigated the possible association between PFOA and blood pressure, the risk of
hypertension, and the risk of pregnancy-induced hypertension and/or pre-eclampsia. A study by Min et
al. (2012) utilizing NHANES data found an increase in hypertension risk among participants with serum
PFOA levels in the 4th quartile. In contrast, no increases in the risk of hypertension were observed in
workers at the Washington Works facility (Steenland et al. 2015), adult community members living near
this facility (Winquist and Steenland 2014a), or adolescent NHANES participants (Geiger et al. 2014a).
There is some epidemiological evidence suggesting that an elevated uric acid level is a risk factor for
hypertension (Johnson et al. 2003; Sündstrom et al. 2005). Several occupational, community, and general
population studies have found increases in uric acid levels and increased risks of hyperuricemia; these
data are discussed in Section 2.10. Overall, the results of these studies are suggestive of a connection
between serum PFOA and increased risk of hyperuricemia.

Several studies have examined the possible associations between PFOA and pregnancy-induced
hypertension/pre-eclampsia. Four studies have evaluated the community living near the Washington
Works facility using different approaches to assess PFOA exposure. Savitz et al. (2012a, 2012b) used
residential history and environmental dispersion of PFOA to estimate serum PFOA levels over time.
Stein et al. (2009) used serum PFOA levels measured in 2005–2006 to assess the risk of pre-eclampsia
occurring prior to the blood sampling. The fourth study (Darrow et al. 2013) primarily used serum PFOA
levels measured in 2005–2006 to assess the association with pregnancy-induced hypertension occurring
after the blood samples were collected. Savitz et al. (2012a) found an increased risk of self-reported preeclampsia in C8 Health Project participants with elevated PFOA levels. Another study of participants in
the C8 Health Project that used measured serum perfluoroalkyl levels found significant increases in the
odds ratios (ORs) for pregnancy-induced hypertension in women with higher PFOA (≥6.9 ng/mL) levels
(Darrow et al. 2013). A third study of highly exposed residents reported a weak association between

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

132
2. HEALTH EFFECTS

serum PFOA and pre-eclampsia in subjects whose serum PFOA levels were above the median (Stein et al.
2009); however, there was no dose-response gradient. Using birth record data and serum PFOA levels
predicted from addresses, Savitz et al. (2012b) found no consistent associations between serum PFOA
and the occurrence of pregnancy-induced hypertension in participants in the C8 Health Project. Similarly,
Stein et al. (2009) did not find increases in the odds of self-reported pre-eclampsia among C8 Health
Project participants categorized by serum PFOA levels. Another study of residents of this area did not
find increases in the risk of pregnancy-induced hypertension among residents living in an area where
PFOA-contaminated water was supplied by the Little Hocking Water Authority (Nolan et al. 2010). A
general population study did not find an association between plasma PFOA and the risk of pre-eclampsia
(Starling et al. 2014a).

Laboratory Animal Studies. No histopathological alterations were seen in the heart from rats exposed
intermittently head-only to up to 84 mg/m3 APFO dusts for 2 weeks (Kennedy et al. 1986).
Administration of APFO in the diet at doses up to approximately 100–110 mg/kg/day to male and female
CD rats or 10 mg/kg/day by gavage to Rhesus monkeys did not cause gross or microscopic alterations in
the heart or aorta (Griffith and Long 1980). Similar negative findings were reported in Cynomolgus
monkeys administered up to 20 mg/kg/day APFO by capsule for 26 weeks (Butenhoff et al. 2002) and in
male and female Sprague-Dawley rats that received doses of up to 15 mg/kg/day APFO for 2 years (3M
1983). No morphological alterations were seen in the heart from male rats dermally exposed to
≤2,000 mg/kg APFO for 2 weeks (Kennedy 1985).
Summary. Cardiovascular toxicity as assessed by deaths from heart disease, risk of heart disease, and
risk of hypertension has been evaluated in workers, community members living near a PFOA facility, and
the general population. In general, occupational exposure studies have not found increases in the risks of
deaths from heart disease or in the risks of ischemic heart disease, cerebrovascular disease, or coronary
disease. Inconsistent results have been found in a small number of studies examining residents living in
areas with high PFOA drinking water contamination or the general population. Studies of hypertension
have also not found associations between serum PFOA and hypertension risk. However, studies of highly
exposed residents provide some suggestive evidence of an association between serum PFOA and
increased risks of pregnancy-induced hypertension/pre-eclampsia. Studies in laboratory animals did not
find histological alterations in the heart following acute-, intermediate-, or chronic-duration oral exposure.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

133
2. HEALTH EFFECTS

PFOS

Epidemiology Studies—Heart Disease. Three studies have evaluated the possible association between
PFOS and heart disease. Melzer et al. (2010) did not find an association between serum PFOS and the
risk of physician-diagnosed coronary artery disease, angina, and/or heart attack among NHANES
participants. In a case-control study (Mattsson et al. 2015), no alterations in the risk of coronary artery
disease were observed. Lin et al. (2013a) found an association between serum PFOS levels and carotid
intima media thickness in a general population study.

Epidemiology Studies—Hypertension. No increases in the risk of hypertension associated with serum
PFOS levels were observed in adolescent NHANES participants (Geiger et al. 2014a). Two studies found
increases in the risk of pregnancy-induced hypertension (Darrow et al. 2013) or pre-eclampsia (Stein et al.
2009) associated with serum PFOS levels among C8 participants. No increase in the risk of preeclampsia was observed in a general population study (Starling et al. 2014b).

Laboratory Animal Studies. Administration of doses of up to 0.75 mg/kg/day PFOS (potassium salt) via
capsule to Cynomolgus monkeys for 26 weeks did not cause any significant gross or microscopic
alterations in the heart or aorta (Seacat et al. 2002). Rats that received up to approximately
1.04 mg/kg/day of PFOS in the diet for 2 years had no significant gross or microscopic changes in the
heart (Butenhoff et al. 2012b; Thomford 2002b).
PFHxS

Epidemiology Studies. Two general population studies examined possible cardiovascular outcomes
associated with PFHxS exposure. No increases in the risk of coronary artery disease (Mattsson et al.
2015) or pre-eclampsia (Starling et al. 2014b) were found.
Laboratory Animal Studies. Dosing of rats with≤10 mg/kg/day PFHxS by gavage for 40–60 days did not
cause morphological alterations in the heart (Butenhoff et al. 2009a; Hoberman and York 2003).
PFNA

Epidemiology Studies. In a general population study, an inverse association between serum PFNA levels
and carotid intima media thickness was observed (Lin et al. 2013a). The investigators suggested that this

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

134
2. HEALTH EFFECTS

finding may be secondary to an interaction between higher serum PFOS levels and lower serum PFNA
levels in the study population. Other general population studies have not found increases in the risk of
coronary heart disease (Mattsson et al. 2015) or pre-eclampsia (Starling et al. 2014b) associated with
increasing serum PFNA levels.

PFDeA

Epidemiology Studies. The risks of coronary artery disease (Mattsson et al. 2015) and pre-eclampsia
(Starling et al. 2014b) were not associated with serum PFDeA levels in general population studies.

Laboratory Animal Studies. Death in female C57BL/6N mice following administration of a single lethal
dose of 160 or 320 mg/kg PFDeA by gavage was associated with mural thrombosis of the left ventricle of
the heart (Harris et al. 1989). Doses ≤80 mg/kg did not cause gross or microscopic alterations in the
heart, assessed 30 days after dosing, but 80 mg/kg significantly decreased relative heart weight (Harris et
al. 1989).
PFUA

Epidemiology Studies. Starling et al. (2014b) found an inverse association between serum PFUA levels
and the risk of pre-eclampsia in pregnant women. No association between serum PFUA levels and
carotid intima artery thickness was observed in a general population study (Lin et al. 2013a). A third
general population study (Mattsson et al. 2015) did not find an increase in the risk of coronary artery
disease associated with serum PFUA levels.

PFHpA

Epidemiology Studies. Mattsson et al. (2015) found an increase in the risk of coronary artery disease in
individuals with serum PFHpA levels in the 3rd quartile; however, the risk was not increased for those
with serum levels in the 4th quartile.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

135
2. HEALTH EFFECTS

PFBuS

Laboratory Animal Studies. No morphological alterations were reported in the heart or aorta from rats
dosed with ≤900 mg/kg/day PFBuS by gavage for 28 days (3M 2001) or ≤600 mg/kg/day PFBuS for
90 days (Lieder et al. 2009a).

PFBA

Laboratory Animal Studies. PFBA administered to rats by gavage in doses of up to 184 mg/kg/day for
5 days, 150 mg/kg/day for 28 days, or 30 mg/kg/day for 90 days did not induce gross or microscopic
alterations in the heart (3M 2007a; Butenhoff et al. 2012a; van Otterdijk 2007a, 2007b).
PFDoA

Epidemiology Studies. No increase in the risk of coronary heart disease associated with serum PFDoA
levels was found in a general population study (Mattsson et al. 2015).
2.6

GASTROINTESTINAL

Overview. Available epidemiology data on the potential of perfluoroalkyls to induce gastrointestinal
effects are limited to two studies of workers at a PFOS facility that found mixed results on the possible
association between PFOS and colon polyps; summaries of these studies are presented in the Supporting
Document for Epidemiological Studies for Perfluoroalkyls, Table 4. Studies examining ulcerative colitis
are discussed in Section 2.14, Immunological. Laboratory animal studies have examined the
gastrointestinal tract for morphological alterations following inhalation, oral, or dermal exposure to PFOA
(Tables 2-1, 2-3, and 2-6), oral exposure to PFOS (Table 2-4), and oral exposure to other perfluoroalkyls
(Table 2-5); the NOAELs and LOAELs are presented in Figures 2-4, 2-6, and 2-7. Studies on PFOA and
PFBuS have reported some signs of gastrointestinal irritation following gavage administration. Most
studies did not report histological alterations in the gastrointestinal tract following exposure to PFOA,
PFOS, PFHxS, or PFBA.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

136
2. HEALTH EFFECTS

PFOA

Laboratory Animal Studies. Stomach irritation was reported in male rats exposed head-only to
≥380 mg/m3 APFO dusts for 4 hours (Kennedy et al. 1986). No histopathological alterations were seen in
the stomach, small intestine, or large intestine from male rats exposed intermittently nose-only to up to
84 mg/m3 APFO dusts for 2 weeks (Kennedy et al. 1986).

No significant gross or microscopic alterations of the gastrointestinal tract were observed in male or
female rats exposed to approximately 100–110 mg/kg/day APFO through the diet for 90 days (Griffith
and Long 1980). Similar observations were reported in male and female rats exposed to 15 mg/kg/day
APFO via the diet for 2 years (3M 1983). The same investigators also reported that emesis occurred in
Rhesus monkeys exposed to lethal doses (30 and 100 mg/kg/day) of APFO by gavage for 90 days
(Griffith and Long 1980). In another intermediate-duration study in which Cynomolgus monkeys were
exposed to up to 20 mg/kg/day APFO administered via a capsule for 26 weeks, no treatment-related
alterations in the gastrointestinal tract were observed at termination (Butenhoff et al. 2002).

Intermittent application of up to 2,000 mg/kg/day APFO to the skin of male rats for up to 2 weeks did not
result in gross or microscopic alterations in the gastrointestinal tract (Kennedy 1985).
PFOS

Epidemiology Studies. There are limited data available on the potential of PFOS to induce
gastrointestinal damage. A study of current, retired, or former workers employed for at least 1 year at a
PFOS-based fluorochemical manufacturing facility in Decatur, Alabama found no association between
self-reported incidence of gastric ulcer or colon polyps and having worked in a job with either low
(estimated serum PFOS levels of 390–890 ng/mL) or high (estimated PFOS serum levels of 1,300–
1,970 ng/mL) exposure to PFOS, as compared to workers with no direct workplace exposure (estimated
serum PFOS levels of 110–290 ng/mL) (Grice et al. 2007). A second study of workers at the Decatur
facility found an increase in the risk ratio episodes of care for benign colonic polyps in workers with high
potential exposure to PFOS (Olsen et al. 2004a).

Laboratory Animal Studies. Unpublished data summarized by OECD (2002) indicate that distension of
the small intestine was observed in rats exposed to lethal concentrations of airborne PFOS dusts (1,890–
45,970 mg/m3) for 1 hour. Treatment of rats with up to approximately 1.04 mg/kg/day PFOS via the diet

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

137
2. HEALTH EFFECTS

for 2 years did not induce morphological alterations in the gastrointestinal tract (Butenhoff et al. 2012b;
Thomford 2002b).
PFHxS

Laboratory Animal Studies. No morphological alterations were observed in the gastrointestinal tract of
rats administered ≤10 mg/kg/day PFHxS via gavage for 42–56 days (Butenhoff et al. 2009a; Hoberman
and York 2003).
PFBuS

Laboratory Animal Studies. Necrosis of individual squamous cells and hyperplasia and hyperkeratosis
were observed in the limiting ridge of the forestomach of male and female rats administered
600 mg/kg/day PFBuS via gavage for 90 days (Lieder et al. 2009a); these lesions were likely due to
irritation from the repeated gavage administration with PFBuS. In another study, no morphological
alterations were observed in the gastrointestinal tract of rats administered ≤900 mg/kg/day PFBuS via
gavage for 28 days (3M 2001).
PFBA

Laboratory Animal Studies. Administration of PFBA to rats by gavage in doses of up to 184 mg/kg/day
for 5 days, 150 mg/kg/day for 28 days, or 30 mg/kg/day for 90 days did not cause morphological
alterations in the gastrointestinal tract (3M 2007a; Butenhoff et al. 2012a; van Otterdijk 2007a, 2007b).

2.7

HEMATOLOGICAL

Overview. A small number of epidemiology studies have evaluated hematological endpoints in workers
exposed to PFOA or PFOS and in a community exposure study; these studies did not find alterations in
hematological indices. Details of these studies are presented in the Supporting Document for
Epidemiological Studies for Perfluoroalkyls, Table 5. Laboratory animal studies have evaluated potential
alterations in hematological endpoints for a variety of perfluoroalkyl compounds (Tables 2-1, 2-3, 2-4,
2-5, and 2-6). Some laboratory animal studies have reported alterations in hematological indices
following exposure to higher doses of PFOA, PFOS, PFHxS, PFUA, PFBuS, PFBA, or PFHxA.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

138
2. HEALTH EFFECTS

PFOA

Epidemiology Studies. Information on effects on hematological parameters is available from a study of
residents in the Little Hocking water district in southeastern Ohio where there was significant
environmental exposure to PFOA via the water supply (Emmett et al. 2006b). No significant correlations
between any of the hematology parameters evaluated (including hemoglobin, hematocrit, red blood cell
indices, white cell count, and platelet count) and serum PFOA were observed, whether the analysis
included all of the individuals as a group or separate analyses were done for adults or children. In an
occupational study, the investigators reported no alterations in blood counts in workers, with a range of
serum PFOA levels of 5–9,550 ng/mL (Sakr et al. 2007b).

Laboratory Animal Studies. No treatment-related hematological alterations were reported in male rats
exposed intermittently nose-only to up to 84 mg/m3 APFO dusts for 2 weeks (Kennedy et al. 1986). The
specific parameters evaluated included erythrocyte counts, hemoglobin concentration, hematocrit, and
differential leukocyte counts.

No significant hematological alterations were reported in male and female rats orally dosed with
approximately 100–110 mg/kg/day APFO in diet for 90 days (Griffith and Long 1980). Similar results
were reported in Cynomolgus monkeys treated daily with up to 20 mg/kg/day APFO administered via a
capsule (Butenhoff et al. 2002; Thomford 2001) or in Rhesus monkeys dosed daily by gavage with up to
30 mg/kg/day (Griffith and Long 1980). In a 2-year dietary study in rats dosed with 1.5 or 15 mg/kg/day
APFO, hematology tests performed at various times during the study showed changes in treated groups
consisting of decreases in red blood cell counts, hemoglobin concentration, and hematocrit that were not
always dose-related or consistent among sexes and were within acceptable ranges for the rat (3M 1983).

Hematology tests (erythrocyte count, hemoglobin concentration, hematocrit, total and differential
leukocyte count, and red cell indices) conducted in blood from rats following intermittent dermal
exposure to ≤2,000 mg/kg/day APFO for 2 weeks showed inconsistent alterations or changes of unlikely
biological significance (Kennedy 1985).
PFOS

Epidemiology Studies. Two occupational exposure studies (Olsen et al. 1998a, 2003a) have examined
the potential association between serum PFOS and hematological parameters (including hematocrit,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

139
2. HEALTH EFFECTS

hemoglobin, red blood cells, white blood cells, and platelets) in workers at 3M facilities in Decatur,
Alabama and Antwerp, Belgium; mean measured levels of serum PFOS ranged from 800 to 2,440 ng/mL.
No consistent alterations in hematological parameters were observed at either facility or at the different
measuring time points.

Laboratory Animal Studies. Treatment of male and female rats with approximately 1.5–1.8 mg/kg/day
PFOS (potassium salt) in the diet for 4 weeks did not result in significant alterations in hematological
parameters (Seacat et al. 2003). Oral dosing with 1.3–1.6 mg/kg/day for 14 weeks resulted in a
significant increase (45%) in non-segmented neutrophils (Seacat et al. 2003). The biological significance
of this finding was not discussed by the investigators. In a 4-week study, oral administration of up to
2 mg/kg/day PFOS to Cynomolgus monkeys had no effect on hematological parameters (Thomford
2002a). In Cynomolgus monkeys dosed with 0, 0.03, 0.15, or 0.75 mg/kg/day PFOS (potassium salt)
administered via a capsule for 26 weeks and subjected to comprehensive hematological tests during the
study, the only significant effect was a 9% decrease in hemoglobin in 0.75 mg/kg/day males at
termination (Seacat et al. 2002). The investigators considered this a treatment-related effect, but not
biologically significant given that the value was within the published range and there was no evidence of
blood in the stools. No significant hematological effects were reported in a 2-year study in rats dosed
with approximately 1.04 mg/kg/day PFOS in the diet (Butenhoff et al. 2012b; Thomford 2002b).
PFHxS
Laboratory Animal Studies. Treatment of male rats with doses ≥0.3 mg/kg/day PFHxS by gavage for at
least 42 days significantly increased prothrombin time (Butenhoff et al. 2009a; Hoberman and York
2003). Doses ≥1 mg/kg/day significantly decreased hemoglobin concentration, whereas ≥3 mg/kg/day
decreased erythrocyte count and hematocrit; the decrease in hemoglobin (<5%) was not considered
adverse at 1 mg/kg/day. Oral treatment of female rats with up to 10 mg/kg/day PFHxS did not
significantly alter hematological parameters (Butenhoff et al. 2009a; Hoberman and York 2003).
PFUA

Laboratory Animal Studies. Treatment of rats with 1.0 mg/kg/day PFUA via gavage for 41–46 days
resulted in significant hematological changes (Takahashi et al. 2014). Effects in males included
decreased mean corpuscular volume (MCV) (5%), mean corpuscular hemoglobin (MCH) (5%), activated
partial thromboplastin time (APTT) (16–25%), and fibrinogen (19–33%), and increased platelet counts

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

140
2. HEALTH EFFECTS

(13%) and white blood cells (7%). In females, there were increases in MCV (10%) and MCH (10%) and
a decrease in fibrinogen (32%). The NOAEL was 0.3 mg/kg/day.
PFBuS

Laboratory Animal Studies. A 90-day exposure to PFBuS resulted in significant decreases in
hemoglobin and hematocrit levels in males orally administered 200 or 600 mg/kg/day, and a decrease in
erythrocyte levels was observed in males administered 600 mg/kg/day; the NOAEL was 60 mg/kg/day
(Lieder et al. 2009a). In contrast, no hematological alterations were observed in rats administered
900 mg/kg/day PFBuS for 28 days (3M 2001).
PFBA

Laboratory Animal Studies. Administration of PFBA by gavage to rats in doses of up to 184 mg/kg/day
for 5 days or up to 150 mg/kg/day for 28 days did not result in significant alterations in hematological
parameters (3M 2007a; Butenhoff et al. 2012a; van Otterdijk 2007a). Oral doses of 30 mg/kg/day, but
not 6 mg/kg/day, for 90 days resulted in significant reductions in red blood cell counts, hemoglobin, and
hematocrit, and an increase in red cell distribution width in male rats (Butenhoff et al. 2012a; van
Otterdijk 2007b). This dose level also caused a reduction in MCH and reduced MCH concentration in
male rats. The lower hemoglobin and hematocrit observed in males were still detected at the end of a
3-week recovery period. These hematological effects were considered minor and not evidence of an
adverse effect on red blood cell turnover by the investigator based on lack of alterations in bone marrow
or the spleen.

PFHxA

Laboratory Animal Studies. In a 104-week gavage study, treatment of female Sprague-Dawley rats
(60 or 70/dose) with 200 mg/kg/day PFHxA resulted in significant decreases in mean red blood cell
counts (8.1%) and hemoglobin (5.2%), as well as significant increases in reticulocyte counts at weeks 25
and 51 (23.6 and 53.6%, respectively) (Klaunig et al. 2015). There were no alterations in hematological
parameters in males. The NOAEL was 100 mg/kg/day.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

141
2. HEALTH EFFECTS

2.8

MUSCULOSKELETAL

Overview. Several epidemiology studies have evaluated possible associations between perfluoroalkyls
and bone mineral density, risk of bone fractures, and risk of osteoarthritis; the results of these studies are
summarized in Table 2-9, with more detailed descriptions presented in the Supporting Document for
Epidemiological Studies for Perfluoroalkyls, Table 6. Several cross-sectional community and general
population studies have found associations between serum PFOA and the risk of osteoarthritis,
particularly in participants under the age of 55 years. However, associations were not found in a study of
mostly male workers. Mixed results were found in studies of PFOS, with studies finding a decreased risk
of osteoarthritis, increased risk in women under 50 years of age, or no association. One general
population study found increased risks of osteoarthritis associated with serum PFHxS and PFNA. The
data provide some suggestive evidence of a relationship between serum perfluoroalkyls and osteoarthritis.
Assessing whether there is a link between perfluoroalkyl exposure and osteoarthritis is complicated by the
lack of mechanistic data to support this association and it is noted that there are a number of factors that
contribute to the osteoarthritis risk, and that some of these factors may be affected by perfluoroalkyls,
including elevations in uric acid levels. No morphological alterations were noted in bone or skeletal
muscle in laboratory animals following exposure to PFOA, PFOS, PFBuS, or PFBA; these data are
summarized in Tables 2-1, 2-3, 2-4, and 2-5 and Figures 2-4, 2-6, 2-7, and 2-8.
PFOA

Epidemiology Studies. Several studies have examined the possible association between serum PFOA
levels and the risk of osteoarthritis; the possible mechanisms associated with these findings have not been
elucidated. In an occupational study (80% male), no association between cumulative serum PFOA levels
and the risk of osteoarthritis was found (Steenland et al. 2015). Innes et al. (2011) examined adult
participants in the C8 Health Project and found that the odds of reporting osteoarthritis were higher in
participants with serum PFOA levels in the 2nd, 3rd, and 4th quartiles compared to participants in the
1st quartile. When segregated by age and BMI, the strongest associations between serum PFOA levels
and osteoarthritis were found in subjects under 55 years of age and in nonobese (BMI <30) subjects.
Increases in the risk of osteoarthritis associated with serum PFOA levels were observed in female
NHANES participants (Khalil et al. 2016; Uhl et al. 2013); there were no associations in men. When
stratified by age, the associations were found in women 20–49 years of age, but not in older women (50–
84 years old) (Uhl et al. 2013). Two studies of adult NHANES participants found no associations

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

142
2. HEALTH EFFECTS

Table 2-9. Summary of Skeletal Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

PFOA
Steenland et al. 2015

Cumulative exposure

Osteoarthritis risk

NS (p=0.92 for trend), no lag
NS (p=0.13 for trend), 10-year lag

13.6–28.0 ng/mL (2nd PFOA
quartile)

Osteoarthritis risk
(physician diagnosed)

Occupational (n=3,713 workers)
Innes et al. 2011
Community (C8) (n=49,432 adults)
Khalil et al. 2016

3.7 ng/mL (mean PFOA)

General population (NHANES)
(n=1,914 participants)
Lin et al. 2014
General population (NHANES)
(n=2,339 participants)

4.70 and 3.31 ng/mL
(geometric mean PFOA in
males and females)

OR 1.16 (1.03–1.31)*, 2nd quartile
OR 1.22 (1.02–1.45)*, 2nd quartile
participants <55 years of age
Total femur neck mineral β -0.017 (-0.033 to -0.001)*, women
density
β 0.001 (-0.025–0.022), men
Osteoarthritis risk
OR 1.84 (1.17–2.90; p=0.008)*, per ln(women)
PFOA increase
Total lumbar spine bone NS (p>0.01), premenopausal women,
mineral density
postmenopausal women, men
Total hip bone mineral
NS (p>0.01), premenopausal women,
density
postmenopausal women, men
All fracture types
OR 0.98 (0.75–1.28), premenopausal
women
OR 1.53 (0.63–3.74), postmenopausal
women
OR 0.84 (0.67–1.07), men
Hip fracture
OR 1.59 (0.57–4.46), premenopausal
women
OR 0.48 (0.06–4.16), postmenopausal
women
OR 0.64 (0.39–1.06), men
Wrist fracture
OR 1.07 (0.65–1.77), premenopausal
women
OR 1.21 (0.46–3.13), postmenopausal
women
OR 1.12 (0.75–1.70), men

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

143
2. HEALTH EFFECTS

Table 2-9. Summary of Skeletal Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
Spine fracture

Uhl et al. 2013

Community (C8) (n=49,432 adults)
Khalil et al. 2016

Osteoarthritis risk (selfreported)

≥29.4 ng/mL (4th PFOS
quartile)

Osteoarthritis risk
(physician diagnosed)

OR 0.76 (0.68–0.85)*, 4th quartile

12.7 ng/mL (mean PFOS)

Total femur neck mineral
density
Osteoarthritis risk
(women)
Total lumbar spine bone
mineral density

β -0.016 (-0.029 to -0.002)*, women
β -0.013 (-0.024 to -0.002)*, men
OR 1.14 (0.68–1.94; p=0.619), per
ln-PFOS increase
β -0.022 (-0.038 to -0.007)*,
premenopausal women
NS (p>0.01), postmenopausal women
NS (p>0.01), men
NS (p>0.01), premenopausal women,
postmenopausal women, men
OR 0.97 (0.75–1.24), premenopausal
women
OR 1.59 (0.88–2.86), postmenopausal
women
OR 0.92 (0.73–1.16), men

General population (NHANES)
(n=1,914 participants)
Lin et al. 2014
General population (NHANES)
(n=2,339 participants)

OR 1.83 (0.59–5.61), premenopausal
women
OR 0.84 (0.46–1.53), postmenopausal
women
OR 1.54 (0.85–2.79), men
OR 1.98 (1.24–3.19)*, 4th quartiles
females
OR 0.82 (0.40–1.70), 4th quartile males
OR 4.95 (1.27–19.4)*, 4th quartile
women 20–49 years of age
OR 1.33 (0.82–1.16), 4th quartile
women 50–84 years of age

>5.89 ng/mL (4th PFOA
quartile)

General population (NHANES) (n=1,888 male
and 1,921 female adults)

PFOS
Innes et al. 2011

Resultc

19.23 and 12.09 ng/mL
(geometric mean PFOS in
males and females)

Total hip bone mineral
density
All fracture types

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

144
2. HEALTH EFFECTS

Table 2-9. Summary of Skeletal Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
Hip fracture

Wrist fracture

Spine fracture

PFHxS
Khalil et al. 2016

Resultc
OR 1.12 (0.62–2.03), premenopausal
women
OR 0.83 (0.23–3.00), postmenopausal
women
OR 1.07 (0.76–1.52), men
OR 1.04 (0.63–1.72), premenopausal
women
OR 1.22 (0.61–2.45), postmenopausal
women
OR 1.09 (0.72–1.66), men
OR 0.52 (0.15–1.86), premenopausal
women
OR 1.12 (0.26–4.78), postmenopausal
women
OR 1.27 (0.67–2.42), men
β -0.014 (-0.074 to -0.014)*, women
β -0.026 (-0.065–0.013), men
OR 1.64 (1.14–2.38; p=0.008)*, per
ln-PFHxS increase

2.5 ng/mL (mean PFHxS)

Total femur bone mineral
density
Osteoarthritis risk
(women)

1.9 ng/mL (mean PFNA)

Total femur bone mineral β -0.040 (-0.077 to -0.003)*, women
density
β 0.007 (-0.031–0.045), men

General population (NHANES)
(n=1,914 participants)
PFNA
Khalil et al. 2016
General population (NHANES) (n=1,914
participants)

Osteoarthritis risk
(women)

OR 1.45 (1.02–2.05; p=0.001)*, per
ln-PFNA increase.

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 6 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

NHANES = National Health and Nutrition Examination Survey; NS = not significant; OR = odds ratio; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

145
2. HEALTH EFFECTS

between serum PFOA and bone mineral density of the total femur (Khalil et al. 2016), hip (Lin et al.
2014), or lumbar spine (Khalil et al. 2016; Lin et al. 2014); however, an inverse association was found in
the neck portion of the femur in the Khalil et al. (2016) study. Additionally, Lin et al. (2014) did not find
associations between serum PFOA levels and the risk of bone fractures (total fractures, hip fractures,
wrist fractures, or spine fractures) in premenopausal women, postmenopausal women, or men.
Laboratory Animal Studies. In male rats exposed head-only to up to 84 mg/m3 APFO dusts for up to
2 weeks, examinations of the sternebrae were unremarkable (Kennedy et al. 1986). Similarly, no gross or
microscopic alterations were reported in the sternum from rats following dietary exposure to 100–
110 mg/kg/day APFO for 90 days (Griffith and Long 1980) or in the femur, sternum, or thigh skeletal
muscle from Cynomolgus monkeys dosed with up to 20 mg/kg/day APFO administered via a capsule for
26 weeks (Butenhoff et al. 2002). In utero exposure to PFOA resulted in morphometrical alterations in
the femur (increases in the periosteal area) and decreases in bone mineral density in the tibia of 13- or
17-month-old mice (Koskela et al. 2016). No alterations in biomechanical properties were found. The
investigators noted that PFOA concentrations were 4–5 times higher in the PFOA-exposed animals than
in controls.
PFOS

Epidemiology Studies. Several epidemiology studies have evaluated the potential of PFOS to induce
skeletal damage. In the participants of the C8 Health Study, a decreased risk of osteoarthritis was found
in participants with serum PFOS levels in the 2nd, 3rd, and 4th quartiles (Innes et al. 2011). In contrast, Uhl
et al. (2013) found an increased risk of osteoarthritis in NHANES participants with serum levels of
>20.97 ng/mL. When categorized by sex and age, the osteoarthritis risk was approximately 5 times
higher in women aged 20–49 years with serum PFOS levels in the 4th quartile. Another study of
NHANES participants (Khalil et al. 2016) did not find an increased risk of osteoarthritis in women.
However, the study did find an inverse association between serum PFOS and femur neck bone mineral
density, but no associations with total femur or lumbar spine bone mineral density.

Laboratory Animal Studies. Treatment of monkeys with up to 0.75 mg/kg/day PFOS (potassium salt)
administered via a capsule for 26 weeks had no significant effect on the gross or microscopic appearance
of the femur, sternum, or thigh skeletal muscle (Seacat et al. 2002). Similar observations were made in
rats treated with up to 1.04 mg/kg/day PFOS in the diet for 2 years (Butenhoff et al. 2012b; Thomford
2002b).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

146
2. HEALTH EFFECTS

PFHxS

Epidemiology Studies. A study of NHANES participants found an increase in the risk of osteoarthritis
among women that was associated with serum PFHxS levels (Khalil et al. 2016). An inverse association
between serum PFHxS (fourth quartile) and total femur bone mineral density was also found in women.
There were no associations between serum PFHxS and femur neck or lumbar spine bone mineral density
(Khalil et al. 2016).
PFNA

Epidemiology Studies. Khalil et al. (2016) found an increase in the risk of osteoarthritis in women
NHANES participants that was associated with serum PFNA levels. Increasing serum PFNA levels did
not result in alterations in bone mineral density of the lumbar spine or femur neck, but was inversely
associated with total femur bone mineral density in women with serum PFNA levels in the fourth quartile.
PFBuS

Laboratory Animal Studies. Treatment of rats with up to 900 mg/kg/day PFBuS by gavage for 28 or
90 days did not induce morphological alterations in skeletal muscle (3M 2001; Lieder et al. 2009a).
PFBA

Laboratory Animal Studies. PFBA administered to rats by gavage in doses of up to 184 mg/kg/day for
5 days did not induce morphological alterations in skeletal muscle (3M 2007a). Administration of
150 mg/kg/day PFBA for 28 days or 30 mg/kg/day for 90 days did not induce gross or microscopic
alterations in bone (femur and sternum) or skeletal muscle (Butenhoff et al. 2012a; van Otterdijk 2007a,
2007b).
2.9

HEPATIC

Overview. Epidemiology studies on perfluoroalkyls have examined three potential hepatic outcomes:
liver disease, alterations in serum enzyme and bilirubin levels, and alterations in serum lipid levels.
Summaries of the epidemiology studies examining these outcomes are presented in Tables 2-10, 2-11, and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

147
2. HEALTH EFFECTS

2-12, with more detailed descriptions presented in the Supporting Document for Epidemiological Studies
for Perfluoroalkyls, Table 7. There are limited epidemiology data on potential associations between
serum perfluoroalkyls and risk of liver disease. Occupational exposure and community studies did not
find increased risk of liver disease associated with PFOA or PFOS. As assessed by serum enzyme and
bilirubin levels, the epidemiology studies provide suggestive evidence of liver damage. Increases in
aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyl transferase
(GGT) levels and decreases in serum bilirubin levels have been reported in occupational, community,
and/or general population studies. Although there is considerable variability across studies, the evidence
is adequate for PFOA, PFOS, and PFHxS, particularly for ALT levels; consistent results were not found
for PFNA. The results of available epidemiology studies suggest associations between increases in serum
lipids, particularly total cholesterol and LDL cholesterol, and serum PFOA, PFOS, PFNA, and PFDeA.
For PFHxS, PFUA, PFHpA, PFBuS, PFBA, and PFDoA, there are too few studies or the results are too
inconsistent to determine if they also would affect serum lipid levels at environmental exposure levels.

Numerous animal studies have evaluated the hepatotoxicity of perfluoroalkyls following inhalation, oral,
and dermal exposure; summaries of these studies are presented in Tables 2-1, 2-2, 2-3, 2-4, 2-5, and 2-6
and the NOAEL and LOAEL values are graphically presented in Figures 2-4, 2-5, 2-6, 2-7, and 2-8. The
results of these studies provide strong evidence that the liver is a sensitive target of PFOA, PFOS, PFHxS,
PFNA, PFUA, PFBuS, and PFBA toxicity. Observed effects in rodents include increases in liver weight;
hepatocellular hypertrophy, hyperplasia, and necrosis; and decreases in serum cholesterol and triglyceride
levels. As discussed in greater detail in Section 2.20, these effects are believed to be initiated by PPARα;
however, studies in PPARα-null mice suggest that other mechanisms are also involved. Increases in liver
weight have also been observed in monkey studies for PFOA and PFOS; these studies have also found
alterations in serum lipid levels and hepatocellular hypertrophy (PFOS only).

To address concern over the relevance of liver enlargement in rodents to human health risk, the European
Society of Toxicologic Pathology (ESTP) convened an expert panel to define what constitutes an adverse
hepatic effect and whether hepatic effects induced by nuclear hormone receptors such as PPARα,
constitutive androstane receptor (CAR), or pregnane X receptor (PXR) are rodent-specific adaptive
reactions; the findings of the panel are summarized by Hall et al. (2012). As discussed by Hall et al.
(2012), criteria were established for determining whether increases in liver organ weight and liver cell
hypertrophy observed in studies of rodents exposed to agents inducing enzyme induction can be
considered adaptive responses and of little relevance to humans. According to the ESTP criteria,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

148
2. HEALTH EFFECTS

Table 2-10. Summary of Liver Disease in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

PFOA
Anderson-Mahoney et al. 2008

NR

Liver problems (selfreported)

SPR 1.01 (0.64–1.59)

Liver disease

Community (C8) (n=28,831)

Cumulative
16.5 ng/mL (median
2005/2006)

Steenland et al. 2015

Cumulative

Occupational (n=3,713)
PFOS
Alexander et al. 2003

NR

Liver cirrhosis deaths

SMR 0.81 (0.10–2.94)

1,300–1,970 ng/mL (high
potential workers)

Liver disease
Cholelithiasis
Cholecystitis

OR 1.21 (0.56–2.60)
OR 0.91 (0.57–1.46)
OR 1.15 (0.65–2.06)

Community (n=566)
Darrow et al. 2016

HR 0.97 (0.92–1.03), no lag per ln
increase in PFOA
HR 0.98 (0.93–1.04), 10-year lag
Enlarged liver, fatty liver, HR 0.97 (0.91–1.04), no lag per ln
or cirrhosis
increase in PFOA
HR 1.00 (0.94–1.07), 10-year lag
Non-hepatitis liver disease NS (p=0.86), no lag
risk
NS (p=0.40), 10-year lag

Occupational (n=2,083)
Grice et al. 2007
Occupational (n=1,400)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

149
2. HEALTH EFFECTS

Table 2-10. Summary of Liver Disease in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Olsen et al. 2004a

NR

Cholelithiasis or acute
cholecystitis

RREpC 8.6 (1.1–>100)*
RREpC 25 (2.1–>100)*, workers with
≥10 years high exposure potential
RREpC 1.2 (0.2–8.6)
RREpC 1.6 (0.8–2.9)
RREpC 2.6 (1.2–5.5)*, workers with
≥10 years high exposure potential

Occupational (n=652 exposed, n=659 for nonexposed)

Liver disease
Biliary duct disorders

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 7 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living near
PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

HR = hazard ratio; NR = not reported; NS = not significant; OR = odds ratio; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; RREpC = risk ratio
episode of care; SMR = standardized mortality ratio; SPR = standard prevalence ratio

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

150
2. HEALTH EFFECTS

Table 2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humansa
Reference and study populationb
PFOA
Costa et al. 2009
Occupational (n=37 current workers;
n=16 former workers; n=107 non-exposed
workers)

Serum perfluoroalkyl level

Outcome evaluated

Resultc

12,930 ng/mL (mean PFOA
current workers)
6,810 ng/mL (mean former
workers)

AST

NS (p>0.05) (34 current workers)

ALT

NS (p>0.05) (34 current workers)
Association (p<0.01)* (56 current,
former, non-exposed workers)
NS (p>0.05) (34 current workers)
Association (p<0.01)* (56 current,
former, non-exposed workers)
Inverse association (p<0.01)*
(56 current, former, non-exposed
workers)
NS (p=0.32)
NS (p=0.80)
NS (p=0.81)

GGT

Total bilirubin

Gilliland and Mandel 1996
Occupational (n=115)
Olsen et al. 2000
Occupational (n=111, 80, and 74 in 1993,
1995, and 1997)

NR (serum fluorine levels used
as surrogate for serum PFOA)

ALT
AST
GGT

5,000, 6,400, and 6,400 ng/mL ALT
(mean PFOA in 1993, 1995, and
1997)
AST
Workers divided into three
groups: 0–<1,000, 1,000–
<10,000, and ≥10,000 ng/mL
GGT

Total bilirubin

Direct bilirubin

***DRAFT FOR PUBLIC COMMENT***

NS (p=0.82, 0.30, 0.73) differences
between exposure groups for each
measurement period
NS (p=0.33, 0.45, 0.83) differences
between exposure groups for each
measurement period
NS (p=0.24, 0.41, 0.78) differences
between exposure groups for each
measurement period
NS (p=0.48, 0.11, 0.58) differences
between exposure groups for each
measurement period
NS (p=0.82, 0.05, 0.74) differences
between exposure groups for each
measurement period

 PERFLUOROALKYLS

151
2. HEALTH EFFECTS

Table 2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Olsen and Zobel 2007

2170 ng/mL (mean 8th PFOA
GGT
decile)
Elevated GGT
12,150 ng/mL (mean 10th PFOA Total bilirubin
decile)
ALT
Elevated ALT
AST

Association (p=0.05)*
OR 1.0 (0.3–2.9), 10th decile
Inverse association (p=0.001)*
NS (p=0.06)
OR 1.2 (0.5–3.4), 10th decile
NS (p=0.55)

1,130 ng/mL (mean PFOA)

Total bilirubin
AST
ALT
GGT
GGT
AST

Association (p=0.006)*
Association (p=0.009)*
NS (p>0.05)
NS (p>0.05)
Association (p=0.016)*
NS (p=0.317)

ALT
Bilirubin
AST
ALT

NS (p=0.124)
NS (p=0.590)
Association (p=0.02)*
NS (p=0.38)

ALT

Association (p<0.0001 for trend)*,
cumulative levels
Association (p<0.0001 for trend)*,
2005/2006 levels
NS (p=0.1021), cumulative levels
NS (p=0.1552), 2005/2006 levels
Inverse association (p=0.0029 for
trend)*, cumulative levels
Inverse association (p=0.0036 for
trend)*, 2005/2006 levels

Occupational (n=552)

Sakr et al. 2007a
Occupational (n=454)

Sakr et al. 2007b

428 ng/mL (mean PFOA)

Occupational (n=1,025)

Wang et al. 2012

2,157.74 ng/mL (mean PFOA)

Occupational (n=55)
Darrow et al. 2016
Community (C8) (n=28,831)

Cumulative
16.5 ng/mL (median PFOA in
2005/2006)

Outcome evaluated

GGT
Bilirubin

***DRAFT FOR PUBLIC COMMENT***

Resultc

 PERFLUOROALKYLS

152
2. HEALTH EFFECTS

Table 2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Emmett et al. 2006b

354 ng/mL (median PFOA)

ALT
Abnormal ALT
AST
Abnormal AST
GGT
Abnormal GGT
ALT
Abnormal ALT
GGT
Abnormal GGT
Direct bilirubin
Abnormal bilirubin
ALT
AST

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Inverse association (p=0.03)*
NS (p>0.05)
NS (p>0.05)
Correlation (p<0.001)*
OR 1.19 (1.03–1.37)*, 3rd decile
Correlation (p<0.001)*
NS (p=0.213 for trend)
NS (p>0.05)
NS (p=0.496 for trend)
NS (p=0.05)
NS (p=0.22)

ALT
Elevated ALT
AST
Elevated AST

Association (p<0.001)*
Association (p=0.007 for trend)*
Association (p<0.01)*
NS (p=0.058 for trend).

Community (n=371)

Gallo et al. 2012

NR

Community (C8) (n=46,452)

Wang et al. 2012

378.30 ng/mL (mean PFOA)

Community (n=132)
Gleason et al. 2015

3.7 ng/mL (median PFOA)

General population (NHANES) (n=4,333)

Lin et al. 2010
General population (NHANES) (n=2,216)
Yamaguchi et al. 2013
General population (n=608)

GGT
Elevated GGT
Total bilirubin
Elevated bilirubin
5.05 and 4.06 ng/mL (geometric ALT
mean PFOA in males and
GGT
females)
Total bilirubin
2.1 ng/mL (mean PFOA)
ALT
AST
GGT

***DRAFT FOR PUBLIC COMMENT***

Association (p<0.01)*
Association (p=0.042 for trend)*
Association (p<0.01)*
Association (p<0.001 for trend)*
Association (p=0.005)*
Association (p=0.019)*
NS (p=0.645)
Association (p=0.02)*
Association (p=0.001)*
Association (p=0.03)*

 PERFLUOROALKYLS

153
2. HEALTH EFFECTS

Table 2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humansa
Reference and study populationb
PFOS
Grice et al. 2007
Occupational (n=1,400)
Olsen et al. 1999
Occupational (n=178 in 1995; n=149 in
1997)

Olsen et al. 2003a

Serum perfluoroalkyl level

Outcome evaluated

Resultc

1,300–1,970 ng/mL (high
potential workers)

Cholelithiasis

OR 0.91 (0.57–1.46)

Cholecystitis

OR 1.15 (0.65–2.06)

2,440 and 1,930 ng/mL (mean
PFOS in 1995 in DeCatur and
Antwerp)
1,960 and 1,480 ng/mL (mean
PFOS in 1997 in DeCatur and
Antwerp)

AST

2460 ng/mL (median 4th PFOS
quartile)

AST
ALT

NS (p=0.14 for trend), 1995
NS (p=0.67 for trend), 1997
NS (p=0.38 for trend), 1995
NS (p=0.46 for trend), 1997
NS (p=0.71 for trend), 1995
NS (p=0.34 for trend), 1997
NS (p>0.05), no adjustments
Higher levels (p<0.05)*, males only
with no adjustments
OR 2.1 (0.6–7.3)
Difference (p<0.05)*, females only
with no adjustments
OR 2.0 (0.7–5.8)
Correlation (p<0.001)*
OR 1.19 (1.04–1.37)*, 5th decile
NS (p>0.05)
Association (p=0.047 for trend)*
Correlation (p<0.001)*
Association (p=0.015 for trend)*

Occupational (n=518)

ALT
GGT

Risk of abnormal ALT
GGT

Gallo et al. 2012
Community (C8) (n=46,452)

NR

Risk of abnormal GGT
ALT
Abnormal ALT
GGT
Abnormal GGT
Direct bilirubin
Abnormal bilirubin

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

154
2. HEALTH EFFECTS

Table 2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Gleason et al. 2015

11.3 ng/mL (median PFOS)

ALT
Elevated ALT

NS (p>0.01)
NS (p=0.370 for trend)

AST
Elevated AST
GGT
Elevated GGT
Total bilirubin
Elevated bilirubin

NS (p>0.01)
NS (p=0.438 for trend)
NS (p>0.01)
NS (p=0.654 for trend)
NS (p>0.01)
Association (p=0.028 for trend)*

General population (NHANES) (n=4,333)

Lin et al. 2010
General population (NHANES) (n=2,216)
Yamaguchi et al. 2013
General population (n=608)
PFHxS
Gleason et al. 2015
General population (NHANES) (n=4,333)

Lin et al. 2010
General population (NHANES) (n=2,216)

27.39 and 22.20 ng/mL
ALT
(geometric mean PFOS in males GGT
and females)
Total bilirubin
5.8 ng/mL (mean PFOS)
ALT
AST
GGT

NS (p=0.066)
NS (p=0.808)
NS (p=0.223)
Association (p=0.03)*
Association (p=0.01)*
Association (p=0.03)*

1.8 ng/mL (median PFHxS)

Association (p<0.01)*
NS (p=0.484 for trend)
Association (p<0.001)*
NS (p=0.230 for trend)
NS (p>0.01)
NS (p=0.415 for trend)

ALT
Elevated ALT
AST
Elevated AST
GGT
Elevated GGT

Total bilirubin
Elevated bilirubin
2.29 and 1.72 ng/mL (geometric ALT
mean PFHxS in males and
GGT
females)
Total bilirubin

***DRAFT FOR PUBLIC COMMENT***

Association (p<0.01)*
Association (p=0.041 for trend)*
NS (p=0.691)
NS (p=0.898)
NS (p=0.063)

 PERFLUOROALKYLS

155
2. HEALTH EFFECTS

Table 2-11. Summary of Alterations in Serum Hepatic Enzymes and Bilirubin Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

PFNA
Mundt et al. 2007

NR

ALT

NS, longitudinal analysis

AST
GGT
Bilirubin
ALT
Elevated ALT
AST

NS, longitudinal analysis
NS, longitudinal analysis
NS, longitudinal analysis
Association (p<0.001)*
NS (p=0.042 for trend)
NS (p>0.01)

Occupational (n=592)

Gleason et al. 2015
General population (NHANES) (n=4,333)

Lin et al. 2010
General population (NHANES) (n=2,216)

1.4 ng/mL (median PFNA)

Elevated AST
GGT
Elevated GGT
Total bilirubin
Elevated bilirubin
0.89 and 0.72 ng/mL (geometric ALT
mean PFNA in males and
GGT
females)
Total bilirubin

NS (p=0.516 for trend)
Association (p<0.01)*
NS (p=0.126 for trend)
NS (p>0.01)
NS (p=0.614 for trend)
NS (p=0.131)
NS (p=0.857)
NS (p=0.053)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 7 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

ALT = alanine aminotransferase; AST = aspartate aminotransferase; GGT = gamma-glutamyl transferase; OR = odds ratio; NHANES = National Health and
Nutrition Examination Survey; NR = not reported; NS = not significant; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

156
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

NR

Total cholesterol
Non-HDL cholesterol
HDL cholesterol
LDL cholesterol
Total triglycerides
12,930 ng/mL (mean PFOA in Total cholesterol
current workers)
6,810 ng/mL (mean PFOA in
former workers)
HDL cholesterol
Triglycerides
NR (serum fluorine levels
Total cholesterol
used as surrogate for serum Total LDL
PFOA)
Total HDL

Association (p=0.03)*
Association (p=0.03)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p=0.005)* (current
workers)
Association (p<0.05)* (56 current,
former, non-exposed workers)
NS (p>0.05) (34 current workers)
NS (p>0.05)
NS (p=0.62)
NS (p=0.87)

5,000, 6,400, and
Total cholesterol
6,400 ng/mL (mean PFOA in
1993, 1995, and 1997)
LDL cholesterol
Workers divided into three
groups: 0–<1,000, 1,000–
<10,000, and ≥10,000 ng/mL
HDL cholesterol

NS (p=0.45, 0.48, 0.08) differences
between exposure groups for each
measurement period
NS (p=0.84, 0.96, 0.11) differences
between exposure groups for each
measurement period

PFOA
Costa 2004
Occupational (n=35)

Costa et al. 2009
Occupational (n=37 current workers; n=16
former workers; n=107 non-exposed workers)

Gilliland and Mandel 1996
Occupational (n=115)
Olsen et al. 2000
Occupational (n=111, 80, and 74 in 1993,
1995, and 1997)

Triglycerides

***DRAFT FOR PUBLIC COMMENT***

NS (p=0.66)

NS (p=0.32, 0.70, 0.40) differences
between exposure groups for each
measurement period
NS (p=0.77, 0.07, 0.13) differences
between exposure groups for each
measurement period

 PERFLUOROALKYLS

157
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Olsen and Zobel 2007

2,170 ng/mL (mean of
8th PFOA decile)
12,150 ng/mL (mean of
10th PFOA decile)

Occupational (n=552)

Sakr et al. 2007a

1,130 ng/mL (mean PFOA)

Occupational (n=454)

Sakr et al. 2007b

428 ng/mL (mean PFOA)

Occupational (n=1,025)

Steenland et al. 2015
Occupational (n=3,713)
Wang et al. 2012

Cumulative PFOA

2,157.74 ng/mL (mean
PFOA)

Occupational (n=55)

Emmett et al. 2006b

354 ng/mL (median PFOA)

Resultc

Total cholesterol
NS (p=0.20)
Elevated total cholesterol OR 1.1 (0.5–2.6), 10th decile
LDL cholesterol
Elevated LDL- cholesterol
HDL cholesterol
Decreased HDL
cholesterol
Triglycerides
Elevated triglycerides
Total cholesterol

NS (p=0.81)
OR 1.2 (0.5–2.8), 10th decile
Association (p=0.01)*
OR 1.8 (0.7–4.8), 10th decile

LDL cholesterol
HDL cholesterol
Triglycerides
Total bilirubin
Total cholesterol
LDL cholesterol
VLDL cholesterol

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p=0.006)*
Association (p=0.002)*
Association (p=0.008)*
Association (p=0.031)*

HDL cholesterol
Triglycerides
Elevated cholesterol

NS (p=0.680)
NS (p=0.384)
NS (p=0.56), no lag
NS (p=0.62), 10-year lag

Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides
Total cholesterol
Abnormal cholesterol

NS (p=0.36)
NS (p=0.43)
Inverse association (p=0.01)*
NS (p=0.37)
NS (p>0.05)
NS (p>0.05)

Community (n=371)

***DRAFT FOR PUBLIC COMMENT***

Association (p=0.0001)*
OR 1.8 (0.8–4.4), 10th decile
Association (p=0.011)*

 PERFLUOROALKYLS

158
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Fitz-Simon et al. 2013

140.1 and 68.2 ng/mL (mean Total cholesterol
PFOA at first and second
examinations)
LDL cholesterol

-1.65% (-0.32 to -2.97)*, 50%
decrease in PFOA
-3.58% (-1.47 to -5.66)*, 50%
decrease in PFOA
-1.33% (0.21 to -2.85), 50% decrease
in PFOA
0.78% (5.34 to -3.58), 50% decrease
in PFOA
Association (p<0.001)*, children
5th quintile
Association (p<0.001)*, adolescents
5th quintile
OR 1.1 (1.0–1.3), 2nd quintile

Community (C8) (n=560 adults)

HDL cholesterol
Triglycerides
Frisbee et al. 2010
Community (C8) (n=12,476 children and
adolescents)

77.7 ng/mL (mean PFOA in
children)
61.8 ng/mL (mean PFOA in
adolescents)

Total cholesterol

Abnormal cholesterol
LDL cholesterol

Abnormal LDL levels

Steenland et al. 2009b
Community (C8) (n=46,294)

80.3 ng/mL (mean PFOA)
13.2–26.5 ng/mL (2nd PFOA
quartile)

Association (p=0.001)*, children
5th quintile
Association (p=0.004)*, adolescents
5th quintile
OR 1.2 (1.0–1.5), 2nd quintile

HDL cholesterol

NS (p=0.88), children 5th quintile
NS (p=0.20), adolescents 5th quintile

Triglycerides

NS (p=0.1), children 5th quintile
NS (p=0.1), adolescents 5th quintile

Total cholesterol
Abnormal cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides

Association (p<0.001 for trend)*
OR 1.21 (1.12–1.31)*, 2nd quartile
Association (p<0.05 for trend)*
NS (p>0.05)
Association (p<0.05 for trend)*

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

159
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Wang et al. 2012
Community (n=132)

378.30 ng/mL (mean PFOA)

Total cholesterol
LDL cholesterol

NS (p=0.85)
NS (p=0.97)

Triglycerides
HDL cholesterol
Hypercholesterolemia

NS (p=0.73)
NS (p=0.39)
HR 1.24 (1.15–1.33)*, 2nd quintile for
cumulative exposure

Total cholesterol

Association (p=0.01)*

Total cholesterol
High cholesterol levels
Non HDL cholesterol
LDL cholesterol
HDL cholesterol
Total cholesterol
Elevated cholesterol

NS (p=0.22)
OR 1.5 (0.86–2.62), 4th quartile
NS (p=0.13)
NS (p=0.63)
NS (p=0.96)
Association (p=0.015)*
OR 0.55 (0.09–3.31)

LDL cholesterol
Elevated LDL
HDL cholesterol
Elevated HDL
Triglycerides
Elevated triglyceride

Association (p=0.022)*
OR 0.71 (0.14–3.49)
NS (p=0.260)
OR 0.67 (0.13–3.51)
NS (p=0.298)
OR 1.97 (0.59–6.55)

Winquist and Steenland 2014a
Community (C8) (n=28,541)
Eriksen et al. 2013

142–<234 ng/mL (estimated
2nd quintile for cumulative
PFOA)
7.1 ng/mL (mean PFOA)

General population (n=753)
Fisher et al. 2013
General population (n=2,368)

2.46 ng/mL (mean PFOA)

Fu et al. 2014a

1.43 ng/mL (median PFOA)

General population (n=133)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

160
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Geiger et al. 2014b

4.2 ng/mL (mean PFOA)

Association (p=0.0170 for trend)*
OR 1.44 (1.11–1.88, p=0.0253 for
trend)*, log transformed PFOA
Association (p=0.0027 for trend)*

General population (NHANES) (n=815 12–
18-year-old adolescents)

Maisonet et al. 2015a
General population (n=111 for 7-year-old and
n=88 for 15-year-old girls

Total cholesterol
Elevated cholesterol
LDL cholesterol

1.1–3.1, 3.2–4.4, and 4.5–
16.4 ng/mL (maternal PFOA
for 1st, 2nd, and 3rd tertiles)

Elevated LDL
HDL cholesterol
Decreased HDL
Triglycerides
Elevated triglycerides
Total cholesterol in 7 year
olds

NS (p=0.0539 for trend)
NS (p=0.1769 for trend)
NS (p=0.1493 for trend)
NS (p=0.9943 for trend)
NS (p=0.5975 for trend)
Association (β 13.75, 0.05–27.45)*,
1st tertile
NS, 2nd and 3rd tertiles
Total cholesterol in
Association (β 17.19, 0.405–33.93)*,
15 year olds
1st tertile
NS, 2nd and 3rd tertiles
LDL cholesterol in 7 year Association (β 14.01, 3.26–24.76)*,
olds
1st tertile
NS, 2nd and 3rd tertiles
LDL cholesterol in 15 year Association (β 14.261, 0.25–28.26)*,
olds
1st tertile
NS, 2nd and 3rd tertiles
HDL cholesterol
NS (β -0.40, -1.82–1.01), 3rd tertile
7 year olds
NS (β -0.520, -2.10–1.06), 3rd tertile
15 year olds
Triglycerides
NS (β -0.020, -0.068–0.029),
3rd tertile, 7 year olds
NS (β -0.013, -0.051–0.025),
3rd tertile, 15 year olds

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

161
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Nelson et al. 2010

4.6 ng/mL (mean PFOA)

Total cholesterol
LDL cholesterol

NS (p=0.07)
NS (p=0.84)

Non-HDL cholesterol

General population (NHANES) (n=860)

Skuladottir et al. 2015
General population (n=854 pregnant women)
Starling et al. 2014a

HDL cholesterol

Association (p=0.05)*
β 1.38 (0.12–2.65), per ng/mL
increase in PFOA
NS (p=0.34)

4.1 ng/mL (mean PFOA)

Total cholesterol

Association (p=0.01 for trend)*

2.25 ng/mL (50th PFOA
percentile)

Total cholesterol

NS (β 2.58; -4.32–9.47), per ln-unit
increase in PFOA
NS (β 0.35 -3.97–8.48), per ln-unit
increase in PFOA
Association (β 3.42 0.56–6.28)*,
4th quartile
NS (β 0.00 (-0.07–0.06), per ln-unit
increase in PFOA
NS (p=0.91), normal weight children
Association (p=0.002)*, obese
children

General population (n=854 pregnant women)

LDL cholesterol
HDL cholesterol
Triglycerides

Timmermann et al. 2014

9.3 ng/mL (median PFOA)

General population (n=499 children, 8–10 years
old)
Zeng et al. 2015
1.1 and 0.92 ng/mL (mean
PFOA in boys and girls)
General population (n=225 adolescents, 12–
15 years old)

Triglycerides

Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides

***DRAFT FOR PUBLIC COMMENT***

Association (p=0.001)*
Association (p=0.002)*
NS (p=0.06)
Association (p<0.001)*

 PERFLUOROALKYLS

162
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb
PFOS
Olsen et al. 1999
Occupational (n=178 in 1995; n=149 in 1997)

Serum perfluoroalkyl level Outcome evaluated

Resultc

2,440 and 1,930 ng/mL
(mean PFOS in 1995 in
DeCatur and Antwerp)
1,960 and 1,480 ng/mL
(mean in 1997 in DeCatur
and Antwerp)

NS (p=0.96 for trend), 1995
Association (p=0.006 for trend)*,
1997
NS (p=0.87 for trend), 1995
Association (p=0.01 for trend)*,
1997
Inverse association (p=0.04 for
trend)*, 1995
NS (p=0.34) 1997
NS (p=0.35 for trend), 1995
NS (p=0.67 for trend), 1997
NS (p>0.05), no adjustments
Association (p=0.04)*, with
adjustments
NS (p>0.05), no adjustments
Higher levels (p<0.05)*, males only
with no adjustments
Association (p=0.01)*, with
adjustments
Association (p<0.001)*, children
5th quintile
Association (p<0.001)*, adolescents
5th quintile
OR 1.3 (1.1–1.4)*, 2nd quintile
Association (p=0.002)*, children
5th quintile
Association (p<0.001)*, adolescents
5th quintile
OR 1.2 (1.0–1.5)*, 2nd quintile

Total cholesterol

LDL cholesterol

HDL cholesterol

Triglycerides
Olsen et al. 2003a

2,460 ng/mL (median 4th
PFOS quartile)

Total cholesterol

Occupational (n=518)
HDL cholesterol
Triglycerides

Frisbee et al. 2010
Community (C8) (n=12,476 children and
adolescents)

23.6 ng/mL (mean PFOS in
children)
21.9 ng/mL (mean PFOS in
adolescents)

Total cholesterol

Abnormal cholesterol
LDL cholesterol

Abnormal LDL levels

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

163
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
HDL cholesterol

Triglycerides
Steenland et al. 2009b
Community (C8) (n=46,294)

22.4 ng/mL (mean PFOS)
Total cholesterol
13.3–19.5 ng/mL (2nd quartile) Abnormal cholesterol
LDL cholesterol
HDL cholesterol

Châtaeu-Degat et al. 2010

25.7 ng/mL (mean PFOS)

36.1 ng/mL (mean PFOS)

Total cholesterol

8.04 ng/mL (mean PFOS)

Total cholesterol
High cholesterol levels
Non HDL cholesterol
LDL cholesterol
HDL cholesterol

NS (p=0.35)
OR 1.36 (0.87–2.12), 4th quartile
NS (p=0.14)
NS (p=0.42)
NS (p=0.33)

Triglycerides

Eriksen et al. 2013

Association (p=0.007)*, children
5th quintile
Association (p=0.001)*, adolescents
5th quintile
NS (p=0.1), children 5th quintile
NS (p=0.1), adolescents 5th quintile
Association (p<0.001 for trend)*
OR 1.14 (1.05–1.23)*, 2nd quartile
Association (p<0.05 for trend)*
NS (p>0.05)
Association (p<0.05 for trend)*
NS (p=0.086)
NS (p=0.242)
Association (p<0.001)*, men
Association (p=0.001)*, women
NS (p=0.162), men
Inverse association (p=0.040)*.
women
Association (p=0.02)*

General population (n=723)

Triglycerides
Total cholesterol
LDL cholesterol
HDL cholesterol

Resultc

General population (n=753)
Fisher et al. 2013
General population (n=2,368)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

164
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Fu et al. 2014a

1.47 ng/mL (median PFOS)

Total cholesterol
Elevated cholesterol

NS (p=0.287)
OR 2.27 (0.47–10.92)

LDL cholesterol
Elevated LDL
HDL cholesterol
Elevated HDL
Triglycerides
Elevated triglyceride

NS (p=0.357)
OR 2.27 (0.50–10.37)
NS (p=0.260)
OR 0.29 (0.06–1.50)
NS (p=0.711)
OR 1.26 (0.41–3.90)

Total cholesterol
Elevated cholesterol

NS (p=0.0512 for trend)
OR 1.35 (1.11–1.64, p=0.0183 for
trend)*, log transformed PFOS
Association (p=0.0081 for trend)*
OR 1.48 (1.15–1.90, p=0.0178 for
trend)*, log transformed PFOS
NS (p=0.9703)
NS (p=0.9873 for trend)
NS (p=0.1104 for trend)
NS (p=0.2418 for trend)
NS (β -0.10, -0.73–0.54), 7 year olds
Association (β -0.77, -1.40
to -0.13)*, 15 year olds
NS (β 0.02, -0.48–0.53), 7 year olds
Association (β -0.54, -1.08
to -0.003)*, 15 year olds
NS (β -0.04, -0.33–0.25), 7 year olds
NS (β -0.18, -0.47–0.12), 15 year olds
NS (β -0.004, -0.015–0.006), 7 year
olds
NS (β -0.004, -0.011–0.004), 15 year
olds

General population (n=133)

Geiger et al. 2014b

17.7 ng/mL (mean PFOS)

General population (NHANES) (n=815 12–
18-year-old adolescents)

Maisonet et al. 2015b
General population (n=111 for 7-year-old and
n=88 for 15-year-old girls)

LDL cholesterol
Elevated LDL

23.5–94.5 ng/mL (3rd tertile
maternal PFOS)

HDL cholesterol
Decreased HDL
Triglycerides
Elevated triglycerides
Total cholesterol

LDL cholesterol

HDL cholesterol
Triglycerides

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

165
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Nelson et al. 2010

25.3 ng/mL (mean PFOS)

Total cholesterol
LDL cholesterol

Association (p=0.01)*
NS (p=0.27)

22.3 ng/mL (mean PFOS)

Non-HDL cholesterol
HDL cholesterol
Total cholesterol

Association (p=0.02)*
NS (p=0.78)
Association (p=0.01 for trend)*

13.03 ng/mL (50th PFOS
percentile)

Total cholesterol
LDL cholesterol

Association (p<0.05)*
NS (β 6.48, -0.07–13.03), per ln-unit
increase in PFOS
Association (β 4.39, 2.37–6.42)*, per
ln-unit increase in PFOS
NS (β -0.02, -0.09–0.04), per ln-unit
increase in PFOS
NS (p=0.78), normal weight children
Association (p=0.002)*, obese
children

General population (NHANES) (n=860)

Skuladottir et al. 2015
General population (n=854 pregnant women)
Starling et al. 2014a
General population (n=854 pregnant women)

HDL cholesterol
Triglycerides
Timmermann et al. 2014

41.5 ng/mL (median PFOS)

General population (n=499 children, 8–10 years
old)
Zeng et al. 2015
32.4 and 34.2 ng/mL (mean
PFOS in boys and girls)
General population (n=225 children, 12–
15 years old)
PFHxS
Fisher et al. 2013
General population (n=2,368)

2.18 ng/mL (mean PFHxS)

Triglycerides

Total cholesterol

Association (p<0.001)*

LDL cholesterol
HDL cholesterol
Triglycerides

Association (p<0.001)*
NS (p=0.72)
Association (p=0.05)*

Total cholesterol
High cholesterol levels

Association (p=0.005)*
OR 1.27 (1.11–1.45)*, 4th quartile

Non HDL cholesterol
LDL cholesterol
HDL cholesterol

Association (p=0.002)*
Association (p=0.02)*
NS (p=0.67)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

166
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Nelson et al. 2010

2.6 ng/mL (mean PFHxS)

Total cholesterol
LDL cholesterol

NS (p=0.07)
NS (p=0.10)

Non-HDL cholesterol
HDL cholesterol
Total cholesterol

General population (NHANES) (n=860)

2.1 and 2.1 ng/mL (mean
PFHxS in boys and girls)

Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides

Association (p=0.04)*
NS (p=0.11)
NS (β 3.00, -1.75–7.76), per ln-unit
increase in PFHxS
NS (β 1.92, -2.50–6.33), per ln-unit
increase in PFHxS
Association (β 1.46; 0.19–2.73)*, per
ln-unit increase in PFHxS
NS (β -0.01, -0.05–0.03), per ln-unit
increase in PFHxS
NS (p=0.23)
NS (p=0.17)
NS (p=0.54)
NS (p=0.15)

Mundt et al. 2007

NR

Occupational (n=592)
Fu et al. 2014a

Total cholesterol
Triglycerides

NS, longitudinal analysis
NS, longitudinal analysis

0.37 ng/mL (median PFNA)

Total cholesterol

Association (p=0.002)*

Elevated cholesterol
LDL cholesterol
Elevated LDL
HDL cholesterol
Lowered HDL
Triglycerides

OR 1.03 (0.24–4.46)
Association (p=0.004)
OR 2.51 (0.59–10.74)
NS (p=0.191)
OR 1.06 (0.20–5.57)
NS (p=0.460)

Elevated triglycerides

OR 0.80 (0.26–2.49)

Starling et al. 2014a

0.60 ng/mL (50th PFHxS
percentile)

General population (n=854 pregnant women)

LDL cholesterol
HDL cholesterol
Triglycerides

Zeng et al. 2015
General population (n=225 children, 12–
15 years old)
PFNA

General population (n=133)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

167
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Nelson et al. 2010

1.3 ng/mL (mean PFNA)

Total cholesterol
LDL cholesterol

Association (p=0.04)*
NS (p=0.08)

Non-HDL cholesterol
HDL cholesterol
Total cholesterol

Association (p=0.04)*
NS (p=0.31)
NS (β 0.01, -5.98–6.00), per ln-unit
increase in PFNA
NS (β -2.15, -7.31–3.02), per ln-unit
increase in PFNA
Association (β 2.84; 0.97–4.71)*, per
ln-unit increase in PFNA
NS (β -0.02, -0.07–0.03), per ln-unit
increase in PFNA
Association (p=0.04)*
Association (p=0.05)*
NS (p=0.37)
Association (p=0.007)*

General population (NHANES) (n=860)

Starling et al. 2014a

0.39 ng/mL (50th PFNA
percentile)

General population (n=854 pregnant women)

LDL cholesterol
HDL cholesterol
Triglycerides

Zeng et al. 2015
General population (n=225 children, 12–
15 years old)

0.8 and 0.9 ng/mL (mean
PFNA in boys and girls)

Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides

PFDeA
Fu et al. 2014a
General population (n=133)

0.19 ng/mL (median PFDeA) Total cholesterol
Elevated cholesterol
LDL cholesterol
Elevated LDL
HDL cholesterol
Elevated HDL
Triglycerides
Elevated triglycerides

***DRAFT FOR PUBLIC COMMENT***

Association (p=0.048)*
OR 3.84 (0.87–16.95)
NS (p=0.251)
OR 2.17 (0.52–9.04)
Association (p=0.007)*
OR 2.21 (0.49–10.07)
NS (p=0.317)
OR 0.51 (0.17–1.58)

 PERFLUOROALKYLS

168
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Starling et al. 2014a

0.09 ng/mL (50th PFDeA
percentile)

Total cholesterol
LDL cholesterol

NS (β 1.84, -2.12–5.79), per ln-unit
increase in PFDeA
NS (β 0.19, -3.30–3.69), per ln-unit
increase in PFDeA
Association (β 2.54, 1.22–3.87)*, per
ln-unit increase in PFDeA
NS (β -0.03, -0.07–0.01), per ln-unit
increase in PFDeA
NS (p=0.74)
NS (p=0.85)

HDL cholesterol
Triglycerides

NS (p=0.47)
NS (p=0.92)

Total cholesterol
Elevated cholesterol
LDL cholesterol
Elevated LDL

NS (p=0.184)
OR 3.70 (0.76–18.03)
NS (p=0.270)
OR 4.16 (0.96–18.00)

HDL cholesterol
Elevated HDL
Triglycerides
Elevated triglycerides
Total cholesterol

NS (p=0.279)
OR 0.54 (0.11–2.57)
NS (p=0.755)
OR 0.74 (0.25–2.21)
NS (β 0.89, -3.28–5.06), per ln-unit
increase in PFUA
NS (β -2.36, -5.97–1.25), per ln-unit
increase in PFUA
Association (β 4.05, 2.75–5.35)*, per
ln-unit increase in PFUA
NS (β -0.04, -0.08–0.00), per ln-unit
increase in PFUA

General population (n=854 pregnant women)

Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides

Zeng et al. 2015

1.0 and 1.0 ng/mL (mean
PFDeA in boys and girls)

General population (n=225 children, 12–
15 years old)
PFUA
Fu et al. 2014a

0.26 ng/mL (median PFUA)

General population (n=133)

Starling et al. 2014a
General population (n=854 pregnant women)

0.22 ng/mL (50th PFUA
percentile)

LDL cholesterol
HDL cholesterol
Triglycerides

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

169
2. HEALTH EFFECTS

Table 2-12. Summary of Serum Lipid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

PFHpA
Fu et al. 2014a

0.04 ng/mL (median PFHpA) Total cholesterol

NS (p>0.05)

LDL cholesterol
HDL cholesterol
Triglycerides

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Total cholesterol
LDL cholesterol

Association (p=0.04)*
NS (p=0.14)

HDL cholesterol
Triglycerides

NS (p=0.15)
NS (p=0.81)

Total cholesterol
LDL cholesterol
HDL cholesterol

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Triglycerides

NS (p>0.05)

Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides

NS (p=0.37)
NS (p=0.44)
NS (p=0.68)
NS (p=0.40)

General population (n=133)

PFBuS
Zeng et al. 2015

0.5 and 0.4 ng/mL (mean
PFBuS in boys and girls)

General population (n=225 children, 12–
15 years old)
PFBA
Fu et al. 2014a

0.11 ng/mL (median PFBA)

General population (n=133)

PFDoA
Zeng et al. 2015
General population (n=225 children, 12–
15 years old)

4.5 and 4.4 ng/mL (mean
PFDoA in boys and girls)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 7 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

HDL = high density lipoprotein; LDL = low density lipoprotein; NHANES = National Health and Nutrition Examination Survey; NR = not reported; NS = not
significant; OR = odds ratio; PFBA = perfluorobutyric acid; PFBuS = perfluorobutane sulfonic acid; PFDeA = perfluorodecanoic acid;
PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFUA = perfluoroundecanoic acid; VLDL = very low-density lipoprotein

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

170
2. HEALTH EFFECTS

increases in liver weight without histological evidence, such as (1) degenerative or necrotic changes
including hepatocyte necrosis, inflammation, and steatotic vascular degeneration; (2) biliary/oval cell
proliferation, degeneration, fibrosis, and cholestasis; or (3) necrosis and degeneration of other resident
cells within the liver, are not considered adverse or relevant for human risk assessment. In the absence of
histological changes, increases in liver organ weight are not considered relevant for human risk
assessment unless at least two of the following three parameters are present: (1) at least 2–3 times
increase in ALT levels; (2) biologically significant change in other biomarkers of hepatobiliary damage
(alkaline phosphatase, AST, GGT, etc.); or (3) biologically significant change in another clinical
pathology marker indicating liver dysfunction (albumin, bilirubin, bile acids, coagulation factors,
cholesterol, triglycerides, etc.). ATSDR has adapted the criteria from Hall et al. (2012) for determining
the adversity of the liver effects reported in the rodent perfluoroalkyl studies. Doses associated with
increases in liver weight and hepatocellular hypertrophy were not considered adverse effect levels unless
hepatocellular degenerative or necrotic changes or evidence of biliary or other liver cell damage were also
present. The lowest doses associated with the liver weight increases and hepatocellular hypertrophy are
noted in the LSE tables even though the dose levels are considered NOAELs.
PFOA

Epidemiology Studies—Liver Disease. Three studies of highly exposed populations have examined
possible associations between PFOA and increased risk of liver disease. In workers, no association
between estimated cumulative serum PFOA levels and the risk of non-hepatitis liver disease was observed
(Steenland et al. 2015). Similarly, two studies of residents living near the Washington Works PFOA
facility reported no increases in liver disease. In a study by Anderson-Mahoney et al. (2008), no
significant increases in self-reported liver problems were found in residents primarily served by the
Lubeck Public Water Service District or Little Hocking Water District; the study did not measure serum
PFOA levels. In a C8 Health Project study that included workers at the Washington Works facility,
estimated cumulative serum PFOA levels were not associated with any liver disease or enlarged liver,
fatty liver, or cirrhosis (Darrow et al. 2016).

Epidemiology Studies—Hepatic Serum Enzymes and Bilirubin Levels. The possible association
between PFOA exposure and hepatic enzymes has been examined in seven occupational exposure studies
that have found inconsistent results. A small study of Italian perfluoroalkyl workers did not find
associations between serum PFOA and ALT, AST, or GGT activities when only current workers were
examined (Costa et al. 2009). In analysis of all workers (current, former, and non-exposed workers),

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

171
2. HEALTH EFFECTS

associations between serum PFOA levels and ALT and GGT activities were found; total bilirubin was
also inversely associated with serum PFOA. Another small study of workers at a fluorochemical facility
in China found an association between serum PFOA and AST activity, but not ALT activity (Wang et al.
2012). Gilliland and Mandel (1996) did not find associations between serum fluorine levels (used as a
surrogate for serum PFOA) and ALT, AST, or GGT levels in workers. In a follow-up study of this
facility, there were no differences between AST, ALT, GGT, or total bilirubin levels between workers in
three exposure groups (Olsen et al. 2000); the mean serum PFOA levels in this study ranged from
5,000 to 6,400 ng/mL at three time points and the serum PFOA levels in the lowest exposure group
ranged from 0 to <1,000 ng/mL. Increases in GGT and decreases in total bilirubin levels associated with
increases in serum PFOA were observed in a study of workers exposed to high levels of PFOA and PFOS
(Olsen and Zobel 2007); ALT activity was not affected. In a cross-sectional study of active workers at a
PFOA facility, a modest but statistically significant positive association between serum PFOA and GGT
activity was found (Sakr et al. 2007b). No associations were found for bilirubin levels or ALT and AST
activities.

The possible associations between serum PFOA and serum enzyme and bilirubin levels were examined in
two longitudinal occupational exposure studies. Sakr et al. (2007a) examined the relationship between
serum PFOA and liver enzymes in a longitudinal study of 454 workers who had two or more
measurements of serum PFOA from 1979 until the study was conducted. The average length of
employment among workers with multiple PFOA measurements was 11 years, and, on average,
10.8 years elapsed between their first and last serum PFOA measurement. The means of the first and last
PFOA measurement were 1,040 and 1,160 ng/mL, respectively. After adjustment for potential
confounders, serum PFOA was associated with AST activity, but not ALT, GGT, or total bilirubin. The
second study included 179 workers involved in the demolition of 3M perfluoroalkyl manufacturing
facilities examined over a mean period of 164 days (Olsen et al. 2012). In workers with prior exposure to
PFOA who had a decrease in serum PFOA levels during the study period, there was a significant increase
in ALT levels. An increase in serum PFOA levels did not significantly alter AST or total bilirubin levels.

Community and general population exposure studies have also examined possible associations between
serum PFOA levels and alterations in serum hepatic enzyme and bilirubin levels. As with the
occupational exposure studies, several studies of populations living near PFOA facilities have found
inconsistent results. Darrow et al. (2016) found associations between ALT and bilirubin (inverse
association) and cumulative and 2005/2006 serum PFOA levels in participants of the C8 Health Project
(6.5% of the participants also worked at the facility); there were no associations with GGT activity. Gallo

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

172
2. HEALTH EFFECTS

et al. (2012) also reported a significant correlation between serum PFOA levels and ALT activity in C8
Health Project participants. Unlike the Darrow et al. (2016) study, a significant correlation between
serum PFOA levels and GGT activity, but no correlation with direct bilirubin levels, was found. An
earlier study of residents in the same area, as well as a study of residents near a facility in China, did not
find associations between serum PFOA and ALT, AST, or GGT (Emmett et al. 2006b; Wang et al. 2012).

More consistent results were found in three general population studies. In studies utilizing data from
NHANES, Gleason et al. (2015) and Lin et al. (2010) reported associations between serum PFOA levels
and ALT, AST, and GGT activities; total bilirubin was also found to be associated with serum PFOA in
the Gleason et al. (2015) study, but not in the Lin et al. (2010) study. A general population study
conducted in Japan (Yamaguchi et al. 2013) also found associations between serum PFOA levels and
AST, ALT, and GGT activities.

Although a number of epidemiology studies have found associations between serum PFOA and serum
hepatic enzyme and bilirubin levels, many of the investigators noted that liver biomarker levels were
typically within the normal range. Four studies examining the risk of having biomarker levels outside of
the normal range provide useful information for evaluating the health impact of the enzyme level
alterations. For ALT, Gallo et al. (2012) and Gleason et al. (2015) found increased risks of abnormal
levels in C8 and NHANES participants, respectively. In contrast, Olsen and Zobel (2007) and Emmett et
al. (2006b) did not find increased risks of abnormal ALT levels in workers and C8 participants,
respectively. No alterations in the risk of abnormal AST levels associated with elevated serum PFOA
levels were observed in NHANES participants (Gleason et al. 2015). Emmett et al. (2006b) found a
decrease in the risk of abnormal AST levels with increasing serum PFOA levels in community members.
Associations between the risk of elevated GGT and serum PFOA were found in the study conducted by
Gleason et al. (2015), but not in the Olsen and Zobel (2007), Gallo et al. (2012), or Emmett et al. (2006b)
studies. Similarly, Gleason et al. (2015) reported an association between serum PFOA and the risk of
elevated bilirubin levels, whereas Gallo et al. (2012) did not find this association in the higher exposed
population.

Epidemiology Studies—Serum Lipids. Occupational, community, and general population studies have
examined the possible associations between serum PFOA levels and serum lipid levels; the results of
these studies are presented in Table 2-12. Summaries of the changes in serum total cholesterol and LDL
cholesterol levels, as well as the risk associated with elevated serum cholesterol and LDL cholesterol
levels, are presented in Figures 2-9, 2-10, 2-11, and 2-12.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

173
2. HEALTH EFFECTS

Figure 2-9. Serum Total Cholesterol Levels Relative to Serum PFOA Levels
(Presented as percent change in cholesterol levels)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

174
2. HEALTH EFFECTS

Figure 2-10. Risk of Abnormal Cholesterol Levels Relative to PFOA Levels (Presented as Adjusted Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

175
2. HEALTH EFFECTS

Figure 2-11. Serum LDL Cholesterol Levels Relative to Serum PFOA Levels
(Presented as percent change in LDL cholesterol levels)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

176
2. HEALTH EFFECTS

Figure 2-12. Risk of Abnormal LDL Cholesterol Levels Relative to PFOA Levels (Presented as Adjusted Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

177
2. HEALTH EFFECTS

A study of workers at a manufacturing facility in Italy found higher total cholesterol and non- highdensity lipoprotein (HDL)-cholesterol levels (non-HDL cholesterol was estimated by subtracting HDL
cholesterol from total cholesterol) in the PFOA-exposed workers, as compared to levels in workers who
were not exposed to PFOA (Costa 2004). A second study at this facility (Costa et al. 2009) also found an
association between serum PFOA levels and total cholesterol levels, but no association with HDL
cholesterol levels. No associations were found for HDL cholesterol or triglyceride levels. In another
small study of workers at a fluorochemical facility in China (Wang et al. 2012), no associations between
serum PFOA and total cholesterol, LDL cholesterol, or triglyceride levels were observed; the study did
find an inverse association between serum PFOA and HDL cholesterol levels.

Several studies have examined workers at 3M facilities in Cottage Grove, Minnesota, Decatur, Alabama,
and/or Antwerp, Belgium; workers at these facilities were also exposed to high levels of PFOS. Gilliland
and Mandel (1996) examined workers at the Cottage Grove facility in 1990 and found no associations
between serum fluorine levels (used as a surrogate for PFOA) and total cholesterol, LDL cholesterol, or
HDL cholesterol. In a follow-up to this study, Olsen et al. (2000) examined workers in 1993, 1995, and
1997; only 17 workers were examined at all three time periods, 21 workers were examined in 1995 and
1997, and 68 workers were examined in 1993 and 1995. The study did not adjust for the use of
cholesterol-lowering medication. When workers were categorized by blood PFOA levels (0–<1,000,
1,000–<10,000, and >10,000 ng/mL), no significant differences in serum cholesterol, LDL cholesterol,
HDL cholesterol, or triglyceride levels were found at any of the monitoring periods. A study in workers
at the three 3M facilities, most of whom were not taking cholesterol-lowering medications, did not find
associations between serum PFOA levels and total cholesterol or LDL cholesterol levels; however, serum
PFOA levels were associated with elevated triglyceride levels and inversely associated with HDL
cholesterol levels (Olsen and Zobel 2007). The study did not find increases in the risk of elevated total
cholesterol (≥200 mg/dL), elevated LDL cholesterol (≥130 mg/dL), elevated triglyceride (≥150 mg/dL),
or decreased HDL cholesterol (≤40 mg/dL) levels in workers with serum PFOA levels in the highest
deciles. In addition to these cross-sectional studies, two longitudinal studies were conducted at these
facilities. Using data for 174 workers with medical surveillance data in 2000 and 1997 and/or 1995,
Olsen et al. (2003a) found that serum PFOA was a significant predictor of cholesterol and triglyceride
levels, which was primarily due to 21 workers at the Antwerp facility (mean serum level 8,400 ng/mL)
whose serum PFOA levels increased over time. In a longitudinal study, Olsen et al. (2012) examined
workers (none of the subjects reported using cholesterol-lowering medication) involved in the demolition
of 3M perfluoroalkyl manufacturing facilities; serum PFOA and lipid levels were measured prior to the
demolition and after demolition (mean time interval of 164 days). The mean baseline serum PFOA levels

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

178
2. HEALTH EFFECTS

were 881 ng/mL in 14 3M workers with prior PFOA or PFOS exposure and 28.9 ng/mL in the remaining
165 workers. Among the 119 workers whose serum PFOA/PFOS levels (mean increase 50.9 ng/mL)
increased during the observation period, there was a significant increase in HDL cholesterol levels, but no
change in total cholesterol or non-HDL cholesterol levels. No significant alterations in serum lipid levels
were observed in the 55 workers whose serum PFOA/PFOS levels decreased during the observation
period. In workers whose baseline levels of PFOA and PFOS were <15 and <50 ng/mL, respectively,
there were no significant differences between pre- and post-exposure serum lipid levels.

Investigators have also examined workers at the DuPont Washington Works facility in West Virginia. In
a cross-sectional study, Sakr et al. (2007b) found associations between serum PFOA levels and total
cholesterol, LDL cholesterol, and very-low-density lipoprotein (VLDL) cholesterol levels in all subjects
and in a subset of subjects not taking cholesterol-lowering medication. The study did not find any
association between serum PFOA and HDL cholesterol or triglyceride levels. In a second study,
Steenland et al. (2015) did not find an association between estimated serum PFOA levels and the
occurrence of elevated cholesterol levels that required medication. In a longitudinal study of workers
who had at least two serum PFOA measurements between 1979 and 2004, Sakr et al. (2007a) found a
positive association between serum PFOA and total cholesterol levels; no associations with triglycerides,
LDL cholesterol, or HDL cholesterol were found. Total cholesterol levels increased 1.06 mg/dL for each
1,000 ng/mL increase in serum PFOA.

Several studies have been conducted of residents living near the Washington Works facility. A study by
Emmett et al. (2006b) of adults and children living in a community serviced by the Little Hocking Water
Authority did not find an association between serum PFOA levels and total cholesterol levels; the study
included an adjustment for the use of cholesterol-lowering medication. Four larger-scale studies of
participants in the C8 Science Panel studies found associations between serum PFOA levels and serum
lipid levels (Fitz-Simon et al. 2013; Frisbee et al. 2010; Steenland et al. 2009b; Winquist and Steenland
2014a). Positive associations between serum PFOA levels and total cholesterol and LDL cholesterol
were found in a study of over 12,000 children and adolescents, with mean serum PFOA levels of
32.6 ng/mL in children (aged 1.0–11.9 years) and 26.3 ng/mL in adolescents (aged 12.0–17.9 years)
(Frisbee et al. 2010). Serum PFOA was also positively associated with triglyceride levels. Additionally,
there was an increased risk of elevated cholesterol (≥170 mg/dL) in subjects with serum PFOA levels in
the 4th or 5th quintiles. Increased odds of high LDL cholesterol (≥110 mg/dL) were also observed for the
5th PFOA quintile (OR 1.4, 95% CI 1.2–1.7). The investigators noted that the dose-response relationship
between serum PFOA and serum lipids was nonlinear, with greater increases in lipids observed at the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

179
2. HEALTH EFFECTS

lower serum PFOA levels. Similar findings were reported in a study of >46,000 adults with a median
serum PFOA level of 26.6 ng/mL; the study excluded subjects who reported taking cholesterol-lowering
medication (Steenland et al. 2009b). Associations were found between serum PFOA levels and total
cholesterol, LDL cholesterol, and non-HDL cholesterol; a positive association between serum PFOA and
triglycerides was also found. No associations between serum PFOA levels and HDL cholesterol levels
were found. Increased risks of having high cholesterol (≥240 mg/dL) were found in subjects with serum
PFOA levels in the 2nd, 3rd, and 4th quartiles. The investigators noted that the odds of high cholesterol
from the 1st to the 5th quartile were approximately 40% for PFOA, which may be important given that the
Framingham study found that the risk of coronary heart disease was about 1.8 times higher in subjects
with total cholesterol levels >240 mg/dL as compared to subjects with levels <200 mg/dL. Steenland et
al. (2009b) also found an association between serum PFOA levels and total cholesterol levels in a study
of 10,746 adults taking cholesterol-lowering medication. Using both groups of subjects (taking or not
taking cholesterol-lowering medication), the investigators analyzed whether taking cholesterol-lowering
medication was associated with lower serum PFOA levels, which may be indicative of reverse causality.
Although serum PFOA levels were significantly lower in subjects taking cholesterol-lowering
medication, the difference between the groups was low (4%). Using estimated cumulative serum PFOA
levels as the exposure metric, Winquist and Steenland (2014a) found increased risks of
hypercholesterolemia at cumulative exposure levels ≥142 ng/mL. In a longitudinal study by Fitz-Simon
et al. (2013), adults participating in the C8 Health Project and not taking cholesterol-lowering medication
were examined twice, with an average of 4.4 years between examinations. Mean serum PFOA levels
were 74.8 ng/mL at the first examination and 30.8 ng/mL at the second examination. In subjects whose
serum PFOA levels halved between examinations, there was a 3.6% decrease in LDL cholesterol levels
and 1.7% decrease in total cholesterol levels. However, there were very small changes in LDL
cholesterol and total cholesterol levels in subjects whose serum PFOA levels decreased by >64% and
there were slight increases in LDL cholesterol and total cholesterol levels in subjects whose serum PFOA
levels fell by <50%. Changes in PFOA levels were not associated with changes in HDL cholesterol or
triglyceride levels. Similarly, Wang et al. (2012) found no associations between serum PFOA levels and
total cholesterol, HDL cholesterol, LDL cholesterol, or triglycerides in a study of adults living near a
PFOA manufacturing facility in China; the mean serum PFOA level was 378.30 ng/mL and did not
include an adjustment for the use of cholesterol-lowering medication.

General population studies were conducted in the United States, Canada, Denmark, Norway, Japan,
China, and Taiwan; these studies have examined possible associations between serum PFOA levels and
serum lipid levels in children, adolescents, pregnant women, and adults. In a study of 8–10-year-old

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

180
2. HEALTH EFFECTS

children (median serum PFOA of 9.3 ng/mL), Timmermann et al. (2014) found an association between
serum PFOA and triglyceride levels among obese children; this association was not found among normal
weight children. In a study of adolescents (12–18 years of age) participating in NHANES (mean serum
PFOA level of 4.2 ng/mL), Geiger et al. (2014b) found associations between serum PFOA and total
cholesterol and LDL cholesterol levels; no associations were found for HDL cholesterol or triglycerides.
The study also found increased risks of elevated total cholesterol levels (>170 mg/dL) associated with
serum PFOA levels. No alterations in the risk of elevated LDL cholesterol or triglycerides or decreased
HDL cholesterol were found. Associations between serum total cholesterol, LDL cholesterol, and
triglycerides have also been observed in a study of Taiwanese adolescents (12–15 years of age, median
PFOA level of 9.3 ng/mL) (Zeng et al. 2015); no association was found for HDL cholesterol. A fourth
study found associations between maternal PFOA levels and total cholesterol and LDL cholesterol in
7- and 15-year-old girls, but no associations for girls whose maternal PFOA levels were in the 2nd or
3rd tertiles (Maisonet et al. 2015b). No associations were found for HDL cholesterol or triglyceride
levels.

Studies in adults have found mixed results for serum lipids. Using NHANES data for adults not taking
cholesterol-lowering medication (mean serum PFOA level of 4.6 ng/mL), Nelson et al. (2010) found an
association between serum PFOA levels and non-HDL cholesterol levels; no associations were found for
total cholesterol, LDL cholesterol, or HDL cholesterol. Associations between serum PFOA levels and
total cholesterol levels were also found in a study of Danish adults not taking cholesterol-lowering
medication (mean serum PFOA level of 7.1 ng/mL) (Eriksen et al. 2013). A study in Chinese adults
(median PFOA level of 1.43 ng/mL) also found associations between serum PFOA and total cholesterol
and LDL cholesterol, with no associations for HDL cholesterol or triglycerides (Fu et al. 2014a). This
study did not find increased risks of elevated total cholesterol, LDL cholesterol, or triglycerides or
decreased HDL cholesterol associated with serum PFOA. A study of pregnant women in Denmark also
found an association between serum PFOA (mean serum PFOA level of 4.1 ng/mL at gestation week 30)
and total cholesterol levels (Skuladottir et al. 2015). No associations between serum PFOA levels and
total cholesterol, LDL cholesterol, or non-HDL cholesterol levels were found in Canadian adults not
taking cholesterol-lowering medication with a geometric mean serum PFOA level of 2.46 ng/mL (Fisher
et al. 2013). In a second study of pregnant women (median PFOA level of 2.25 ng/mL at gestation week
18), no associations between plasma PFOA and total cholesterol, LDL cholesterol, or triglycerides were
found (Starling et al. 2014a). The study did find an association between plasma PFOA and HDL
cholesterol.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

181
2. HEALTH EFFECTS

A number of epidemiology studies have reported associations between serum PFOA levels and serum
lipid levels; the most consistently found alteration was for increased serum total cholesterol levels.
Associations between serum PFOA and serum cholesterol levels have been observed in occupational
(Costa 2004; Costa et al. 2009; Sakr et al. 2007a, 2007b), community (Fitz-Simon et al. 2013; Frisbee et
al. 2010; Steenland et al. 2009b; Winquist and Steenland 2014a), and general population (Eriksen et al.
2013; Fu et al. 2014a; Geiger et al. 2014b; Skuladottir et al. 2015; Zeng et al. 2015) studies, whereas
other investigators have not found associations in worker populations (Gilliland and Mandel 1996; Olsen
et al. 2000; Olsen and Zobel 2007; Steenland et al. 2015; Wang et al. 2012), community populations
(Emmett et al. 2006b; Wang et al. 2012), or general populations (Fisher et al. 2013; Nelson et al. 2010;
Starling et al. 2014a). Longitudinal studies conducted in workers and highly exposed residents strengthen
the interpretation of this association between serum PFOA and serum lipid levels. Serum PFOA levels
were found to be a significant predictor of serum cholesterol levels in workers examined at least twice in
a ≥5-year period (Olsen et al. 2003a; Sakr et al. 2007a). Similarly, a study of highly-exposed residents
examined twice with approximately 4 years between examinations found that there was a 1.7% decrease
in serum total cholesterol levels in subjects whose serum PFOA levels decreased by 50% between
examinations (Fitz-Simon et al. 2013). As noted in Steenland et al. (2010a), there is considerable
variation in the strength of the association between PFOA and serum cholesterol, with the greatest
changes in serum cholesterol occurring at lower PFOA levels. The change in cholesterol levels per
ng/mL change in serum PFOA ranged from 0.0007, calculated from data from the Olsen et al. (2000)
occupational exposure study, to 2.0 calculated from data from the Nelson et al. (2010) general population
study; the mean serum PFOA levels in these studies were ~22,000 and 4 ng/mL respectively. In a clinical
trial (results only available as an abstract), administration of APFO to patients with advanced refractory
solid tumors at doses of 50–1,200 mg weekly for 6 weeks resulted in decreases in serum LDL cholesterol
levels (it is unclear if this effect was observed at all dose levels) (MacPherson et al. 2011). These results
are similar to those observed in laboratory animals, suggesting that the dose-response curve may be
biphasic. Steenland et al. (2010a) and Frisbee et al. (2010) suggested that this may be due to a steep doseresponse curve at low PFOA levels, which flattens out at higher PFOA levels and may be indicative of
saturation. A similar pattern was also observed in the risks of elevated cholesterol per increases in serum
PFOA levels (Figure 2-12). Several investigators have explored whether PFOA and cholesterol could be
jointly affected or whether the associations were due to reverse causality (i.e., increased cholesterol
resulted in increased serum PFOA levels). Butenhoff et al. (2012c) explored the issues of whether PFOA
distributes into serum lipoprotein fractions, and whether increases in serum lipoproteins would result in
increases in serum PFOA. They concluded that there was limited distribution to plasma lipoproteins, and
that this was not a non-causal factor. The Steenland et al. (2009b) study found slightly lower serum

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

182
2. HEALTH EFFECTS

PFOA levels (4%) among individuals taking cholesterol medication, as compared to those not taking
medication and noted that this was primarily a function of the large sample size. This finding does not
support reverse causality.

Laboratory Animal Studies. Information from inhalation studies in animals is limited. Head-only
exposure of male rats to 810 mg/m3 APFO dusts for 4 hours caused liver enlargement, but
microscopically, the liver tissue appeared normal (Kennedy et al. 1986). Exposure head-only of male rats
to 0, 1, 7.6, or 84 mg/m3 APFO dusts 6 hours/day, 5 days/week for 2 weeks resulted in significant
increases in absolute and relative liver weight at 7.6 and 84 mg/m3 on exposure day 10; in rats from the
84 mg/m3 group, absolute and relative liver weights were still significantly increased 28 days after
exposure ceased (Kennedy et al. 1986). The activities of serum enzymes markers of liver function were
unremarkable except for alkaline phosphatase, which was significantly increased in the 7.6 and 84 mg/m3
groups immediately after exposure on day 10 and remained elevated in the 84 mg/m3 group on day 14 of
recovery. Histopathological changes were restricted to the 7.6 and 84 mg/m3 groups and consisted of
panlobular and centrilobular hepatocellular hypertrophy and necrosis. Panlobular hepatocellular
hypertrophy was seen only after the 10th exposure, but was limited to the centrilobular hepatocytes 14 or
28 days after exposure terminated, and was absent 42 days following cessation of exposure. Inhalation
exposure of pregnant rats to 25 mg/m3 APFO dusts 6 hours/day during GDs 6–15 induced an 18%
increase in absolute liver weight (Staples et al. 1984); no significant effect was reported in rats exposed to
≤10 mg/m3.
Nose-only exposure of male CD rats to 67 mg/m3 ammonium perfluorononanoate dusts for 4 hours
induced significant increases (28–37%) in absolute and relative liver weight, assessed 5 and 12 days after
exposure (Kinney et al. 1989). Histopathological examinations were not conducted in this study.

The liver is the main target organ for perfluoroalkyl compounds in animals following short- or long-term
oral exposures. The hepatic response to exposure to many perfluoroalkyl compounds, particularly in
rodents, is initiated by the activation of the nuclear hormone receptor, PPARα, which triggers a
characteristic sequence of morphological and biochemical events characterized by liver hypertrophy and
alteration of a wide range of enzymes, particularly those involved in lipid metabolism. It appears that
PFOA can also damage the liver via a method independent of PPARα. The most sensitive liver effect
observed in rats and mice after acute oral exposure to PFOA is an increase in liver weight (Cook et al.
1992; Das et al. 2017; Eldasher et al. 2013; Haughom and Spydevold 1992; Ikeda et al. 1985; Iwai and
Yamashita 2006; Kawashima et al. 1995; Kennedy 1987; Liu et al. 1996; Loveless et al. 2006; Pastoor et

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

183
2. HEALTH EFFECTS

al. 1987; Permadi et al. 1992, 1993; Qazi et al. 2012; White et al. 2009; Wolf et al. 2007, 2008a; Xie et al.
2003; Yahia et al. 2010; Yang et al. 2001, 2002b). In rats orally administered 50 mg/kg/day PFOA for 1,
3, or 7 days, a 10% increase in liver weight was observed after the first dose; however, the relative liver
weight was not significantly different from controls (Pastoor et al. 1987). After 3 days of exposure, the
relative liver weight was significantly higher (36%) than controls. Similarly, in mice, exposure to
390 mg/kg/day PFOA in the diet resulted in a significant increase in liver weight after 5 days of exposure,
but not after 2 days of exposure (Permadi et al. 1992). The lowest LOAELs for increased relative liver
weight in rats were 4.7 mg/kg/day in a 7-day study (Kawashima et al. 1995) and 2 mg/kg/day in a 14-day
study (Liu et al. 1996); these studies also identified NOAELs of 2.4 and 0.2 mg/kg/day, respectively. In
mice, the lowest LOAEL for increases in liver weight was 1 mg/kg/day PFOA administered in the diet for
10 days (Yang et al. 2001) or administered via gavage for 7 days (Eldasher et al. 2013; Wolf et al. 2008a).
Pastoor et al. (1987) noted that oral administration of 50 mg/kg/day PFOA to rats for 7 days resulted in a
2-fold increase in absolute and relative liver weight, but no significant change in total deoxyribonucleic
acid (DNA), indicating that the hepatomegaly represented hypertrophy rather than hyperplasia. Few
acute-duration studies included histological examinations of the liver. Centrilobular and midzonal
hypertrophy was observed in mice administered 1 or 3 mg/kg/day PFOA via gavage for 7 days;
panlobular hypertrophy with cytoplasmic vacuolation was observed at 10 mg/kg/day (Wolf et al. 2008a).
Qazi et al. (2010a) reported hepatocellular hypertrophy in mice exposed to 3.5 mg/kg/day PFOA in the
diet for 10 days. Elcombe et al. (2010) reported hepatocellular hypertrophy in rats orally exposed to
18 mg/kg/day for 7 days, but not after 1 day of exposure. Increases in steatosis and triglyceride levels
were observed in the livers of mice administered 10 mg/kg/day for 7 days (Das et al. 2017). A related
liver effect was the finding of reduced serum cholesterol and triacylglycerol levels in rats administered
16 mg/kg/day PFOA in the diet for 7 days (Haughom and Spydevold 1992) and decreases in serum
cholesterol and triglyceride levels in rats administered 18 mg/kg/day PFOA via gavage for 7 days
(Elcombe et al. 2010).

Similar to the acute-duration studies, intermediate-duration oral exposure to PFOA resulted in increases in
absolute and relative liver weights in rats (Biegel et al. 2001; Butenhoff et al. 2004b; Griffith and Long
1980; Perkins et al. 2004) and mice (Abbott et al. 2007; Ahmed and Abd Ellah 2012; Albrecht et al. 2013;
Griffith and Long 1980; Kennedy 1987; Lau et al. 2006; Son et al. 2008; Wolf et al. 2007). The lowest
dose resulting in increases in liver weight in rats was 0.96 mg/kg/day, observed following gavage
administration of APFO for 28 days (Loveless et al. 2008); the lowest dose in mice was 0.5 mg/kg/day,
observed in two 28-day studies using APFO (Kennedy 1987; Son et al. 2008). No significant alterations
in liver weight were observed in rats administered 0.29 mg/kg/day for 28 days (Loveless et al. 2008) or in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

184
2. HEALTH EFFECTS

mice exposed to 0.2 mg/kg/day for 21 days (Kennedy 1987). Hepatocellular hypertrophy was the
predominant histopathological alteration in rats (Cui et al. 2009; Griffith and Long 1980; Loveless et al.
2008; Perkins et al. 2004) and mice (Albrecht et al. 2013; Filgo et al. 2015a; Griffith and Long 1980;
Loveless et al. 2008; Tan et al. 2013); the severity of the hypertrophy was dose-related (Filgo et al. 2015a;
Loveless et al. 2008). At higher doses, focal necrosis was observed (29 mg/kg/day in rats and
0.96 mg/kg/day in mice exposed for 28 days) (Loveless et al. 2008). Fatty changes were observed in rats
administered 20 mg/kg/day for 28 days (Cui et al. 2009) and mice administered 9.6 mg/kg/day (Loveless
et al. 2008). No significant alterations in liver weight or histopathology were observed in rats allowed to
recover for 8 weeks following a 13-week exposure to 0.6–6.5 mg/kg/day (Perkins et al. 2004).
Intermediate-duration exposure to PFOA also resulted in decreases in serum HDL cholesterol levels in
rats and mice administered ≥0.29 or 0.96 mg/kg/day, respectively, for 28 days (Loveless et al. 2008).
Serum cholesterol levels were decreased in rats administered 0.29 or 0.96 mg/kg/day (no changes were
observed at higher doses) and in mice administered 9.6 or 29 mg/kg/day (Loveless et al. 2008). Similarly,
serum triglyceride levels were decreased in rats administered 0.29–9.6 mg/kg/day and in mice
administered 9.6 or 29 mg/kg/day (Loveless et al. 2008). In a study of mice fed a western-type diet,
increases in plasma cholesterol levels were observed after 6 weeks of dietary exposure to 0.55 mg/kg/day
in BALB/c or C57BL/6 mice (Rebholz et al. 2016). The results of this study suggest that diet (fat intake
and/or cholesterol levels) may influence the response to PFOA and may account for some of the
differences observed in humans and rats fed a standard diet, which is typically low in fat.

Chronic exposure of rats to PFOA resulted in hepatomegalocytosis, hepatocellular necrosis, and portal
mononuclear cell infiltration after a 1-year exposure to 15 mg/kg/day in the diet (3M 1983). A 2-year
exposure to 15 mg/kg/day resulted in hepatomegalocytosis, cystoid degeneration, and portal mononuclear
cell infiltration (3M 1983). The study also found significant increases in ALT and AST levels in male
rats exposed to 1.5 mg/kg/day. A second chronic exposure study found significant increases in relative
liver weight in rats exposed to 13.6 mg/kg/day in the diet for 2 years; no non-neoplastic lesions were
noted in the liver (Biegel et al. 2001).

Studies in monkeys suggest that longer-term exposure may also result in liver toxicity. Significant
increases in absolute and relative liver weight were observed in Cynomolgus monkeys exposed to
20/30 mg/kg/day administered via capsules for 26 weeks (Butenhoff et al. 2002). A significant increase
in absolute, but not relative, liver weight was also observed in monkeys administered 3 or 10 mg/kg/day.
However, no histological alterations were observed in the livers at the doses tested. Similarly, no
histological alterations were observed in the livers of Cynomolgus monkeys administered 2 or

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

185
2. HEALTH EFFECTS

20 mg/kg/day via capsules for 30 days (Thomford 2001) or Rhesus monkeys administered 3 or
10 mg/kg/day via gavage for 90 days (Griffith and Long 1980). Significant increases in serum
triglyceride levels were observed in the 10 and 20/30 mg/kg/day groups; the increases were statistically
significant at only some of the time points (Butenhoff et al. 2002). At 10 mg/kg/day, increases in serum
triglyceride levels at 4, 10, and 14 weeks of exposure were significantly higher than pre-treatment levels.
Increases in cholesterol levels were only observed in the 20/30 mg/kg/day group after 13 weeks of
exposure, but not after 26 weeks. No alterations in serum cholesterol or triglyceride levels were observed
in the Thomford (2001) study.
Several studies have examined PPARα-null mice to assess whether PFOA-induced liver effects can also
occur via a mechanism not involving peroxisome proliferation. Similar to wild-type mice, exposure to
PFOA resulted in significant increases in liver weight (Abbott et al. 2007; Das et al. 2017; Minata et al.
2010; Wolf et al. 2008a; Yang e al. 2002b). Abbott et al. (2007) found that the effect level was slightly
higher in PPARα-null mice than wild-type mice (3 versus 1 mg/kg/day) following oral exposure on
GDs 1–17 (liver weights measured at weaning). Wolf et al. (2008a) and Minata et al. (2010) reported the
same effect level (1 or 5 mg/kg/day, respectively) in PPARα-null mice and wild-type mice administered
PFOA via gavage for 7 days or 4 weeks. Wolf et al. (2008a) found dose-related increases in
hepatocellular cytoplasmic vacuoles at ≥1 mg/kg/day and suggested that the increase in liver weight was
due to the accumulation of PFOA in the hepatocytes rather than a toxic response. Hepatocyte
proliferation was also observed at 10 mg/kg/day. Unlike the Wolf et al. (2008a) study, the Minata et al.
(2010) 4-week study reported hepatocellular hypertrophy and microvesicular steatosis in the PPARα-null
mice (no incidence data were provided and it is unclear at what dose levels these effects were found);
cytoplasmic vacuolation was also reported in the hepatocytes. Filgo et al. (2015a) also reported
hepatocellular hypertrophy in PPARα-null mice; the LOAEL was 3 mg/kg/day, which was higher than the
LOAEL of 0.3 mg/kg/day found in wild-type mice. Minata et al. (2010) also reported cholangiopathy in
both the wild-type and PPARα-null mice, but noted that the effect was more intensive in the PPARα-null
mice. No significant alterations in steatosis or triglyceride accumulation were observed in PPARα-null
mice administered 10 mg/kg/day for 7 days, but were observed in wild-type mice (Das et al. 2017).
Additionally, significant decreases in serum total cholesterol levels at 5.2 and 10.2 mg/kg/day and
increases at 20.7 mg/kg/day were observed in the PPARα-null mice; significant decreases in total
cholesterol were observed in the wild-type mice at 10.2 and 20.7 mg/kg/day doses. Serum triglyceride
levels were increased in both strains at 5.2 and 10.2 mg/kg/day doses and in the PPARα-null mice at
20.7 mg/kg/day (Minata et al. 2010).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

186
2. HEALTH EFFECTS

Intermittent application of 20, 200, or 2,000 mg/kg APFO to the skin of rats for 2 weeks resulted in the
presence of one or more foci of coagulative necrosis in the livers from all treated groups (Kennedy 1985).
The Kupffer cells within the foci of hepatocellular necrosis contained large vesicular nuclei and were
markedly increased in number. At 2,000 mg/kg/day, these changes were seen in three out of five rats
killed on the 10th day of exposure, in three out of five rats killed on recovery day 14, and in one out of
five rats killed on recovery day 42. This lesion occurred in two out of five rats from the 20 mg/kg/day
dose group killed on day 10 of exposure. Serum ALT activity appeared elevated at termination of
exposure in a dose-related manner, but without achieving statistical significance. A similar trend was
seen for AST activity, but achieving statistical significance in the high-dose group. The blood
concentrations of organofluorine on the 10th day of exposure were 10.2, 52.4, 79.2, and 117.8 µg/mL in
the control, low-, mid-, and high-dose groups, respectively. A study in mice reported that application of
6.25 mg/kg/day PFOA on the dorsal surface of each ear for 4 days resulted in a 52% increase in absolute
liver weight (Fairley et al. 2007); no significant effect occurred after application of 2.5 mg/kg/day.

Summary. Epidemiology studies examining the hepatotoxicity of PFOA have examined three
outcomes—risk of liver disease, evidence of hepatocellular damage (as measured by alterations in serum
hepatic enzymes and bilirubin levels), and alterations in serum lipid levels (total cholesterol, LDL
cholesterol, HDL cholesterol, and triglycerides)—among workers, residents living near a PFOA
manufacturing facility with high levels of drinking water contamination, and the general population.
Exposure to PFOA does not appear to be associated with increased risks of liver disease in workers or
highly exposed community members. The epidemiology studies have found associations between serum
PFOA levels and increases in serum ALT, AST, and GGT enzyme levels and decreases in serum bilirubin
levels. However, the results have not been consistently found, and serum enzyme levels were typically
within the normal range. Four studies examined the risk of serum enzyme levels outside of the normal
range; the results were mixed for the risk of elevated ALT, with two studies finding an increased risk and
two studies finding no association. A number of occupational, community, and general population
studies have found associations between serum PFOA levels and serum total cholesterol levels; several
studies have also found no associations. Studies examining the change in cholesterol levels per change in
serum PFOA levels have found greater increases in serum cholesterol levels associated with serum PFOA
levels at the lower range of PFOA levels and the dose-response curve suggests a biphasic relationship.
Positive associations have also been observed for LDL cholesterol, although associations have not been
consistently found. In general, no consistent associations were found between serum PFOA and HDL
cholesterol or triglyceride levels. Studies in laboratory animals have found strong associations between
PFOA exposure and hepatotoxicity. Liver effects have been observed in rats exposed to airborne APFO

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

187
2. HEALTH EFFECTS

dusts; in rats, mice, and monkeys following oral exposure for acute-, intermediate-, or chronic-durations;
and in rats following dermal exposure. The observed effects typically include increases in liver weight,
hepatocellular hypertrophy, and decreases in serum cholesterol and triglyceride levels. Other effects that
have been observed include hyperplasia, necrosis, and fatty degeneration. Available evidence suggests
that the increased liver weight, hypertrophy, and serum lipid alterations are likely due to PPARα initiation
and therefore, may not be relevant to humans. However, other mechanisms of liver toxicity are also
involved, as evidenced by liver effects observed in PPARα-null mice (Das et al. 2017; Minata et al. 2010;
Wolf et al. 2008a). In contrast to the results observed in epidemiology studies, an experimental study in
humans exposed to PFOA (MacPherson et al. 2011) and human exposure to other PPARα agonists, such
as fibrates (Staels et al. 1998), suggest that hypolipidemic effects, similar to those observed in rodents,
may occur in humans exposed to PFOA, although humans may not be as sensitive as rodents.
PFOS

Epidemiology Studies—Liver Disease. Several studies have examined the possible association between
PFOS exposure and liver diseases. No increases in deaths from cirrhosis of the liver were found in
workers at the 3M facility in Decatur, Alabama (Alexander et al. 2003). Another study of workers at this
facility found no significant alterations in the episodes of care for all liver disorders or all biliary duct
disorders (Olsen et al. 2004a). However, among workers with at least 10 years of high potential exposure
to PFOS, there were significant increases in episodes of care for cholelithiasis or acute cholecystitis and
for all biliary tract disorders. A third study of workers at a PFOS facility in Decatur, Alabama did not
find increases in cholelithiasis, cholecystitis, or liver disease (including cirrhosis and hepatitis) (Grice et
al. 2007).

Epidemiology Studies—Hepatic Serum Enzymes and Bilirubin Levels. A series of studies conducted by
Olsen and associates evaluated liver function (as assessed by serum liver enzymes) in workers at several
3M facilities involved in PFOS production. Using health data collected in 1995 and 1997, Olsen et al.
(1999) did not find associations between serum PFOS and serum ALT, AST, or GGT enzymes at PFOS
levels <6,000 ng/mL; a positive association with total bilirubin levels was found. No conclusions were
drawn from the few workers with serum PFOS ≥6,000 ng/mL due to their small number (seven in 1995
and five in 1997 data). Similarly, no association of ALT, AST, or GGT and serum PFOA levels were
observed in groups of workers at these facilities examined in 1993 (111 subjects), 1995 (80 subjects),
and/or 1997 (74 subjects) (Olsen et al. 2000). A subsequent evaluation of workers from the same plants,
but that included women and a longitudinal analysis of the workers, reported that, after adjusting for

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

188
2. HEALTH EFFECTS

potential confounding factors, there were no substantial changes in hepatic parameters (Olsen et al.
2003a). GGT levels in females and ALT levels in males with PFOS levels in the 4th quartile were
significantly elevated in comparisons between individuals with serum PFOS levels in the 4th quartile to
those with levels in the 1st quartile; however, there were no statistical adjustments for potential
confounders. In contrast to these findings in workers, Gallo et al. (2012) reported significant increases in
the risks of elevated ALT, GGT, and bilirubin levels in a study of C8 participants. Conflicting results
have been found in general populations studies. Studies using the NHANES data set (Gleason et al. 2015;
Lin et al. 2010) did not find associations between serum PFOS and ALT, AST, GGT, or total bilirubin
levels. No increases in the risk of elevated levels of ALT, AST, or GGT were found (Gleason et al.
2015), although there was an increased risk of elevated total bilirubin levels. In a study of adults in Japan
(Yamaguchi et al. 2013), significant correlations between serum PFOS and ALT, AST, and GGT levels
were found.

Epidemiology Studies—Serum Lipids. Occupational, community, and general population studies have
examined possible associations between serum PFOS levels and serum lipids; these data are summarized
in Table 2-12. A graphical presentation of differences in total cholesterol and LDL cholesterol levels
relative to serum PFOS levels and the risks of elevated total cholesterol and LDL cholesterol are
presented in Figures 2-13, 2-14, 2-15, and 2-16.

In the Olsen occupational studies, significantly higher serum total cholesterol levels were found in
workers with serum PFOS levels between 3,000 and 6,000 ng/mL (Olsen et al. 1999, 2003a). However,
the studies found mixed results for associations between serum PFOS and other serum lipids, with one
study finding an association with LDL cholesterol (Olsen et al. 1999) and the other finding an association
with triglycerides (Olsen et al. 2003a). Longitudinal analysis was conducted using data for 174 workers
with medical surveillance data in 2000 and 1997 and/or 1995 (Olsen et al. 2003a). No significant
differences in serum PFOS levels were observed across the three time periods, and serum PFOS level was
not a significant predictor of cholesterol or triglyceride levels.

Two large-scale studies of participants in the C8 Science Panel studies found associations between serum
PFOS levels and serum lipid levels (Frisbee et al. 2010; Steenland et al. 2009b). Associations between
serum PFOS levels and total cholesterol, LDL cholesterol, and HDL cholesterol were found in a study of
over 12,000 children and adolescents; the mean serum PFOS levels were 20.7 ng/mL in children (aged
1.0–11.9 years) and 19.3 ng/mL in adolescents (aged 12.0–17.9 years) (Frisbee et al. 2010). Similar

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

189
2. HEALTH EFFECTS

Figure 2-13. Serum Total Cholesterol Levels Relative to Serum PFOS Levels
(Presented as percent change in cholesterol levels)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

190
2. HEALTH EFFECTS

Figure 2-14. Risk of Abnormal Cholesterol Levels Relative to PFOS Levels (Presented as Adjusted Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

191
2. HEALTH EFFECTS

Figure 2-15. Serum LDL Cholesterol Levels Relative to Serum PFOS Levels
(Presented as percent change in LDL cholesterol levels)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

192
2. HEALTH EFFECTS

Figure 2-16. Risk of Abnormal LDL Cholesterol Levels Relative to PFOS Levels (Presented as Adjusted Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

193
2. HEALTH EFFECTS

findings were reported in a study of adults with a median serum PFOS level of 19.6 ng/mL; the study
excluded subjects who reported taking cholesterol-lowering medication (Steenland et al. 2009b).
Associations were found between serum PFOS and total cholesterol, LDL cholesterol, and triglyceride
levels, but not with HDL cholesterol. Participants with serum PFOS levels in the 2nd, 3rd, and 4th quartiles
also had elevated risks of high cholesterol levels. The investigators noted that the odds of high
cholesterol from the 1st to the 5th quartile was approximately 50% for PFOS, which may be important
given that the Framingham study found that the risk of coronary heart disease was about 1.8 times higher
in subjects with total cholesterol levels >240 mg/dL as compared to subjects with levels <200 mg/dL.
Steenland et al. (2009b) also examined over 10,000 participants who were taking cholesterol-lowering
medication; an association between serum PFOS and total cholesterol levels was found in this group.
Using both groups of subjects (taking or not taking cholesterol-lowering medication), the investigators
analyzed whether taking cholesterol medication was associated with lower serum PFOA or PFOS levels,
which may be indicative of reverse causality; no differences in serum PFOS levels were found between
the two groups.

General population studies were conducted in the United States, Canada, and several European and Asian
countries; these studies have found mixed results for associations between serum PFOS levels and serum
lipids. Some studies have found associations between serum PFOS levels and serum total cholesterol
(Nelson et al. 2010; Skuladottir et al. 2015; Starling et al. 2014a) and HDL cholesterol (Châtaeu-Degat et
al. 2010); inverse associations between serum PFOS and HDL cholesterol (Starling et al. 2014a) and
triglycerides (Châtaeu-Degat et al. 2010) were also found. However, other studies in adults have not
found associations between serum PFOS and total cholesterol (Châtaeu-Degat et al. 2010; Eriksen et al.
2013; Fisher et al. 2013; Fu et al. 2014a), non-HDL cholesterol (Fisher et al. 2013), LDL cholesterol
(Châtaeu-Degat et al. 2010; Fisher et al. 2013; Fu et al. 2014a; Starling et al. 2014a), HDL cholesterol
(Fisher et al. 2013; Fu et al. 2014a), or triglycerides (Fu et al. 2014a; Starling et al. 2014a). Additionally,
two studies did not find increased risks of elevated cholesterol levels (Fisher et al. 2013; Fu et al. 2014a).
Several of these studies controlled for use of cholesterol-lowering medication (Châtaeu-Degat et al. 2010;
Eriksen et al. 2013; Fisher et al. 2013; Nelson et al. 2010). Overall, studies of children and adolescents
have found associations for serum lipid levels. Geiger et al. (2014b) found increases in the risk of
elevated cholesterol and LDL cholesterol in children and adolescents aged 12–18 years; an association
between serum PFOS and LDL cholesterol levels was also found. Zeng et al. (2015) found associations
between serum PFOS and serum total cholesterol, LDL cholesterol, and triglyceride levels in children
aged 12–15 years. Timmermann et al. (2014) also found an association between serum PFOS and
triglycerides only in obese Danish children (8–10 years of age), but not in normal weight children. In

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

194
2. HEALTH EFFECTS

contrast, Maisonet et al. (2015b) found an inverse association between maternal serum PFOS and total
cholesterol and LDL cholesterol in 15-year-old girls; no association was found when the girls were
7 years of age.

Laboratory Animal Studies. Unpublished data summarized by OECD (2002) indicate that inhalation
exposure of rats to lethal concentrations (1,890–45,970 mg/m3) of PFOS dusts for 1 hour resulted in
varying discoloration of the liver.

Consistent with the results for PFOA, acute-duration oral exposure of rats to PFOS resulted in increases
in liver weight (Elcombe et al. 2012b; Era et al. 2009; Haughom and Spydevold 1992), hepatocellular
hypertrophy (Elcombe et al. 2012b), and decreases in serum cholesterol and/or triglyceride levels
(Elcombe et al. 2012a, 2012b; Haughom and Spydevold 1992). The lowest adverse effect level for
increased liver weight, hypertrophy, and decreased serum cholesterol was 1.79 mg/kg/day in rats exposed
to PFOS in the diet for 7 days (Elcombe et al. 2012b); however, a similar study by this group did not find
significant alterations in liver weight or ALT, AST, or serum cholesterol levels after 7 days of exposure to
1.72 mg/kg/day (Elcombe et al. 2012a). Likewise, in mice, increases in liver weight (Fuentes et al. 2006;
Qazi et al. 2009b, 2010a; Wan et al. 2011), hepatocellular hypertrophy (Qazi et al. 2010a), and decreases
in serum cholesterol levels (Qazi et al. 2010a) were observed following acute exposure to PFOS. The
lowest LOAEL for liver weight was 3 mg/kg/day in mice administered PFOS via gavage on GDs 6–18
(Fuentes et al. 2006); no effects were observed at 1.5 mg/kg/day. The only acute-duration mouse study
that included histopathological examination of the liver and measurement of serum cholesterol levels
identified a LOAEL of 8.5 mg/kg/day in mice exposed to PFOS in the diet for 10 days (Qazi et al. 2010a).

Intermediate-duration exposure to PFOS resulted in increased liver weight in rats (Cui et al. 2009; Curran
et al. 2008; Elcombe et al. 2012a; Seacat et al. 2003; Thibodeaux et al. 2003) and mice (Bijland et al.
2011; Thibodeaux et al. 2003; Wan et al. 2011, 2014b; Xing et al. 2016; Yahia et al. 2008), hepatocellular
hypertrophy in rats (Cui et al. 2009; Curran et al. 2008; Elcombe et al. 2012a; Seacat et al. 2003),
decreased serum cholesterol levels in rats (Curran et al. 2008; Elcombe et al. 2012a; Luebker et al. 2005b;
Seacat et al. 2003), decreased total cholesterol, triglyceride, non-HDL cholesterol, and HDL cholesterol
levels in mice (Bijland et al. 2011), and increased serum AST and GGT levels in mice (Xing et al. 2016).
A mouse study (Bijland et al. 2011) also showed dramatic decreases in the hepatic production of VLDL
and HDL (Bijland et al. 2011). Another mouse study (Lee et al. 2015b) did not find increases in hepatic
lipid levels in dams, although there were alterations in fetal livers. Only one of the intermediate-duration
mouse studies included histopathological examination of the liver. Xing et al. (2016) reported

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

195
2. HEALTH EFFECTS

cytoplasmic vacuolization, focal necrosis, and hepatocellular hypertrophy in mice exposed to PFOS via
gavage for 30 days; however, the study did not report incidence; the lowest dose tested was
2.5 mg/kg/day. The lowest adverse effect level for liver effects in rats was 0.14 mg/kg/day for a
significant increase in relative liver weight in female rats, but not male rats, exposed to PFOS in the diet
for 28 days (Curran et al. 2008). This study also found significant decreases in serum cholesterol levels
and increases in absolute and relative liver weights in males and females at 2.98 mg/kg/day and
hepatocellular hypertrophy at 5.89 mg/kg/day. Seacat et al. (2003) reported increases in liver weight,
hepatocellular hypertrophy, and decreased serum cholesterol levels in rats following a 14-week dietary
exposure to 1.33 mg/kg/day; however, no significant alterations in liver weight or liver histopathology
were observed in rats exposed to 1.77 mg/kg/day PFOS in the diet for 4 weeks (Seacat et al. 2003). In
contrast, Elcombe et al. (2012a) reported increases in liver weight, hepatocellular hypertrophy, and
decreased serum cholesterol in rats exposed to 1.54 mg/kg/day PFOS in the diet for 28 days.

Data on the chronic toxicity of PFOS to the liver in rodents are limited to a study in rats (Butenhoff et al.
2012b; Thomford 2002b). Hepatotoxicity characterized by centrilobular hypertrophy, centrilobular
eosinophilic hepatocytic granules, and centrilobular hepatocytic vacuolation was noted in rats exposed to
PFOS in the diet for 2 years. Among rats sacrificed at the end of the study, significant increases in the
incidence of centrilobular hepatocellular hypertrophy were observed in male and female rats exposed to
≥0.25 mg/kg/day (Thomford 2002b). When animals sacrificed at interim periods (14 or 52 weeks) and
unscheduled deaths were included with animals sacrificed at exposure termination, the incidence of
centrilobular hepatocellular hypertrophy was also increased in males exposed to 0.1 mg/kg/day. At
≥0.1 mg/kg/day, significant increases in the incidences of eosinophilic clear cell altered foci and cystic
hepatocellular degeneration were observed in male rats. An increase in cystic degeneration was observed
in male rats exposed to ≥0.025 mg/kg/day. However, this was mainly due to a high incidence in
unscheduled deaths; among animals sacrificed at exposure termination, the incidence was only increased
in males exposed to 1.04 mg/kg/day. An increased incidence of single cell necrosis was observed in
males and females at 1.04 mg/kg/day (all groups combined). Observations made in a group of rats
exposed to 1.17 mg/kg/day PFOS for 52 weeks and allowed to continue on the control diet for an
additional year showed that hepatotoxicity was not a persistent response, as hepatotoxicity was generally
absent at the end of the recovery period. At termination, electron microscopy showed mild to moderate
smooth endoplasmic reticulum hyperplasia and minimal to mild hepatocellular hypertrophy primarily in
rats dosed with 1.5 mg/kg/day PFOS, the highest dose tested.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

196
2. HEALTH EFFECTS

Treatment of Cynomolgus monkeys with up to 2 mg/kg/day PFOS administered via a capsule for 4 weeks
did not induce gross or microscopic morphological alterations in the liver and did not increase cell
proliferation (Thomford 2002a). In a 26-week study in Cynomolgus monkeys, exposure to
0.75 mg/kg/day PFOS, administered via a capsule resulted in increased absolute liver weight after
183 days of treatment (Seacat et al. 2002). Significant decreases in serum total cholesterol were also
observed at 0.75 mg/kg/day after 91, 153, and 182 days of exposure. On day 182, total cholesterol
decreased to 35 and 53% of predosing values in males and females, respectively. The HDL cholesterol
levels were significantly lower in males at 0.03 and 0.75 mg/kg/day on days 153 and 182 and in females
at 0.15 and 0.75 mg/kg/day on days 153 and 182; the lack of pre-treatment HDL cholesterol
measurements precludes within-group comparisons. Serum bilirubin was significantly lower in males at
0.75 mg/kg/day on days 91, 153, and 182. Light microscopy of liver sections showed centrilobular
vacuolation, hypertrophy, and mild bile stasis in some monkeys exposed to 0.75 mg/kg/day. Electron
microscopy showed lipid-droplet accumulation in some males and females exposed to 0.75 mg/kg/day.
Increased glycogen content was also noted at this dose level. No histological alterations were observed in
the livers of monkeys exposed to 0.75 mg/kg/day for 26 weeks and allowed to recover for 7 months or
1 year. Similarly, serum cholesterol returned to pretreatment levels 36 days post exposure and HDL
cholesterol levels returned to pretreatment levels after 61 days of recovery.

Summary. Epidemiology studies have examined the possible associations between PFOS exposure and
liver disease in workers and hepatocellular damage and alterations in serum lipid levels in workers and
the general population. The available occupational exposure studies or general population studies do not
consistently suggest an association between PFOS exposure and increases in the risk of liver disease or
biliary tract disorders. A small number of occupational exposure studies have not found associations
between serum PFOS levels and increases in ALT, AST, or GGT levels. Overall, the epidemiology
studies suggest an association between serum PFOS levels and increases in serum total cholesterol levels
and possibly serum LDL cholesterol levels. Studies of workers at a PFOS manufacturing facility found
elevated serum total cholesterol levels in workers with high serum PFOS levels; however, a longitudinal
analysis at the same facility did not find that serum PFOS was a significant predictor of cholesterol levels.
Studies of residents living in an area with very high PFOA water levels found increases in serum total
cholesterol levels associated with elevated serum PFOS levels in children, adolescents, and adults. Mixed
results have been found for associations between serum PFOS and increases in serum total cholesterol
levels in general population studies. Associations have been found between serum PFOS levels and
serum LDL-cholesterol levels among non-occupational populations. In laboratory animals, oral exposure
to PFOS results in increases in liver weight, hepatocellular hypertrophy, and decreases in serum lipid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

197
2. HEALTH EFFECTS

levels. A small number of studies also reported focal necrosis and centrilobular hepatocytic
vacuolization. The applicability of the hepatic hypertrophy and serum lipid alterations observed in rodent
studies to humans has been questioned due to species differences in PPARα receptor activation, the
presumed mechanism of action for these effects in rodents.

PFHxS

Epidemiology Studies—Hepatic Serum Enzymes and Bilirubin Levels. Lin et al. (2010) did not find
associations between serum ALT and GGT levels with serum PFHxS levels in a general population study
using the NHANES data set.

Epidemiology Studies—Serum Lipids. Four studies have evaluated the potential association between
serum PFHxS levels and serum lipids in the general population. A study utilizing the NHANES data set
for adults not taking cholesterol-lowering medication reported an association between serum PFHxS and
non-HDL cholesterol, but no associations with total cholesterol, LDL cholesterol, or HDL cholesterol
(Nelson et al. 2010). In a study of Canadian adults not taking cholesterol-lowering medication with a
geometric mean serum PFHxS level of 2.16 ng/mL, associations were found for total cholesterol, LDL
cholesterol, and non-HDL cholesterol (Fisher et al. 2013). The study also found increased odds of having
a high cholesterol level with increasing PFHxS levels. In pregnant women in Norway with median serum
PFHxS levels of 0.60 ng/mL, serum PFHxS levels were associated with serum HDL cholesterol, but not
with total cholesterol, LDL cholesterol, or triglycerides (Starling et al. 2014a). No associations between
serum PFHxS and total cholesterol, LDL cholesterol, HDL cholesterol, or triglyceride levels were found
in a study of Taiwanese children aged 12–15 years (mean serum PFHxS of 2.1 ng/mL) (Zeng et al. 2015).

Laboratory Animal Studies. Acute-duration gavage administration of PFHxS resulted in increases in
liver weight, steatosis, and increases in hepatic triglyceride levels in mice; increases in liver weight and
steatosis were also observed in similarly exposed PPARα-null mice (Das et al. 2017). An intermediateduration study with PFHxS in rats reported that gavage doses ≥3 mg/kg/day induced a significant increase
in absolute and relative liver weight in males (Butenhoff et al. 2009a; Hoberman and York 2003). Light
microscopy revealed minimal to moderate enlargement of centrilobular hepatocytes. Clinical chemistry
tests showed a significant decrease in serum cholesterol at ≥0.3 mg/kg/day and decreased serum
triglycerides at 10 mg/kg/day. None of these alterations were observed in female rats. In male mice,
dietary exposure to PFHxS in a western-type diet resulted in >50% decreases in plasma triglyceride, total
cholesterol, non-HDL cholesterol, and HDL cholesterol levels and approximately 75% decreases in the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

198
2. HEALTH EFFECTS

hepatic production of VLDL (Bijland et al. 2011). Increases in liver weight and hepatic triglyceride
levels were also observed.
PFNA

Epidemiology Studies—Hepatic Serum Enzymes and Bilirubin Levels. A health evaluation of workers
at a U.S. polymer production facility using PFNA did not find alterations in ALT, AST, GGT, or bilirubin
levels related to increases in exposure intensity score in a longitudinal analysis (Mundt et al. 2007).
Associations between serum PFNA and ALT and GGT levels were observed in a NHANES data study
(Gleason et al. 2015); however, another study (Lin et al. 2010) utilizing the NHANES data did not find
associations between serum PFNA and these enzymes. Neither study found associations for AST or total
bilirubin. The Gleason et al. (2015) study also did not find increased risks for elevated ALT, AST, GGT,
or bilirubin.

Epidemiology Studies—Serum Lipids. Longitudinal analysis of serum lipid levels in the occupational
exposure study (Mundt et al. 2007) did not find significant differences in serum total cholesterol or
triglycerides over time. In general population studies, associations have been observed between serum
PFNA levels and total cholesterol levels in adults (Fu et al. 2014a; Nelson et al. 2010) and children (Zeng
et al. 2015); one study in pregnant women did not find an association (Starling et al. 2014a). Several
studies have also found associations with LDL cholesterol (Fu et al. 2014a; Zeng et al. 2015) or non-HDL
cholesterol (Nelson et al. 2010), but others did not find associations for LDL cholesterol (Nelson et al.
2010; Starling et al. 2014a). Most studies did not find an association between serum PFNA and HDL
cholesterol (Fu et al. 2014a; Nelson et al. 2010; Zeng et al. 2015) or triglycerides (Fu et al. 2014a;
Starling et al. 2014a). Exceptions were the Starling et al. (2014a) study of pregnant women, which found
a positive association for HDL cholesterol, and Zeng et al. (2015), which found an association with
triglycerides in children. Fu et al. (2014a) did not find increased risks of elevated cholesterol, LDL
cholesterol, or triglyceride levels or lowered HDL cholesterol levels in adults.

Laboratory Animal Studies. Nine studies have evaluated the hepatic toxicity of PFNA. The observed
effects are consistent with effects observed for other perfluoroalkyl compounds. Alterations in serum
lipid levels consisted of decreases in serum HDL cholesterol levels in rats administered via gavage
≥1 mg/kg/day for 14 days (Fang et al. 2012a) and decreases in serum triglyceride and cholesterol levels in
mice receiving gavage doses of ≥1 mg/kg/day (Wang et al. 2015a). Increases in liver weight were
observed in rats nose-only exposed to ≥67 mg/m3 (Kinney et al. 1989), in mice administered via gavage

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

199
2. HEALTH EFFECTS

10 mg/kg/day PFNA for 7 days (Das et al. 2017), in mice exposed to 0.5 mg/kg/day PFNA in the diet for
14 days (Kennedy 1987), in mice administered ≥0.2 mg/kg/day PFNA via gavage for 14 days (Wang et
al. 2015a), and in the offspring of mice administered via gavage ≥0.83 mg/kg/day PFNA on GDs 1–17 or
1–18 (Das et al. 2015; Wolf et al. 2010). Hepatocellular vacuolation was observed in mice administered
via gavage 5 mg/kg/day for 14 days (Fang et al. 2012b). In PPARα-null mice, increases in liver weight
were observed in non-pregnant mice administered via gavage ≥1.5 mg/kg/day for 18 days, but were not
found in pregnant animals (Wolf et al. 2010). Das et al. (2017) found increases in liver weight, steatosis,
and increases in liver triglyceride levels in PPARα-null mice administered 10 mg/kg/day for 10 days.
PFDeA

Epidemiology Studies—Serum Lipids. Three general population studies have evaluated the potential
relationships between serum PFDeA and serum lipids and reported inconsistent results. Fu et al. (2014a)
found an association between serum PFDeA and total cholesterol in adults; studies in pregnant women
(Starling et al. 2014a) and children (Zeng et al. 2015) did not find associations. Both Fu et al. (2014a)
and Starling et al. (2014a) found positive associations with HDL cholesterol; this was not found in the
Zeng et al. (2015) study. None of the studies found associations between serum PFDeA and LDL
cholesterol or triglycerides (Fu et al. 2014a; Starling et al. 2014a; Zeng et al. 2015). Only the Fu et al.
(2014a) study looked for alterations in the risk of elevated cholesterol, LDL cholesterol, or triglyceride
levels or decreased HDL cholesterol levels, but the study did not find significant increases in the risk.

Laboratory Animal Studies. Hepatic effects observed in laboratory animals exposed to PFDeA include
alterations in liver weight and morphology. Increases in liver weight have been observed in mice
following a single gavage dose of PFDeA; the alterations were observed 2 days after exposure to
40 mg/kg/day (Brewster and Birnbaum 1989) or 30 days after exposure to ≥20 mg/kg/day (Harris et al.
1989). Repeated dietary exposure to 2.4 mg/kg/day PFDeA for 1 week (Kawashima et al. 1995) or
78 mg/kg/day for 10 days (Permadi et al. 1992, 1993) also resulted in increases in liver weight. Oral
doses ≥9.5 mg/kg/day also resulted in increases in hepatic cholesterol levels in rats (Kawashima et al.
1995) and hepatic lipids in mice (Brewster and Birnbaum 1989). These acute doses were also associated
with hepatocellular hypertrophy and evidence of peroxisome proliferation. Thirty days after a single
gavage dose of ≥20 mg/kg/day PFDeA, effects included periportal to panlobular hepatocellular
hypertrophy characterized by swollen hepatocytes with abundant granular eosinophilic cytoplasm and
enlarged and hyperchromatic nuclei (Harris et al. 1989).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

200
2. HEALTH EFFECTS

PFUA

Epidemiology Studies—Serum Lipids. Neither of the two studies (Fu et al. 2014a; Starling et al. 2014a)
examining possible relationships between serum PFUA and serum lipids found associations between
serum PFUA and for total cholesterol, LDL cholesterol, or triglyceride levels. Starling et al. (2014a)
found an association of serum PFUA levels with HDL cholesterol levels; Fu et al. (2014a) did not find an
association for this parameter. No alterations in the risk of abnormal serum lipid levels were found in the
adults examined by Fu et al. (2014a).

Laboratory Animal Studies. Only one animal study was identified that examined the liver following oral
exposure to PFUA. In an intermediate-duration study of rats administered PFUA via gavage, increases in
relative liver weight were observed in males at 0.3 mg/kg/day and in females at 1.0 mg/kg/day, and mild
to moderate centrilobular hepatocellular hypertrophy was observed in males and females at 1.0 mg/kg/day
(Takahashi et al. 2014).

PFHpA

Epidemiology Studies—Serum Lipids. Epidemiology data on PFHpA are limited to a study in adults
conducted by Fu et al. (2014a), which found no associations between serum PFHpA and total cholesterol,
LDL cholesterol, HDL cholesterol, or triglyceride levels.
PFBuS

Epidemiology Studies—Serum Lipids. In the only epidemiology study examining serum lipids and
possible associations with serum PFBuS, Zeng et al. (2015) found an association with total cholesterol
levels in children. No associations were found between serum PFBuS and LDL cholesterol, HDL
cholesterol, or triglycerides.

Laboratory Animal Studies. Treatment of male rats with 900 mg/kg/day PFBuS by gavage for 28 days
induced a significant increase in absolute and relative liver weight (25–30%) relative to controls, which
was no longer detected following a 14-day recovery period (3M 2001). Clinical chemistry tests of liver
function were unremarkable and there were no chemical-related microscopic alterations. The NOAEL for
liver weight effects was 300 mg/kg/day. No alterations in liver weight, serum chemistry parameters
(ALT, AST, cholesterol), or liver morphology were observed in rats administered gavage doses as high as

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

201
2. HEALTH EFFECTS

600 mg/kg/day PFBuS for 90 days (Lieder et al. 2009a). Significant increases in liver weight were
observed at 300 and 1,000 mg/kg/day in a 2-generation study (Lieder et al. 2009b); the alterations were
only observed in male rats. An increase in hepatocellular hypertrophy was also observed in the male P0
and F1 rats administered via gavage 1,000 mg/kg/day. Dietary exposure to mice resulted in decreases in
plasma triglyceride levels and hepatic cholesterol levels, but no alterations in liver weight or plasma
cholesterol, HDL cholesterol, or non-HDL cholesterol (Bijland et al. 2011).
PFBA

Epidemiology Studies—Serum Lipids. Only one epidemiology study examined hepatic outcomes; this
study (Fu et al. 2014a) did not find any associations between serum PFBA levels and total cholesterol,
LDL cholesterol, HDL cholesterol, or triglycerides in adults.

Laboratory Animal Studies. Treatment of rats with up to 184 mg/kg/day PFBA by gavage for 5 days did
not affect liver weight, nor did it cause gross or microscopic morphological alterations in the liver (3M
2007a). In addition, clinical chemistry tests did not indicate altered liver function. Similarly,
administration of approximately 20 mg/kg/day PFBA in the diet to male rats for 2 weeks did not
significantly affect relative liver weight, but the same dose of PFOA induced a 45% increase in liver
weight (Ikeda et al. 1985). Dietary administration of doses of approximately 78 mg/kg/day PFBA to male
mice for 10 days induced a 63% increase in absolute liver weight (Permadi et al. 1992, 1993).

PFBA intermediate-duration studies have consistently found increases in liver weight and histological
alterations. Dosing rats with by gavage for 28 days resulted in significant increases in absolute and
relative liver weight and decreases in serum cholesterol at ≥30 mg/kg/day and hepatocellular hypertrophy
at 150 mg/kg/day (Butenhoff et al. 2012a; van Otterdijk 2007a). Administration of 150 mg/kg/day
induced hepatocyte hypertrophy. These liver effects were no longer detected after a 21-day recovery
period. In a similar 90-day study, administration of 30 mg/kg/day resulted in increased absolute liver
weight and panlobular hepatocyte hypertrophy (Butenhoff et al. 2012a; van Otterdijk 2007b); no liver
effects were observed at 6 mg/kg/day. None of the liver alterations were observed after a 21-day recovery
period.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

202
2. HEALTH EFFECTS

PFDoA

Epidemiology Studies—Serum Lipids. A general population study of adolescents (Zeng et al. 2015) did
not find any associations between serum PFDoA and total cholesterol, LDL cholesterol, HDL cholesterol,
or triglyceride levels.

Laboratory Animal Studies. Dosing of male Sprague-Dawley rats with 10 mg/kg/day PFDoA by gavage
for 14 days induced a 35% increase in total serum cholesterol; doses of 1 or 5 mg/kg/day had no
significant effect (Shi et al. 2007). In a subsequent study, the same group of investigators reported that in
rats dosed via gavage with 1 or 5 mg/kg/day PFDoA, there was a trend for decreased serum triglycerides,
but the differences with controls were not statistically significant (Zhang et al. 2008); at 10 mg/kg/day,
serum triglyceride levels were significantly increased. Liver triglyceride and liver cholesterol levels were
increased at ≥5 mg/kg/day. Single doses ≥1 mg/kg/day significantly induced the expression of PPARα
and PPARγ and their target genes to enhance fatty acid β-oxidation. Absolute liver weight was
significantly reduced in the 5 mg/kg/day group (19%) relative to controls, but this may have been due to a
marked reduction in body weight (shown in Shi et al. [2007], but not in Zhang et al. [2008]).
PFHxA

Laboratory Animal Studies. In a chronic-duration study, gavage administration of 200 mg/kg/day for
2 years resulted in increases in the incidence of hepatocellular necrosis in female rats (Klaunig et al.
2015). At 100 mg/kg/day, decreases in triglyceride levels were observed in male rats.

PFOSA

Laboratory Animal Studies. In the only study examining hepatic effects, Seacat and Luebker (2000)
reported no alterations in liver weight in rats receiving a single gavage dose of 5 mg/kg PFOSA.
2.10 RENAL
Overview. Epidemiology and laboratory animal studies have evaluated the potential of perfluoroalkyls to
be renal toxicants. Human studies have evaluated the risk of kidney disease, alterations in renal function,
damage to the kidney, and alterations in uric acid levels. The results of epidemiology studies evaluating
kidney disease and renal function are summarized in Table 2-13; Table 2-14 contains the studies

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

203
2. HEALTH EFFECTS

Table 2-13. Summary of Renal Outcomes in Humansa
Reference and study populationb
PFOA
Costa et al. 2009

Serum perfluoroalkyl level Outcome evaluated

Resultc

12,930 ng/mL (mean PFOA
in current workers)

Serum urea

NS (p>0.05)

Serum creatinine
Total proteins
α1 globulins, α2
globulins, β globulins, or
γ globulins
α2 globulins

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Occupational
(n =53)

Lundin et al. 2009

Association (p<0.01), current,
former and non-exposed workers.
SMR 5.2 (0.6–18.9)

NR

Nephritis and nephrosis
deaths

Raleigh et al. 2014

NR

Chronic kidney disease
deaths

HR 0.73 (0.21–2.48), 4th quartile

Occupational (n=9,027)
Steenland et al. 2015

Cumulative PFOA

Chronic kidney disease
risk

NS (p=0.92), no lag
NS (p=0.99), 10-year lag

7,800 ng/mL-year (mean
PFOA)

Chronic kidney disease
deaths

SMR 3.79 (1.03–9.71)*, 2nd quartile

Occupational (n=1,084)
Anderson-Mahoney et al. 2008

NR

Kidney disease (selfreported)

SPR 2.26 (1.45–3.51)*

Community (n=566)
Dhingra et al. 2016b

Cumulative PFOA

Chronic kidney disease

354 ng/mL (median PFOA)

Serum creatinine

NS (p=0.80 for trend), no lag
NS (p=0.81 for trend), 5-year lag
NS (p=0.88 for trend), 10-year lag
NS (p=0.30 for trend), 20-year lag
NS (p>0.05)

BUN
Total serum proteins

NS (p>0.05)
NS (p>0.05)

Occupational (n=3,992)

Occupational (n=3,713)
Steenland and Woskie 2012

Community (C8) (n=28,541)
Emmett et al. 2006b
Community (n=371)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

204
2. HEALTH EFFECTS

Table 2-13. Summary of Renal Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Watkins et al. 2013

28.3 ng/mL (median PFOA)

GFR

Inverse association (p=0.02)*

≥4.7 ng/mL (4th PFOA
quartile)

GFR

Inverse association (p<0.01)*,
4th quartile
Association (p<0.01)*

Community (C8) (9,660 children)
Kataria et al. 2015
General population (NHANES)
(n=1,960 adolescents)
Shankar et al. 2011a

Serum uric acid
>5.9 ng/mL (4th PFOA
quartile)

General population (NHANES) (n=4,587)
PFOS
Olsen et al. 1998a
Occupational (n=178 in 1995; n=149 in 1997)

Watkins et al. 2013
Community (C8) (9,660 children)
Kataria et al. 2015
General population (NHANES)
(n=1,960 adolescents)
Shankar et al. 2011a
General population (NHANES) (n=4,587)

2,440 and 1,930 ng/mL
(mean PFOS in 1995 in
DeCatur and Antwerp)
1,960 and 1,480 ng/mL
(mean in 1997 in DeCatur
and Antwerp)
20.0 ng/mL (median PFOS)

GFR
Chronic kidney disease

Inverse association (p<0.001 for
trend)*
OR 1.73 (1.04–2.88)*, 4th quartile

Serum creatinine
BUN

Association (p<0.06)*, 1997 only
NS (p>0.1)

GFR

Inverse association (p=0.0001)*

7.9–12.8 ng/mL (2nd PFOS
GFR
quartile)
≥19.4 ng/mL (4th PFOS
quartile)
>29.5 ng/mL (4th PFOS
GFR
quartile)
11.2–17.8 ng/mL (2nd PFOS Chronic kidney disease
quartile)

Inverse association (p<0.05)*,
2nd quartile

5.2 ng/mL (median PFHxS)

Inverse association (p=0.003)*

Inverse association (p<0.001 for
trend)*
OR 1.82 (1.02–3.27)*, 4th quartile

PFHxS
Watkins et al. 2013

GFR

Community (C8) (9,660 children)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

205
2. HEALTH EFFECTS

Table 2-13. Summary of Renal Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Kataria et al. 2015

≥4 ng/mL (4th PFHxS
quartile)

GFR

NS (p>0.05)

General population (NHANES)
(n=1,960 adolescents)
PFNA
Mundt et al. 2007

NR

BUN
Creatinine

Small, but not clinically significant

Occupational (n=592)
Watkins et al. 2013

1.5 ng/mL (median PFNA)

GFR

Inverse association (p=0.002)*

≥1.5 ng/mL (4th PFNA
quartile)

GFR

NS (p>0.05)

Community (C8) (9,660 children)
Kataria et al. 2015
General population (NHANES)
(n=1,960 adolescents)
aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 8 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

BUN = blood urea nitrogen; GFR = glomerular filtration rate; HR = hazard ratio; NHANES = National Health and Nutrition Examination Survey;
NR = not reported; NS = not significant; OR = odds ratio; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; SMR = standardized mortality ratio

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

206
2. HEALTH EFFECTS

Table 2-14. Summary of Uric Acid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

12,930, ng/mL (mean PFOA Serum uric acid
in current workers)

Association (p=0.039)*

490 ng/mL (median PFOA)

Association (reported by
investigator)*

PFOA
Costa et al. 2009
Occupational (n=53)
Sakr et al. 2007b
Occupational (n=1,025)
Steenland et al. 2010b
Community (n=54,591)
Gleason et al. 2015
General population (NHANES)
(n=4,333 adolescents)
Geiger et al. 2013
General population (NHANES)
(n=1,772 adolescents and adults)
Kataria et al. 2015
General population (NHANES)
(n=1,960 adolescents)
Qin et al. 2016

11.5–20.6 ng/mL (2nd quintile Hyperuricemia risk
PFOA)

OR 1.33 (1.24–1.43)* (2nd quintile)

3.7 ng/mL (median PFOA)

Serum uric acid
Hyperuricemia risk

Association (p<0.001)*
Association (p<0.001)*

4.3 ng/mL (mean PFOA),
>5.4 ng/mL (4th PFOA
quartile)

Serum uric acid
Hyperuricemia risk

Association (p=0.0001)*
OR 1.62 (1.10–2.37)* (4th quartile)

≥4.7 ng/mL (4th PFOA
quartile)

Serum uric acid

Association (p<0.01)*

0.5 ng/mL (median PFOA)

Serum uric acid
Hyperuricemia risk

3.5–5.1 ng/mL (3rd PFOA
quartile)

Serum uric acid
Hyperuricemia risk

Association (p<0.05)*
OR 2.16 (1.29–3.61)* (full cohort)
OR 2.76 (1.37–5.56)* (boys only)
Association (p<0.0001)*
OR 1.90 (1.35–2.69)*, 3rd quartile

General population (n=225 adolescents)
Shankar et al. 2011b

Serum uric acid

General population (NHANES) (n=3,883 adults)
PFOS
Steenland et al. 2010b
17.5–23,26 ng/mL (3rd PFOS Hyperuricemia risk
quintile)
Community (n=54,591)

***DRAFT FOR PUBLIC COMMENT***

OR 1.11 (1.04–1.20)* (3rd quintile)

 PERFLUOROALKYLS

207
2. HEALTH EFFECTS

Table 2-14. Summary of Uric Acid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Gleason et al. 2015

11.3 ng/mL (median PFOS) Serum uric acid

Association (p<0.01)*

General population (NHANES,
n=4,333 adolescents)
Geiger et al. 2013
General population (NHANES)
(n=1,772 adolescents and adults)
Kataria et al. 2015
General population (NHANES)
(n=1,960 adolescents)
Qin et al. 2016
General population (n=225 adolescents)
Shankar et al. 2011b

Hyperuricemia risk

NS (p=0.502)

18.4 ng/mL (mean PFOS),
>25.5 ng/mL (4th PFOS
quartile)

Serum uric acid
Hyperuricemia risk

NS (p=0.0575)
OR 1.65 (1.10–2.49)* (4th quartile)

≥19.4 ng/mL (4th PFOS
quartile)

Serum uric acid

Association (p<0.05)*

28.9 ng/mL (median PFOS) Serum uric acid
Hyperuricemia risk
11.2–17.8 ng/mL (2nd PFOS Serum uric acid
quartile)
Hyperuricemia risk

General population (NHANES) (n=3,883 adults)
PFHxS
Gleason et al. 2015
1.8 ng/mL (median PFHxS)
General population (NHANES) (n=4,333)
Kataria et al. 2015

NS (p>0.05)
OR 1.35 (0.95–1.93) (full cohort)
Association (p=0.0018)*
OR 1.46 (1.11–1.91)*, 2nd quartile

Serum uric acid
Hyperuricemia risk

NS (p>0.01)
NS (p=0.110 for trend)

≥4 ng/mL (4th PFHxS
quartile)

Serum uric acid

NS (p>0.05)

1.3 ng/mL (median PFHxS)

Serum uric acid
Hyperuricemia risk

Association (p<0.05)*
OR 1.39 (0.93–2.07)

NR

Serum uric acid

Small, but not clinically significant

General population (NHANES)
(n=1,960 adolescents)
Qin et al. 2016
General population (n=225 adolescents)
PFNA
Mundt et al. 2007
Occupational (n=592)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

208
2. HEALTH EFFECTS

Table 2-14. Summary of Uric Acid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Gleason et al. 2015

1.4 ng/mL (median PFNA)

Serum uric acid

Association (p<0.001)*

Hyperuricemia risk

NS (p=0.42 for trend)

≥1.5 ng/mL (4th PFNA
quartile)

Serum uric acid

NS (p>0.05)

0.8 ng/mL (median PFNA)

Serum uric acid
Hyperuricemia risk

NS (p>0.05)
OR 1.28 (0.83–1.96)

0.9 ng/mL (median PFDeA) Serum uric acid
Hyperuricemia risk

NS (p>0.05)
OR 1.26 (0.82–1.92)

0.5 ng/mL (median PFBuS)

Serum uric acid
Hyperuricemia risk

NS (p>0.05)
OR 1.23 (0.86–1.75)

2.7 ng/mL (median PFDoA) Serum uric acid
Hyperuricemia risk

NS (p>0.05)
OR 0.93 (0.65–1.34)

General population (NHANES) (n=4,333)
Kataria et al. 2015
General population (NHANES)
(n=1,960 adolescents)
Qin et al. 2016
General population (n=225 adolescents)
PFDeA
Qin et al. 2016
General population (n=225 adolescents)
PFBuS
Qin et al. 2016
General population (n=225 adolescents)
PFDoA
Qin et al. 2016
General population (n=225 adolescents)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

209
2. HEALTH EFFECTS

Table 2-14. Summary of Uric Acid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.2 ng/mL (median PFHxA)

Serum uric acid

NS (p>0.05)

Hyperuricemia risk

OR 1.08 (0.77–1.61)

PFHxA
Qin et al. 2016
General population (n=225 adolescents)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 8 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

NHANES = National Health and Nutrition Examination Survey; NR = not reported; NS = not significant; OR = odds ratio; PFBuS = perfluorobutane sulfonic acid;
PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

210
2. HEALTH EFFECTS

evaluating alterations in uric acid levels. More detailed descriptions of these studies can be found in the
Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 8. Although there are a
couple of studies finding associations between PFOA exposure and kidney disease, the results are not
consistent across study populations. However, there is some indication that perfluoroalkyls may affect
renal function. Decreases in estimated glomerular filtration rate and increases in uric acid levels
associated with serum PFOA or PFOS have been reported in a number of epidemiology studies.
However, these alterations may be due to reverse causality (i.e., increases in serum perfluoroalkyl levels
could be due to a decrease in glomerular filtration and shared renal transporters for perfluoroalkyls and
uric acid. Based on the small number of epidemiology studies or the inconsistency of the results, possible
associations between other perfluoroalkyls (PFHxS, PFNA, PFDeA, PFBuS, or PFDoA) and renal
functions cannot be assessed.

Laboratory animal studies have primarily evaluated kidney morphology; these studies are summarized in
Tables 2-1, 2-3, 2-4, 2-5, and 2-6. The NOAEL and LOAEL values for these studies are illustrated in
Figures 2-4, 2-6, 2-7, and 2-8. In general, the laboratory animal studies have not found evidence of
impaired renal function or morphological damage following exposure to PFOA, PFOS, PFHxS, PFDeA,
PFUA, PFBuS, PFBA, or PFHxA.
PFOA

Epidemiology Studies—Kidney Disease. Several epidemiology studies have examined the possible
association between PFOA exposure and increased risk of kidney disease. In a cohort mortality study of
workers at the DuPont PFOA facility in West Virginia, Steenland and Woskie (2012) found an increase in
deaths from chronic renal disease when compared to DuPont workers at other regional facilities. When
cumulative PFOA exposure was estimated based on the worker’s job history and data from a
biomonitoring survey conducted from 1979 to 2004, there was a significant positive trend for
nonmalignant kidney disease when the workers were divided in estimated cumulative exposure quartiles.
Two studies of workers at the 3M APFO facility in Cottage Grove, Minnesota did not find increases in
deaths from chronic kidney disease (Raleigh et al. 2014) or nephritis and nephrosis (Lundin et al. 2009) as
compared to mortality rates for the state of Minnesota. Similar results were found when chronic kidney
disease deaths were compared to those in cohort of workers in St. Paul Minnesota who worked at a nonAPFO facility (Raleigh et al. 2014). An occupational exposure study (Steenland et al. 2015) and C8
community study (Dhingra et al. 2016b) found no associations between cumulative PFOA exposure and
the risk of chronic kidney disease. Another study of the community living near the Washington Works

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

211
2. HEALTH EFFECTS

facility found a higher prevalence of self-reported kidney disease as compared to rates reported in
NHANES (Anderson-Mahoney et al. 2008).

Epidemiology Studies—Biomarkers of Renal Function. Several biomarkers of renal function have been
evaluated in epidemiology studies; these include BUN, serum creatinine, glomerular filtration rate, and
uric acid levels (discussed in the following section). Kidney function, assessed by levels of BUN and
serum creatinine, was not associated with exposure to PFOA in the occupational exposure studies by
Olsen et al. (2003a) or Costa et al. (2009) or a community exposure study by Emmett et al. (2006b).

Three studies have found inverse associations between serum PFOA and glomerular filtration rate. Using
the NHANES data for the 1999–2008 cycles, Shankar et al. (2011a) found an inverse association between
serum PFOA levels and estimated glomerular filtration rate in adults. The likelihood of chronic kidney
disease, defined as a glomerular filtration rate of <60 mL/minute/1.73 m2, was significantly higher in
adults with the highest serum PFOA (>5.9 ng/mL, OR 1.73, 95% CI 1.04–2.88) levels than in adults with
serum PFOA levels in the lowest quartile. The study also investigated whether the association between
serum PFOA levels and chronic kidney disease was due to reverse causality (i.e., decreased glomerular
filtration leads to a decrease in perfluoroalkyl filtration) and found a stronger negative correlation
between estimated glomerular filtration rate and serum PFOA levels in subjects without chronic kidney
disease, suggesting that it was not due to reverse causality. In another study utilizing NHANES data, an
inverse association was found in adolescents with serum PFOA levels in the 4th quartile (Kataria et al.
2015). Similarly, an inverse association between serum PFOA and glomerular filtration rate was found in
children participating in the C8 Health Project (Watkins et al. 2013). Unlike Shankar et al. (2011a),
Watkins et al. (2013) suggested that the association between serum perfluoroalkyl levels and estimated
glomerular filtration rates may be a consequence of reverse causation because no associations were found
between estimated serum PFOA levels 3 or 10 years prior to enrollment in the study or at the time of
study enrollment and estimated glomerular filtration rates; predicted serum PFOA levels were based on
environmental PFOA levels, self-reported residential history, and PBPK modeling.

Epidemiology Studies—Alterations in Uric Acid Levels. Associations between serum PFOA levels and
serum uric acid levels have been found in several occupational, community, and general population
studies. Costa et al. (2009) and Sakr et al. (2007b) reported associations between serum PFOA levels and
serum uric acid levels in workers with high serum PFOA levels. In adult participants of the C8 Health
Project, positive linear trends between serum uric acid levels and serum PFOA levels were found
(Steenland et al. 2010b). When the subjects were categorized by PFOA levels, significantly increased

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

212
2. HEALTH EFFECTS

risks of hyperuricemia (>6.0 mg/dL for women, >6.8 mg/dL for men) were observed for subjects with
serum PFOA levels in the 2nd, 3rd, 4th, and 5th quintiles (≥11.5 ng/mL). Four studies utilizing NHANES
data have found associations between serum PFOA and serum uric acid levels in adults (Gleason et al.
2015; Shankar et al. 2011b) and adolescents (Geiger et al. 2013; Kataria et al. 2015). A study in
Taiwanese adolescents also found this association between PFOA and uric acid (Qin et al. 2016). Several
studies have also found increases in the risk of hyperuricemia in a highly exposed population (Steenland
et al. 2010b) and the general population (Gleason et al. 2015; Geiger et al. 2013; Qin et al. 2016; Shankar
et al. 2011b). The ORs for the risk of hyperuricemia in these studies are summarized in Figure 2-17.

Laboratory Animal Studies. No gross or microscopic alterations were observed in the kidneys from male
rats following head-only inhalation exposure to up to 84 mg/m3 APFO dusts for 2 weeks (Kennedy et al.
1986). Significantly elevated absolute and relative kidney weight was reported in male rats dosed with
≥3 mg/kg/day PFOA by gavage in water for 70 days (Butenhoff et al. 2004b), but histological evaluation
of the kidney was not conducted in this study. Rats that received much higher doses (100–
110 mg/kg/day) of APFO for 90 days in the diet showed no significant morphological alterations in the
kidneys, and BUN and the urinalysis were unremarkable (Griffith and Long 1980). Also, male mice
dosed with up to 47 mg/kg/day APFO in the drinking water for 21 days showed no morphological
alterations in the kidneys, and BUN and serum creatinine levels were not significantly affected (Son et al.
2008). Treatment of Cynomolgus monkeys with daily doses of up to 20 mg/kg/day APFO, administered
via a capsule, for 26 weeks (Butenhoff et al. 2002) or Rhesus monkeys dosed with up to 10 mg/kg/day by
gavage for 90 days (Griffith and Long 1980) did not cause morphological alterations in the kidneys, and
blood chemistries and urinalyses provided no evidence of alterations in kidney function. In a 2-year
dietary study in rats, relative kidney weight from males dosed with 15 mg/kg/day APFO was significantly
elevated (14%) at the 1-year interim evaluation relative to controls, but gross and microscopic appearance
(at 1 year and at termination), BUN, and urinalyses (several times during the study) were not significantly
affected (3M 1983). No gross or microscopic alterations were seen in the kidneys from rats that received
dermal applications of up to 2,000 mg/kg/day APFO to the shaven skin for 2 weeks (Kennedy 1985).

Summary. Epidemiology studies have examined possible associations between exposure to PFOA and
increases in the risk of kidney disease and alterations in renal function. Mixed results for associations
between serum PFOA and risks of kidney disease have been reported in occupational exposure studies
and studies of highly exposed residents with more studies not finding associations. Several general
population and community studies have found inverse associations between serum PFOA and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

213
2. HEALTH EFFECTS

Figure 2-17. Risk of Hyperuricemia Relative to PFOA Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

214
2. HEALTH EFFECTS

glomerular filtration rate; however, there is suggestive evidence that this association may be due to
reverse causation rather than a direct effect. Associations between serum PFOA levels and serum uric
acid levels have been consistently observed in occupational, community, and general populations.
Laboratory animal studies have not found evidence of alterations in renal function or histological
alterations.

PFOS

Epidemiology Studies—Biomarkers of Renal Function. Three studies have found inverse associations
between serum PFOS levels and glomerular filtration rate in adults (Shankar et al. 2011a), adolescents
(Kataria et al. 2015), and children (Watkins et al. 2013). In the Watkins et al. (2013) study of C8 Health
Project participants, a concentration-related linear trend between decreasing estimated glomerular
filtration rates and increases in serum PFOS levels was observed in children and adolescents 1–<18 years
old. In adolescents 12–19 years of age participating in NHANES, the estimated glomerular filtration rate
was lower in participants with serum PFOA levels in the 2nd, 3rd, and 4th quartiles than those with levels in
the 1st quartile (Kataria et al. 2015). In addition to the inverse association between serum PFOS and
estimated glomerular filtration rate observed in adult NHANES participants, Shankar et al. (2011a) also
found increased risks of chronic kidney disease (defined as a glomerular filtration rate of
<60 mL/minute/1.73 m2) in participants with serum PFOS levels in the 4th quartile.

Epidemiology Studies—Alterations in Uric Acid Levels. In a study of C8 Health Project participants, a
linear trend between serum uric acid levels and serum PFOS levels was found (Steenland et al. 2010b).
When the subjects were categorized by serum PFOS levels, increased risks of hyperuricemia (>6.0 mg/dL
for women, >6.8 mg/dL for men) were observed for subjects with serum PFOS levels in the 3rd, 4th, and
5th quintiles. Similar findings were found in NHANES adult participants (Shankar et al. 2011b). Two
studies of adolescent NHANES participants also found associations between serum PFOS and serum uric
acid levels (Gleason et al. 2015; Kataria et al. 2015); a third study did not find an association (Geiger et
al. 2013). The Geiger et al. (2013) study did find an increased risk of hyperuricemia for adolescents with
serum PFOS levels in the 4th quartile; this was not found in the Gleason et al. (2015) study. A study of
Taiwanese adolescents did not find associations between serum PFOS and uric acid or an increased risk
of hyperuricemia (Qin et al. 2016). The ORs for the risk of hyperuricemia in these studies are
summarized in Figure 2-18.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

215
2. HEALTH EFFECTS

Figure 2-18. Risk of Hyperuricemia Relative to PFOS Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

216
2. HEALTH EFFECTS

Laboratory Animal Studies. No significant morphological alterations or clinical evidence of impaired
kidney function was reported in male and female rats dosed with up to 1.77 mg/kg/day PFOS (potassium
salt) (Seacat et al. 2003) or 5.89 mg/kg/day (Curran et al. 2008) for 4 weeks. Extending the treatment to
14 weeks resulted in an increase in BUN in male (23% increase) and female rats (41% increase), but
histopathology of the kidneys and urinalyses were unremarkable (Seacat et al. 2003). The NOAEL values
were 0.34 and 0.4 mg/kg/day in males and females, respectively. Treatment of Cynomolgus monkeys
with up to 0.75 mg/kg/day PFOS (potassium salt) administered via a capsule for 26 weeks did not cause
morphological alterations in the kidneys, nor did it affect BUN, serum creatinine, or urinary parameters
(Seacat et al. 2002). Similar results were reported in a 4-week study in monkeys dosed with up to
2 mg/kg/day PFOS (Thomford 2002a). A mild increase in BUN was reported in rats treated with
approximately 0.25 or 1.04 mg/kg/day PFOS in the diet for 53 weeks in a 2-year study (Butenhoff et al.
2012b; Thomford 2002b). However, there were no significant gross or microscopic alterations in the
kidneys at week 53 or at termination.

PFHxS

Epidemiology Studies—Biomarkers of Renal Function. A small number of epidemiology studies have
evaluated biomarkers of renal function. In a study of C8 Health Project participants, an inverse
association between serum PFHxS and estimated glomerular filtration rate was observed (Watkins et al.
2013). A study of adolescent participants in NHANES did not find this association (Kataria et al. 2015).
A comparison between the reported median PFHxS level in the Watkins et al. (2013) study (5.2 ng/mL)
and the lower end of the 4th quartile serum PFHxS level in the Kataria et al. (2015) study (≥4 ng/mL)
suggests that the serum PFHxS levels were higher in the Watkins et al. (2013) study.

Epidemiology Studies—Alterations in Uric Acid Levels. In adolescent NHANES participants, no
associations between serum PFHxS levels and serum uric acid levels (Gleason et al. 2015; Kataria et al.
2015) or risk of hyperuricemia (Gleason et al. 2015) were found. A study of Taiwanese adolescents
found an association between serum PFHxS levels and serum uric acid levels, but did not find increased
risks of hyperuricemia (Qin et al. 2016).

Laboratory Animal Studies. Male rats treated by gavage with 10 mg/kg/day PFHxS for at least 42 days
showed a significant increase in BUN levels, but there were no significant gross or microscopic
alterations in the kidneys (Butenhoff et al. 2009a; Hoberman and York 2003); the NOAEL was
3 mg/kg/day. No significant effect on BUN was reported in female rats.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

217
2. HEALTH EFFECTS

PFNA

Epidemiology Studies—Biomarkers of Renal Function. Two epidemiology studies have evaluated the
possible associations between serum PFNA and alterations in renal function biomarkers. In a study of
children participating in the C8 Health Project, an inverse association between serum PFNA and
estimated glomerular filtration rate was observed, but not in adolescents participating in NHANES
(Watkins et al. 2013). Mundt et al. (2007) noted that there were small, but not clinically significant,
alterations in BUN, creatinine, and serum uric acid levels in workers exposed to PFNA.

Epidemiology Studies—Alterations in Uric Acid Levels. Gleason et al. (2015) found an association
between serum PFNA and serum uric acid levels in adolescents; this association was not found in other
studies of adolescents (Kataria et al. 2015; Qin et al. 2016). Studies by Gleason et al. (2015) and Qin et
al. (2016) did not find increases in the risk of hyperuricemia associated with serum PFNA levels.

PFDeA

Epidemiology Studies—Alterations in Uric Acid Levels. Epidemiology studies examining renal
outcomes are limited to a study of Taiwanese adolescents that found no association between serum
PFDeA levels and serum uric acid levels and did not find increased risks of hyperuricemia (Qin et al.
2016).

Laboratory Animal Studies. Administration of a single dose of up to 80 mg/kg PFDeA to female
C57BL/6N mice by gavage did not induce gross or microscopic changes in the kidneys (Harris et al.
1989). However, 2 out of 10 mice that died following administration of a dose of 320 mg/kg showed
mild acute necrosis of the proximal convoluted tubules.
PFUA

Laboratory Animal Studies. Treatment of male and female rats with 1.0 mg/kg/day PFUA via gavage for
41–46 days resulted in significant increases in BUN levels (35–61% in males, 19–45% in females) and
alkaline phosphatase activity (86–140% in males, 83% in females) and significant decreases in total
protein (11% in males, 10–13% in females) and albumin (7% in males) levels (Takahashi et al. 2014); the
NOAEL was 0.3 mg/kg/day.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

218
2. HEALTH EFFECTS

PFBuS

Epidemiology Studies—Alterations in Uric Acid Levels. Serum PFBuS levels were not associated with
serum uric acid levels or increases in the risk of hyperuricemia in a study of adolescents in Taiwan (Qin et
al. 2016).

Laboratory Animal Studies. Treatment of female rats with 900 mg/kg/day PFBuS by gavage for 28 days
caused a significant increase (9–11%) in absolute and relative kidney weight, but caused no significant
alterations in the microscopic appearance of the kidneys (3M 2001). The weight of the kidneys returned
to control levels following a recovery period of approximately 14 days; the NOAEL for kidney weight
effects was 300 mg/kg/day PFBuS. In a 90-day rat study, PFBuS did not result in alterations in kidney
weights, but did result in hyperplasia of the medullary and papillary tubular and ductal epithelial cells in
the inner medullary region at 600 mg/kg/day, but not at 200 mg/kg/day (Lieder et al. 2009a). Minimal to
moderate papillary epithelial tubular/acinar hyperplasia was also observed in a 2-generation rat study at
300 mg/kg/day; the study identified a NOAEL of 100 mg/kg/day (Lieder et al. 2009b).
PFBA

Laboratory Animal Studies. No alterations in renal morphology or clinical indications of impaired renal
function were reported in rats treated with PFBA in doses of up to 184 mg/kg/day for 5 days (3M 2007a),
150 mg/kg/day for 28 days (Butenhoff et al. 2012a; van Otterdijk 2007a), or 30 mg/kg/day by gavage for
90 days (Butenhoff et al. 2012a; van Otterdijk 2007b).

PFDoA

Epidemiology Studies—Alterations in Uric Acid Levels. In adolescents, no associations between serum
PFDoA levels and serum uric acid levels or the risk of hyperuricemia were observed (Qin et al. 2016).
PFHxA

Epidemiology Studies—Alterations in Uric Acid Levels. In adolescents, no associations between serum
PFHxA levels and serum uric acid levels or the risk of hyperuricemia were observed (Qin et al. 2016).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

219
2. HEALTH EFFECTS

Laboratory Animal Studies. In a 2-year gavage study, treatment of female rats with 200 mg/kg/day
PFHxA resulted in mild renal tubular degeneration and mild to severe papillary necrosis (Klaunig et al.
2015); the NOAEL was 100 mg/kg/day. In addition, urinalysis revealed an increased mean urine volume
and reduced specific gravity. There were no histological alternations in the kidneys of males.

2.11 DERMAL
Overview. No studies were located regarding dermal effects in humans. Studies in laboratory animals
have not found dermal effects following head-only inhalation exposure to PFOA (see Table 2-1) or oral
exposure to PFOA, PFOS, or PFBA (see Tables 2-3, 2-4, and 2-5). Dermal exposure to PFOA has
resulted in skin damage (see Table 2-6).
PFOA

In an inhalation head-only exposure study, no histopathological alterations were observed in the
abdominal skin of male rats exposed to ≤84 mg/m3 APFO dusts for 2 weeks (Kennedy et al. 1986).
No microscopic alterations were observed in the skin following oral exposure of rats to ≤100–
110 mg/kg/day APFO via the diet for 90 days (Griffith and Long 1980) or monkeys exposed to up to
20 mg/kg/day PFOA or 0.75 mg/kg/day PFOS for 26 weeks (Butenhoff et al. 2002; Seacat et al. 2002).

Application of a single dose of 5,000 mg/kg of an aqueous paste of APFO to a clipped area of the skin of
rats, and left in place covered for 24 hours produced mild skin irritation (Kennedy 1985); no irritation was
apparent with a dose of 3,000 mg/kg. In a 2-week dermal exposure study, skin irritation was observed in
rats exposed to 200 mg/kg/day (Kennedy 1985). Acute necrotizing dermatitis was observed in two out of
five rats exposed to 2,000 mg/kg/day; this lesion was observed after the 10th treatment. Application of
500 mg/kg APFO to the intact or abraded skin of young rabbits and left covered for 24 hours was nonirritating, as scored according to the Draize procedure immediately after removal of the cover and
48 hours later (Griffith and Long 1980).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

220
2. HEALTH EFFECTS

PFOS

Administration of up to approximately 1.04 mg/kg/day PFOS to rats in the diet for 2 years did not induce
morphological alterations in the skin (Butenhoff et al. 2012b; Thomford 2002b).

PFBA

There were no significant gross or microscopic alterations in the skin of rats receiving gavage doses of
≤150 mg/kg/day PFBA for 28 days or ≤30 mg/kg/day PFBA for 90 days (Butenhoff et al. 2012a; van
Otterdijk 2007a, 2007b).
2.12 OCULAR
Overview. No information was located regarding ocular effects in humans. Ocular irritation has been
observed in laboratory animals following exposure to airborne APFO dust or instillation of PFOA into the
eye (see Tables 2-1 and 2-6). However, ocular effects have not been found following oral exposure to
PFOA, PFOS, PFBuS, or PFBA (see Tables 2-3, 2-4, and 2-5).
PFOA
Rats exposed to 18,600 mg/m3 APFO dusts for 1 hour exhibited a red material around the eyes and
lacrimation during exposure (Griffith and Long 1980). Male rats exposed to ≥810 mg/m3 APFO dusts for
4 hours showed corneal opacity and corrosion, which was confirmed by fluorescein staining (Kennedy et
al. 1986). Examination of the eyes of male rats exposed intermittently to up to 84 mg/m3 APFO for
2 weeks using a bright light and a slit-lamp biomicroscope on days 5 and 9 of exposure did not reveal any
significant exposure-related alterations (Kennedy et al. 1986). Microscopic examination of the eyes from
these rats at termination and following a recovery period of up to 42 days was unremarkable.

In oral exposure studies, examination of the eyes from rats exposed to approximately 100–110 mg/kg/day
APFO in the diet for 90 days did not reveal any significant gross or microscopic alterations (Griffith and
Long 1980). Similar results were reported in rats that received dietary doses up to 15 mg/kg/day APFO
for 2 years (3M 1983) and in monkeys dosed with up to 20 mg/kg/day APFO for 26 weeks (Butenhoff et
al. 2002).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

221
2. HEALTH EFFECTS

No significant gross alterations were observed in the eyes of rats following repeated dermal exposure to
APFO (Kennedy 1985). Microscopic examination of the eyes also did not reveal treatment-related
changes. In a study in rabbits, 0.1 g APFO was instilled once in the conjunctival sac of the right eye and
examinations were conducted after 1, 24, 48, and 72 hours and 5 and 7 days after the application (Griffith
and Long 1980). APFO produced moderate irritation of the eye characterized by iridal and conjunctival
effects. The effects were most pronounced 1 hour after instillation. The irritation was persistent, but by
day 7, it had subsided. In a different experiment in which 0.1 g APFO was instilled for 5 or 30 seconds
before washing with 200 mL of water, there was limited conjunctival irritation, but the effects were
immediate and persistent.
PFOS
No gross or microscopic alterations were observed in the eyes from rats exposed to ≤1.77 mg/kg/day
PFOS in the diet for 4 weeks or ≤1.56 mg/kg/day for 14 weeks (Seacat et al. 2003). Similar findings were
reported in monkeys dosed daily with up to 2 mg/kg/day PFOS administered via a capsule for 4 weeks
(Thomford 2002a) or up to 0.75 mg/kg/day PFOS administered via a capsule for 26 weeks (Seacat et al.
2002), and in rats dosed with up to 1.04 mg/kg/day in the diet for 2 years (Butenhoff et al. 2012b;
Thomford 2002b).
PFBuS
No gross and microscopic alterations were observed in the eyes of rats administered ≤900 mg/kg/day
PFBuS via gavage for 28 days (3M 2001).

PFBA
Examination of the eyes of rats orally exposed to ≤150 mg/kg/day PFBA for 28 days or ≤30 mg/kg/day
for 90 days did not reveal any significant alterations in the eyes (Butenhoff et al. 2012a; van Otterdijk
2007a, 2007b).
2.13 ENDOCRINE
Overview. Epidemiology studies have examined a number of endocrine targets including thyroid gland
and hormones, reproductive hormones, and insulin levels. A discussion of the thyroid effects is included

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

222
2. HEALTH EFFECTS

in this section; the reproductive hormone effects are discussed in Section 2.16, Reproductive, and the
insulin effects (as well as other effects associated with glucose metabolism and utilization) are discussed
in Section 2.18, Other Noncancer. Summaries of results of epidemiology studies evaluating thyroid
outcomes are presented in Table 2-15; more in-depth summaries of the studies are presented in the
Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 9. The results of
community and general population studies suggest an association between serum PFOA and an increased
risk of thyroid disease. There is also limited support of an association between serum PFOS and thyroid
disease. Although some associations between serum PFOA or PFOS and thyroid stimulating hormone
(TSH), triiodothyronine (T3), or thyroxine (T4) levels have been found, the results are not consistent
across studies. A small number of studies have evaluated PFHxS, PFNA, and PFDeA and most studies
have not found consistent associations between serum perfluoroalkyl levels and thyroid hormone levels.

Laboratory animal studies have primarily evaluated potential morphological alterations in endocrine
tissues following oral exposure; these studies are summarized in Tables 2-3, 2-4, and 2-5. Some
alterations in thyroid hormone levels have been observed in laboratory animals exposed to PFOA, PFOS,
or PFDeA. Histopathological alterations have been observed in the thyroid of some laboratory animal
studies for PFHxS and PFBA; the investigators noted that these effects were likely secondary to the
hepatocellular hypertrophy. In general, the pituitary, parathyroid, thyroid, and adrenal glands do not
appear to be sensitive targets following exposure to PFOA, PFOS, or PFBuS.
PFOA

Epidemiology Studies. A number of epidemiology studies have examined the potential of PFOA to
damage the thyroid. Steenland et al. (2015) did not find an association between serum PFOA and the risk
of thyroid disease in male or female workers at the Washington Works facility. The occupational
exposure studies do not suggest an association between serum PFOA and alterations in thyroid hormone
levels. One study (Olsen and Zobel 2007) reported associations between serum PFOA levels and free T4
and T3 levels in workers at 3M facilities; it is noted that the investigators did not consider the results
clinically relevant since the levels were within the normal range. Another study reported an association
between serum PFOA and TSH, but this was only observed at one time point (Olsen et al. 1998b). A
fourth occupational study reported that TSH, T4, and T3 levels were within the reference range (Sakr et
al. 2007b).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

223
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

10,000–<30,000 ng/mL
(PFOA range)

TSH

NS (p=0.09 for trend), 1993 group
Association (p=0.002), 1995 group

TSH

The investigators noted that the levels
were within the reference range

PFOA
Olsen et al. 1998b

Occupational (n=111 in 1993 and n=80 in 1995)
Sakr et al. 2007b
428 ng/mL (mean PFOA)
Occupational (n=1,025)
Steenland et al. 2015

Cumulative PFOA

T4
T3
Thyroid disease risk

Occupational (n=3,713)

Olsen and Zobel 2007

NS (p=0.98 for trend) no lag, males
NS (p=0.55 for trend) 10-year lag,
males
NS (p=0.97 for trend) no lag, females
NS (p=0.27 for trend) 10-year lag,
females
Association (p=0.01)*
NS (p=0.29)
Association (p=0.05)*
NS (p=0.08)

2,210 ng/mL (mean PFOA)

Free T4
T4
T3
TSH

Anderson-Mahoney et al. 2008

NR

Self-reported thyroid
problems

SPR 1.56 (1.22–1.98)*

Community (n=566)
Emmett et al. 2006b

354 ng/mL (median PFOA)

TSH

NS (p>0.05)

Occupational (n=552)

Community (n=371)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

224
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Knox et al. 2011a

52.6, 91.0, 98.6, and
T4
124.3 ng/mL (mean PFOA in
women ≤50 year, men
≤50 years, women
>50 years, men >50 years) T3 uptake

Association (p≤0.0001)*, women
≤50 years
Association (p<0.001), men and
women >50 years
Inverse association (p=0.0001)*
women ≤50 years
Inverse association (p=0.005)*,
women >50 years
Inverse association (p=0.037)*, men
>50 years
OR 1.44 (1.02–2.03)*, per
interquartile shift
OR 1.54 (1.00–2.37)*, per
interquartile shift
OR 0.98 (0.86–1.15), per interquartile
shift
OR 0.81 (0.58–1.15), per interquartile
shift
β -1.1 (-5.3–3.4), 4th quartile
β -0.1 (-1.7–1.4), 4th quartile

Community (n=50,113 adults ≥20 years of age)

Lopez-Espinosa et al. 2012
Community (n=10,725 children aged 1–
17 years)

29.3 and 67.7–2,071 ng/mL
(median and 4th quartile
PFOA)

Thyroid disease
Hypothyroidism
Subclinical
hypothyroidism
Subclinical
hyperthyroidism
TSH
Total T4

Winquist and Steenland 2014b
Community (C8 and occupational) (n=28,541)

114.7–<202.2 ng/mL-year
(2nd quintile cumulative
PFOA)

Functional thyroid
disease

***DRAFT FOR PUBLIC COMMENT***

HR 1.24 (1.02–1.51;p=0.031)*
(women), retrospective analysis
HR 1.01 (0.94–1.07 per log linear
increase in PFOA, p=0.853) (men),
retrospective analysis
NS (p=0.549) (women), prospective
analysis
NS (p=0.087) (men), prospective
analysis

 PERFLUOROALKYLS

225
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
Hyperthyroidism

Hypothyroidism

Bloom et al. 2010
General population (n=31)
Chan et al. 2011
General population (n=94 women with
hypothryoxinemia and 175 matched controls)
Jain 2013

1.33 ng/mL (geometric mean TSH
PFOA)
Free T4

General population (n=633)

NS (p=0.074) (women), retrospective
analysis
NS (p=0.858) (men), retrospective
analysis
NS (p=0.268) (women), prospective
analysis
NS (p=0.760) (men), prospective
analysis
NS (p=0.076) (women), retrospective
analysis
NS (p=0.684) (men), retrospective
analysis
NS (p=0.247) (women), prospective
analysis
HR 1.24 (1.03–1.49)* (men),
prospective analysis
NS (p=0.871)
NS (p=0.896)

1.28 and 1.37 ng/mL
(geometric mean PFOA in
cases and controls)

Hypothyroxinemia risk

OR 0.94 (0.74–1.18)

NR

Total T3
TSH
Free T3
Free T4
Total T4

Association (p=0.013)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Thyroglobulin
TSH
T4

NS (p>0.05)
NS (p=0.4055)
NS (p=0.2221)

General population (NHANES) (n=1,525)

Ji et al. 2012

Resultc

2.74 ng/mL (median PFOA)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

226
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Lewis et al. 2015

1.42–2.55 ng/mL (range of
median PFOA for different
age groups)

Association (p<0.05)*, 12–20-yearold females
Association (p<0.05)*, 20–<40-yearold females
NS (p>0.05)

General population (NHANES) (n=1,682)

TSH
Free T4
Total T4

Melzer et al. 2010
General population (NHANES) (n=3,966)
Raymer et al. 2012
General population (n=256)
Shrestha et al. 2015
General population (n=87 with thyroid disease)

Wang et al. 2013a
General population (n=903 pregnant women)
Wang et al. 2014
General population (n=285 pregnant women)

9.47 and 10.39 ng/mL
(4th PFOA quartile mean in
women and men)
10.4 ng/mL (mean PFOA)

Free T3

Association (p<0.05)*, 60–80-yearold females

Total T3

Association (p<0.05)*, 60–80-yearold females

Thyroid disease risk

OR 1.64 (1.09–2.46)*, females
OR 1.58 (0.74–3.39), males

TSH

NS (p>0.05)

T3
T4
9.17 ng/mL (geometric mean TSH
PFOA)
Free T4
T4
T3
2.13 ng/mL (geometric mean TSH
PFOA)
Elevated TSH risk

NS (p>0.05)
NS (p>0.05)
NS (p=0.176)
NS (p=0.536)
NS (p=0.097)
NS (p=0.208)
NS (p>0.05)

2.39 ng/mL (median PFOA)

TSH
Free T4

NS (p>0.05)
NS (p>0.05)

Total T4
Total T3

NS (p>0.05)
NS (p>0.05)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)

 PERFLUOROALKYLS

227
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Webster et al. 2016

4.2 ng/mL (geometric mean
PFOA)

NS (p>0.05)

General population (NHANES) (n=1,525)

Wen et al. 2013
General population (NHANES) (n=1,181)

TSH

Free T4
Total T4
Free T3
Total T3
4.15 ng/mL (geometric mean TSH
PFOA)
Total T4

Subclinical
hypothyroidism risk
Subclinical
hyperthyroidism risk
TSH
Free T4
Total T4

NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*
NS (p<0.05)
NS (p=0.916), men
NS (p=0.732), women
NS (p=1.0), men
NS (p=0.705), women
NS (p=0.673), men
Association (p=0.035)*, women
NS (p=0.226), men
NS (p=0.341), women
OR1.29 (0.40–4.10), men
OR 7.42 (1.14–48.12, p<0.05), women
OR 0.38 (0.16–0.95, p<0.05)*, men
OR 0.99 (0.13–7.59), women
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Free T3
Total T3

NS (p>0.05)
NS (p>0.05)

1,480–2,440 ng/mL (range of TSH
mean PFOS)
Cortisol

NS (p=0.95)
NS (p=0.45)

1,320 and 800 ng/mL (mean T3
PFOS)

Association (p=0.04)*

Total T3
Thyroglobulin

Yang et al. 2016a
General population (n=157 pregnant women)

PFOS
Olsen et al. 1998a
Occupational (n=327)
Olsen et al. 2003a

1.95 ng/mL (mean PFOA)

Occupational (n=518)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

228
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Knox et al. 2011a

17.3, 24.8, 25.7, and
29.1 ng/mL (mean PFOA in
women ≤50 year, men
≤50 years, women >50
years, men >50 years)

T4

20.0 ng/mL (median PFOS)

Thyroid disease

Subclinical
hypothyroidism
Subclinical
hyperthyroidism
TSH
Total T4
TSH

Association (p<0.0001)*, women ≤50
or >50 years
Association (p=0.0001)*, men ≤50 or
>50 years
Inverse association (p<0.0001)*
women ≤50 years
Inverse association (p=0.0001)*,
women >50 years
Inverse association (p=0.009)*, men
≤50 years
Inverse association (p=0.0001)*,
men >50 years
OR 0.8 (0.62–1.08), per interquartile
shift
OR 0.91 (0.63–1.31), per interquartile
shift
OR 0.99 (0.86–1.13), per interquartile
shift
OR 0.80 (0.62–1.02), per interquartile
shift
β -1.0 (-0.3–2.3), per interquartile shift
β 1.1 (0.6–1.5), per interquartile shift
Association (p=0.03)*, 3rd quartile

19.57 ng/mL (geometric
mean PFOS)

TSH
Free T4

NS (p=0.896)
NS (p=0.623)

7.59 and 7.08 ng/mL
(geometric mean PFOS in
cases and controls)

Hypothyroxinemia risk

OR 0.88 (0.63–1.24)

Community (n=50,113 adults ≥20 years of age)

Lopez-Espinosa et al. 2012
Community (n=10,725 children aged 1–
17 years)

Berg et al. 2015
General population (n=391)
Bloom et al. 2010
General population (n=31)
Chan et al. 2011
General population (n=94 women with
hypothyroxinemia and 175 matched controls)

T3 uptake

Hypothyroidism

8.1–11.0 ng/mL (3rd PFOS
quartile)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

229
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Dallaire et al. 2009

18.28 ng/mL (geometric
mean PFOS)

TSH

Inverse association (p≤0.05)*

T3
T4-binding globulin
Free T4
TSH
Free T3
Total T3

Inverse association (p≤0.05)*
Inverse association (p≤0.01)*
Association (p≤0.05)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Free T4
Total T4
Thyroglobulin
TSH
T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p=0.3537)
NS (p=0.1134)

TSH
Free T4

NS (p>0.05)
Association (p<0.05)*, 20–<40-yearold females
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
OR 1.15 (0.7–1.91, p=0.568), females
OR 1.58 (0.72–3.47, p=0.251), males
OR 2.68 (1.03–6.98, p=0.043)*, males
4th quartile versus combined 1st and
2nd quartiles
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=623)

Jain 2013

NR

General population (NHANES) (n=1,525)

Ji et al. 2012
General population (n=633)
Lewis et al. 2015
General population (NHANES) (n=1,682)

Melzer et al. 2010

7.96 ng/mL (median PFOS)

3.76–11.1 ng/mL (range of
median PFOS for different
age groups)

General population (NHANES) (n=3,966)

57.73 and 50.96 ng/mL
(4th PFOS quartile mean in
women and men)

Raymer et al. 2012

37.4 ng/mL (mean PFOS)

General population (n=256)

Total T4
Free T3
Total T3
Thyroid disease risk

TSH
T3
T4

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

230
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Shrestha et al. 2015

31.6 ng/mL (geometric mean TSH
PFOS)
Free T4
T4
T3
12.77 ng/mL (geometric
TSH
mean PFOS)
Elevated TSH risk

NS (p=0.094)

12.73 ng/mL (median PFOS) TSH
Free T4
Total T4
Total T3

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

13.9 ng/mL (geometric mean TSH
PFOS)
Free T4
Total T4
Free T3
Total T3
14.2 ng/mL (geometric mean TSH
PFOS)
Total T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p<0.05)
NS (p=0.931), men
NS (p=0.358), women
NS (p=0.840), men
NS (p=0.433), women
NS (p=0.404), men
NS (p=0.384), women
NS (p=0.342), men
NS (p=0.061), women
OR 1.98 (1.19–3.28, p<0.05)*, men
OR 3.03 (1.14–8.07, p<0.05)*, women
OR 0.92 (0.19–4.46), men
OR 1.90 (0.33–6.80), women

General population (n=87 with thyroid disease)

Wang et al. 2013a
General population (903 pregnant women)
Wang et al. 2014
General population (285 pregnant women)

Webster et al. 2016
General population (NHANES) (n=1,525)

Wen et al. 2013
General population (NHANES) (n=1,181)

Total T3
Thyroglobulin
Subclinical
hypothyroidism risk
Subclinical
hyperthyroidism risk

***DRAFT FOR PUBLIC COMMENT***

Association (p=0.044)*
Association (p=0.001)*
NS (p=0.287)
Association (p=0.03)*
NS (p>0.05)

 PERFLUOROALKYLS

231
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Yang et al. 2016a

5.08 ng/mL (mean PFOS)

TSH

Inverse association (p<0.01)*

Free T4
Total T4
Free T3
Total T3

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=157 pregnant women)

PFHxS
Bloom et al. 2010
General population (n=31)
Chan et al. 2011
General population (n=94 women with
hypothyroxinemia and 175 matched controls)
Jain 2013

0.75 ng/mL (geometric mean TSH
PFHxS)
Free T4

NS (p=0.956)

1.28 and 1.37 ng/mL
(geometric mean PFHxS in
cases and controls)

Hypothyroxinemia risk

OR 1.12 (0.89–1.41)

NR

TSH
Free T3
Total T3
Free T4
Total T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (NHANES) (n=1,525)

Ji et al. 2012
General population (n=633)
Lewis et al. 2015
General population (NHANES) (n=1,682)

Wang et al. 2013a
General population (n=903 pregnant women)

NS (p=0.567)

Thyroglobulin
1.51 ng/mL (median PFHxS) TSH
T4

NS (p>0.05)
NS (p=0.8144)
NS (p=0.5147)

0.69–1.81 ng/mL (range of
median PFHxS for different
age groups)

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

TSH
Free T4
Total T4
Free T3
Total T3
0.62 ng/mL (geometric mean TSH
PFHxS)
Elevated TSH risk

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)

 PERFLUOROALKYLS

232
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Wang et al. 2014

0.81 ng/mL (median PFHxS) TSH

Association (p<0.05)

General population (n=285 pregnant women)

Webster et al. 2016

1.9 ng/mL (geometric mean
PFHxS)

General population (NHANES) (n=1,525)

Wen et al. 2013

2.0 ng/mL (geometric mean
PFHxS)

General population (NHANES) (n=1,181)

Free T4
Total T4
Total T3
TSH
Free T4
Total T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Free T3
Total T3
TSH

NS (p>0.05)
NS (p>0.05)
NS (p=0.608), men
NS (p=0.720), women
NS (p=0.641), men
Association (p=0.022)*, women
NS (p=0.917), men
Association (p<0.001)*, women
NS (p=0.455), men
NS (p=0.725), women
OR 1.57 (0.76–3.25), men
OR 3.10 (1.22–7.86, p<0.05)*, women
OR 0.56 (0.24–1.20.92), men
OR 12.27 (1.07–4.80.90)*,women
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Total T4
Total T3
Thyroglobulin

Yang et al. 2016a
General population (n=157 pregnant women)

0.63 ng/mL (mean PFHxS)

Subclinical
hypothyroidism risk
Subclinical
hyperthyroidism risk
TSH
Free T4
Total T4
Free T3
Total T3

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

233
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

NR

Investigators noted differences
between groups was small and not
clinically relevant

PFNA
Mundt et al. 2007
Occupational (n=592)
Lopez-Espinosa et al. 2012

1.5 ng/mL (median PFNA)

Community (n=10,725 children aged 1–
17 years)

TSH
T4
T3
Thyroid disease
Hypothyroidism
Subclinical
hypothyroidism
Subclinical
hyperthyroidism
TSH

Bloom et al. 2010
General population (n=31)
Jain 2013

Total T4
0.79 ng/mL (geometric mean TSH
PFNA)
Free T4
NR

General population (NHANES) (n=1,525)

Ji et al. 2012
General population (n=633)
Wang et al. 2013a
General population (n=903 pregnant women)

2.09 ng/mL (median PFNA)

OR 1.05 (0.78–1.41), per interquartile
shift
OR 1.11 (0.77–1.60), per interquartile
shift
OR 0.99 (0.88–1.12), per interquartile
shift
OR 0.78 (0.61–1.01), per interquartile
shift
β -1.1 (CI 0.7–1.5), per interquartile
shift
Β 0.8 (-0.4–2.0), per interquartile shift
NS (p=0.789)
NS (p=0.424)

TSH
Free T3
Total T3
Free T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Total T4
Thyroglobulin
TSH
T4

NS (p>0.05)
NS (p>0.05)
NS (p=0.1354)
NS (p=0.7436)

0.37 ng/mL (geometric mean TSH
PFNA)
Elevated TSH risk

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

234
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Wang et al. 2014

1.51 ng/mL (median PFNA)

TSH

NS (p>0.05)

Free T4
Total T4
Total T3
TSH
Free T4
Total T4

Inverse association (p<0.001)*
Inverse association (p<0.001)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=285 pregnant women)

Webster et al. 2016

1.5 ng/mL (geometric mean
PFNS)

General population (NHANES) (n=1,525)

Wen et al. 2013
General population (NHANES) (n=1,181)

Free T3
Total T3
1.54 ng/mL (geometric mean TSH
PFNA)
Total T4
Total T3
Thyroglobulin

Yang et al. 2016a
General population (n=157 pregnant women)

PFDeA
Berg et al. 2015

0.52 ng/mL (mean)

Subclinical
hypothyroidism risk
Subclinical
hyperthyroidism risk
TSH
Free T4
Total T4
Free T3
Total T3

0.31–2.34 ng/mL (4th PFDeA T3
quartile)

General population (n=391 pregnant women)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p=0.973), men
NS (p=0.407), women
NS (p=0.097), men
NS (p=0.632), women
NS (p=0.063), men
NS (p=0.258), women
NS (p=0.537), men
NS (p=0.395), women
OR 1.30 (0.65–2.60), men
OR 2.54 (0.40–16.05), women
OR 2.41 (0.48–12.04), men
OR 1.91 (0.83–4.38), women
Inverse association (p<0.05)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p=0.03) (4th quartile)

 PERFLUOROALKYLS

235
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Bloom et al. 2010

0.21 ng/mL (geometric mean TSH
PFDeA)
Free T4

NS (p=0.365)

NR

NS (p>0.05)
NS (p>0.05)

General population (n=31)
Jain 2013
General population (NHANES) (n=1,525)

Ji et al. 2012
General population (n=633)
Wang et al. 2013a
General population (903 pregnant women)
Wang et al. 2014
General population (285 pregnant women)

Yang et al. 2016a

TSH
Free T3

Total T3
Free T4
Total T4
Thyroglobulin
0.91 ng/mL (median PFDeA) TSH
T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p=0.2721)
NS (p=0.2176)

0.09 ng/mL (geometric mean TSH
PFDeA)
Elevated TSH risk

NS (p>0.05)
NS (p>0.05)

0.46 ng/mL (median PFDeA) TSH
Free T4
Total T4
Total T3
0.45 ng/mL (mean PFDeA) TSH

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.01)*
Inverse association (p<0.01)*

General population (n=157 pregnant women)

PFUA
Berg et al. 2015

NS (p=0.107)

0.4–0.96 ng/mL (4th PFUA
quartile)

Free T4
Total T4
Free T3
Total T3

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Free T3

Association (p=0.00)*, 4th quartile

General population (n=391)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

236
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Bloom et al. 2010

0.20 ng/mL (geometric mean TSH
PFUA)
Free T4

NS (p=0.527)

1.75 ng/mL (median PFUA)

NS (p=0.5368)
NS (p=0.0642)

General population (n=31)
Ji et al. 2012
General population (n=633)
Wang et al. 2013a
General population (n=903 pregnant women)
Wang et al. 2014

0.20 ng/mL (geometric mean TSH
PFUA)
Elevated TSH risk

NS (p>0.05)
NS (p>0.05)

3.26 ng/mL (median PFUA)

TSH
Free T4
Total T4
Total T3

NS (p>0.05)
Inverse association (p<0.001)*
Inverse association (p<0.001)*
NS (p>0.05)

0.45 ng/mL (mean PFUA)

TSH
Free T4
Total T4
Free T3
Total T3

Inverse association (p<0.05)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=285 pregnant women)

Yang et al. 2016a
General population (n=157 pregnant women)

PFDoA
Ji et al. 2012
General population (n=633)
Wang et al. 2014
General population (285 pregnant women)

TSH
T4

NS (p=0.204)

0.92 ng/mL (median PFDoA) TSH
T4
0.36 ng/mL (median PFDoA) TSH
Free T4
Total T4
Total T3

***DRAFT FOR PUBLIC COMMENT***

NS (p=0.6925)
NS (p=0.7153)
NS (p>0.05)
Inverse association (p<0.001)*
Inverse association (p<0.01)*
NS (p>0.05)

 PERFLUOROALKYLS

237
2. HEALTH EFFECTS

Table 2-15. Summary of Thyroid Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Yang et al. 2016a

0.046 ng/mL (mean PFDoA) TSH

Inverse association (p<0.01)*

General population (n=157 pregnant women)

Me-PFOSA-AcOH
Jain 2013
General population (NHANES) (n=1,525)

NR

Free T4
Total T4
Free T3
Total T3

Inverse association (p<0.05)*
Inverse association (p<0.05)*
Inverse association (p<0.01)*
Inverse association (p<0.01)*

TSH

NS (p>0.05)

Free T3
Total T3
Free T4
Total T4
Thyroglobulin

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 9 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

HR = hazard ratio; NHANES = National Health and Nutrition Examination Survey; NR = not reported; NS = not significant; OR = odds ratio;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFUA = perfluoroundecanoic acid; T3 = triiodothyronine; T4 = thyroxine; TSH = thyroid
stimulating hormone

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

238
2. HEALTH EFFECTS

Three studies of the community affected by the Washington Works facility reported increases in selfreported thyroid disease (Anderson-Mahoney et al. 2008), any type of functional thyroid disease (LopezEspinosa et al. 2012; Winquist and Steenland 2014b), or hypothyroidism (Lopez-Espinosa et al. 2012).
No associations between cumulative serum PFOA and hyperthyroidism or hypothyroidism were found in
retrospective analysis (Winquist and Steenland 2014b). However, in prospective analysis, an association
between cumulative serum PFOA and hypothyroidism was found in men (Winquist and Steenland
2014b). Consistent with the occupational exposure data, no association between serum PFOA and TSH
levels was found (Emmett et al. 2006b; Knox et al. 2011a; Lopez-Espinosa et al. 2012). Increases in
serum PFOA were also associated with increases in T4 levels and decreases in T3 uptake in adults (Knox
et al. 2011a).

A number of studies have examined the thyroid outcomes associated with serum PFOA levels in the
general population. An association between serum PFOA and thyroid disease risk was found in female
NHANES participants, but not in males (Melzer et al. 2010). Another study utilizing NHANES data
(Wen et al. 2013) found an increased risk of subclinical hypothyroidism among women, but not men, and
a decreased risk of subclinical hyperthyroidism among men, but not women. A case-control study of
women did not find that serum PFOA levels were associated with the risk of hypothyroxinemia (Chan et
al. 2011). Although four studies found associations between serum PFOA and T3 levels (Jain 2013;
Lewis et al. 2015; Webster et al. 2016; Wen et al. 2013), four other studies did not find these associations
(Raymer et al. 2012; Shrestha et al. 2015; Wang et al. 2014; Yang et al. 2016a). No associations between
serum PFOA and TSH or T4 (Bloom et al. 2010; Jain 2013; Ji et al. 2012; Lewis et al. 2015; Raymer et
al. 2012; Shrestha et al. 2015; Wang et al. 2013a, 2014; Webster et al. 2016; Wen et al. 2013; Yang et al.
2016a) levels were found in the general population studies, with the exception of the study by Lewis et al.
(2015), which found an association for TSH and T4 levels.
Laboratory Animal Studies. Repeated intermittent head-only exposure of male rats to up to 84 mg/m3
APFO dusts for 2 weeks did not result in significant gross or microscopic alterations in the thyroid or
adrenal gland (Kennedy et al. 1986).

In a 2-generation study in rats, daily treatment of the parental generation with 0, 1, 3, 10, or 30 mg/kg/day
APFO by gavage in water for 70–90 days produced an increased incidence of hypertrophy and/or
vacuolation of the zona glomerulosa of the adrenal gland from high-dose males (Butenhoff et al. 2004b).
The respective incidences were 0/10, 0/10, 0/10, 2/10, and 7/10. This effect was also observed in
F1 generation males treated with the same dose level. No explanation was apparent for this finding. In

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

239
2. HEALTH EFFECTS

rats dosed with up to 15 mg/kg/day APFO in the diet for 2 years, there were no significant morphological
alterations in the adrenals (3M 1983). A study in monkeys treated with APFO also reported effects on the
adrenal glands. Griffith and Long (1980) reported diffuse lipid depletion in the adrenals from Rhesus
monkeys dosed daily for 90 days with 30 mg/kg/day APFO by gavage. This dose level was lethal to
some monkeys; no such effect was seen in monkeys dosed with 10 mg/kg/day.

For the most part, morphological evaluations of other endocrine glands in animals treated with PFOA
have been negative. For example, male and female rats dosed via the diet with approximately 100–
110 mg/kg/day APFO for 90 days showed no gross or microscopic alterations in the pituitary or thyroid
glands (Griffith and Long 1980). Similar observations were reported in the pituitary, thyroid, and
parathyroid glands from male and female rats dosed with up to 15 mg/kg/day APFO in the diet for 2 years
(3M 1983).

Administration of up to 20 mg/kg/day PFOA administered via a capsule to Cynomolgus monkeys for
4 weeks did not significantly alter free T4, total T4, free T3, total T3, or TSH (Thomford 2001). Serum
T4 and total T4 were significantly reduced in Cynomolgus monkeys dosed with 10 mg/kg/day APFO
administered via a capsule for up to 6 months, but were still within the normal range (Butenhoff et al.
2002). No significant changes were seen on serum free T3, total T3, or TSH, or thyroid histology.

The only relevant dermal information is that no morphological alterations were observed in the thyroid of
rats following dermal application of up to 2,000 mg/kg/day APFO for 2 weeks in the Kennedy (1985)
study.

PFOS

Epidemiology Studies. A number of epidemiology studies have examined the risk of thyroid disease and
alterations in thyroid hormone levels to evaluate whether the thyroid gland is a target of PFOS toxicity.
In studies of NHANES participants, no increases in the risk of thyroid disease were observed in men or
women (Lewis et al. 2015; Melzer et al. 2010). Melzer et al. (2010) did find an increase in the risk of
having thyroid disease and currently taking thyroid medication among men, and Wen et al. (2013) found
increased risks of subclinical hypothyroidism among men and women. Although some studies have
found alterations in thyroid hormone levels, the results are not consistent across studies. Associations
between serum PFOS and TSH levels were observed in two general population studies (Berg et al. 2015;
Wang et al. 2014). In contrast, two other studies found inverse associations for TSH (Dallaire et al. 2009;

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

240
2. HEALTH EFFECTS

Yang et al. 2016a). An occupational exposure study (Olsen et al. 1998a) and eight general population
studies (Bloom et al. 2010; Jain 2013; Ji et al. 2012; Lewis et al. 2015; Raymer et al. 2012; Shrestha et al.
2015; Wang et al. 2014; Wen et al. 2013) did not find associations between serum PFOS and TSH levels.
Conflicting results were also reported for T3 levels, with some studies reporting associations (Olsen et al.
2003a), inverse associations (Dallaire et al. 2009), or no association (Jain 2013; Lewis et al. 2015;
Raymer et al. 2012; Shrestha et al. 2015; Wang et al. 2014; Webster et al. 2016; Wen et al. 2013; Yang et
al. 2016a). Most studies did not find an association with T4 levels (Jain 2013; Ji et al. 2012; Lewis et al.
2015; Raymer et al. 2012; Wang et al. 2014; Webster et al. 2016; Wen et al. 2013; Yang et al. 2016a), but
three studies did find associations between T4 levels and serum PFOS (Dallaire et al. 2009; Lewis et al.
2015; Shrestha et al. 2015). In NHANES participants with two indicators of thyroid stress (low iodine
levels and high thyroid peroxidase antibody), serum PFOS levels were significantly (p<0.05) associated
with increases in free and total T3, decreases in free T4, and increases in TSH levels (Webster et al.
2016).

Laboratory Animal Studies. Chang et al. (2008b) conducted a study of thyroid function in rats exposed
to PFOS (potassium salt). Administration of a single dose of 15 mg/kg by gavage in water (only dose
level tested) reduced serum total T4 significantly at 2, 6, and 24 hours after dosing. This effect was
attributed to a PFOS-induced transient increase in tissue availability of thyroid hormones and turnover of
T4 with a resulting reduction in serum total T4. Chang et al. (2008b) concluded that PFOS did not induce
a classical hypothyroid state or alter the hypothalamic-pituitary-thyroid axis. In another acute-duration
study, dosing of pregnant mice with 6 mg/kg/day PFOS (potassium salt) on GDs 6–18 did not affect
maternal serum levels of free or total T3 or T4 (Fuentes et al. 2006).

Changes in thyroid hormones have also been reported following intermediate-duration exposure to PFOS.
For example, in a 2-generation gavage study in which dosing of rats started before mating and continued
through gestation, doses ≥0.4 mg/kg/day (the lowest dose tested) caused a significant and dose-related
reduction in total T4 in maternal serum on postpartum day 5 (Luebker et al. 2005b). Free T4 and TSH
were not significantly affected. Exposure of pregnant rats to ≥1 mg/kg/day PFOS on GDs 2–20 induced
significant reductions in total T4 and free T4 and less marked reductions in T3 during pregnancy,
particularly on GD 7 (Thibodeaux et al. 2003); however, serum TSH values were not significantly altered.
A similar study in pregnant mice reported a decrease in total T4 on GD 6 in mice dosed with
20 mg/kg/day PFOS on GDs 1–17 (Thibodeaux et al. 2003). No alterations in total T4 were reported in
mice dosed with 15 mg/kg/day. No information was provided regarding other thyroid hormones.
Decreases in T4 levels were observed in male and female rats exposed to PFOS in the diet for 28 days

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

241
2. HEALTH EFFECTS

(Curran et al. 2008); T3 levels were decreased in female rats exposed to 50 or 100 mg/kg/day and in male
rats at 100 mg/kg/day. No histological alterations were observed in the thyroid. Another study with
PFOS found no thyroid histological effects in rats exposed to 10.3 mg/kg/day for 1 day, 8.17 mg/kg/day
for 7 days, or 7.34 mg/kg/day for 28 days (Elcombe et al. 2012a). Exposure of rats to ≥0.27 mg/kg/day
PFOS in drinking water for 91 days resulted in decreases in total T4 levels (Yu et al. 2009a), but no
changes in T3 or TSH levels (highest dose tested was 2.37 mg/kg/day). Curran et al. (2008) suggested
that the apparent decreases in T4 levels, in the absence of TSH alterations and histological alterations in
the thyroid, may be a result of measurement error when analog assays (chemiluminometric immunoassay
and radioimmunoassay) are used due to binding interference. In a study in Cynomolgus monkeys, T3
was numerically lower than controls in one female and one male monkey dosed with 2 mg/kg/day PFOS
by capsule for 4 weeks (Thomford 2002a). However, it is difficult to determine whether the effect was
treatment-related based on only two animals. In a 26-week study in Cynomolgus monkeys, the highest
dose of PFOS tested, 0.75 mg/kg/day, induced a significant increase in serum TSH (approximately twice
control value, but still within the reference range) and a decrease in total T3 at termination, but not at
earlier time points; variations in other thyroid hormones, including T4, were inconsistent regarding dose
and over time (Seacat et al. 2002). The clinical relevance of the lowered total T3 values was not apparent
since there was no indication of a clinical hypothyroid response, and thyroid histology was not altered by
treatment with PFOS.

Examination of the adrenal glands from rats dosed with up to 1.77 mg/kg/day PFOS via the diet for 4 or
14 weeks did not show any significant gross or microscopic alterations (Seacat et al. 2003). No
significant gross or microscopic lesions were reported in the adrenals, thyroid and parathyroid, or
pituitary gland from rats dosed with up to 1.04 mg/kg/day PFOS in the diet for 2 years (Butenhoff et al.
2012b; Thomford 2002b).
PFHxS

Epidemiology Studies. Ten general population studies have evaluated possible associations between
serum PFHxS levels and alterations in thyroid hormone levels. With the exception of a study of pregnant
women, which found an association between serum PFHxS levels and TSH levels (Wang et al. 2014), and
a study of NHANES participants, which found associations between serum PFHxS and total T4 and T3 in
women (Wen et al. 2013), the epidemiology studies did not find associations for TSH, T3, or T4 (Bloom
et al. 2010; Jain 2013; Ji et al. 2012; Lewis et al. 2015; Wang et al. 2013a, 2014; Webster et al. 2016;
Yang et al. 2016a). Chan et al. (2011) did not find an increase in the risk of hypothyroxinemia associated

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

242
2. HEALTH EFFECTS

with serum PFHxS levels, but Wen et al. (2013) found increases in the risk of subclinical hypothyroidism
and subclinical hyperthyroidism among women, but not men.

Laboratory Animal Studies. Hypertrophy and hyperplasia of the follicular cells were observed in the
thyroids of male rats treated with ≥3 mg/kg/day PFHxS for at least 42 days (Butenhoff et al. 2009a;
Hoberman and York 2003). The NOAEL was 1 mg/kg/day. The investigators noted that the observed
changes in rats are consistent with the known effects of inducers of microsomal enzymes where the
hepatocellular hypertrophy results in a compensatory hypertrophy and hyperplasia of the thyroid due to an
increase in plasma turnover of T4 and associated stimulation of TSH. Neither thyroid hormones nor TSH
were measured in the study.
PFNA

Epidemiology Studies. Inverse associations between serum PFNA levels and T4 levels (Wang et al.
2014) and TSH levels (Yang et al. 2016a) have been reported in general population studies. However,
several other studies have not found alterations in TSH, T4, or T3 levels associated with serum PFNA
levels (Bloom et al. 2010; Jain 2013; Ji et al. 2012; Lopez-Espinosa et al. 2012; Wang et al. 2013a;
Webster et al. 2016; Wen et al. 2013; Yang et al. 2016a). The investigators for an occupational exposure
study reported that differences in TSH, T4, and T3 levels were small and clinically insignificant in groups
of workers exposed to low levels, high levels, or no PFNA (Mundt et al. 2007). No associations were
found for thyroid disease, hypothyroidism, or subclinical hypo- or hyperthyroidism among residents
living near the Washington Works PFOA facility (Lopez-Espinosa et al. 2012) or in NHANES
participants (Wen et al. 2013).

PFDeA

Epidemiology Studies. Most general population studies did not find associations between serum PFDeA
levels and TSH, T3, or T4 levels (Bloom et al. 2010; Ji et al. 2012; Wang et al. 2013a, 2014; Yang et al.
2016a). The exceptions were studies by Berg et al. (2015) and Wang et al. (2014), which found
associations with T3 levels in pregnant women, and a study by Yang et al. (2016a), which found an
inverse association with TSH levels in pregnant women.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

243
2. HEALTH EFFECTS

Laboratory Animal Studies. Administration of a single dose of 80 mg/kg PFDeA to female C57BL/6N
mice by gavage resulted in 2- and 4-fold increases in serum T3 and T4, respectively, relative to controls
30 days after dosing (Harris et al. 1989).
PFUA

Epidemiology Studies. Inverse associations between serum PFUA and serum TSH (Yang et al. 2016a) or
T4 (Wang et al. 2014) have been reported in pregnant women; another study found an association
between PFUA and T3 levels (Berg et al. 2015). However, other general population studies have not
found association between PFOA and TSH, T4, or T3 levels (Bloom et al. 2010; Ji et al. 2012; Wang et
al. 2013a, 2014; Yang et al. 2016a).
PFBuS

Laboratory Animal Studies. Treatment of rats with up to 900 mg/kg/day PFBuS by gavage for 28 days
did not alter the gross or microscopic appearance of the adrenal, pituitary, or thyroid/parathyroid glands
(3M 2001). Levels of thyroid hormones in serum were not available in this study.
PFBA

Laboratory Animal Studies. Treatment of rats with up to 184 mg/kg/day PFBA by gavage for 5 days did
not affect the gross or microscopic morphology of the adrenal, thyroid, or pituitary glands (3M 2007a).
Treatment with ≥30 mg/kg/day for 28 or 90 days significantly increased the incidence of hyperplasia/
hypertrophy of the follicular epithelium of the thyroid gland (Butenhoff et al. 2012a; van Otterdijk 2007a,
2007b). These changes were not observed following a 3-week recovery period. Van Otterdijk (2007a,
2007b; Butenhoff et al. 2012a) suggested that the thyroid lesion likely reflected an increase in T4
producing follicular cells in response to feedback mechanisms from the increased turnover of T4 by the
hypertrophic hepatocytes. None of these studies measured thyroid hormones or TSH in serum.
PFDoA

Epidemiology Studies. Three general population studies have evaluated the effect of PFDoA on thyroid
hormone levels. Wang et al. (2014) reported inverse associations between serum PFDoA and free T4 and
total T4 in pregnant women; no associations were found for TSH or total T3. In another study of

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

244
2. HEALTH EFFECTS

pregnant women (Yang et al. 2016a), inverse associations were found for TSH, free T4, total T4, free T3,
and total T3. The third study (Ji et al. 2012) found no associations between serum PFDoA and TSH or
T4.
Me-PFOSA-AcOH

Epidemiology Studies. In the only available study examining thyroid outcomes, Jain (2013) found no
associations between serum Me-PFOSA-AcOH levels and TSH, T3, or T4 levels in NHANES
participants.
2.14 IMMUNOLOGICAL
Overview. Epidemiology studies have evaluated three categories of altered immune response related to
exposure to perfluoroalkyls: immunosuppression (altered antibody response, infectious disease
resistance), hypersensitivity (asthma, wheezing, eczema, atopic dermatitis, allergies), and autoimmunity.
A summary of epidemiology studies evaluating immunological endpoints is presented in Table 2-16;
more detailed descriptions of individual studies are presented in the Supporting Document for
Epidemiological Studies for Perfluoroalkyls, Table 10. In general, the epidemiology studies identify the
immune system as a target of perfluoroalkyl toxicity. The strongest evidence of the immunotoxicity of
perfluoroalkyls in humans comes from epidemiology studies finding associations evaluating the antibody
response to vaccines. Associations have been found for PFOA, PFOS, PFHxS, and PFDeA. There is also
some limited evidence for decreased antibody response for PFNA, PFUA, and PFDoA, although many of
the studies did not find associations for these compounds. In general, decreases in disease resistance have
not been found for PFOA, PFOS, PFHxS, or PFNA. Epidemiology studies have also found increases in
the asthma diagnosis that were associated with serum PFOA levels. There is also marginal evidence for
PFOS, PFHxS, PFNA, PFDeA, PFBuS, and PFDoA; the evidence was considered marginal due to the
small number of studies evaluating the outcome and/or conflicting study results. There are limited data of
effects on autoimmunity; epidemiology studies provide suggestive evidence of an association between
serum PFOA and the risk of ulcerative colitis. The small number of studies investigating immunotoxicity
following exposure to PFHpA did not find associations.

Laboratory animal studies have also evaluated immunosuppression (disease resistance, antibody response,
NK cell activity, delayed-type hypersensitivity response, monocyte phagocytosis), hypersensitivity
(airway resistance, local lymph node assay), and autoimmunity. In addition, laboratory animal studies

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

245
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

NR

OR 1.1 (0.6–1.8)
OR 1.1 (0.6–1.9)

PFOA
Ashley-Martin et al. 2015

IL-33/TSLP (cord blood)
IgE (cord blood)

General population (1,258 women)
Buser and Scinicariello 2016

3.59 and 3.27 ng/mL
Food allergies
(geometric mean 2005–2006 Food sensitization
General population (NHANES) (n=637 and and 2007–2010)
701 adolescents in 2005–2006 and 2007–
>4.47 ng/mL (4th quartile)
2010)
Dalsager et al. 2016
General population (n=359 1–4-year-old
children)

Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Zhu et al. (2016)

2.04–10.12 ng/mL (maternal
3rd tertile PFOA)

Risk of number of days
above the median
Fever
Cough
Nasal discharge
Diarrhea
Vomiting
Risk of number of days
Fever
Cough
Nasal discharge
Diarrhea
Vomiting
Asthma diagnosis
IgE

OR 9.09 (3.52–24.90)*, 4th quartile
NS (p=0.74 for trend)

OR 1.97 (1.07–3.62)*, 3rd tertile
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

OR 1.12 (0.82–1.54), 3rd tertile
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
1.5 and 1.0 ng/mL (mean
OR 2.67 (1.49–4.79)*, 3rd quartile
serum PFOA levels in the
Association (p<0.05)*, asthmatics
asthmatic and non-asthmatic
NS (p>0.05), non-asthmatics
children, respectively; serum
Absolute
eosinophil
counts
Association (p<0.05)*, asthmatics
levels were not reported for
NS (p>0.05), non-asthmatics
full cohort)
Eosinophil cationic protein Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

246
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Fei et al. 2010

5.6 ng/mL (maternal PFOA)

General population. (n=1,400 pregnant
women and young children)
Grandjean et al. 2012; Mogensen et al.
2015a

Risk of hospitalization for
IRR 0.96 (0.87–1.06) for trend
infectious disease in young IRR 1.21 (1.04–1.42)* for trend, girls
children
IRR 0.83 (0.73–0.95)* for trend, boys

4.1 and 4.4 ng/mL (median
PFOA at age 5 and 7 years)

Tetanus antibody levels at
age 5

General population (n=456 and
n=464 children 5 and 7 years of age)

3.20 ng/mL (geometric mean Tetanus antibody levels at
maternal PFOA)
age 7

Grandjean et al. 2017
General population (n=516 children
examined at age 7 and 13 years)

Resultc

NS, maternal PFOA
NS, PFOA at age 5
NS, maternal PFOA
β -35.8% (-51.9 to -14.2)*, per 2-fold
increase in PFOA levels at age 5
NS, PFOA at age 7

Diphtheria antibody levels
at age 5

NS, maternal PFOA
NS (p=0.69), PFOA at age 5

Diphtheria antibody levels
at age 7

NS, maternal PFOA
β -25.2% (-42.9 to -2.0)*, per 2-fold
increase in PFOA levels at age 5
β -25.4% (-40.9 to -5.8)*, per 2-fold
increase in PFOA levels, PFOA at
age 7
NS (p=0.637), PFOA at age 7
NS (p=0.856), PFOA at age 13
Full cohort
NS (p=0.742), PFOA at age 7
NS (p=0.129), PFOA at age 13
Cohort restricted to children without
possible unscheduled booster vaccines
NS (p=0.480), PFOA at age 7
Association (p=0.029)*, PFOA at
age 13

4.4 and 2.0 ng/mL (median
Tetanus antibody levels at
PFOA at age 7 and 13 years) age 13
Diphtheria antibody levels
at age 13

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

247
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Granum et al. 2013

1.1 ng/mL (mean maternal
PFOA)

Inverse association (p=0.001)*

General population (n=56 children age
3 years)

Humblet et al. 2014
General population (NHANES)
(n=1,877 adolescents)
Kielsen et al. 2016
General pop. (n=12 adults)
Looker et al. 2014
Community (C8) (n=411)

4.3 and 4.0 ng/mL (median
PFOA in asthmatics and
nonasthmatics)
1.69 ng/mL (median PFOA)

Rubella antibody levels

Hemophilus influenza type NS (p>0.05)
B antibody levels
Tetanus antibody levels
Asthma diagnosis
Atopic eczema
Eczema and itchiness
Number of episodes of otitis
media
Number of episodes of
common cold
Number of episodes of
gastroenteritis
Asthma episode in last
12 months
Current asthma
Wheezing
Diphtheria antibody levels

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Tetanus antibody levels

NS (p=0.970), unadjusted.

33.74 ng/mL (geometric
Seroprotection from
mean)
influenza A H3N2 virus
13.8–31.5 ng/mL (2nd quartile) Seroprotection from
influenza A H1N1 virus
Seroprotection from
influenza type B virus
Cold or flu infection
Frequency of colds

***DRAFT FOR PUBLIC COMMENT***

Association (p<0.001)*
Association (p=0.048)*
OR 1.18 (1.01–1.39)*, per doubling
PFOA
NS (p=0.26)
NS (p=0.98)
NS (p=0.250), unadjusted

OR 0.34 (0.14–0.83)*, 2nd quartile
NS (p=0.02)
NS (p=0.68)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

248
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Okada et al. 2012

1.3 ng/mL (maternal median
PFOA)

General population (n=343 pregnant
women)

Okada et al. 2014

2.67 ng/mL (maternal mean
PFOA)

General population (n=2,603 infants)

Smit et al. 2015
General population (n=1,024 children)

Steenland et al. 2013
Community (C8) (28,441)

0.97 and 1.79 ng/mL
(maternal mean PFOA in
Ukraine and Greenland
cohorts)

Cumulative

Cord IgE levels
Males
Females
Infant food allergy
Eczema

Resultc
NS (p>0.05)
Inverse association (p<0.05)*
OR 1.67 (0.52–5.37)
OR 0.96 (0.23–4.02)

Wheezing
Otitis media
Risk of allergic diseases
Males
Females
Eczema
Males
Females

OR 1.27 (0.27–6.05)
OR 1.51 (0.45–5.12)

Ever having asthma
Ever having eczema
Current eczema
Ever having wheezing
Current wheezing
Ulcerative colitis
Rheumatoid arthritis

OR 0.80 (0.62–1.04), whole cohort
OR 0.97 (0.81–1.17), whole cohort
OR 1.01 (0.79–1.29), whole cohort
OR 0.91 (0.76–1.10), whole cohort
OR 0.97 (0.71–1.33), whole cohort
OR 1.76 (1.04–2.99)*, 2nd quartile
NS (p>0.05)

Crohn’s disease
Type I diabetes
Lupus
Multiple sclerosis

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

***DRAFT FOR PUBLIC COMMENT***

OR 0.63 (0.63–1.37), 4th quartile
OR 0.64 (0.42–0.97)*, 4th quartile
OR 0.75 (0.48–1.18), 4th quartile
OR 0.65 (0.39–1.09), 4th quartile

 PERFLUOROALKYLS

249
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Steenland et al. 2015

Cumulative

NS (p=0.27), with no lag
NS (p=0.53), with 10-year lag
Positive categorical trend (p=0.05)*,
with no lag
RR 6.57 (1.47–29.40)*, with 10-year lag
Positive categorical trend (p=0.04)*,
with no lag
NS (95% CI included unity)
NS (95% CI included unity), whole cohort
β -6.6% (-11.7 to -1.5)*, per 2-fold
increase in PFOA levels, seropositive
subcohort
NS (95% CI included unity), whole cohort
β -8.9% (-14.6 to -2.9)*, per 2-fold
increase in PFOA levels, seropositive
subcohort
OR 1.35 (1.10–1.66)*
OR 1.28 (0.81–2.04)
OR 0.94 (0.51–1.73)
OR 1.12 (0.85–1.47)

Asthma

Occupational (n=3,713)
Ulcerative colitis
Rheumatoid arthritis
Stein et al. 2016a
General population (NHANES)
(n=1,191 adolescents)

4.13 ng/mL (geometric mean) Measles antibody titers
Mumps antibody titers

Rubella antibody titers

Stein et al. 2016a
General population (NHANES)
(n=640 adolescents)

3.59 ng/mL (geometric mean) Rhinitis
Current asthma
Wheeze
Allergy
Allergic sensitization
Plants
Dust mites
Pets
Cockroach or shrimp
Rodents
Mold
Food

***DRAFT FOR PUBLIC COMMENT***

OR 0.88 (0.67–1.15)
OR 0.93 (0.75–1.16)
OR 1.17 0.81–1.68)
OR 0.79 (0.55–1.13)
OR 1.65 (0.59–4.60
OR 1.21 (0.85–1.72)
OR 1.02 (0.60–1.73)

 PERFLUOROALKYLS

250
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Stein et al. 2016b

2.28 ng/mL (geometric mean) Seroconversion
Hemagglutinin
Immunohistochem.
Serum cytokine levels
Serum chemokine levels

NS (p=0.07 for trend)
NS (p=0.27 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)

General population (n=78 adults receiving
influenza vaccine)

Wang et al. 2011
General population (n=244 children aged
2 years)
Zhu et al. 2016
General population (n=231 asthmatic and
225 non-asthmatic children)

1.71 ng/mL (median cord
PFOA)

1.51 and 1.00 ng/mL (mean in Asthma diagnosis
asthmatics and nonasthmatics)

This is the same group of children
evaluated by Dong et al. (2013)

PFOS
Ashley-Martin et al. 2015
General population (1,258 women)
Buser and Scinicariello 2016

Nasal cytokine levels
Nasal chemokine levels
Serum IgA levels
Serum IgE levels
Cord blood IgE levels
Atopic dermatitis

NR

14.98 and 8.74 ng/mL
(geometric mean PFOS
General population (NHANES) (n=637 and 2005–2006 and 2007–2010)
701 adolescents in 2005–2006 and 2007–
9.17–13.75 ng/mL
2010)
(3rd quartile)

NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p=0.870)
Association (p=0.047)*
NS (p>0.05)
OR 4.24 (1.91–9.42)*, males 4th quartile
OR 3.68 (1.43–9.48)*, females 4th
quartile

T-helper cytokines
IL-4
IL-5
IFN-γ
IL-2
Serum IgE

Association (p=0.001 for trend)*
Association (p=0.004 for trend)*
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)

IL-33/TSLP (cord blood)
IgE (cord blood)

1.1 (0.6–1.9)
OR 1.1 (0.6–1.9)

Food allergies

OR 2.43 (1.05–5.59)*, 3rd quartile
(trend not significant, p=0.27)
NS (p=0.49 for trend)

Food sensitization

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

251
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Dalsager et al. 2016

10.19–25.10 ng/mL (maternal Risk of number of days
3rd tertile PFOS)
above the median
Fever
Cough
Nasal discharge
Diarrhea
Vomiting

OR 2.35 (1.34–4.11)*, 3rd tertile
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=359 1–4-year-old
children)

Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Zhu et al. (2016)

Fei et al. 2010
General population (n=1,400 pregnant
women and young children)
Grandjean et al. 2012; Mogensen et al.
2015a
General population (n=456 and
n=464 children 5 and 7 years of age)

Risk of number of days
Fever
Cough
Nasal discharge
Diarrhea
Vomiting
45.5 and 33.4 ng/mL (mean Asthma diagnosis
serum PFOS levels in the
Asthma severity
asthmatic and non-asthmatic
children, respectively; serum IgE
levels were not reported for
Absolute eosinophil counts
full cohort)

OR 1.65 (1.24–2.18)*, 3rd tertile
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
OR 2.63 (1.48–4.69)*, 4th quartile
Association (p=0.045 for trend)*
Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Eosinophil cationic protein Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
35.3 ng/mL (maternal PFOS) Risk of hospitalization for
IRR 1.00 (0.91–1.09) for trend
infectious disease in young IRR 1.18 (1.03–1.36)* for trend, girls
children
IRR 0.90 (0.80–1.12) for trend, boys
17.3 and 15.5 ng/mL (median Tetanus antibody levels at
PFOS at age 5 and 7 years) age 5

NS, maternal PFOS
β -28.5% (-45.5 to -6.1)*, per 2-fold
increase in PFOS levels at age 5

27.3 ng/mL (geometric mean Tetanus antibody levels at
maternal PFOS)
age 7

NS, maternal PFOS
NS, PFOS at ages 5 and 7

Diphtheria antibody levels
at age 5

***DRAFT FOR PUBLIC COMMENT***

NS, maternal PFOS
NS, PFOS at age 5

 PERFLUOROALKYLS

252
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
Diphtheria antibody levels
at age 7

Grandjean et al. 2017

15.3 and 6.7 ng/mL (median
at age 7 and 13 years)

Tetanus antibody levels at
age 13

General population (n=516 children
examined at age 7 and 13 years)

Granum et al. 2013
General population (n=56 children age 3
years)

Humblet et al. 2014
General population (NHANES)
(n=1,877 adolescents)

5.6 ng/mL (mean maternal
PFOS)

Diphtheria antibody levels
at age 13
Rubella antibody levels
Hemophilus influenza
type B antibody levels
Tetanus antibody levels
Asthma diagnosis
Atopic eczema

Eczema and itchiness
Number of episodes of otitis
media
Number of episodes of
common cold
Number of episodes of
gastroenteritis
17.0 and 16.8 ng/mL (median Asthma episode in last
PFOS in asthmatics and non- 12 months
asthmatics)
Current asthma
Wheezing

***DRAFT FOR PUBLIC COMMENT***

Resultc
β -27.6% (-45.8 to -3.3)*, per 2-fold
increase in PFOS levels at age 5
β -30.3% (47.3 to -7.8)*, per 2-fold
increase in PFOS levels at age 7
Full cohort
NS (p=0.240), PFOS at age 7
NS (p=0.237), PFOS at age 13
Cohort restricted to children without
possible unscheduled booster vaccines
Association (p=0.043)*, PFOS at
age 7
NS (p=0.144), PFOS at age 13
NS (p=0.07), age 7
NS (p=0.454), age 13
Inverse association (p=0.007)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p=0.501)
NS (p=0.367)
NS (p=0.13), per doubling PFOS
NS (p=0.24)
NS (p=0.08)

 PERFLUOROALKYLS

253
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Kielsen et al. 2016

9.52 ng/mL (median PFOS)

Inverse association (p=0.044)*,
unadjusted
NS (p=0.420), unadjusted

General population (n=12 adults)
Looker et al. 2014

Tetanus antibody levels
8.32 ng/mL (geometric mean Response to influenza A
PFOS)
H3N2 virus vaccine

NS (p>0.05)

Response to influenza A
H1N1 virus vaccine

NS (p>0.05)

Response to influenza
type B virus vaccine

NS (p>0.05)

Cold or flu infection
Frequency of colds
Cord IgE levels
Infant food allergy
Eczema
Wheezing
Otitis media

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
OR 3.72 (0.81–17.10)
OR 0.87 (0.15–5.08)
OR 2.68 (0.39–18.30)
OR 1.40 (0.33–6.00)

Community (C8) (n=411)

Okada et al. 2012

5.2 ng/mL (maternal median
PFOS)

General population (n=343 infants)

Okada et al. 2014

5.56 ng/mL (maternal mean
PFOS)

General population (n=2,603 infants)

Smit et al. 2015
General population (n=1,024 children)

Diphtheria antibody levels

4.88 and 20.6 ng/mL
(maternal mean PFOS in
Ukraine and Greenland
cohorts)

Risk of allergic diseases
Males
Females
Eczema
Males
Females
Ever having asthma
Ever having eczema
Current eczema
Ever having wheezing
Current wheezing

***DRAFT FOR PUBLIC COMMENT***

OR 0.95 (0.65–1.37), 4th quartile
OR 0.79 (0.53–1.17), 4th quartile
OR 0.98 (0.63–1.53), 4th quartile
OR 0.84 (0.51–1.38), 4th quartile
OR 0.86 (0.67–1.10), whole cohort
OR 0.98 (0.88–1.18), whole cohort
OR 1.05 (0.82–1.33), whole cohort
OR 0.83 (0.69–1.00), whole cohort
OR 0.60 (0.38–0.92)*, Ukraine cohort
OR 0.91 (0.62–1.36) Greenland
OR 0.76 (0.56–1.01), whole cohort

 PERFLUOROALKYLS

254
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Stein et al. 2016a

20.8 ng/mL (geometric mean Measles antibody titers
PFOS)
Mumps antibody titers

NS (95% CI included unity)

General population (NHANES)
(n=1,191 adolescents)

Rubella antibody titers

Stein et al. 2016a
General population (NHANES)
(n=640 adolescents)

Stein et al. 2016b
General population (n=78 adults receiving
influenza vaccine)

15.0 ng/mL (geometric mean) Rhinitis
Current asthma
Wheeze
Allergy
Allergic sensitization
Plants
Dust mites
Pets
Cockroach or shrimp
Rodents
Mold
Food
5.22 ng/mL (geometric mean) Seroconversion
Hemagglutinin
Immunohistochemistry
Serum cytokine levels
Serum chemokine levels
Nasal cytokine levels
Nasal chemokine levels
Serum IgA levels

***DRAFT FOR PUBLIC COMMENT***

β -7.4% (-12.8 to -1.7)*, per 2-fold
increase in PFOS levels, whole cohort
β -5.9% (-9.9 to -1.6)*, per 2-fold
increase in PFOS levels, seropositive
subcohort
NS (95% CI included unity), whole cohort
β -13.3% (-19.9–6.2)*, per 2-fold
increase in PFOS levels, seropositive
subcohort
OR 1.16 (0.90–1.50)
OR 1.20 (0.88–1.63)
OR 0.76 (0.45–1.19)
OR 1.05 (0.80–1.37)
OR 0.17 (0.53–0.97)*
OR 1.00 (0.73–1.38)
OR 0.83 (0.56–1.22)
OR 0.67 (0.48–0.93)*
OR 0.85 (0.29–2.45)
OR 1.33 (1.06–1.69)*
OR 0.74 (0.39–1.40)
NS (p=0.81 for trend)
NS (p=0.12 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)

 PERFLUOROALKYLS

255
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Wang et al. 2011

5.50 ng/mL (median cord
PFOS)

Serum IgE levels

NS (p=0.179)

Cord blood IgE levels
Atopic dermatitis

Association (p=0.017)*
NS (p>0.05)

General population (n=244 children aged
2 years)
Zhu et al. 2016
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Dong et al. (2013)
PFHxS
Ashley-Martin et al. 2015
General population (1,258 women)
Buser and Scinicariello 2016

45.86 and 33.9 ng/mL (mean Asthma diagnosis
in asthmatics and nonasthmatics)
T-helper cytokines
IL-4
IL-5
IFN-γ
IL-2

NR

General population (n=359 1–4-year-old
children)

NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)

Serum IgE

NS (p>0.05 for trend)

IL-33/TSLP (cord blood)
IgE (cord blood)

1.0 (0.7–1.4)
OR 1.0 (0.7–1.4)

2.09 and 2.19 ng/mL
Food allergies
(geometric mean 2005–2006
General population (NHANES) (n=637 and and 2007–2010)
Food sensitization
701 adolescents in 2005–2006 and 2007–
>4.00 ng/mL (4th quartile)
2010)
Dalsager et al. 2016

OR 4.38 (2.02–9.50)*, males 4th quartile
NS (p=0.899 for trend), females

0.32 ng/mL (maternal median Risk of number of days
PFHxS)
above the median
Fever
Cough
Nasal discharge
Diarrhea
Vomiting

***DRAFT FOR PUBLIC COMMENT***

OR 3.06 (1.35–6.93)*, 4th quartile (trend
not significant, p=0.11)
NS (p=0.72 for trend)

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

256
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Zhu et al. (2016)

Serum perfluoroalkyl level Outcome evaluated
Risk of number of days
Fever
Cough
Nasal discharge
Diarrhea
Vomiting
3.9 and 2.1 ng/mL (mean
Asthma diagnosis
serum PFHxS levels in the
Asthma severity
asthmatic and non-asthmatic
children, respectively; serum IgE
levels were not reported for
Absolute eosinophil counts
full cohort)
Eosinophil cationic protein

Grandjean et al. 2012; Mogensen et al.
2015a

0.6 and 0.5 ng/mL (median
Tetanus antibody levels at
PFHxS at age 5 and 7 years) age 5

General population (n=456 and n=464
children 5 and 7 years of age)

4.41 ng/mL (geometric mean Tetanus antibody levels at
maternal PFHxS)
age 7

Grandjean et al. 2017
General population (n=516 children
examined at age 7 and 13 years)

0.5 and 0.4 ng/mL (median
PFHxS at age 7 and
13 years)

Diphtheria antibody levels
at age 5
Diphtheria antibody levels
at age 7
Tetanus antibody levels at
age 13
Diphtheria antibody levels
at age 13

***DRAFT FOR PUBLIC COMMENT***

Resultc
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
OR 2.94 (1.65–5.25)*, 3rd quartile
NS (p=0.722 for trend)
NS (p>0.05), asthmatics
NS (p>0.05), non-asthmatics
Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
NS, maternal PFHxS
-19.0% (-29.8 to -6.6)*, per 2-fold
increase in PFHxS levels at age 5
NS, maternal PFHxS
β -19.7% (-31.6 to -5.7)*, per 2-fold
increase PFHxS levels at age 5
β -22.3% (-36.3 to -5.2)*, per 2-fold
increase PFHxS levels at age 7
NS, maternal PFHxS
NS, PFHxS at age 5
NS, maternal PFHxS
NS, PFHxS at age 5 or 7
NS (p=0.334), PFHxS at age 7
NS (p=0.568), PFHxS at age 13
NS (p=0.264), PFHxS at age 7
NS (p=0.583), PFHxS at age 13

 PERFLUOROALKYLS

257
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Granum et al. 2013

0.3 ng/mL (mean maternal
PFHxS levels)

Rubella antibody levels

Inverse association (p=0.008)*

Hemophilus influenza
type B antibody levels

NS (p>0.05)

Tetanus antibody levels
Asthma diagnosis
Atopic eczema
Eczema and itchiness
Number of episodes of otitis
media
Number of episodes of
common cold
Number of episodes of
gastroenteritis
Asthma episode in last
12 months
Current asthma
Wheezing
Diphtheria antibody levels

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Tetanus antibody levels

NS (p=0.390), unadjusted

General population (n=56 children age
3 years)

Humblet et al. 2014
General population (NHANES)
(n=1,877 adolescents)
Kielsen et al. 2016
General population (n=12 adults)
Okada et al. 2014
General population (n=2,603 infants)

2.2 and 2.0 ng/mL (median
PFHxS in asthmatics and
nonasthmatics)
0.37 ng/mL (median PFHxS)

0.324 ng/mL (maternal mean Risk of allergic diseases
PFHxS)
Males
Females
Eczema
Males
Females

***DRAFT FOR PUBLIC COMMENT***

NS (p=0.078)
Association (p=0.007)*
NS (p=0.66), per doubling PFHxS
NS (p=0.99)
NS (p=0.92)
NS (p=0.055), unadjusted

OR 0.81 (0.56–1.16), 4th quartile
OR 1.13 (0.75–1.69), 4th quartile
OR 0.78 (0.51–1.19), 4th quartile
OR 0.82 (0.49–1.36), 4th quartile

 PERFLUOROALKYLS

258
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Smit et al. 2015

1.53 and 2.14 ng/mL
(maternal mean PFHxS in
Ukraine and Greenland
cohorts)

OR 0.91 (0.69–1.18), whole cohort

General population (n=1,024 children)

Stein et al. 2016a
General population (NHANES)
(n=1,191 adolescents)

Stein et al. 2016a
General population (NHANES)
(n=640 adolescents)

Stein et al. 2016b
General population (n=78 adults receiving
influenza vaccine)

Ever having asthma

Ever having eczema
Current eczema
Ever having wheezing
Current wheezing
2.47 ng/mL (geometric mean Measles antibody titers
PFHxS)
Mumps antibody titers
Rubella antibody titers

2.09 ng/mL (geometric mean Rhinitis
PFHxS)
Current asthma

1.1 ng/mL (geometric mean
PFHxS)

Wheeze
Allergy
Allergic sensitization
Plants
Dust mites
Pets
Cockroach or shrimp
Rodents
Mold
Food
Seroconversion
Hemagglutinin
Immunohistochem.
Serum cytokine levels
IFN-γ
IFN-α2
TNF-α
IP-10

***DRAFT FOR PUBLIC COMMENT***

OR 1.03 (0.86–1.24), whole cohort
OR 0.93 (0.73–1.20), whole cohort
OR 0.96 (0.79–1.17), whole cohort
OR 0.93 (0.68–1.27), whole cohort
NS (95% CI included unity)
NS (95% CI included unity)
NS (95% CI included unity), whole cohort
β -6.0% (-9.6 to -2.2)*, per 2-fold
increase PFHxS levels, seropositive
subcohort
OR 0.81 (0.57–1.16)
OR 0.98 (0.51–1.87)
OR 0.99 (0.68–1.44)
OR 0.83 (0.59–1.17)
OR 0.93 (0.62–1.39)
OR 1.01 (0.84–1.22)
OR 0.96 (0.71–1.30)
OR 0.72 (0.56–0.93)
OR 0.81 (0.54–1.21)
OR 0.98 (0.65–1.47)
OR 1.03 (0.74–1.42)
NS (p=0.22 for trend)
NS (p=0.34 for trend)
Association (p=0.05 for trend)*
NS (p=0.09 for trend)
Association (p=0.04 for trend)*
NS (p=0.59 for trend)

 PERFLUOROALKYLS

259
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Zhu et al. 2016
General population (n=231 asthmatic and
225 non-asthmatic children)

Serum perfluoroalkyl level Outcome evaluated

3.86 and 2.10 ng/mL (mean
PFHxS in asthmatics and
non-asthmatics)

PFNA
Buser and Scinicariello 2016

0.93 and 1.13 ng/mL
(geometric mean PFNA
General population (NHANES) (n=637 and 2005–2006 and 2007–2010)
701 adolescents in 2005–2006 and 2007–
>1.36 ng/mL (4th quartile)
2010)
Dalsager et al. 2016
General population (n=359 1–4-year-old
children)

Serum chemokine levels

NS (p>0.05 for trend)

Nasal cytokine levels
Nasal chemokine levels
Serum IgA levels
Asthma diagnosis

NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
OR 2.97 (1.33–6.64)*, males 4th quartile
OR 5.02 (2.05–12.30)*, females
4th quartile

T-helper cytokines
IL-4
IL-5
IFN-γ
IL-2
Serum IgE

This is the same group of children
evaluated by Dong et al. (2013)

Resultc

Food allergies
Food sensitization

0.56–0.81 and 0.82–
Risk of number of days
3.64 ng/mL (maternal 2nd and above the median
Fever
3rd tertile PFNA)
Cough
Nasal discharge
Diarrhea
Vomiting
Risk of number of days
Fever
Cough
Nasal discharge
Diarrhea
Vomiting

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p=0.28 for trend)
OR 0.51 (0.28–0.92)*, 4th quartile (trend
not significant, p=0.15)

NS (p>0.05)
NS (p>0.05)
OR 0.53 (0.31–0.92)*, 2nd tertile
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
OR 1.12 (0.84–1.49), 3rd tertile
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

260
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Dong et al. 2013

1.1 and 0.9 ng/mL (mean
serum PFNA levels in the
asthmatic and non-asthmatic
children, respectively; serum
levels were not reported for
full cohort)

OR 2.56 (1.41–4.65)*, 4th quartile

General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Zhu et al. (2016)

Grandjean et al. 2012
General population (n=456 and
n=464 children 5 and 7 years of age)

Grandjean et al. 2017
General population (n=516 children
examined at age 7 and 13 years)
Granum et al. 2013
General population (n=56 children age
3 years)

Asthma diagnosis
Asthma severity
IgE

NS (p=0.217 for trend)
Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Absolute eosinophil counts Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Eosinophil cationic protein Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
1.00 ng/mL (geometric mean Tetanus antibody levels at NS, maternal PFNA
PFNA at age 5 years)
age 5
NS, PFNA at age 5
Tetanus antibody levels at NS, maternal PFNA
0.60 ng/mL (geometric mean age 7
NS, PFNA at age 5
PFNA at age 7 years)
Diphtheria antibody levels NS, maternal PFNA
at age 5
β -16.1% (-28.8 to -1.0)*, per 2-fold
increase PFNA levels at age 5
Diphtheria antibody levels
at age 7
1.1 and 0.7 ng/mL (median
Tetanus antibody levels at
PFNA at age 7 and 13 years) age 13
Diphtheria antibody levels
at age 13
0.3 ng/mL (mean maternal
Rubella antibody levels
PFNA levels)
Hemophilus influenza
type B antibody levels
Tetanus antibody levels
Asthma diagnosis
Atopic eczema
Eczema and itchiness
Number of episodes of otitis
media

***DRAFT FOR PUBLIC COMMENT***

NS, maternal PFNA
NS, PFNA at age 5
NS (p=0.075), PFNA at age 7
NS (p=0.394), PFNA at age 13
NS (p=0.243), PFNA at age 7
NS (p=0.693), PFNA at age 13
Inverse association (p=0.007)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

261
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Humblet et al. 2014
General population (NHANES)
(n=1,877 adolescents)
Kielsen et al. 2016
General population
(n=12 adults)
Okada et al. 2014

Serum perfluoroalkyl level Outcome evaluated

Association (p=0.035)*

0.9 and 0.8 ng/mL (median
PFNA in asthmatics and
nonasthmatics)

Number of episodes of
common cold
Number of episodes of
gastroenteritis
Asthma episode in last
12 months
Current asthma

0.66 ng/mL (median PFNA)

Wheezing
Diphtheria antibody levels

NS (p=0.94)
Inverse association (p=0.004)*,
unadjusted
NS (p=0.250), unadjusted

Tetanus antibody levels
1.36 ng/mL (maternal mean
PFNA)

General population (n=1,024 children)

Stein et al. 2016a
General population (NHANES)
(n=1,191 adolescents)
Stein et al. 2016a
General population (NHANES)
(n=640 adolescents)

NS (p=0.883)
NS (p=0.92), per doubling PFNA
NS (p=0.97)

Risk of allergic diseases
Males
Females
Eczema
Males
Females

OR 0.96 (0.61–1.52), 4th quartile
OR 0.63 (0.38–1.02), 4th quartile

0.62 and 0.73 ng/mL
(maternal mean PFNA in
Ukraine and Greenland
cohorts)

Ever having asthma
Ever having eczema
Current eczema
Ever having wheezing
Current wheezing

OR 0.90 (0.70–1.14), whole cohort
OR 0.94 (0.78–1.14), whole cohort
OR 1.03 (0.82–1.30), whole cohort
OR 0.91 (0.75–1.09), whole cohort
OR 0.90 (0.66–1.23), whole cohort

0.765 ng/mL (geometric
mean)

Measles antibody titers
Mumps antibody titers
Rubella antibody titers

NS (95% CI included unity)
NS (95% CI included unity)
NS (95% CI included unity)

0.929 ng/mL (geometric
mean)

Rhinitis
Current asthma

OR 1.24 (0.97–1.60)
OR 1.26 (0.79–2.01)

Wheeze
Allergy

OR 0.99 (0.58–1.68)
OR 1.12 (0.85–1.47)

General population (n=2,603 infants)

Smit et al. 2015

Resultc

***DRAFT FOR PUBLIC COMMENT***

OR 0.95 (0.66–1.38), 4th quartile
OR 0.55 (0.36–0.82)*, 4th quartile

 PERFLUOROALKYLS

262
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Stein et al. 2016b
General pop. (n=78 adults receiving
influenza vaccine)

Wang et al. 2011
General population (n=244 children aged
2 years)
Zhu et al. 2016
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Dong et al. (2013)

Serum perfluoroalkyl level Outcome evaluated
Allergic sensitization
Plants
Dust mites
Pets
Cockroach or shrimp
Rodents
Mold
Food
0.77 ng/mL (geometric mean Seroconversion
PFNA)
Hemagglutinin
Immunohistochem.
Serum cytokine levels
Serum chemokine levels

2.30 ng/mL (median cord
PFNA)

Nasal cytokine levels
Nasal chemokine levels
Serum IgA levels
Serum IgE levels
Cord blood IgE levels
Atopic dermatitis

1.07 and 0.87 ng/mL (mean Asthma diagnosis
PFNA in asthmatics and nonasthmatics)
T-helper cytokines
IL-4
IL-5
IFN-γ
IL-2
Serum IgE

***DRAFT FOR PUBLIC COMMENT***

Resultc
OR 0.96 (0.74–1.23)
OR 1.05 (0.78–1.41)
OR 1.26 (0.85–1.87)
OR 0.86 (0.60–1.24)
OR 2.25 (0.83–6.10)
OR 1.31 (0.83–2.06)
OR 0.91 (0.55–1.50)
NS (p=0.33 for trend)
NS (p=0.40 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p=0.837)
NS (p=0.908)
NS (p>0.05)
OR 3.33 (1.46–7.58)*, males 4th quartile
NS (p=0.142 for trend), females
Association (p=0.031 for trend)*
Association (p=0.011 for trend)*
NS (p>0.05 for trend)
NS (p>0.05 for trend)
Association (p=0.008 for trend)*

 PERFLUOROALKYLS

263
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.27 ng/mL (median maternal Symptoms of infection
serum PFDeA level

NS (p>0.05)

1.2 and 1.0 ng/mL (mean
serum PFDeA levels in the
asthmatic and non-asthmatic
children, respectively; serum
levels were not reported for
full cohort)

Asthma diagnosis
Asthma severity

OR 3.22 (1.75–5.94)*, 4th quartile
Association (p=0.005 for trend)*

IgE

Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics

PFDeA
Dalsager et al. 2016
General population (n=359 children aged
1–4 years)
Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Zhu et al. (2016)

Absolute eosinophil counts Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Eosinophil cationic protein

Grandjean et al. 2012
General population (n=456 and
n=464 children 5 and 7 years of age)

Grandjean et al. 2017
General population (n=516 children
examined at age 7 and 13 years)

0.28 ng/mL (geometric mean Tetanus antibody levels at
PFDeA at age 5 years)
age 5
0.28 ng/mL (geometric mean Tetanus antibody levels at
maternal PFDeA)
age 7
Diphtheria antibody levels
at age 5
Diphtheria antibody levels
at age 7
0.4 and 0.3 ng/mL (median at Tetanus antibody levels at
age 7 and 13 years)
age 13
Diphtheria antibody levels
at age 13

***DRAFT FOR PUBLIC COMMENT***

Association (p<0.05)*, asthmatics
Association (p<0.05)*, non-asthmatics
NS, maternal PFDeA
β -19.9% (-33.1 to -3.9), per 2-fold
increase PFDeA levels at age 5
NS, maternal PFDeA
β -22.3 (-35.8 to -5.8), per 2-fold
increase PFDeA levels at age 5
NS, maternal PFDeA
NS, PFDeA at age 5
NS, maternal PFDeA
NS, PFDeA at age 5
Association (p=0.022)*, PFDeA at
age 7
NS (p=0.258), PFDeA at age 13
Association (p=0.008)*, PFDeA at
age 7
NS (p=0.726), PFDeA at age 13

 PERFLUOROALKYLS

264
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Kielsen et al. 2016

0.30 ng/mL (median PFDeA) Diphtheria antibody levels

Inverse association (p=0.009)*,
unadjusted
NS (p=0.130), unadjusted

General population (n=12 adults)
Okada et al. 2014
General population (n=2,603 infants)

Smit et al. 2015
General population (n=1,024 children)

Zhu et al. 2016
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Dong et al. (2013)

PFUA
Kielsen et al. 2016
General population (n=12 adults)

Tetanus antibody levels
0.563 ng/mL (maternal mean Risk of allergic diseases
PFDeA)
Males
Females
Eczema
Males
Females
0.16 and 0.42 ng/mL
Ever having asthma
(maternal mean PFDeA in
Ever having eczema
Ukraine and Greenland
Current eczema
cohorts)
Ever having wheezing
Current wheezing

OR 1.13 (0.78–1.64), 4th quartile
OR 0.70 (0.47–1.04), 4th quartile
OR 0.93 (0.60–1.44), 4th quartile
OR 0.78 (0.49–1.25), 4th quartile
OR 0.92 (0.73–1.16), whole cohort
OR 0.88 (0.73–1.06), whole cohort
OR 0.95 (0.75–1.20), whole cohort
OR 0.85 (0.70–1.01), whole cohort
OR 0.76 (0.56–1.04), whole cohort

1.24 and 1.02 ng/mL (mean in Asthma diagnosis
asthmatics and nonasthmatics)
T-helper cytokines
IL-4
IL-5
IFN-γ
IL-2
Serum IgE

OR 3.45 (1.51–7.88)*, males 4th quartile
OR 3.68 (1.43–9.48)*, females 4th
quartile

0.21 ng/mL (median PFUA)

Inverse association (p=0.036)*,
unadjusted
Inverse association (p=0.039)*,
unadjusted

Diphtheria antibody levels
Tetanus antibody levels

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
NS (p>0.05 for trend)
Association (p=0.002 for trend)*

 PERFLUOROALKYLS

265
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Okada et al. 2014

1.50 ng/mL (maternal mean
PFUA)

General population (n=2,603 infants)

Smit et al. 2015
General population (n=1,024 children)

0.16 and 0.68 ng/mL
(maternal mean PFUA in
Ukraine and Greenland
cohorts)

Risk of allergic diseases
Males
Females
Eczema
Males
Females
Ever having asthma
Ever having eczema
Current eczema
Ever having wheezing
Current wheezing

PFHpA
Kielsen et al. 2016
General population (n=12 adults)
Smit et al. 2015
General population (n=1,024 children)

Resultc
OR 1.13 (0.79–1.63), 4th quartile
OR 0.58 (0.39–0.86)*, 4th quartile
OR 1.16 (0.75–10.81), 4th quartile
OR 0.50 (0.30–0.81)*, 4th quartile
OR 0.96 (0.77–1.21), whole cohort
OR 0.95 (0.79–1.15), whole cohort
OR 1.07 (0.85–1.34), whole cohort
OR 0.84 (0.70–1.00), whole cohort
OR 0.87 (0.65–1.17), whole cohort

0.12 ng/mL (median PFHpA) Diphtheria antibody levels
Tetanus antibody levels

NS (p=0.750), unadjusted
NS (p=0.280), unadjusted

0.03 and 0.05 ng/mL
(maternal mean PFHpA in
Ukraine and Greenland
cohorts)

OR 0.93 (0.71–1.22), whole cohort
OR 0.93 (0.78–1.11), whole cohort
OR 0.90 (0.70–1.15), whole cohort
OR 1.03 (0.84–1.25), whole cohort
OR 0.62 (0.40–0.97)*, Ukraine cohort
OR 1.24 (0.79–1.93), Greenland cohort
OR 0.88 (0.64–1.20), whole cohort

Ever having asthma
Ever having eczema
Current eczema
Ever having wheezing
Current wheezing

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

266
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

PFBuS
Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)
This is the same group of children
evaluated by Zhu et al. (2016)

Zhu et al. 2016
General population (n=231 asthmatic and
225 non-asthmatic children)

0.5 and 0.5 ng/mL (mean
serum PFBuS levels in the
asthmatic and non-asthmatic
children, respectively; serum
levels were not reported for
full cohort)

0.53 and 0.48 ng/mL (mean
serum PFBuS in asthmatics
and non-asthmatics)

This is the same group of children
evaluated by Dong et al. (2013)

OR 1.90 (1.08–3.37)*, 4th quartile
NS (p=0.092 for trend)
NS (p>0.05), asthmatics
NS (p>0.05), non-asthmatics
Absolute eosinophil counts Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Eosinophil cationic protein NS (p>0.05), asthmatics
NS (p>0.05), non-asthmatics
Asthma diagnosis
OR 2.59 (1.14–5.87)*, males 4th quartile
NS (p=0.505 for trend), females
T-helper cytokines
IL-4
NS (p>0.05 for trend)
IL-5
Association (p=0.023 for trend)*
IFN-γ
NS (p>0.05 for trend)
IL-2
NS (p>0.05 for trend)
Serum IgE
NS (p>0.05 for trend)
Asthma diagnosis
Asthma severity
IgE

PFDoA
Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)

Kielsen et al. 2016
General population (n=12 adults)

OR 1.81 (1.02–3.23)*, 4th quartile
Association (p=0.024 for trend)*
Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Absolute eosinophil counts Association (p<0.05)*, asthmatics
NS (p>0.05), non-asthmatics
Eosinophil cationic protein Association (p<0.05)*, asthmatics
Association (p<0.05)*, non-asthmatics
0.039 ng/mL (median PFDoA) Diphtheria antibody levels Inverse association (p=0.038)*,
unadjusted
Tetanus antibody levels
Inverse association (p=0.038)*,
unadjusted
5.8 and 4.5 ng/mL (mean
serum PFDoA levels in the
asthmatic and non-asthmatic
children, respectively; serum
levels were not reported for
full cohort)

Asthma diagnosis
Asthma severity
IgE

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

267
2. HEALTH EFFECTS

Table 2-16. Summary of Immunological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Okada et al. 2014

0.188 ng/mL (maternal mean
PFDoA)

General population (n=2,603 infants)

Smit et al. 2015
General population (n=1,024 children)

0.04 and 0.13 ng/mL
(maternal mean PFDoA in
Ukraine and Greenland
cohorts)

Risk of allergic diseases
Males
Females
Eczema
Males
Females
Ever having asthma
Ever having eczema
Current eczema
Ever having wheezing
Current wheezing

PFHxA
Dong et al. 2013
General population (n=231 asthmatic and
225 non-asthmatic children)

0.3 and 0.2 ng/mL (mean
serum PFHxA levels in the
asthmatic and non-asthmatic
children, respectively; serum
levels were not reported for
full cohort)

Resultc
OR 0.93 (0.65–1.34), 4th quartile
OR 0.58 (0.39–0.85)*, 4th quartile
OR 1.00 (0.64–1.55), 4th quartile
OR 0.73 (0.45–1.18), 4th quartile
OR 1.03 (0.81–1.30), whole cohort
OR 0.90 (0.75–1.08), whole cohort
OR 0.88 (0.70–1.14), whole cohort
OR 0.97 (0.80–1.16), whole cohort
OR 0.87 (0.64–1.18), whole cohort

OR 1.60 (0.90–2.86), 4th quartile
NS (p=0.854)
NS (p>0.05), asthmatics
NS (p>0.05), non-asthmatics
Absolute eosinophil counts NS (p>0.05), asthmatics
NS (p>0.05), non-asthmatics
Eosinophil cationic protein NS (p>0.05), asthmatics
NS (p>0.05), non-asthmatics
Asthma diagnosis
Asthma severity
IgE

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 10 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk and bold indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

CI = confidence interval; IFN-α-2 = interferon-α2; IFN-γ = interferon-γ; IgA = immunoglobulin A; IgE = immunoglobulin E; IP-10 = interferon-γ-inducible protein 10;
IRR= incidence risk ratio; NHANES = National Health and Nutrition Examination Survey; NR = not reported; NS = not significant; OR = odds ratio;
PFBuS = perfluorobutane sulfonic acid; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid;
PFUA = perfluoroundecanoic acid; RR= relative risk; TNF-α =tumor necrosis factor-α

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

268
2. HEALTH EFFECTS

have examined secondary outcomes (lymphoid organ weights, lymphocyte counts or subpopulations,
lymphocyte proliferation, cytokine levels, serum antibody levels, histological alterations in immune
organs). Summaries of the laboratory animal studies are presented in the LSE tables for PFOA, PFOS,
and other perfluoroalkyls (Tables 2-3, 2-4, 2-5, and 2-6); the NOAEL and LOAEL values are presented in
Figures 2-6, 2-7, and 2-8. Studies in laboratory animals identify the immune system as a sensitive target
of toxicity following exposure to PFOA and PFOS. The observed effects include impaired responses to
T-dependent antigens, impaired response to infectious disease, and secondary outcomes (decreases in
spleen and thymus weights and in the number of thymic and splenic lymphocytes). A small number of
studies evaluated the immunotoxicity of other perfluoroalkyls and most did not evaluate immune
function. No alterations in spleen or thymus organ weights or morphology were observed in studies on
PFHxS, PFBA, and PFDeA. A study on PFNA found decreases in spleen and thymus weights and
alterations in splenic lymphocyte phenotypes.

The National Toxicology Program (NTP 2016b) concluded that exposure to PFOA or PFOS is presumed
to be an immune hazard to humans based on a high level of evidence that PFOA and PFOS suppressed
the antibody response from animals and a moderate level of evidence from studies in humans. It was
noted that the strongest evidence is for suppression of the antibody response and increased
hypersensitivity (PFOA only).
PFOA

Epidemiology Studies—Immunosuppression Outcomes. Studies evaluating the immunosuppressive
effects of PFOA have examined disease resistance and antibody responses. One study found associations
between maternal serum PFOA and the number of episodes of the common cold and other respiratory
tract infections and the number of episodes of gastroenteritis with vomiting or diarrhea in 3-year-old
children (Granum et al. 2013). Another study found an association between maternal PFOA and the risk
of having a greater number of days with a fever greater than the median (Dalsager et al. 2016), although
there was no increase in the number of days with a fever. However, other studies have not found
associations between PFOA levels and the frequency of the common cold or flu in adults (Looker et al.
2014), between maternal PFOA levels and otitis media in 1.5–3-year-old children (Granum et al. 2013;
Okada et al. 2012), between maternal PFOA and the risk of hospitalization for infectious diseases in
young children (Fei et al. 2010), or between maternal PFOA and the risk of number of days with cough,
nasal discharge, diarrhea, or vomiting (Dalsager et al. 2016).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

269
2. HEALTH EFFECTS

Several studies have evaluated the antibody response to vaccination in adults and children; the changes in
the response to antibody levels relative to serum PFOA levels are graphically presented in Figure 2-19. In
adults, decreases in antibody response against influenza A H3N2 virus were associated with increasing
serum PFOA levels; however, there were no associations with two other strains of influenza virus
(influenza A H1N1 and influenza B) (Looker et al. 2014). Another study of adults also did not find an
altered immune response to influenza A H1N1 virus (Stein et al. 2016b). A small-scale study of 12 adults
did not find significant alterations in the response to diphtheria or tetanus booster vaccines associated
with serum PFOA levels (Kielsen et al. 2016). Increasing current serum PFOA levels were associated
with lower antibody levels for mumps and rubella, but not for measles, in a cross-sectional study of
adolescents (Stein et al. 2016a). A series of prospective studies by Grandjean and associates (Grandjean
et al. 2012, 2017; Mogensen et al. 2015a) evaluated tetanus and diphtheria antibody levels in children at
5, 7, and 13 years of age. Diphtheria antibody levels at age 7 and 13 were inversely associated with
serum PFOA levels at age 5 and 7 (Grandjean et al. 2012; Mogensen et al. 2015a) and with serum PFOA
at age 13 (Grandjean et al. 2017), respectively. Decreases in tetanus antibody levels at age 7 were
associated with increases in serum PFOA levels at age 5, but not at age 7 (Grandjean et al. 2012;
Mogensen et al. 2015a) and tetanus antibody levels were not associated with serum PFOA at age 7 or
13 (Grandjean et al. 2017). In studies comparing maternal serum PFOA with antibody levels in children,
no associations were found for tetanus antibodies at age 3 (Granum et al. 2013), age 5 (Grandjean et al.
2012), or age 7 (Grandjean et al. 2012) or for diphtheria at age 5 or 7 (Grandjean et al. 2012). It is noted
that Grandjean and associates also found an inverse association between serum polychlorinated biphenyls
(PCBs) and serum antibody concentrations against tetanus and diphtheria in children living in the Faroe
Islands (Heilmann et al. 2010). Lower levels of rubella antibodies at age 3 were associated with
increasing maternal PFOA (Granum et al. 2013).

NTP (2016b) concluded that there is moderate confidence that exposure to PFOA is associated with
suppression of the antibody response based on the available human studies. NTP (2016b) also concluded
that there is low confidence that exposure to PFOA is associated with increased incidence of infectious
disease (or lower ability to resist or respond to infectious disease).

Epidemiology Studies—Hypersensitivity Outcomes. Of the different types of hypersensitivity effects,
the most widely studied endpoint is asthma; the possible association between exposure to PFOA and
asthma has been studied in occupational, community, and general population studies. Several studies
have found associations between current serum PFOA levels and diagnosis of asthma in children (Dong et
al. 2013; Humblet et al. 2014) and adults (Anderson-Mahoney et al. 2008; Zhu et al. 2016).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

270
2. HEALTH EFFECTS

Figure 2-19. Antibody Responses Relative to Serum PFOA Levels in Epidemiology Studies
(Presented as percent difference in antibody concentration per 2-fold increase in serum PFOA)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

271
2. HEALTH EFFECTS

However, other studies have found no association between estimated cumulative serum PFOA levels and
incidence of asthma being treated with medication in workers (Steenland et al. 2015) or asthma in the
general population (Stein et al. 2016a). In children, no associations between maternal serum PFOA levels
and asthma-related health outcomes were observed in 3-year-old children (Granum et al. 2013) or 5–
9-year-old children (Smit et al. 2015), or between current PFOA levels and current asthma in adolescents
(Stein et al. 2016a). However, the Stein et al. (2016a) study did find an association with rhinitis in
adolescents. No associations between maternal PFOA and wheezing were found in infants up to
18 months of age (Okada et al. 2012), infants 12 or 24 months of age (Okada et al. 2014), children 3 years
of age (Granum et al. 2013), or children 5–9 years of age (Smit et al. 2015) or between current serum
PFOA levels and wheezing in adults (Stein et al. 2016a). The ORs for asthma diagnosis relative to serum
PFOA levels are graphically presented in Figure 2-20. No associations between maternal PFOA and
eczema were found in infants up to 18 months of age (Okada et al. 2012), children 3 years of age
(Granum et al. 2013), or children 5–9 years of age (Smit et al. 2015). Similarly, no association was found
between cord blood PFOA and atopic dermatitis in children 2 years of age (Wang et al. 2011).

No association between current serum PFOA levels and risks of allergy or allergic sensitization were
found in adults (Stein et al. 2016a). Two studies examining the possible association between current
serum PFOA levels in adults and food allergies have found mixed results, with one study finding an
association (Buser and Scinicariello 2016) and one not finding an association (Stein et al. 2016a); a study
in infants did not find an association between the risk of food allergy and maternal serum PFOA levels
(Okada et al. 2012). It is noted that IgE levels, which were used to assess food allergies, is not a sensitive
measure of clinical food allergy. No association was found for food sensitization (Buser and Scinicariello
2016).

Associations between serum PFOA and IgE, eosinophil counts, and eosinophil cationic protein levels
were observed in asthmatic children (9–16 years of age), but not in non-asthmatic children (Dong et al.
2013; Zhu et al. 2016). Significantly higher IL-4 and IL-5 levels were observed in male children with
asthma with the highest PFOA levels (Zhu et al. 2016). Two studies found associations between PFOA
and IgE levels in infants. An inverse association was found between maternal PFOA and IgE levels in
female infants but not in male infants (Okada et al. 2012), whereas Wang et al. (2011) found a correlation
between cord blood PFOA and child IgE levels in males only or in males and females combined. A third
study did not find an association between cord blood PFOA and IgE levels in infants (Ashley-Martin et

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

272
2. HEALTH EFFECTS

Figure 2-20. Risk of Asthma Diagnosis Relative to PFOA Levels (Presented as Adjusted Odds Ratios)

1age

5–9 years
age 9–16 years
3females age 9–16 years
4age 10–15 years
5age 12–19 years
6adult
2males,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

273
2. HEALTH EFFECTS

al. 2015). NTP (2016b) concluded that there is low confidence that exposure to PFOA during childhood
is associated with increased hypersensitivity responses.

Associations between serum PFOA and IgE, eosinophil counts, and eosinophil cationic protein levels
were observed in asthmatic children (9–16 years of age), but not in non-asthmatic children (Dong et al.
2013; Zhu et al. 2016). Significantly higher IL-4 and IL-5 levels were observed in male children with
asthma with the highest PFOA levels (Zhu et al. 2016). Two studies found associations between PFOA
and IgE levels in infants. An inverse association was found between maternal PFOA and IgE levels in
female infants but not in male infants (Okada et al. 2012), whereas Wang et al. (2011) found a correlation
between cord blood PFOA and child IgE levels in males only or in males and females combined. A third
study did not find an association between cord blood PFOA and IgE levels in infants (Ashley-Martin et
al. 2015). NTP (2016b) concluded that there is low confidence that exposure to PFOA during childhood
is associated with increased hypersensitivity responses.

Epidemiology Studies—Autoimmune Outcomes. There are limited data that can be used to evaluate the
possible association between PFOA exposure and the risk of autoimmune diseases. Significant increases
in the risk of ulcerative colitis were observed in an occupational exposure study (Steenland et al. 2015)
and a C8 Science Panel study (Steenland et al. 2013). Although both studies found consistent results, it
should be noted that the community study also included participants with occupational exposure to
PFOA. The occupational study also found an association between PFOA exposure and rheumatoid
arthritis; this was not observed in the community study. The community study (Steenland et al. 2013)
also found no associations for other autoimmune diseases (Crohn’s disease, Type I diabetes, lupus, and
multiple sclerosis). A third study examined neural- and non-neural-specific antibodies and found no
associations with cord blood PFOA or current serum PFOA in 7-year-old children (Osuna et al. 2014).

NTP (2016b) concluded that there is low confidence that exposure to PFOA is associated with ulcerative
colitis.

Laboratory Animal Studies. The results of several mouse studies support the epidemiology data
suggesting that exposure to PFOA can result in immunosuppression. Significant alterations in IgM levels
in response to T-dependent antigens, such as sheep red blood cells (sRBCs) or horse red blood cells were
observed in acute and intermediate oral mouse studies (DeWitt et al. 2008, 2009, 2016; Loveless et al.
2008; Yang et al. 2002a); the lowest-adverse-effect level was 3.75 mg/kg/day in mice exposed to PFOA
in the drinking water for 15 days (DeWitt et al. 2008). Rats appear to be less sensitive than mice; no

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

274
2. HEALTH EFFECTS

alterations in IgM levels were observed rats administered PFOA via gavage for 28 days (Loveless et al.
2008). In a mouse developmental toxicity study, exposure to PFOA on GDs 6–17 was not associated
with alterations in IgM or IgG levels in the offspring (Hu et al. 2010). Limited data suggest that
alterations in NK cells or delayed type hypersensitivity are not sensitive endpoints for PFOA in laboratory
animals. Exposure of male rats to 50 mg/kg/day PFOA by gavage for 14 days did not significantly affect
the numbers of T cells, NK cells, or helper T cells (Iwai and Yamashita 2006), and tests for delayed-type
hypersensitivity response in mice challenged with bovine serum albumin following exposure to
30 mg/kg/day PFOA via drinking water for 15 days were negative (DeWitt et al. 2008).
Two studies have evaluated hypersensitivity in mice. Application of ≥18.8 mg/kg/day PFOA to the
dorsal surface of the ears of mice and subsequently injected with ovalbumin resulted in a significant
increase in serum total IgE compared to mice exposed only to ovalbumin (Fairley et al. 2007).
Ovalbumin-specific airway hyperreactivity also increased in mice co-exposed to ovalbumin and 25 mg/kg
PFOA relative to mice exposed to ovalbumin alone. The investigators suggested that PFOA exposure
may increase the IgE response to environmental allergens (Fairley et al. 2007). In contrast to the results
of the dermal study, no increases in airway hyperresponsiveness were observed in ovalbumin-sensitized
mice exposed in utero and post-weaning to PFOA in the diet (Ryu et al. 2014). In nonsensitized mice,
PFOA did induce airway hyperresponsiveness in 12-week-old pups.

Numerous studies have evaluated secondary outcomes in monkeys, rats, and mice. In the spleen and
thymus, exposure to PFOA resulted in decreases in organ weight, decreases in the number of cells, and/or
atrophy (DeWitt et al. 2008; Loveless et al. 2008; Qazi et al. 2009a, 2012; Son et al. 2009; Yang et al.
2000, 2001, 2002b). Acute exposure resulted in decreases in absolute thymus weight at 11.5 mg/kg/day
(Yang et al. 2001), decreases in spleen weight at 30 mg/kg/day (Qazi et al. 2012; Yang et al. 2000), and
severe thymic atrophy at 30 mg/kg/day (Qazi et al. 2012; Yang et al. 2000). Exposure of male rats to
50 mg/kg/day PFOA by gavage for 14 days did not significantly affect the absolute or relative spleen
weight nor did it alter lymphocyte subsets (Iwai and Yamashita 2006).
Decreases in relative spleen weight were observed at ≥0.96 mg/kg/day PFOA, and absolute spleen weight
and absolute and relative thymus weights were decreased at 9.6 and 29 mg/kg/day (Loveless et al. 2008).
The lowest-adverse-effect levels for spleen and thymus weight changes identified in mouse intermediate
studies were 3.75 mg/kg/day PFOA for decreases in absolute spleen weight (DeWitt et al. 2008) and
9.6 mg/kg/day for decreases in absolute and relative thymus weight (Loveless et al. 2008). In rats, no

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

275
2. HEALTH EFFECTS

alterations in spleen weight were observed following chronic exposure to 15 mg/kg/day in the diet (3M
1983).

Decreases in the number of splenic and thymic lymphocytes were observed in mice administered via
gavage ≥9.6 mg/kg/day PFOA for 28 days (Loveless et al. 2008). In contrast, administration of
29 mg/kg/day PFOA by gavage for 28 days did not result in alterations in the number of splenic or thymic
lymphocytes in rats (Loveless et al. 2008). A 10-day exposure of mice to 3.0 mg/kg/day PFOA resulted
in decreases in the number of bone marrow B-lymphoid cells (Qazi et al. 2012); a decrease in bone
marrow myeloid cells was also observed at 30 mg/kg/day. Examination of the B-lymphoid cell
subpopulations showed decreases in pro/pre B cells, immature B cells, and early mature B cells, with the
greatest reductions observed for pro/pre B cells. When mice were allowed to recover for 10 days
following a 10-day exposure to 30 mg/kg/day PFOA in the diet, only a partial recovery of B-lymphoid
cells was observed. Significant increases in CD4-CD8- and CD4-CD8+ thymic lymphocytes were
observed in mice exposed to 47.21 mg/kg/day for 21 days; increases in CD4+CD8+ lymphocytes were
observed at 17.63 and 47.21 mg/kg/day (Son et al. 2009). Similarly, there were decreases in splenic
CD4-CD8- lymphocytes at 47.21 mg/kg/day and CD4-CD8+ lymphocytes at ≥0.49 mg/kg/day and
increases in splenic CD4+CD8- lymphocytes at 17.63 and 47.21 mg/kg/day.

Two studies examined the immune response to mitogens in mice exposed to PFOA. Marked decreases in
total leukocytes, lymphocytes, and neutrophils levels and increases in tumor necrosis factor-α (TNF-α)
and interleukin-6 (IL-6) were observed in the peritoneal cavity, bone marrow, and spleen cells in response
to lipopolysaccharide (LPS) stimulation in mice exposed to approximately 40 mg/kg/day PFOA for
10 days (Qazi et al. 2009a). Exposure of splenic lymphocytes isolated from PFOA-exposed mice to
concavalin A (ConA) or LPS resulted in decreases in lymphocyte proliferation (Yang et al. 2002a).

A number of studies have evaluated the potential of PFOA to induce histological alterations in immune
organs. In monkeys, administration of approximately 20 mg/kg/day PFOA administered via a capsule to
Cynomolgus monkeys for 4 or 26 weeks did not affect the gross or microscopic morphology of the spleen
(Butenhoff et al. 2002; Thomford 2001). Administration via gavage of 30 mg/kg/day PFOA to Rhesus
monkeys for 90 days induced atrophy of lymphoid follicles in the spleen and lymph nodes and slight to
moderate hypocellularity of the bone marrow (Griffith and Long 1980). No histological alterations were
observed in the spleen or thymus of rats exposed intermittently to ≤84 mg/m3 APFO dusts for 2 weeks
(Kennedy et al. 1986), ≤29 mg/kg/day administered via gavage for 28 days (Loveless et al. 2008), or
dermal doses of ≤2000 mg/kg/day for 2 weeks (Kennedy 1985) or in the spleen and mesenteric lymph

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

276
2. HEALTH EFFECTS

nodes of rats exposed to ≤110 mg/kg/day PFOA in the diet for 90 days (Griffith and Long 1980) or
≤15 mg/kg/day PFOA in the diet for 2 years (3M 1983).
Studies in wild-type mice and PPARα-null mice demonstrate that PFOA-induced immunomodulation
results from PPARα-dependent and -independent mechanisms (DeWitt et al. 2016; Yang et al. 2002b).
Exposure to 30 or 33 mg/kg/day PFOA resulted in decreases in spleen weight, thymus weight, number of
splenic lymphocytes, number of thymic lymphocytes, and CD4+ and CD8+ splenic and thymic
lymphocytes in wild-type mice. Similar exposures of PPARα knockout mice did not result in alterations
in spleen weight, number of splenic lymphocytes, or their phenotypes. Although decreases in thymus
weight, number of thymic lymphocytes, and their phenotypes were observed in the knockout mice, the
magnitudes of the changes were lower in the knockout mice than in the wild-type mice. However, similar
responses were observed in T-cell-dependent antibody responses. Exposure to 30 mg/kg/day PFOA
resulted in 16 and 14% decreases in the response to sRBCs in wild-type and knockout mice, respectively
(DeWitt et al. 2016).

In a systematic review of the available laboratory animal data, NTP (2016b) concluded that there is high
confidence that exposure to PFOA is associated with suppression of the antibody response, very low
confidence that PFOA is associated with the ability to respond to infectious disease, and moderate
confidence that PFOA is associated with increased hypersensitivity.

Summary. Epidemiology studies have evaluated several aspects of immunotoxicity including
immunosuppression, hypersensitivity, and autoimmunity. A number of general populations studies have
found significant inverse associations between serum PFOA levels and antibody responses to vaccines.
However, no consistent associations were found between serum PFOA and disease resistance, as
measured by episodes of the common cold, cough, fever, or hospitalization for infectious disease. In tests
of hypersensitivity, there is some evidence of an association between serum PFOA and asthma diagnosis
in children and adults, although this finding was not consistent across studies; increased risk of allergy or
allergic sensitization does not appear to be associated with serum PFOA. Based on the findings of an
occupational exposure and community exposure study, there is some suggestive association between
serum PFOA and an increased risk of ulcerative colitis, but not for other autoimmune diseases. Animal
studies suggest that the immune system is a sensitive target of PFOA toxicity. A number of studies in
mice have demonstrated evidence of immunosuppression and increased hypersensitivity. Laboratory
animal studies have also found secondary immune outcomes in the spleen and thymus, which included
decreases in organ weight and decreases in the number of lymphocytes.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

277
2. HEALTH EFFECTS

PFOS

Epidemiology Studies—Immunosuppression Outcomes. Several epidemiology studies have evaluated
the potential of PFOS to cause immunosuppression. In studies that evaluated infectious disease
resistance, no alterations in the risk of otitis media were observed in infants monitored through 18 months
or 3 years of age (Granum et al. 2013; Okada et al. 2012), common cold or other upper respiratory
infections (Granum et al. 2013), gastroenteritis with vomiting or diarrhea (Granum et al. 2013),
hospitalizations due to infectious diseases in children (Fei et al. 2010), or symptoms of infection such as
nasal discharge, cough, diarrhea, or vomiting in children (Dalsager et al. 2016). However, associations
between the number of days with symptoms of infection and maternal PFOS levels were observed in
children (Dalsager et al. 2016).

Other studies evaluating immunosuppression found significant alterations in the response to vaccines; the
changes in the response to antibody levels relative to serum PFOS levels are graphically presented in
Figure 2-21. In children receiving a tetanus vaccination at age 5, there were associations between serum
PFOS levels at age 5 and tetanus antibody levels at age 5 (Grandjean et al. 2012) and between serum
PFOS levels at age 7 and tetanus antibody levels at age 7 when the analysis was restricted to children who
were not likely to have had a booster vaccine after age 5 (Grandjean and Budtz-Jorgensen 2013).
However, no associations were found between tetanus antibody levels at age 5 and maternal PFOS or
child PFOS levels (Grandjean et al. 2012), between tetanus antibody levels at age 7 and maternal PFOS or
child PFOS levels at age 5 or 7 (Grandjean et al. 2012; Mogensen et al. 2015a), or between tetanus
antibody levels at age 14 and child PFOS levels at age 13 (Grandjean et al. 2017). Similarly, diphtheria
antibody levels at age 7 were significantly associated with serum PFOS levels at age 5 and 7 (Grandjean
et al. 2012; Mogensen et al. 2015a), but antibody levels at age 5 were not associated with maternal PFOS
or child PFOS at age 5 (Grandjean et al. 2012) and antibody levels at age 13 were not associated with
child PFOS levels at age 7 or 13 (Grandjean et al. 2017). In another study of children (Granum et al.
2013), decreased rubella antibody levels were associated with higher maternal PFOS levels, but no
associations were found for tetanus or Haemophilus influenza type B antibodies. In adolescents, recent
serum PFOS levels were inversely associated with mumps and rubella antibody levels, but not with
measles antibody levels (Stein et al. 2016a). In studies in adults, recent PFOS levels were inversely
associated with diphtheria antibody levels 30 days after booster administration (Kielsen et al. 2016), but
not with tetanus antibody levels 30 days after booster administration (Kielsen et al. 2016) or influenza
types A H3N2, A H1N1, or B antibody levels 21 days post-vaccination (Looker et al. 2014).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

278
2. HEALTH EFFECTS

Figure 2-21. Antibody Responses Relative to Serum PFOS Levels in Epidemiology Studies
(Presented as percent difference in antibody concentration per 2-fold increase in serum PFOS)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

279
2. HEALTH EFFECTS

NTP (2016b) concluded that there is moderate confidence that exposure to PFOS is associated with
suppression of the antibody response and that there is low confidence that exposure to PFOS is associated
with increased incidence of infectious disease (or lower ability to resist or respond to infectious disease).

Epidemiology Studies—Hypersensitivity Outcomes. Several studies examined the risk of
hypersensitivity associated with serum PFOS in children and adolescents; however, the results are
inconsistent. In two case-control studies, increased risks of asthma were observed. Dong et al. (2013)
reported an increased risk of asthma diagnosis and increased severity of asthma episodes in children with
PFOS levels in the 4th quartile. Zhu et al. (2016) also reported an association between asthma diagnosis
and serum PFOS levels in the 4th quartile; however, the association was only significant in males.
Prospective and cross-sectional studies in children (Granum et al. 2013) did not find an association
between maternal PFOS levels and the risk of asthma diagnosis in 3 year olds, between maternal PFOS
and asthma diagnosis in children in adolescence (Humblet et al. 2014; Stein et al. 2016a). Data
evaluating associations between serum PFOS and the risk of asthma diagnosis are presented in
Figure 2-22.

No associations between maternal PFOS and eczema, atopic dermatitis, or wheezing have been found in
children (Granum et al. 2013; Okada et al. 2012, 2014; Smit et al. 2015; Wang et al. 2011). Similarly, no
associations between recent serum PFOS levels in adolescents and food allergies or sensitizations (Buser
and Scinicariello 2016; Stein et al. 2016a) or maternal PFOS levels and food allergies in infants (Okada et
al. 2012) were observed. However, in a cross-sectional study of adolescents, recent PFOS levels were
associated with mold allergies and inversely associated with the risk of plant or cockroach or shrimp
allergies (Stein et al. 2016a). In related studies, cord blood PFOS levels were associated with an increase
in cord IgE levels, but not in infant serum IgE levels (Wang et al. 2011). Two other studies did not find
associations between maternal PFOS levels and cord IgE levels (Ashley-Martin et al. 2015; Okada et al.
2012).

NTP (2016b) concluded that there is very low confidence that exposure to PFOS is associated with
changes in the hypersensitivity response in children.

Laboratory Animal Studies. A limited number of laboratory animal studies examined PFOS-induced
immunosuppression. Guruge et al. (2009) reported an impaired response to an influenza A virus
challenge in mice administered 0.025 mg/kg/day PFOS via gavage for 21 days (Guruge et al. 2009).
Several studies have found an impaired response to sRBCs (Dong et al. 2009, 2011; Peden-Adams et al.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

280
2. HEALTH EFFECTS

Figure 2-22. Risk of Asthma Diagnosis Relative to PFOS Levels (Presented as Adjusted Odds Ratios)
Ever having asthma
Asthma Diagnosis

Reference

Current or Episode in last 12m

Exposure Metric

Outcome

Smit et al. 2015

1 SD change in maternal PFOS

Ever having asthma¹

Zhu et al. 2016

2nd quartile

Asthma diagnosis²

​

3rd quartile

Asthma diagnosis²

​

4th quartile

Asthma diagnosis²

​

2nd quartile

Asthma diagnosis³

​

3rd quartile

Asthma diagnosis³

​

4th quartile

Asthma diagnosis³

2nd quartile

Asthma diagnosis⁴

​

3rd quartile

Asthma diagnosis⁴

​

4th quartile

Asthma diagnosis⁴

Dong et al. 2013

Humblet et al. 2014 doubling PFOS
​
Stein et al. 2016a

doubling PFOS

Asthma episode in last 12 months⁵
Current asthma⁵

shift from 25th to 75th percentile Current asthma⁵
-1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5

OR (95% CI)
1age

5–9 years
age 9–16 years
3females age 9–16 years
4age 10–15 years
5age 12–19 years
2males,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

281
2. HEALTH EFFECTS

2008); however, decreases in NK cell activity were observed at higher doses (0.83–2.08 mg/kg/day)
(Dong et al. 2009). Qazi et al. (2009a) reported several alterations in parameters associated with the
innate immune system in mice exposed to approximately 40 mg/kg/day PFOS in the diet for 10 days.
These alterations included marked decreases in total leukocyte and lymphocyte levels and increases in
TNF-α and IL-6 levels in the peritoneal cavity and bone marrow in response to LPS stimulation; no
alterations were observed in mice exposed to a 20-fold lower dose.

No alterations in spleen or thymus weights were observed in mice exposed to 0.025 mg/kg/day PFOS
(Guruge et al. 2009); at a higher dose (0.42 mg/kg/day), significant decreases in relative spleen and
thymus weights were observed (Dong et al. 2009; Zheng et al. 2009). Decreases in splenic and thymic
cellularity were also observed at ≥0.42 mg/kg/day PFOS (Dong et al. 2009; Qazi et al. 2009b, 2012;
Zheng et al. 2009). Bone marrow cells (B-lymphoid and myeloid cells) were also significantly decreased
in mice exposed to 30 mg/kg/day PFOS for 10 days (Qazi et al. 2012). Within the B-lymphoid cell
population, there were decreases in the number of pro/pre B cells and immature cells (Qazi et al. 2012).
Significant alterations in all splenic T cell CD4 and CD8 subpopulations were observed at
≥0.00331 mg/kg/day PFOS (Peden-Adams et al. 2008) and thymic lymphocyte phenotypes were altered
at 0.42 mg/kg/day PFOS (Dong et al. 2009).

Rats treated with 1.77 mg/kg/day PFOS for 4 weeks, 6.34 mg/kg/day for 28 days, 1.56 mg/kg/day for
14 weeks, or 1.04 mg/kg/day for 2 years did not show significant morphological alterations in the spleen,
thymus, or mesenteric lymph nodes (Butenhoff et al. 2012b; Lefebvre et al. 2008; Seacat et al. 2003;
Thomford 2002b). Similar findings were reported in Cynomolgus monkeys dosed with up to
2 mg/kg/day for 4 weeks or up to 0.75 mg/kg/day PFOS for 26 weeks (Seacat et al. 2002; Thomford
2002a).

In a systematic review of the available laboratory animal data, NTP (2016b) concluded that there is high
confidence that exposure to PFOS is associated with suppression of the antibody response, moderate
confidence that PFOS is associated with the ability to respond to infectious disease, and low confidence
that PFOS is associated with increased hypersensitivity.

Summary. A number of epidemiology studies have examined the potential immunotoxicity of PFOS.
The database provides convincing evidence of an association between serum PFOS levels and
immunosuppression, particularly impaired antibody responses to vaccines in adults and children. Mixed
results have been observed in studies evaluating infectious disease resistance. Similarly, inconsistent

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

282
2. HEALTH EFFECTS

results have been examined in studies evaluating associations between serum PFOS and hypersensitivity
outcomes, such as asthma; no associations were found for eczema, dermatitis, food allergies/
sensitizations. Laboratory animal studies, particularly studies in mice, provide strong evidence of the
immunotoxicity of PFOS. The strongest evidence comes from studies reported impaired antibody
responses resulting in oral exposure to relatively low doses of PFOS. Other immune effects include
decreased response to infectious disease, decreases in spleen and thymus weights, and decreases in
splenic and thymic cellularity and bone marrow cells.
PFHxS

Epidemiology Studies—Immunosuppression Outcomes. Several epidemiology studies have examined
the potential of PFHxS to suppress the immune system. Altered antibody responses relative to serum
PFHxS levels are graphically presented in Figure 2-23. Inverse associations were observed between
tetanus antibody levels in 5 and 7 year olds and serum PFHxS levels at age 5 or 7 years (Grandjean et al.
2012; Mogensen et al. 2015a); but there were no associations between serum PFHxS levels at age 7 or 13
and tetanus antibody levels at age 13 (Grandjean et al. 2017). No associations were found between
maternal PFHxS levels and tetanus antibody levels in the children. These studies found no associations
between diphtheria antibody levels at ages 5, 7, or 13 and serum PFHxS levels in the mother or in the
children. A study in 3-year-old children found an inverse association between maternal PFHxS levels and
rubella antibody levels, but no association with influenza type B or tetanus antibody levels (Granum et al.
2013). In adolescents, serum PFHxS levels were also inversely associated with rubella antibody titers in
a seropositive subcohort (Stein et al. 2016a); no associations were found for measles or mumps antibody
titers. Another study in adolescents did not find associations between recent serum PFHxS levels and
tetanus or diphtheria antibody levels (Kielsen et al. 2016). A study in adults did not find associations
between PFHxS levels and response to influenza vaccine; some alterations in serum cytokine levels were
observed, but chemokine and IgA levels were not altered (Stein et al. 2016b).

In general, the available studies do not suggest an association between serum PFHxS and decreased
infectious disease resistance. No alterations in the frequency of fever, cough, nasal discharge, otitis
media, diarrhea, or vomiting were observed in children (Dalsager et al. 2016; Granum et al. 2013); one
study found an association between maternal PFHxS levels and the number of episodes of gastroenteritis
in children (Granum et al. 2013).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

283
2. HEALTH EFFECTS

Figure 2-23. Antibody Responses Relative to Serum PFHxS Levels in Epidemiology Studies
(Presented as percent difference in antibody concentration per 2-fold increase in serum PFHxS)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

284
2. HEALTH EFFECTS

Epidemiology Studies—Hypersensitivity Outcomes. Data evaluating associations between serum PFHxS
and the risk of asthma diagnosis are presented in Figure 2-24. No associations were observed between
asthma diagnosis, wheezing, and/or eczema in children and maternal serum PFHxS levels (Granum et al.
2013; Smit et al. 2015) or with recent PFHxS levels in adolescents (Humblet et al. 2014; Okada et al.
2014). In contrast, case-control studies in asthmatic children did find associations between recent PFHxS
serum levels and asthma diagnosis (Dong et al. 2013; Zhu et al. 2016), but no association with asthma
severity (Dong et al. 2013). Dong et al. (2013) also reported associations between serum PFHxS levels
and eosinophil counts and eosinophil cationic protein levels in asthmatic children, but not in nonasthmatics. No associations were found with IgE levels in either case-control study (Dong et al. 2013;
Zhu et al. 2016) or in a study measuring cord blood IgE (Ashley-Martin et al. 2015).

An increased risk of food allergies associated with serum PFHxS levels, but not increased sensitivity to
foods, was found in adolescents (Buser and Scinicariello 2016). Another study found no associations
between serum PFHxS levels and allergic sensitization to plants, dust mites, pets, cockroaches/shrimp,
rodents, mold, or food in adolescents (Stein et al. 2016a).

Laboratory Animal Studies. In the only available study evaluating immunotoxicity for PFHxS,
Butenhoff et al. (2009a; Hoberman and York 2003) did not find histological alterations in the spleen,
thymus, or lymph nodes of rats administered 10 mg/kg/day PFHxS via gavage for 42–56 days.
PFNA

Epidemiology Studies—Immunosuppression Outcomes. Most studies examining a possible association
between serum PFNA levels and immunosuppression have not found associations. No associations were
found between maternal or child PFNA levels and tetanus antibody levels at ages 3, 5, 7, or 13
(Grandjean et al. 2012, 2017; Granum et al. 2013) or in adults (Kielsen et al. 2016). Some studies have
found associations between serum PFNA and diphtheria antibody levels, but the results were not
consistent. Grandjean and associates found a significant inverse association between diphtheria
antibodies levels at age 5 (Grandjean et al. 2012) and serum PFNA levels at age 5, but not for antibody
levels at age 13 and PFNA levels at age 7 or 13 (Grandjean et al. 2017). Kielsen et al. (2016) also
reported an inverse association (unadjusted for potential confounders) between serum PFNA and
diphtheria antibody levels in a small study of adults. An inverse association between maternal serum
PFNA and rubella antibody levels was observed in children (Granum et al. 2013), but there was no

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

285
2. HEALTH EFFECTS

Figure 2-24. Risk of Asthma Diagnosis Relative to PFHxS Levels (Presented as Adjusted Odds Ratios)

1age

5–9 years
age 9–16 years
3females age 9–16 years
4age 10–15 years
5age 12–19 years
2males,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

286
2. HEALTH EFFECTS

association for influenza type B antibody levels. Similarly, no associations were found between recent
PFNA serum levels and measles, mumps, or rubella antibody titers in adolescents (Stein et al. 2016a).
Data evaluating associations between serum PFNA and altered antibody response are presented in
Figure 2-25.

Overall, the epidemiology data do not suggest an association between decreased infectious disease
resistance and serum PFNA levels. No alterations in the risk of increased number of days with fever,
cough, nasal discharge, diarrhea, or vomiting were observed in children (Dalsager et al. 2016), although
the study did find a significant increase in the number of days above the median for nasal discharge.
Another study found that the number of episodes of the common cold in children was associated with
maternal serum PFNA; no associations were found for otitis media or gastroenteritis (Granum et al.
2013).

Epidemiology Studies—Hypersensitivity Outcomes. Case-control studies of asthmatic children have
reported associations between serum PFNA and asthma diagnosis (Dong et al. 2013; Zhu et al. 2016), but
no association with asthma severity (Dong et al. 2013). However, cross-sectional or retrospective studies
(Humblet et al. 2014; Smit et al. 2015; Stein et al. 2016a) have not found associations. Data evaluating
associations between serum PFNA and the risk of asthma diagnosis are presented in Figure 2-26. No
associations were found in adolescents between food allergies (Buser and Scinicariello 2016), allergies
(Stein et al. 2016a), or allergic sensitizations to plants, dust mites, pets, cockroach/shrimp, rodents, mold,
or food (Stein et al. 2016a). However, inverse associations between serum PFNA and food sensitizations
were observed in adolescents (Buser and Scinicariello 2016) and between maternal serum PFNA and
allergic diseases in infants (Okada et al. 2014). No increases in the risk of other hypersensitivity effects
(wheezing, eczema, or atopic dermatitis) were observed (Humblet et al. 2014; Okada et al. 2014; Smit et
al. 2015; Stein et al. 2016a; Wang et al. 2011).

Laboratory Animal Studies. Administration of PFNA for 14 days resulted in decreases in thymus and/or
spleen weights at ≥3 mg/kg/day in rats and mice (Fang et al. 2008, 2009, 2010); at 1 mg/kg/day, an
increase in thymus weight was observed in rats (Fang et al. 2009). Fang et al. (2009) reported increases
in the ratio of thymic cortex to medulla in rats presumably administered ≥3 mg/kg/day PFNA. In the
spleen, there were decreases in the percentage of F4/80+ and CD49b+ cells at ≥1 mg/kg/day and in
CD11c+ cells at ≥3 mg/kg/day (Fang et al. 2008). Increases in pro-inflammatory cytokines were
observed in the serum at ≥3 mg/kg/day (Fang et al. 2009) and spleen at 5 mg/kg/day (Fang et al. 2010).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

287
2. HEALTH EFFECTS

Figure 2-25. Antibody Responses Relative to Serum PFNA Levels in Epidemiology Studies
(Presented as percent difference in antibody concentration per 2-fold increase in serum PFNA)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

288
2. HEALTH EFFECTS

Figure 2-26. Risk of Asthma Diagnosis Relative to PFNA Levels (Presented as Adjusted Odds Ratios)

1age

5–9 years
age 9–16 years
3females age 9–16 years
4age 10–15 years
5age 12–19 years
2males,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

289
2. HEALTH EFFECTS

No alterations were observed in the response of splenic T lymphocytes to ConA at 5 mg/kg/day (Fang et
al. 2008).

Two weeks after a single intraperitoneal administration of 46 mg/kg PFNA to male and female B57BL/6J
mice, a number of immunological alterations included significant decreases relative spleen weight and
splenic leukocyte counts, alterations in splenic T-lymphoctye phenotypes (increased ratios of CD4+ and
CD8+ cells), a decrease in viable thymic cells, a marked decrease in CD4+CD8+ thymic lymphocytes and
an increase in CD4+ and CD8+ thymic lymphocytes, and increased levels of tumor necrosis factor-α in
response to exposure to the LPS (Rockwell et al. 2013). Similar effects were observed 4 weeks postexposure (Rockwell et al. 2017). Comparison of the results 2 weeks post-exposure to 4 weeks postexposure showed a partial recovery in spleen weight and specific thymic lymphocyte subpopulations, but
no recovery of the ratio of specific splenic lymphocytes, thymocyte viability, or response to LPS
(Rockwell et al. 2017). Some sex-related differences were noted, with females appearing to be more
sensitive than males (Rockwell et al. 2017).

PFDeA

Epidemiology Studies—Immunosuppression Outcomes. Studies examining possible associations
between serum PFDeA levels and response to vaccines have reported mixed results; see Figure 2-27 for a
graphical presentation of the antibody response relative to PFDeA levels. Inverse associations were
observed between serum PFDeA levels at age 5 and tetanus antibody levels at ages 5 and 7 (Grandjean et
al. 2012) and serum PFDeA levels at age 7 and antibody levels at age 13 (Grandjean et al. 2017).
Similarly, diphtheria antibody levels at age 13 were inversely associated with serum PFDeA levels at
age 7 (Grandjean et al. 2017), but no associations were observed at other time periods (Grandjean et al.
2012). In adults, diphtheria antibody levels were inversely associated with serum PFDeA levels, but there
was no association for tetanus antibody levels (Kielsen et al. 2016); this study did not adjust for potential
confounders. In the only study examining the possible association between serum PFDeA levels and
infectious disease resistance, no association was found between maternal serum PFDeA levels and
symptoms of infection in children aged 1–4 years (Dalsager et al. 2016).

Epidemiology Studies—Hypersensitivity Outcomes. In case-control studies, associations between
asthma diagnosis and asthma severity were observed in children (Dong et al. 2013; Zhu et al. 2016);
associations with serum IgE levels, absolute eosinophil counts, and eosinophil cationic protein levels

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

290
2. HEALTH EFFECTS

Figure 2-27. Antibody Responses Relative to Serum PFDeA Levels in Epidemiology Studies
(Presented as percent difference in antibody concentration per 2-fold increase in serum PFDeA)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

291
2. HEALTH EFFECTS

were also observed. A cross-sectional study of children did not find associations between maternal
PFDeA levels and asthma, eczema, or wheezing in children (Smit et al. 2015). Another cross- sectional
study found no association between allergic diseases or eczema in infant and maternal PFDeA levels
(Okada et al. 2014). Data evaluating associations between serum PFDeA and the risk of asthma diagnosis
are presented in Figure 2-28.

Laboratory Animal Studies. A single gavage dose of 80 mg/kg PFDeA did not significantly alter relative
thymus weight in female C57BL/6N mice, but it caused a 28% decrease in relative spleen weight 30 days
after dosing (Harris et al. 1989). Lethal doses (160 and 320 mg/kg) induced atrophy and lymphoid
depletion in both the thymus and spleen.
PFUA

Epidemiology Studies. Three epidemiology studies have evaluated the potential immunotoxicity of
PFUA in humans. Kielsen et al. (2016) reported inverse associations between serum PFUA (unadjusted
for potential confounders) and diphtheria and tetanus antibody levels in adults. No significant
associations between maternal PFUA levels and the risk of asthma diagnosis, eczema, or wheezing were
observed in children (Smit et al. 2015). The third study (Okada et al. 2012) found inverse associations
between maternal serum PFUA and risk of allergies or eczema in female infants, but not in males.
PFHpA

Epidemiology Studies. In general, the two available human immunotoxicity studies did not find
associations between serum PFHpA levels and diphtheria or tetanus antibody levels in adults (Kielsen et
al. 2016) or risk of asthma diagnosis, eczema, or wheezing in children (Smit et al. 2015). The Smit et al.
(2015) study did find an inverse association between maternal PFHpA levels and current wheezing in one
subcohort; however, this was not observed in the other subcohort with higher mean maternal PFHpA
levels.
PFBuS

Epidemiology Studies. The epidemiology database for PFBuS consists of two case-control studies in
asthmatic children (Dong et al. 2013; Zhu et al. 2016). Both studies reported increases in asthma
diagnosis, but no association with serum IgE levels.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

292
2. HEALTH EFFECTS

Figure 2-28. Risk of Asthma Diagnosis Relative to PFDeA Levels (Presented as Adjusted Odds Ratios)

1age

5–9 years
age 9–16 years
3females age 9–16 years
4age 10–15 years
2males,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

293
2. HEALTH EFFECTS

Laboratory Animal Studies. No significant histological alterations were observed in spleen, thymus, or
lymph nodes of rats administered via gavage 900 mg/kg/day for 28 days (3M 2001).
PFBA

Laboratory Animal Studies. No significant gross or microscopic alterations were reported in the spleen,
thymus, or mesenteric lymph nodes from rats dosed with PFBA by gavage in doses of up to
184 mg/kg/day for 5 days, 150 mg/kg/day for 28 days, or 30 mg/kg/day for 90 days (3M 2007a;
Butenhoff et al. 2012a; van Otterdijk 2007a, 2007b).
PFDoA

Epidemiology Studies. Four epidemiology studies examining potential immunotoxic endpoints were
identified. Kielsen et al. (2016) found inverse associations between recent serum PFDoA levels (not
adjusted for potential confounders) and diphtheria and tetanus antibody levels in adults. Associations
between serum PFDoA levels and the risk of asthma diagnosis, severity of asthma, serum IgE levels,
absolute eosinophil counts, and eosinophil cationic protein levels were observed in a case-control study of
asthmatic children (Dong et al. 2013). A cross-sectional study of children did not find associations
between maternal serum PFDoA levels and risk of asthma diagnosis, eczema, or wheezing (Smit et al.
2015). Another study did not find associations between maternal serum PFDoA levels and the risk of
allergic disease or eczema in infants (Okada et al. 2014).
PFHxA

Epidemiology Studies. In the only epidemiology study examining potential immunotoxic endpoints,
Dong et al. (2013) found no associations between serum PFHxA levels in asthmatic and nonasthmatic
children and asthma diagnosis, asthma severity, or IgE levels.
2.15 NEUROLOGICAL
Overview. There are limited data on the neurotoxicity of perfluoroalkyls in humans or laboratory
animals; epidemiology data come from two studies examining memory and animal studies primarily
evaluated for morphological alterations; the results of these human studies are summarized in Table 2-17
with more detailed descriptions in the Supporting Document for Epidemiological Studies for

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

294
2. HEALTH EFFECTS

Table 2-17. Summary of Neurological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

14.1–27.0 ng/mL (2nd PFOA Memory loss (selfquintile)
reported)

OR 0.88 (0.79–0.97)*, 2nd quintile

PFOA
Gallo et al. 2013
Community (C8) (n=21,024 older adults;
>50 years of age)
Power et al. 2013
General population (NHANES) (n=1,766 older
adults aged 60–<85 years)
PFOS
Gallo et al. 2013
Community (C8) (n=21,024 older adults;
>50 years of age)
Power et al. 2013
General population (NHANES) (n=1,766 older
adults aged 60–<85 years)
PFHxS
Gallo et al. 2013
Community (C8) (n=21,024 older adults;
>50 years of age)
Power et al. 2013
General population (NHANES) (n=1,766 older
adults aged 60–<85 years)
PFNA
Gallo et al. 2013

4.08 ng/mL (median PFOA)

Difficulty remembering or OR 0.92 (0.78–1.09)
periods of confusion (selfreported)

20.5–27.1 ng/mL (3rd PFOS
quintile)

Memory loss (selfreported)

OR 0.86 (0.78–0.96)*, 3rd quintile

22.63 ng/mL (median PFOS) Difficulty remembering or OR 0.90 (0.78–1.03)
periods of confusion (selfreported)

5.7–232.6 ng/mL (5th PFHxS Memory loss (selfquintile)
reported)

OR 0.89 (0.79–0.99)*, 5th quintile

2.05 ng/mL (median PFHxS) Difficulty remembering or OR 0.93 (0.82–1.06)
periods of confusion (selfreported)

1.0–1.2 ng/mL (2nd PFNA
quintile)

Memory loss (selfreported)

Community (C8) (n=21,024 older adults;
>50 years of age)

***DRAFT FOR PUBLIC COMMENT***

OR 0.86 (0.78–0.96)*, 2nd quintile

 PERFLUOROALKYLS

295
2. HEALTH EFFECTS

Table 2-17. Summary of Neurological Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Power et al. 2013

1.01 ng/mL (median PFNA)

General population (NHANES) (n=1,766 older
adults aged 60–<85 years)

Resultc

Difficulty remembering or OR 0.91 (0.79–1.04)
periods of confusion (selfreported)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 11 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

OR = odds ratio; NHANES = National Health and Nutrition Examination Survey; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

296
2. HEALTH EFFECTS

Perfluoroalkyls, Table 11. The potential to induce neurodevelopmental effects (including the risk of
attention deficit hyperactivity disorder [ADHD]) has been more widely studied; these data are discussed
in Section 2.17, Developmental. There are limited epidemiology data on the neurotoxicity of
perfluoroalkyls. The two available studies found decreases in the risk of memory loss associated with
serum PFOA, PFOS, PFHxS, and PFNA.

The results of the laboratory animal studies are presented in Tables 2-1, 2-3, 2-4, 2-5, and 2-6 and in
Figures 2-4, 2-6, 2-7, and 2-8. No morphological alterations in the brain and nerves were observed in
studies of PFOA, PFOS, PFBuS, or PFBA. No alterations in neurological function tests were observed in
studies of PFHxS, PFHxA, PFBA, or PFDoA. Impaired learning and memory was observed in a study of
PFOS and decreases in grip strength were observed in a study of PFUA.
PFOA

Epidemiology Studies. Gallo et al. (2013) found a decreased risk of self-reported memory loss in older
adult (>50 years of age) C8 participants with serum PFOA levels in the 2nd, 3rd, 4th, or 5th quintiles. When
the participants were categorized by diabetic status, the risk of memory loss was higher among the
diabetics than nondiabetics (p=0.014). In sensitivity analyses, the association between serum PFOA
levels and memory impairment was compared within and across water districts. Within a water district,
the association between serum PFOA and memory impairment was significant, but there was no
association between the geometric mean concentration of PFOA in a district and memory impairment. In
a second study, no association between serum PFOA and self-reported difficulty remembering or periods
of confusion was found in NHANES participants aged 60–<85 years (Power et al. 2013).
Laboratory Animal Studies. Exposure of rats to 18,600 mg/m3 APFO dusts for 1 hour induced excessive
salivation. Intermittent, head-only exposure of male rats exposed to up to 84 mg/m3 APFO dusts for
2 weeks did not reveal gross or microscopic alterations in the brain (Kennedy et al. 1986).

A small number of studies have examined the potential toxicity of perfluoroalkyls to the nervous system
in animals, but comprehensive testing has not been conducted. No alterations in performance on a novel
recognition test were observed in rats administered a single 50 mg/kg dose of PFOA (Kawabata et al.
2017). No overt signs of neurotoxicity or altered response to stimuli were observed in rats and mice
administered up to 1,000 mg/kg PFOA via gavage and observed for 14 days (Sato et al. 2009). Exposure
of rats to up to approximately 110 mg/kg/day PFOA via the diet for 90 days did not induce gross or

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

297
2. HEALTH EFFECTS

microscopic alterations in the brain, spinal cord, or peripheral nerves (Griffith and Long 1980). Similar
results were reported in rats fed a diet that provided approximately 15 mg/kg/day PFOA for 2 years (3M
1983). Rhesus monkeys exposed to doses of PFOA that caused lethality (≥30 mg/kg/day by gavage)
showed signs of hypoactivity and prostration, but examination of the brain did not reveal treatmentrelated alterations (Griffith and Long 1980). Treatment of Cynomolgus monkeys with doses of up to
20 mg/kg/day PFOA administered via a capsule did not induce morphological alterations in the brain or
sciatic nerve (Butenhoff et al. 2002).

Similarly, no gross or microscopic alterations were reported in the brain from rats dermally exposed to
APFO in the Kennedy (1985) study.
PFOS

Epidemiology Studies. Two studies have examined the influence of serum PFOS levels on self-reported
memory in older adults. Gallo et al. (2013) found an inverse association between serum PFOS levels and
the risk of memory loss in C8 Health Study participants. No association for difficulty remembering or
periods of confusion was found in the second study of NHANES participants (Power et al. 2013).

Laboratory Animal Studies. No histological alterations were observed in the brain, spinal cord, and/or
sciatic nerve of rats administered a single gavage dose of up to 500 mg/kg PFOS (Sato et al. 2009), rats
treated with up to 1.6–1.8 mg/kg/day PFOS for 4 or 14 weeks (Seacat et al. 2003), rats exposed to
8.5 mg/kg/day PFOS in the diet for 13 weeks (Kawamoto et al. 2011), rats exposed to 1.04 mg/kg/day
PFOS in the diet for 2 years (Butenhoff et al. 2012b; Thomford 2002b), or Cynomolgus monkeys dosed
with up to 0.75 mg/kg/day PFOS for 26 weeks (Seacat et al. 2002). However, ultrasonic stimulation
resulted in bursts of locomotion immediately followed by tonic convulsions in mice administered
125 mg/kg PFOS and rats administered 250 mg/kg PFOS (Sato et al. 2009); the effect was observed 1–
7 days postexposure and frequently resulted in death. Similarly, tonic convulsions following ultrasonic
stimulation were observed in rats exposed to 8.5 mg/kg/day PFOS in the diet for 6 weeks (Kawamoto et
al. 2011); this effect was not observed at ≤2.0 mg/kg/day. Impaired spatial learning and memory,
assessed using the Morris water maze test, was observed in mice administered 2.15 or 10.75 mg/kg/day
PFOS, but not 0.43 mg/kg/day, for 3 months (Long et al. 2013). Similarly, impaired performance on
retention tasks, as assessed by the water maze test, was observed in mice administered 3 or 6 mg/kg/day
PFOS for 4 weeks (Fuentes et al. 2007c). Histopathological examination of the hypothalamus in male
Sprague-Dawley rats administered PFOS via gavage for 28 days revealed degeneration of gonadotropic

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

298
2. HEALTH EFFECTS

cells of the pituitary gland at ≥1.0 mg/kg/day and dense chromatin, condensed ribosomes, and loss of
morphology in the hypothalamus at ≥3.0 mg/kg/day (López-Doval et al. 2014).
PFHxS

Epidemiology Studies. A decrease in the risk of self-reported memory loss was observed in older adult
participants of the C8 Health Study who had serum PFHxS levels in the 5th quintile (Gallo et al. 2013).
No association between serum PFHxS levels and self-reported difficulty remembering or periods of
confusion was reported in a study of NHANES participants (Power et al. 2013).

Laboratory Animal Studies. In a reproductive study in rats dosed with PFHxS, a functional observational
battery (FOB) and motor activity tests were conducted in males on exposure days 36 and 39 and in
females on postpartum day 17 (Butenhoff et al. 2009a; Hoberman and York 2003). The battery assessed
autonomic functions, reactivity and sensitivity to stimuli, excitability, gait and sensorimotor coordination,
limb grip strength, and abnormal clinical signs. No significant alterations were reported in males or
females dosed with up to 10 mg/kg/day PFHxS.
PFNA

Epidemiology Studies. Self-reported memory loss was shown to be inversely associated with serum
PFNA levels in a study of older C8 Health Study participants (Gallo et al. 2013). Another study of
NHANES participants did not find an association with self-reported difficulty remembering or periods of
confusion (Power et al. 2013).

PFUA

Laboratory Animal Studies. In the only study located for PFUA, a decrease in grip strength was
observed in male and female rats administered 1.0 mg/kg/day PFUA for 41–46 days and allowed to
recover for 14 days (Takahashi et al. 2014). No other alterations in performance on FOB tests were
found.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

299
2. HEALTH EFFECTS

PFBuS

Laboratory Animal Studies. A significant decrease in tail flick latency to a thermal stimulus was
observed in all groups of male rats administered via gavage PFBuS for 28 days. However, other tests of
sensory reactivity to stimuli, grip strength, and motor activity were not affected (3M 2001), and the
significance of this isolated finding is difficult to ascertain. Gross and microscopic examination of the
brain, spinal cord, and sciatic nerve did not show any significant alterations. In a 90-day study, no
significant alterations in motor activity or performance on functional observation tests were observed in
rats at PFBuS doses as high as 600 mg/kg/day (Lieder et al. 2009a).
PFBA

Laboratory Animal Studies. Administration of up to 184 mg/kg/day PFBA by gavage for 5 consecutive
days to rats had no significant effect on the gross or microscopic morphology of the brain or spinal cord
(3M 2007a). In a 28-day gavage study, male rats dosed with 150 mg/kg/day, but not 30 mg/kg/day,
showed a delay in bilateral pupillary reflex at the end of the treatment period (Butenhoff et al. 2012a; van
Otterdijk 2007a). Results from other tests, including hearing ability, static righting reflex, grip strength,
and motor activity, were comparable between groups, and histological examinations of the brain
(including the optic nerve), spinal cord, and sciatic nerve were unremarkable. In a 90-day study, pupillary
reflex tests conducted in weeks 8 and 12 showed delayed dilation under dark conditions in rats dosed with
30 mg/kg/day (2/40 in controls versus 7/39 in high-dose rats; p=0.071 according to the Fisher Exact Test)
(Butenhoff et al. 2012a; van Otterdijk 2007b). Since no abnormalities were recorded during a 3-week
recovery period, and there were no histopathological alterations in the eyes, the effect was not considered
biologically significant by the investigators. Tests for hearing ability, static righting reflex, grip strength,
and motor activity showed no associations with treatment with PFBA. In addition, there were no
significant gross or microscopic alterations in the brain, spinal cord, or sciatic nerve.
PFDoA

Laboratory Animal Studies. Single-dose administration of 50 mg/kg resulted in impaired performance
on a novel object recognition test, but did not result in alterations in other tests of memory, anxiety, or
open field activity (Kawabata et al. 2017).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

300
2. HEALTH EFFECTS

PFHxA

Laboratory Animal Studies. Treatment with 100 or 200 mg/kg/day PFHxA in male and female SpragueDawley rats via gavage for 104 weeks had no effect on locomotion or performance in the FOB test
(Klaunig et al. 2015).

2.16 REPRODUCTIVE
Overview. A number of epidemiology studies have evaluated the reproductive toxicity of perfluoroalkyls;
summaries of these studies are presented in the Supporting Document for Epidemiological Studies for
Perfluoroalkyls, Table 12. These studies have evaluated the following categories of reproductive
outcomes: alterations in reproductive hormone levels; effects on sperm; effects on menopause onset,
menstrual cycle length, endometriosis, and breastfeeding duration; and effects on fertility. Overviews of
the studies examining these specific endpoints are presented in Tables 2-18, 2-19, 2-20, and 2-21,
respectively. In addition to these reproductive outcomes, several epidemiology studies have evaluated the
influence of perfluoroalkyls on sexual maturation; these data are discussed in Section 2.17,
Developmental. Although some studies examining reproductive hormone levels have found associations
with PFOA, PFOS, or PFUA levels, the findings are not consistent across studies or there are too few
studies to interpret the results. Alterations in reproductive hormone levels have not been found in studies
of PFHxS, PFNA, PFDeA, or PFOSA. Some associations between serum perfluoroalkyls (PFOA, PFOS,
PFHxS, PFNA, PFDeA) levels and sperm parameters have been found; often, only one sperm parameter
was altered and it is difficult to assess the adversity of this alteration. There is some suggestive evidence
of an association between serum PFOA, PFOS, PFHxS, or PFNA levels and an increased risk of early
menopause; however, this may be due to reverse causation since an earlier onset of menopause would
result in a decrease in the removal of perfluoroalkyls in menstrual blood. Epidemiology studies of PFOA
and PFOS provide evidence of impaired fertility (increased risks of longer time to pregnancy and
infertility); there is also some evidence for PFHxS and PFNA, but the results are not consistent across
studies. The small number of studies evaluating fertility for PFDeA, PFUA, PFOSA, Me-PFOSA-AcOH,
and Et-PFOSA-AcOH did not find associations.

Studies in laboratory animals have evaluated the potential histological alterations in reproductive tissues,
alterations in reproductive hormones, and impaired reproductive functions. Summaries of these studies

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

301
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0–80,000 ng/mL (PFOA
range)

Estradiol

Association (p=0.01 for trend)*,
1993
NS (p=0.58 for trend), 1995
NS (p=0.66 and 0.56 for trend), 1993
and 1995
NS (p=0.21 and 0.18 for trend), 1993
and 1995
NS (p=0.07 and 0.85 for trend), 1993
and 1995
NS (p=0.15 or 0.82 for trend), 1993
and 1995
Association (p=0.017)*, males

Testosterone

Association (p=0.034)*, males

11.3–19.8 ng/mL (2nd PFOA
quintile)

Estradiol concentration

NS (p>0.05), menopausal and
perimenopausal subgroups

3.61 and 2.31 ng/mL (mean
PFOA in nulliparous and
parous women)

Follicular estradiol

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts

PFOA
Olsen et al. 1998b
Occupational (n=111 males in 1993 and
80 males in 1995)

Prolactin

Estradiol
17α-Hydroxyprogesterone
Bound testosterone
Free testosterone.

Sakr et al. 2007b
Occupational (n=1,025)
Knox et al. 2011
Community (C8) (n=25,957 women)
Barrett et al. 2015
General population (n=178 women)

428 ng/mL (mean PFOA)

Luteal progesterone

Joensen et al. 2013
General population (n=247 young men; mean
age 19.6 years)

3.46 ng/mL (mean PFOA)

Total testosterone
Free testosterone
Free androgen index
LH
Estradiol
SHBG

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

FSH

NS (p>0.05)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

302
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Raymer et al. 2012

10.4 ng/mL (mean PFOA)

Estradiol

NS (p=0.751)

Prolactin
FSH
LH
Free testosterone
Total testosterone
SHBG

NS (p=0.349)
NS (p=0.581)
Correlation (p=0.011)*
Correlation (p=0.015)*
NS (p=0.440)
NS (p=0.39 for trend)

General population (n=256 men)

Specht et al. 2012
General population (n=604 men)
Tsai et al. 2015
General population (n=540 adolescents and
young adults aged 12–30 years)
Vested et al. 2013
General population (n=169 males aged 19–
21 years)

PFOS
Olsen et al. 1998a
Occupational (n=327)

1.3–4.8 (range of PFOA
means of different sites)

2.74 ng/mL (geometric mean SHGB
PFOA)
FSH
Testosterone

Association (p<0.05)*, females 12–
17 years old
NS (p>0.05)
NS (p>0.05)

3.8 ng/mL (median maternal
PFOA)

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p=0.03)*
Association (p=0.01)*

Testosterone
Estradiol
Inhibin B
SHBG
Free antigen index
LH
FSH

1,480–2,440 ng/mL (range of DHEAS
PFOS means at different time FSH
periods)
17-HP
LH
Prolactin
SHBG
Free testosterone

***DRAFT FOR PUBLIC COMMENT***

NS (p=0.60)
NS (p=0.91)
NS (p=0.99)
NS (p=0.69)
NS (p=0.25)
NS (p=0.77)
NS (p=0.90)

 PERFLUOROALKYLS

303
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
Bound testosterone

Knox et al. 2011
Community (C8) (n=25,957 women)
Barrett et al. 2015
General population (n=178 women)

Estradiol
11.9–17.0 and 17.1–
Estradiol concentration
22.4 ng/mL (2nd and 3rd PFOS Perimenopausal
quintiles)
subgroup
Menopausal subgroup
16.44 and 14.18 ng/mL
Follicular estradiol
(mean PFOS in nulliparous
and parous women)

8.46 ng/mL (mean PFOS)

General population (n=247 young men; mean
age 19.6 years)

Raymer et al. 2012
General population (n=256 men)

37.4 ng/mL (mean PFOS)

NS (p=035)
NS (p=0.14), after removal of 1 outlier
Inverse association (p=0.0001)*

Total testosterone
Free testosterone
Free androgen index
LH
Estradiol
SHBG

Inverse association (p=0.007)*
Inverse association (β -0.013, 95%
CI -0.023 to -0.001)*, whole cohort
Inverse association (β -0.025, 95%
CI -0.043 to -0.007)*, nulliparous
subcohort
NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
Inverse association (p<0.05)*
Inverse association (p<0.05)*
Inverse association (p<0.05)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

FSH
Estradiol
Prolactin
FSH
LH
Free testosterone

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Total testosterone

NS (p>0.05)

Luteal progesterone

Joensen et al. 2013

Resultc

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

304
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Tsai et al. 2015

7.78 ng/mL (geometric mean SHGB
PFOS)
FSH

NS (p>0.05)

General population (n=540 male and female
adolescents and young adults aged 12–
30 years)
Vested et al. 2013
General population (n=169 males aged 19–
21 years)

PFHxS
Barrett et al. 2015
General population (n=178 women)

Joensen et al. 2013
General population (n=247 young men; mean
age 19.6 years)

Testosterone
21.2 ng/mL (median maternal Testosterone
PFOS)
Estradiol
Inhibin B
SHBG
Free antigen index
LH

1.22 and 1.65 ng/mL (mean
PFHxS in nulliparous and
parous women)

0.81 ng/mL (mean PFHxS)

Inverse association (p<0.05)*, males
12–17 years old
Inverse association (p<0.05)*,
females 12–17 years old
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

FSH

NS (p>0.05)

Follicular estradiol

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts

Luteal progesterone

Total testosterone

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (p>0.05)

Free testosterone
Free androgen index
LH
Estradiol
SHBG
FSH

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

305
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.67 and 0.60 ng/mL (mean
PFNA in nulliparous and
parous women)

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (p>0.05)
NS (p>0.05)

PFNA
Barrett et al. 2015
General population (n=178 women)

Follicular estradiol

Luteal progesterone

Joensen et al. 2013
General population (n=247 young men; mean
age 19.6 years)

Tsai et al. 2015
General population (n=540 male and female
adolescents and young adults aged 12–
30 years)
PFDeA
Barrett et al. 2015
General population (n=178 women)

1.23 ng/mL (mean PFNA)

Total testosterone
Free testosterone

Free androgen index
LH
Estradiol
SHBG
FSH
1.10 ng/mL (geometric mean SHBG
PFNA)
FSH

0.25 and 0.24 ng/mL (mean
PFDoA in nulliparous and
parous women)

NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Testosterone

NS (p>0.05)

Follicular estradiol

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts

Luteal progesterone

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

306
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Joensen et al. 2013

0.38 ng/mL (mean PFDeA)

Total testosterone

NS (p>0.05)

Free testosterone
Free androgen index
LH
Estradiol
SHBG
FSH

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=247 young men; mean
age 19.6 years)

PFUA
Barrett et al. 2015
General population (n=178 women)

0.40 and 0.42 ng/mL (mean
Follicular estradiol
PFUA in nulliparous and parous
women)
Luteal progesterone

Tsai et al. 2015
General population (n=540 males and females
aged 12–30 years)

5.84 ng/mL (geometric mean
PFUA)

SHBG
FSH
Testosterone

***DRAFT FOR PUBLIC COMMENT***

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (p>0.05)
Inverse association (p<0.05)*,
females 12–17 years old
NS (p>0.05)

 PERFLUOROALKYLS

307
2. HEALTH EFFECTS

Table 2-18. Summary of Alterations in Reproductive Hormone Levels in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

PFOSA
Barrett et al. 2015
General population (n=178 women)

0.25 and 0.23 ng/mL (mean
PFOSA in nulliparous and
parous women)

Follicular estradiol

Luteal progesterone

NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts
NS (95% CI included unity), whole
cohort and parous and nulliparous
subcohorts

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 12 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

CI = confidence interval; DHEAS = dihydroepiandrosterone sulfate; FSH = follicle stimulating hormone; LH = luteinizing hormone; NS = not significant;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid;
PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid; SHBG = sex hormone binding globulin

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

308
2. HEALTH EFFECTS

Table 2-19. Summary of Male Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

4.6 and 5.3 ng/mL (median
PFOA in Michigan and
Texas)

Sperm viability
Sperm count
Sperm motility
↑ curvilinear velocity
Other parameters
Sperm morphology
↑ percentage of sperm
head acrosome area
↓ percentage sperm
with coiled tails
Other parameters
Sperm volume
Sperm concentration
Sperm count

NS (p>0.05)
NS (p>0.05)

Percentage progressive
motile sperm
Sperm morphology

NS (p>0.05)

1.91–5.19 ng/mL (range of
PFOA means)

Y-X chromosome ratio

NS (p>0.05)

10.4 ng/mL (mean PFOA)

Semen volume

NS (p>0.05)

Semen pH
Sperm motility
Sperm concentration
↑ percent motile sperm
Sperm concentration
Sperm volume

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*
NS (p>0.05)
NS (p>0.05)

Sperm count
Sperm morphology

NS (p>0.05)
NS (p>0.05)

PFOA
Buck Louis et al. 2015
General population (n=96 Michigan and 366 in
Texas)

Joensen et al. 2013

3.46 ng/mL (mean PFOA)

General population (n=247 young men; mean
age of 19.6 years)

Kvist et al. 2012
General population (n=588 men)
Raymer et al. 2012
General population (n=256 men)

Toft et al. 2012
General population (n=588 males)

3.8 ng/mL (median PFOA)

***DRAFT FOR PUBLIC COMMENT***

Association (p<0.05)*
NS (p>0.05)
Association (p<0.05)*
Association (p<0.05)*
NS (p<0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

NS (p>0.05)

 PERFLUOROALKYLS

309
2. HEALTH EFFECTS

Table 2-19. Summary of Male Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Vested et al. 2013

3.8 ng/mL (median maternal Sperm concentration
PFOA)
Total sperm count
Semen volume
Percentage progressive
spermatozoa
Percentage
morphologically normal
spermatozoa
Mean testicular volume

Inverse association (p=0.01)*

19.15 and 21.6 ng/mL
(median PFOS in Michigan
and Texas)

NS (p>0.05)
NS (p>0.05)

General population (n=169 males aged 19–
21 years)

PFOS
Buck Louis et al. 2015
General population (n=96 Michigan and 366 in
Texas)

Joensen et al. 2013
General population (n=247 young men; mean
age of 19.6 years)

Kvist et al. 2012
General population (n=588 men)
Raymer et al. 2012
General population (n=256 men)

8.46 ng/mL (mean PFOS)

Sperm viability
Sperm count
Sperm motility
↑ distance travelled
Other parameters
Sperm morphology
Sperm volume
Sperm concentration

Sperm count
Percentage progressive
motile sperm
8.20–51.65 ng/mL (range of Y-X chromosome ratio
mean PFOS)
51.65 ng/mL (mean for
Greenland subcohort)
37.4 ng/mL (mean PFOS)
Semen volume
Semen pH
Sperm motility
Sperm concentration

***DRAFT FOR PUBLIC COMMENT***

Inverse association (p=0.001)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

NS (p>0.05)

Association (p<0.05)*
NS (p>0.05)
NS (p<0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*, whole cohort
Inverse association (p=0.044 for
trend)*, Greenland subcohort
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

310
2. HEALTH EFFECTS

Table 2-19. Summary of Male Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Toft et al. 2012

18.4 ng/mL (median PFOS)

Percent motile sperm

NS (p>0.05)

21.2 ng/mL (median
maternal PFOS)

Sperm concentration
Sperm volume
Sperm count
Percent normal sperm
Sperm concentration
Total sperm count

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Inverse association (p<0.05)*
NS (p>0.05)
NS (p>0.05)

Semen volume
Percentage progressive
spermatozoa
Percentage
morphologically normal
spermatozoa

NS (p>0.05)
NS (p>0.05)

Mean testicular volume

NS (p>0.05)

Sperm volume
Sperm concentration
Sperm count
Percentage progressive
motile sperm
Sperm morphology
Percent motile sperm

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Sperm concentration
Sperm volume
Sperm count
Percent normal sperm

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Inverse association (p<0.05)*

General population (n=588 males)

Vested et al. 2013
General population (n=169 males aged 19–
21 years)

PFHxS
Joensen et al. 2013

0.81 ng/mL (mean PFHxS)

General population (n=247 young men; mean
age of 19.6 years)

Toft et al. 2012
General population (n=588 males)

1.1 ng/mL (median PFHxS)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)

NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

311
2. HEALTH EFFECTS

Table 2-19. Summary of Male Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

1.0 and 1.65 ng/mL (median Sperm viability
PFNA in Michigan and
Sperm count
Texas)
Sperm motility
Sperm morphology
↑ percentage of normal
sperm
↓ percentage sperm
with coiled tails
Other parameters
1.23 ng/mL (mean PFNA)
Sperm volume

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

PFNA
Buck Louis et al. 2015
General population (n=96 Michigan and 366 in
Texas)

Joensen et al. 2013
General population (n=247 young men; mean
age of 19.6 years)

Toft et al. 2012

1.2 ng/mL (median PFNA)

General population (n=588 males)

PFDeA
Buck Louis et al. 2015
General population (n=96 Michigan and 366 in
Texas)

0.3 and 0.5 ng/mL (median
PFDeA in Michigan and
Texas)

Association (p<0.05)*
Association (p<0.05)*
NS (p<0.05)
NS (p>0.05)

Sperm concentration
Sperm count
Percentage progressive
motile sperm
Sperm morphology
Percent motile sperm
Sperm concentration
Sperm volume
Sperm count

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Percent normal sperm

NS (p>0.05)

Sperm viability
Sperm count
Sperm motility
Sperm morphology
↑ sperm head length
↓ percentage sperm
with coiled tails
Other parameters

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Association (p<0.05)*
Association (p<0.05)*
NS (p<0.05)

 PERFLUOROALKYLS

312
2. HEALTH EFFECTS

Table 2-19. Summary of Male Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Joensen et al. 2013

0.38 ng/mL (mean PFDeA)

Sperm volume

NS (p>0.05)

Sperm concentration
Sperm count
Percentage progressive
motile sperm
Sperm morphology

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=247 young men; mean
age of 19.6 years)

Me-PFOSA-AcOH
Buck Louis et al. 2015

0.4 and 0.25 ng/mL (median
Sperm viability
Me-PFOSA-AcOH in Michigan Sperm count
General population (n=96 Michigan and 366 in and Texas)
Sperm motility
Texas)
Sperm morphology
↑ percentage of
neck/midpiece
abnormalities
↑ immature sperm
Other parameters

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*

Association (p<0.05)*
NS (p<0.05)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 12 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane sulfonamide) acetic acid; NS = not significant; PFDeA = perfluorodecanoic acid; PFHxS = perfluorohexane
sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

313
2. HEALTH EFFECTS

Table 2-20. Summary of Female Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

>2,130 ng/mL∙year
(cumulative PFOA exposure
5th quintile)

Menopause age

HR 1.11 (0.97–1.26, p=0.14),
5th quintile

Early menopause risk

>4,670 ng/mL∙year
(cumulative PFOA exposure
Community (C8) (n=3,334, prospective analysis) 5th quintile)

Menopause age
(cumulative)
Menopause age
(measured)

NS (p=0.45), 5-year lag
NS (p=0.58), 10-year lag
NS (p=0.57), 15-year lag
NS (p=0.20), 20-year lag
HR 1.10 (0.84–1.43, p=0.51),
5th quintile
HR 1.12 (0.86–1.45, p=0.40),
5th quintile

PFOA
Dhingra et al. 2016a
Community (C8) (n=8,759; retrospective
analysis)

Dhingra et al. 2016a

>80.8 ng/mL (measured 5th
PFOA quintile)
Knox et al. 2011b

11.3–19.8 ng/mL (2nd PFOA
quintile)

Community (C8) (n=25,957)

Buck Louis et al. 2012
General population (n=473)

2.65 and 2.15 ng/mL
(geometric mean PFOA in
women with or without
endometriosis)

Early menopause risk
(menopausal subgroup)

OR 1.5 (1.1–2.1)*, 2nd quintile

Early menopause risk
(perimenopausal
subgroup)
Endometriosis

OR 1.4 (1.1–1.8)*, 2nd quintile

Risk of moderate to
severe endometriosis

Fei et al. 2010
General population (n=1,347 pregnant women)

OR 1.89 (1.17–3.06)*, without parity
adjustment
OR 1.62 (0.99–2.66), with parity
adjustment
OR 2.58 (1.18–5.64)*, without parity
adjustment
OR 1.86 (0.81–4.24) with parity
adjustment

3.91–5.20 ng/mL (2nd quartile Breastfeeding duration
for maternal PFOA)
≤3 months

OR 1.98 (1.17–3.24)*, 2nd quartile

Breastfeeding duration
≤6 months

OR 1.88 (1.31–2.72)*, 2nd quartile

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

314
2. HEALTH EFFECTS

Table 2-20. Summary of Female Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Lyngsø et al. 2014

1.5 ng/mL (median PFOA)

General population (n=1,623 pregnant women)
Romano et al. 2016

5.5–7.6 ng/mL (maternal
3rd quartile PFOA)

General population (n=336 women)
Taylor et al. 2014
General population (n=2,151 women)
Vagi et al. 2014
General population (n=52 cases and
50 controls)
PFOS
Knox et al. 2011b
Community (C8) (n=25,957)

Buck Louis et al. 2012
General population (n=473)

Fei et al. 2010
General population (n=1,347 pregnant women)

>2.5–4.4 and >4.4 ng/mL
(2nd and 3rd PFOA tertiles)

Resultc

Irregular menstrual cycle OR 1.4 (0.9–2.2)
Long menstrual cycle
Short menstrual cycle
Breastfeeding duration
≤3 months
Breastfeeding duration
≤6 months
Menopause
Hysterectomy

OR 1.7 (1.1–2.6)*
OR 0.7 (0.3–1.5)
RR 1.63 (1.16–2.28)*, 3rd quartile
RR 1.38 (1.06–1.79)*, 3rd quartile
HR 1.36 (1.05–1.75)*, 3rd tertile
HR 1.83 (1.31–2.56)*, 2nd tertile

4.1 and 2.3 ng/mL (geometric Polycystic ovary
mean PFOA for cases and
syndrome risk
controls)

OR 6.93 (1.79–29.92, p=0.003)*,
3rd tertile

11.9–17.0 and 17.1–
Early menopause risk
22.4 ng/mL (2nd and 3rd PFOS (menopausal subgroup)
quintiles)
Early menopause risk
(perimenopausal
subgroup)
7.20 and 6.11 ng/mL
Endometriosis
(geometric mean PFOS in
women with or without
endometriosis)
Risk of moderate to
severe endometriosis

OR 1.5 (1.1–2.1)*, 2nd quintile

3.91–5.20 ng/mL (2nd quartile Breastfeeding duration
for maternal PFOA)
≤3 months

OR 1.89 (1.19–3.01)*, 4th quartile

Breastfeeding duration
≤6 months

OR 1.56 (1.10–2.22)*, 2nd quartile

***DRAFT FOR PUBLIC COMMENT***

OR 1.1 (1.1–1.8)*, 3rd quintile

OR 1.39 (0.98–1.98), without parity
adjustment
OR 1.25 (0.87–1.80), with parity
adjustment
OR 1.86 (1.05–3.30)*, without parity
adjustment
OR 1.50 (0.82–2.74) with parity
adjustment

 PERFLUOROALKYLS

315
2. HEALTH EFFECTS

Table 2-20. Summary of Female Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Lyngsø et al. 2014

8.0 ng/mL (median PFOS)

General population (n=1,623 pregnant women)
Romano et al. 2016
General population (n=336 women)
Taylor et al. 2014
General population (n=2,151 women)
Vagi et al. 2014
General population (n=52 cases and
50 controls)
PFHxS
Buck Louis et al. 2012
General population (n=473)

Irregular menstrual cycle OR 1.0 0.6–1.6)

Long menstrual cycle
Short menstrual cycle
13.9 ng/mL (maternal median Breastfeeding duration
PFOS)
≤3 months
Breastfeeding duration
≤6 months
>9–18.4 and >18.4 ng/mL
Menopause
(2nd and 3rd PFOS tertiles)
Hysterectomy

OR 0.7 (0.4–1.2)
OR 0.7 (0.3–1.5)
NS (p=0.065 for trend)

8.2 and 4.9 ng/mL (geometric Polycystic ovary
mean PFOS for cases and
syndrome risk
controls)

OR 5.79 (1.58–24.12, p=0.005)*,
3rd tertile

0.48 and 0.43 ng/mL
(geometric mean PFHxS in
women with or without
endometriosis)

OR 1.14 (0.58–2.24), without parity
adjustment
OR 0.85 (0.42–1.73), with parity
adjustment
OR 2.12 (0.85–5.27), without parity
adjustment
OR 1.24 (0.47–3.31) with parity
adjustment
NS (p=0.124 for trend)

Endometriosis

Risk of moderate to
severe endometriosis

Romano et al. 2016

1.5 ng/mL (maternal median
PFHxS)

General population (n=336 women)
Taylor et al. 2014
General population (n=2,151 women)

Resultc

>0.9–1.8 and >1.8 ng/mL
(2nd and 3rd PFHxS tertiles)

Breastfeeding duration
≤3 months
Breastfeeding duration
≤6 months
Menopause
Hysterectomy

***DRAFT FOR PUBLIC COMMENT***

NS (p=0.111 for trend)
HR 1.16 (0.91–1.48), 3rd tertile
HR 1.44 (1.12–1.85)*, 2nd tertile

NS (p=0.087 for trend)
HR 1.42 (1.08–7.87)*, 2nd tertile
HR 2.22 (1.66–2.98)*, 2nd tertile

 PERFLUOROALKYLS

316
2. HEALTH EFFECTS

Table 2-20. Summary of Female Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Vagi et al. 2014

1.1 and 0.7 ng/mL (geometric Polycystic ovary
mean PFHxS for cases and syndrome risk
controls)

OR 1.20 (0.35–4.07), 3rd tertile

0.69 and 0.58 ng/mL
(geometric mean PFNA in
women with or without
endometriosis)

OR 2.20 (1.02–4.75)*, without parity
adjustment
OR 1.99, 0.91–4.33), with parity
adjustment
OR 1.21 (0.35–4.19), without parity
adjustment
OR 0.99 (0.27–3.65) with parity
adjustment

General population (n=52 cases and
50 controls)
PFNA
Buck Louis et al. 2012
General population (n=473)

Endometriosis

Risk of moderate to
severe endometriosis

Romano et al. 2016

0.9 ng/mL (maternal median
PFNA)

General population (n=336 women)
Taylor et al. 2014
General population (n=2,151 women)
Vagi et al. 2014
General population (n=52 cases and
50 controls)

>0.80–1.5 and >1.5 ng/mL
(2nd and 3rd PFNA tertiles)

Breastfeeding duration
≤3 months

NS (p=0.591 for trend)

Breastfeeding duration
≤6 months
Menopause

NS (p=0.349 for trend)

Hysterectomy

HR 1.39 (1.08–1.80)*, 2nd tertile

1.2 and 0.9 ng/mL (geometric Polycystic ovary
mean PFNA for cases and
syndrome risk
controls)

***DRAFT FOR PUBLIC COMMENT***

HR 1.47 (1.14–1.90)*, 3rd tertile

OR 2.25 (0.67–8.00), 3rd tertile

 PERFLUOROALKYLS

317
2. HEALTH EFFECTS

Table 2-20. Summary of Female Reproductive Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.20 and 0.18 ng/mL
(geometric mean PFDeA in
women with or without
endometriosis)

OR 2.95 (0.72–12.1), without parity
adjustment
OR 2.60 (0.62–10.9), with parity
adjustment
OR 0.72 (0.06–8.09), without parity
adjustment
OR 0.58 (0.04–7.42) with parity
adjustment

PFDeA
Buck Louis et al. 2012
General population (n=473)

Endometriosis

Risk of moderate to
severe endometriosis

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 12 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

HR = hazard ratio; OR = odds ratio; NS = not significant; PFDeA = perfluorodecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic
acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; RR = risk ratio

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

318
2. HEALTH EFFECTS

Table 2-21. Summary of Fertility Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

2.0 ng/ml (median maternal
PFOA)

FR 1.00 (0.99–1.01), per 0.1 ng/mL
OR 1.00 (0.98–1.01), per 0.1 ng/mL

PFOA
Bach et al. 2015a
General population (n=1,372 pregnant women)
Bach et al. 2015b, 2015c
General population (n=440 pregnant women)

Bach et al. 2015b, 2015c
(re-analysis of Fei et al. 2009, 2012 data)
General population (n=1,161 pregnant women)

Fei et al. 2009
General population (n=1,240 pregnant women)
Fei et al. 2012
(re-analysis of Fei et al. 2009 data)
General population (n=1,240 pregnant women)

Jørgensen et al. 2014a, 2014b
General population (n=938 pregnant women)
Vélez et al. 2015
General population (n=1,743 pregnant women)

Fecundability
Infertility risk

5.6–7.7 ng/mL (4th PFOA
quartile)

Fecundability
Parous subgroup
Nulliparous subgroup
Infertility
Parous subgroup
Nulliparous subgroup
nd
4.1–5.4 ng/mL (2 PFOA
Fecundability
quartile)
Parous subgroup
Nulliparous subgroup
7.2–41.5 ng/mL (4th PFOA
Infertility
quartile)
Parous subgroup
Nulliparous subgroup
nd
3.91–5.20 ng/mL (2 PFOA Fecundability
quartile, maternal)
Infertility
3.91–5.20 and ≥6.97 ng/mL Fecundability
(2nd and 4th PFOA quartiles,
Parous subgroup
Nulliparous subgroup
maternal)
Infertility
Parous subgroup
Nulliparous subgroup
1.65 ng/mL (median PFOA)

Fecundability
Primiparous subgroup
Infertility
1.7 ng/mL (median maternal Fecundability
PFOA)
Infertility

***DRAFT FOR PUBLIC COMMENT***

FR 0.86 (0.63–1.19), 4th quartile
FR 0.74 (0.48–1.13), 4th quartile
FR 0.99 (0.64–1.54), 4th quartile
OR 1.67 (0.70–4.00), 4th quartile
OR 1.74 (0.46–6.55), 4th quartile
OR 1.56 (0.55–4.42), 4th quartile
FR 0.78 (0.65–0.94)*, 2nd quartile
FR 0.76 (0.59–0.96)*, 2nd quartile
FR 0.74 (0.56–0.98)*, 4th quartile
OR 1.91 (1.16–3.13)*, 2nd quartile
OR 2.30 (1.09–4.87)*, 2nd quartile
OR 1.48 (0.80–2.75), 4th quartile
FOR 0.72 (0.57–0.90)*, 2nd quartile
OR 2.06 (1.22–3.51)*, 2nd quartile
FOR 0.61 (0.46–0.80)*, 2nd quartile
FOR 0.63 (0.39–1.04), 4th quartile
OR 3.39 (1.75–6.53)*, 2nd quartile
OR 1.30 (0.52–3.21; p=0.082 for
trend), 4th quartile
FR1.04 (0.87–1.25)
FR 1.31 (1.03–1.68)*
OR 1.11 (0.74–1.66)
FOR 0.89 (0.83–0.94, p<0.001)*
OR 1.31 (1.11–1.53, p=0.001)*

 PERFLUOROALKYLS

319
2. HEALTH EFFECTS

Table 2-21. Summary of Fertility Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Vestergaard et al. 2012

5.58 and 5.61 ng/mL
(median PFOA in women
General population (n=222 nulliparous couples) with no pregnancy and
pregnant)
Whitworth et al. 2012b
1.66–2.24, 2.25–3.02, and
≥3.02 ng/mL (2nd, 3rd, and
General population (n=416 subfecund pregnant 4th PFOA quartiles)
women and 474 controls)
PFOS
Bach et al. 2015a
8.3 ng/ml (median maternal
PFOS)
General population (n=1,372 pregnant women)
10.85–36.10 ng/mL
(4th PFOS quartile)
Bach et al. 2015b, 2015c
36.3–103.8 ng/mL (4th PFOS
quartile)
General population (n=440 pregnant women)

Bach et al. 2015b, 2015c
(re-analysis of Fei et al. 2009, 2012 data)
General population (n=1,161)

Fei et al. 2009
General population (n=1,240 pregnant women)

27.0–34.2, 34.3–43.8, and
43.9–106.7 ng/mL (2nd, 3rd,
and 4th PFOS quartiles)

Resultc

Fecundability

OR 1.18 (0.78–1.78)

Not becoming pregnant
within first six cycles

OR 1.21 (0.67–2.18)

Infertility
OR 1.6 (1.1–2.3)*, 2nd quartile
Parous subgroup
OR 2.4 (1.4–4.1)*, 3rd quartile
Primiparous subgroup
OR 0.5 (0.2–1.2), 4th quartile

Fecundability
Infertility risk

FR 1.09 (0.92–1.30), 4th quartile
OR 0.71 (0.47–1.07), 4th quartile

Fecundability
Parous subgroup
Nulliparous subgroup
Infertility
Parous subgroup
Nulliparous subgroup
Fecundability
Parous subgroup
Nulliparous subgroup
Infertility
Parous subgroup
Nulliparous subgroup

FR 0.96 (0.75–1.24), 4th quartile
FR 1.04 (0.70–1.55), 4th quartile
FR 0.97 (0.62–1.51), 4th quartile
OR 1.03 (0.54–2.00), 4th quartile
OR 0.70 (0.16–3.11), 4th quartile
OR 1.23 (0.452–3.39), 4th quartile
FR 0.79 (0.66–0.95)*, 2nd quartile
FR 0.90 (0.70–1.14), 4th quartile
FR 0.68 (0.52–0.91)*, 3rd quartile
OR 1.65 (1.01–2.68)*, 2nd quartile
OR 1.60 (0.78–3.28), 4th quartile
OR 2.71 (1.38–5.30)*, 3rd quartile

26.1–33.3 ng/mL (2nd PFOS Fecundability
quartile, maternal)
Infertility

***DRAFT FOR PUBLIC COMMENT***

OR 0.70 (0.56–0.87)*, 2nd quartile
OR 1.70 (1.01–2.86)*, 2nd quartile

 PERFLUOROALKYLS

320
2. HEALTH EFFECTS

Table 2-21. Summary of Fertility Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Fei et al. 2012
(re-analysis of Fei et al. 2009 data)

26.1–33.3 and 33.4–43.2
ng/mL (2nd and 3rd PFOS
quartiles, maternal)

NS (p=0.32 for trend)
FOR 0.63 (0.43–0.91)*, 3rd quartile

General population (n=1,240 pregnant women)

Jørgensen et al. 2014a, 2014b
General population (n=938 pregnant women)
Vélez et al. 2015
General population (n=1,743 pregnant women)

10.60 ng/mL (median PFOS) Fecundability
Infertility

NS (p=0.26 for trend)
OR 2.50 (1.16–5.37, p=0.36 for
trend)*, 3rd quartile
FR 0.90 (0.76–1.07)
OR 1.39 (0.93–2.07)

4.7 ng/mL (median maternal Fecundability
PFOS)
Infertility

FOR 0.96 (0.91–1.02, p=0.17)
OR 1.14 (0.98–1.34, p=0.09)

Vestergaard et al. 2012

35.75 and 36.29 ng/mL
(median PFOS in women
General population (n=222 nulliparous couples) with no pregnancy and
pregnant)
Whitworth et al. 2012b
10.34–16.60 and
≥16.61 ng/mL (3rd and
General population (n=416 subfecund women
4th PFOS quartile)
and 474 controls)
PFHxS
Bach et al. 2015a
0.5 ng/ml (median maternal
PFHxS)
General population (n=1,372 pregnant women)
Jørgensen et al. 2014a, 2014b
1.94 ng/mL (median PFHxS)
General population (n=938 pregnant women)
Vélez et al. 2015
General population (n=1,743 pregnant women)

Fecundability
Parous subgroup
Nulliparous subgroup
Infertility
Parous subgroup
Nulliparous subgroup

1 ng/mL (median maternal
PFHxS)

Fecundability

NS (p=0.29)

Not becoming pregnant
within first six cycles

OR 0.98 (0.54–1.77)

Infertility
OR 1.4 (1.0–2.0)*, 3rd quartile
Parous subgroup
OR 2.1 (1.2–3.8)*, 4th quartile
Primiparous subgroup
OR 0.7 (0.4–1.3), 4th quartile

Fecundability
Infertility risk

FR 1.00 (0.99–1.01), per 0.1 ng/mL
OR 0.98 (0.93–1.03), per 0.1 ng/mL

Fecundability
Infertility

FR 0.97 (0.85–1.11)
OR 0.99 (0.73–1.33)

Fecundability
Infertility

FOR 0.91 (0.86–0.97, p=0.002)*
OR 1.27 (1.09–1.48, p=0.003)*

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

321
2. HEALTH EFFECTS

Table 2-21. Summary of Fertility Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Vestergaard et al. 2012

1.12 and 1.22 ng/mL
(median PFHxS in women
General population (n=222 nulliparous couples) with no pregnancy and
pregnant)
PFNA
Bach et al. 2015a
0.8 ng/mL (median maternal)
General population (n=1,372 pregnant women)
Jørgensen et al. 2014a, 2014b

0.64 ng/mL (median PFNA)

General population (n=938 pregnant women)
Vestergaard et al. 2012

0.45 and 0.51 ng/mL
(median PFNA in women
General population (n=222 nulliparous couples) with no pregnancy and
pregnant)
PFDeA
Bach et al. 2015a
0.3 ng/mL (median maternal
PFDeA)
General population (n=1,372 pregnant women)
Vestergaard et al. 2012
0.10 and 0.11 ng/mL
(median PFDeA in women
General population (n=222 nulliparous couples) with no pregnancy and
pregnant)
PFUA
Bach et al. 2015a
General population (n=1,372 pregnant women)
PFOSA
Vestergaard et al. 2012

Fecundability

OR 1.33 (1.01–1.75)*

Not becoming pregnant
within first six cycles

OR 0.67 (0.37–1.20)

Fecundability
Infertility risk

FR 1.00 (0.98–1.02), per 0.1 ng/mL
OR 0.99 (0.95–1.03), per 0.1 ng/mL

Fecundability
Primiparous subgroup
Infertility
Fecundability
Not becoming pregnant
within first six cycles

FR 0.80 (0.69–0.94)*
FR 0.99 (0.88–1.22)
OR 1.53 (1.08–2.15)*
OR 1.17 (0.88–1.54)
OR 0.67 (0.37–1.25)

Fecundability
Infertility risk

FR 1.00 (0.97–1.03), per 0.1 ng/mL
OR 0.99 (0.92–1.07), per 0.1 ng/mL

Fecundability
Not becoming pregnant
within first six cycles

OR 1.15 (0.89–1.49)
OR 0.61 (0.33–1.12)

0.3 ng/mL (median maternal Fecundability
PFUA)
Infertility risk

0.10 and 0.11 ng/mL
(median PFOSA in women
General population (n=222 nulliparous couples) with no pregnancy and
pregnant)

Resultc

Fecundability
Not becoming pregnant
within first six cycles

***DRAFT FOR PUBLIC COMMENT***

FR 1.01 (0.98–1.03), per 0.1 ng/mL
OR 0.98 (0.92–1.04), per 0.1 ng/mL

OR 1.01 (0.86–1.18)
OR 0.81 (0.45–1.46)

 PERFLUOROALKYLS

322
2. HEALTH EFFECTS

Table 2-21. Summary of Fertility Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Me-PFOSA-AcOH
Vestergaard et al. 2012

0.45 and 0.39 ng/mL
Fecundability
(median Me-PFOSA-AcOH Not becoming pregnant
General population (n=222 nulliparous couples) in women with no pregnancy within first six cycles
and pregnant)
Et-PFOSA-AcOH
Vestergaard et al. 2012
2.12 and 1.79 ng/mL
Fecundability
(median Et-PFOSA-AcOH in Not becoming pregnant
General population (n=222 nulliparous couples) women with no pregnancy
within first six cycles
and pregnant)

OR 0.99 (0.82-–1.18)
OR 1.23 (0.68–2.20)

OR 0.79 (0.62–1.00)
OR 1.53 (0.85–2.77)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 12 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; FOR = fecundability odds ratio; FR = fecundability ratio (probability of conceiving during a
given menstrual cycle); OR = odds ratio; Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane sulfonamide) acetic acid; NS = not significant;
PFDeA = perfluorodecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

323
2. HEALTH EFFECTS

are presented in Tables 2-1, 2-3, 2-4, 2-5, and 2-6 and in Figures 2-4, 2-6, 2-7, and 2-8. Multigeneration
studies on PFOA, PFOS, and PFBuS have not found alterations in reproductive parameters in animals;
similarly, no effect on fertility was observed for PFHxS. An increase in the incidence of Leydig cell
hyperplasia (reclassified as gonodal stromal hyperplasia) has been observed in animals exposed to PFOA;
one study for PFDoA reported ultrastructural alterations in the testes. Studies on PFOS, PFHxS, PFBuS,
and PFBA have not found histological alterations. Delays in mammary gland development have been
observed in mice exposed to PFOA; this effect has also been observed in perinatally exposed mice (see
Section 2.17, Developmental).
PFOA

Epidemiology Studies—Reproductive Hormone Levels. Two studies have evaluated potential effects of
PFOA exposure on reproductive hormone levels in workers (Olsen et al. 1998b; Sakr et al. 2007b). Sakr
et al. (2007b) found associations between serum PFOA and estradiol and testosterone levels male workers
at the Washington Works facility. In contrast, Olsen et al. (1998b) did not find associations between
serum PFOA and estradiol or testosterone in male workers at the 3M Cottage Grove facility. The study
did find an association with prolactin levels, but this was only found in workers examined in 1993, but
not in those examined in 1995. In a general population study of men aged 30–66 years of age,
correlations were found between serum PFOA levels and free testosterone levels and LH levels; no
correlations were found for estradiol, prolactin, follicle stimulating hormone (FSH), or total testosterone
levels (Raymer et al. 2012). Another study of similar aged men did not find an association between
serum PFOA and sex hormone binding globulin levels (Specht et al. 2012). Studies of young men
(median age 19 years) (Joensen et al. 2013) or adolescents and young men (12–30 years of age) (Tsai et
al. 2015) did not find associations between serum PFOA and reproductive hormone levels. A third study
(Vested et al. 2013) found an association between LH and FSH levels and maternal serum PFOA levels in
young adult males; other hormones were not affected.

Two studies of women (Barrett et al. 2015; Knox et al. 2011b) did not find associations with estradiol
levels or luteal progesterone levels. A third study of adolescent and young women (Tsai et al. 2015)
found an association between serum PFOA and sex hormone binding globulin levels in adolescents (12–
17 years), but not in young adults; no associations with FSH or testosterone were observed in either
group.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

324
2. HEALTH EFFECTS

Epidemiology Studies—Effects on Sperm. Six general population studies have evaluated the potential
alterations in sperm parameters associated with PFOA exposure. Although some associations have been
found, the results are not consistent across studies. Buck Louis et al. (2015) reported an increase in
curvilinear velocity and some alterations in sperm morphology that were associated with serum PFOA
levels. Toft et al. (2012) found a PFOA-related increase in the percentage motile sperm in men with
serum PFOA levels in the 3rd tertile. Vested et al. (2013) reported inverse associations between maternal
serum PFOA levels and sperm concentration and total sperm count in young adults; no alterations in
motility or morphology were observed. Other studies did not find alterations in sperm viability, count,
concentration, motility, or morphology (Buck Louis et al. 2015; Joensen et al. 2013; Raymer et al. 2012;
Toft et al. 2012) or the Y-X chromosome ratio (Kvist et al. 2012).

Epidemiology Studies—Effects on Menstrual Cycle Length, Menopause Onset, Endometriosis, and
Breastfeeding Duration. One study examined possible associations between serum PFOA levels and
alterations in menstrual cycle length (Lyngsø et al. 2014). An increased risk of a long menstrual cycle
(≥32 days) was observed in women with serum PFOA levels in the 3rd tertile and when serum PFOA was
used as a continuous variable. No alterations in the risk of having a short menstrual cycle (≤24 days) or
irregular menstrual cycles (≥7 days difference between cycles) were observed. Four studies have
evaluated the risk of early menopause. In a study of C8 Health Study participants, increases in the risk of
early menopause was observed in perimenopausal (>42–≤51 years of age) and menopausal (>51–
≤65 years of age) women with serum PFOA levels in the 2nd, 3rd, 4th, and 5th quintiles (Knox et al. 2011b).
An increase in menopause risk was also observed in a cross-sectional study of NHANES participants with
serum PFOA levels in the 3rd tertile (Taylor et al. 2014). Taylor et al. (2014) also found a higher risk of
hysterectomy among women with serum PFOA levels in the 2nd and 3rd tertiles. An analysis by Taylor et
al. (2014) of possible reverse causality showed that serum PFOA levels increased after menopause. In
contrast, no alterations in the risk of early menopause or age of menopause were associated with
cumulative serum PFOA levels in retrospective and prospective studies of C8 Health Study participants
(Dhingra et al. 2016a); age of menopause was also not associated with measured serum PFOA levels in
the prospective study (Dhingra et al. 2016a). Buck Louis et al. (2012) showed that the risk of
endometriosis and the risk of moderate-to-severe endometriosis were associated with serum PFOA levels;
however, adjustment for parity resulted in confidence intervals that included unity. A case-control study
(Vagi et al. 2014) found an increased risk of polycystic ovary syndrome among women with serum PFOA
levels in the 3rd tertile.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

325
2. HEALTH EFFECTS

Two studies have evaluated a possible link between maternal PFOA levels and breastfeeding duration.
Both studies found increases in the risk of breastfeeding ≤3 or 6 months that were associated with
maternal PFOA levels (Fei et al. 2010; Romano et al. 2016). When the women were segregated by parity,
the associations were only found in multiparous women (Fei et al. 2010). This finding may be due to
reverse causality since shorter breastfeeding would likely result in lower maternal transfer of PFOA.

Epidemiology Studies—Effects on Fertility. Several general population studies have examined the
possible association between female serum PFOA levels and decreased fertility or infertility; the results
are graphically presented in Figures 2-29 and 2-30, respectively. With the exception of the Vestergaard et
al. (2012) study, all of the women were pregnant; thus, couples with unresolved infertility are
underrepresented in these analyses. No studies were identified that used paternal serum PFOA levels as
the biomarker of exposure. Most of the studies evaluated two aspects of fertility: fecundability, which is
a measure of time to pregnancy, and risk of infertility, which is typically time to pregnancy of
>12 months.

In a study of pregnant women participating in the Danish National Birth Cohort study, a decrease in
fecundability and an increase in infertility were observed in women with serum PFOA (measured at
gestation week 12) levels ≥3.91 ng/mL (Fei et al. 2009). When the women were categorized by parity,
decreased fecundability OR and increased infertility OR were only found in the parous group; the ORs for
the nulliparous women included unity (Fei et al. 2009). A second re-analysis of these data (Bach et al.
2015a) using a different statistical approach confirmed the results of the whole group and the parous
subgroup; this re-analysis also found a decrease in the fecudability risk among the nulliparous women. In
another set of women participating in the Danish National Birth Cohort study (Bach et al. 2015b), no
alterations in fecudability or infertility risk were observed in the whole cohort or when the women were
categorized into parous and nulliparous subcohorts. It was noted that the median serum PFOA levels in
this second study (4.0 ng/mL) were lower than the levels in the larger study (5.4 ng/mL). A decrease in
fecundability and an increase in infertility risk were also observed in a Canadian study of pregnant
women (Vélez et al. 2015). An increase in infertility risk was also found in a Norwegian study of
subfecund pregnant women with serum PFOA levels in the three highest quartiles (Whitworth et al.
2012b); when the women were categorized based on parity, the infertility risk was only elevated in the
parous women with serum PFOA levels in the 3rd and 4th quartiles. A multinational study also found an
alteration in fecudability (Jørgensen et al. 2014a); however, this study found that higher serum PFOA
levels resulted in a decrease in the time to pregnancy (fecundability ratio >1) among primiparous women.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

326
2. HEALTH EFFECTS

Figure 2-29. Fecundability Relative to PFOA Levels (Presented as Adjusted Fecundability Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

327
2. HEALTH EFFECTS

Figure 2-30. Infertility Relative to PFOA Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

328
2. HEALTH EFFECTS

Other studies have not found alterations in fecudability and/or infertility (Bach et al. 2015a; Jørgensen et
al. 2014a; Vestergaard et al. 2012).

Laboratory Animal Studies. Examination of the testes and epididymides of rats exposed intermittently
head-only to up to 84 mg/m3 APFO dusts for 2 weeks did not reveal any gross or microscopic treatmentrelated alterations (Kennedy et al. 1986).

Several studies have been conducted in rats to examine whether induction of Leydig cells tumors could be
due to an endocrine-related mechanism. In a 14-day gavage study in which rats were dosed with up to
50 mg/kg/day PFOA, testes weight was not significantly affected and microscopic examination did not
reveal any significant alterations (Cook et al. 1992). However, the weight of the accessory sex organ unit
(ventral and dorsal lateral prostate, seminal vesicles, and coagulating glands) was significantly decreased
in rats dosed with 25 mg/kg/day PFOA (17% decrease) and 50 mg/kg/day PFOA (18% decrease) relative
to controls and to a pair-fed group. There was also a trend for reduced serum and interstitial fluid
testosterone in PFOA-treated rats; serum LH was not altered and estradiol was significantly increased
(63%) at ≥10 mg/kg/day. Challenge experiments conducted with human chorionic gonadotropin,
gonadotropin-releasing hormone, or naloxone suggested that the decrease in serum testosterone was due
to a lesion at the level of the testes. Serum levels of progesterone and 17α-hydroxyprogesterone were not
altered by 50 mg/kg/day PFOA, but androstenedione levels were reduced 2-fold. The data suggested that
the decrease in serum testosterone may be due to a decrease in the conversion of 17α-hydroxyprogesterone to androstenedione, and this could be attributed to the elevated serum levels of estradiol.
The decrease in weight of the accessory sex organ unit could also be attributed to the elevated estradiol
serum levels. In a subsequent study from the same group of investigators, rats dosed with 25 mg/kg/day
PFOA for 14 days showed a significant increase in estradiol in serum and in testicular interstitial fluid
relative to controls (Biegel et al. 1995). Treatment with PFOA for 14 days significantly increased
aromatase activity in the liver (aromatase converts testosterone to estradiol), but not in testes, muscle, or
adipose tissue, suggesting that PFOA increases serum estradiol by inducing aromatase activity in the
liver. Treatment with PFOA also increased testicular interstitial fluid transforming growth factor α
(TGFα). Collectively, the results were consistent with the hypothesis that increased estradiol levels
ultimately produce Leydig cell hyperplasia and adenoma by acting as a mitogen or enhancing growth
factor secretion. A study of the dose-response relationship for PFOA and serum estradiol reported a
significant increase in serum estradiol in rats dosed with ≥2 mg/kg/day, which was well correlated with
total hepatic aromatase activity (Liu et al. 1996). Significant increases in serum estradiol were also

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

329
2. HEALTH EFFECTS

reported during the first year of treatment of male rats with 13.6 mg/kg/day PFOA in a 2-year dietary
study (Biegel et al. 2001).

Significant increases in the incidence of Leydig cell hyperplasia were observed in rats exposed to
13.6 mg/kg/day PFOA in the diet for 2 years (Biegel et al. 2001). Another 2-year study found an
increased incidence of vascular mineralization in the testes of rats exposed to 15 mg/kg/day PFOA in the
diet; no effects were observed at 1.5 mg/kg/day (3M 1983). In female rats, increases in the incidence of
tubular hyperplasia of the ovaries were observed following a 2-year exposure to 1.5 mg/kg/day (3M
1983). A peer review of the histological slides from 3M (1983) concluded that the more current
nomenclature for the tubular hyperplasia was gonadal stromal hyperplasia (Mann and Frame 2004).
Additionally, the peer reviewers substantially disagreed with the incidence of lesions in the 1.5 mg/kg/day
group and slightly disagreed with the incidence in the 15 mg/kg/day group. Based on the incidence
reported by the peer reviewers, no statistically significant increases in the occurrence of gonadal stromal
hyperplasia were observed in either group; a significant increase in grade 3 and above lesions were
observed in the 15 mg/kg/day group.

In a 2-generation reproduction study in which male and female rats were dosed with up to 30 mg/kg/day
PFOA by gavage in water for 70 days before mating and until sacrifice, there were no effects on estrous
cycling, sperm number and quality, mating and fertility, or histopathology of the reproductive organs
assessed in the parental and F1 generations (Butenhoff et al. 2004b). Intermediate-duration studies of rats
and monkeys also did not find gross or microscopic alterations in the sex organs at termination;
Cynomolgus monkeys were dosed with up to 20 mg/kg/day PFOA for 4 or 26 weeks (Butenhoff et al.
2002; Thomford 2001), Rhesus monkeys with up to 100 mg/kg/day PFOA for 13 weeks (Griffith and
Long 1980), and rats with up to approximately 100–110 mg/kg/day PFOA for 13 weeks (Griffith and
Long). Serum levels of estradiol and estriol were not significantly altered in the 4-week study conducted
by Thomford (2001), but estrone was reduced in monkeys dosed with 2 and 20 mg/kg/day PFOA; no
possible explanation was discussed. In the 26-week study (Butenhoff et al. 2002), no treatment-related
alterations were reported in serum estrone, estriol, estradiol, or testosterone, indicating that the reduced
serum estrone levels in the 4-week study was transitory. In 2-year dietary studies in rats, doses of
13.6 mg/kg/day PFOA significantly increased the incidence of Leydig cell hyperplasia (Biegel et al.
2001), whereas 15 mg/kg/day increased the incidence of vascular mineralization in the testes and
1.5 mg/kg/day increased the incidence of tubular hyperplasia in the ovaries (3M 1983).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

330
2. HEALTH EFFECTS

A study in pregnant mice dosed with 5 mg/kg/day PFOA (only dose level tested) reported that the
mammary gland showed changes suggesting substantial delay (possibly up to 10 days) in gland
differentiation on PND 20 and alterations in milk protein gene expression on PND 20 (White et al. 2007).
Subsequent studies by this group support the finding of delayed mammary gland differentiation. On
PND 1, the mammary glands of mice administered 5 mg/kg/day on GDs 8–17 appeared immature; the
morphology was similar to that seen in late pregnancy prior to parturition and the initiation of nursing
(White et al. 2009). Another study found that the normal weaning-induced mammary gland involution
was compromised on PND 22 in mice exposed to 1 mg/kg/day on GDs 1–17 or 0.001 mg/kg/day
administered on GD 7–PND 22 (White et al. 2011); the investigators noted that the mammary gland
structure was similar to mammary gland tissue at or near the peak of lactation (PND 10). Necrosis was
observed in the placenta of mice administered via gavage 10 or 25 mg/kg/day PFOA on GDs 11–16 (Suh
et al. 2011); no alterations were observed at 2 mg/kg/day.

No gross or microscopic alterations were reported in the testes from rats dermally exposed to
2,000 mg/kg/day APFO (Kennedy 1985).

Summary. Epidemiology studies have examined a several types of reproductive endpoints. Due to
inconsistent results, the available data are not suitable for determining whether there are associations
between serum PFOA and reproductive hormones or effects on sperm. There is some suggestive
evidence that increases in serum PFOA levels can result in earlier onset of menopause; however, this is
based on the findings of two studies (a third study did not find an association). Several general
population studies found associations between serum PFOA and impaired fertility (increased time to
pregnancy and/or infertility). The results of a multi-generational study in rats do not suggest that the
reproductive system is a sensitive target of PFOA toxicity. Additionally, histological alterations have not
been observed in monkeys or rats following intermediate and/or chronic oral exposure.
PFOS

Epidemiology Studies—Reproductive Hormone Levels. In an occupational exposure study of workers at
3M Decatur and Antwerp facilities (Olsen et al. 1998a) and a general population study (Raymer et al.
2012), no associations between serum PFOS and reproductive hormones were found. Studies in
adolescent and young adult males have found inverse associations between serum PFOS levels and total
and free testosterone levels (Joensen et al. 2013), free androgen index (Joensen et al. 2013), and FSH

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

331
2. HEALTH EFFECTS

levels (Tsai et al. 2015). Another study of young men did not find alterations in reproductive hormone
levels (Vested et al. 2013).

In a study of females participating in the C8 Health Studies, serum PFOS levels were inversely associated
with estradiol levels in both perimenopausal and menopausal women (Knox et al. 2011b). An inverse
association with follicular estradiol levels was also observed in a general population study (Barrett et al.
2015); when segregated by parity, the inverse association was only found in nulliparous women. An
inverse association between serum PFOS levels and testosterone levels was observed in adolescent
females; no association was found in older females (Tsai et al. 2015).

Epidemiology Studies—Effects on Sperm. The available general population data do not provide
evidence that PFOS damages sperm. One study (Buck Louis et al. 2015) found an association for one
measure of sperm motility (distance travelled) but not for other measures. Another study (Toft et al.
2012) found an inverse association between serum PFOS levels and percentage of normal sperm. Other
studies have not found alterations in sperm viability, count, motility, volume, or morphology (Buck Louis
et al. 2015; Joensen et al. 2013; Raymer et al. 2012; Toft et al. 2012; Vested et al. 2013). A multinational
study (Kvist et al. 2012) found a nonlinear association between serum PFOS and Y-X chromosome ratio;
however, when categorized by country, the only significant trend was a negative trend in the Greenland
cohort. It is noted that these are studies of individuals exposed to background levels of PFOS, involved a
single measurement of PFOS, and are not adequate for establishing causality.

Epidemiology Studies—Effects on Menstrual Cycle Length, Menopause Onset, Endometriosis, and
Breastfeeding Duration. No alterations in the risk of irregular, short, or long menstrual cycle lengths
associated with serum PFOS levels were observed in a study of pregnant women (Lyngsø et al. 2014). A
study of C8 Health Study participants found increases in the risk of early menopause in perimenopausal
and menopausal women with serum PFOS levels in the ≥3rd and ≥2nd quintiles, respectively (Knox et al.
2011b). In contrast, a study of NHANES participants did not find an association between serum PFOS
and the risk of early menopause (Taylor et al. 2014). The risk of endometriosis was not associated with
serum PFOS levels (Buck Louis et al. 2012). However, there was a greater risk of having moderate to
severe endometriosis; adjusting for parity decreased the risk and the CIs included unity. General
population studies found increases in the risk of having a hysterectomy in women having serum PFOS
levels in the 2nd and 3rd tertiles (Taylor et al. 2014) and the risk of having polycystic ovary syndrome in
women with serum PFOS levels in the 3rd tertile (Vagi et al. 2014). Most of these endpoints were only

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

332
2. HEALTH EFFECTS

examined in one study and the evidence is inconclusive to determine whether there is an association
between PFOS exposure and these female reproductive outcomes.
Maternal serum PFOS levels have been associated with increases in the risk of breastfeeding for ≤3 or
6 months (Fei et al. 2010; Romano et al. 2016). When the women were segregated by parity, the
associations were only found in multiparous women (Fei et al. 2010). This finding may be due to reverse
causality since longer breastfeeding would likely result in lower maternal PFOS levels.

Epidemiology Studies—Effects on Fertility. Several general population studies have evaluated whether
there is a possible link between serum PFOS and time-to-pregnancy (as measured using a fecundability
ratio) or infertility; graphical presentations of potential associations between fecundability and infertility
relative to serum PFOA levels are presented in Figures 2-31 and 2-32, respectively. A couple of studies
have found associations, but most have not found associations. Fei et al. (2009) found decreases in
fecundability and increases in infertility risk among pregnant women with serum PFOS levels in the top
three quartiles. When the women were categorized by parity (Fei et al. 2012), the decrease in
fecundability and increase in infertility risk were only observed in nulliparous women with serum PFOS
levels in the 3rd and 4th quartiles; no alterations were observed among parous women. A re-analysis of
these data using a different statistical approach (Bach et al. 2015b) resulted in the same findings of
associations between PFOS and fecundability and infertility. Whitworth et al. (2012b) also found an
increased risk of infertility among subfecund women with serum PFOS levels in the 3rd quartile;
categorizing by parity resulted in increases in only parous women with serum PFOS levels in the
4th quartile. In contrast, other studies have not found alterations in fecundability or fertility associated
with maternal serum PFOS levels (Bach et al. 2015a, 2015b; Jørgensen et al. 2014a; Vélez et al. 2015;
Vestergaard et al. 2012).

Laboratory Animal Studies. A significant decrease in serum testosterone levels and epididymal sperm
count was observed in mice administered 10 mg/kg/day PFOS for 21 days (Wan et al. 2011). No
alterations were observed in mice administered 5 mg/kg/day PFOS or in mice administered 5 or
10 mg/kg/day PFOS for 14 days. No alterations in reproductive performance (number of litters, gestation
length, number of implantation sites, or potential resorptions) were observed in rats administered
1 mg/kg/day PFOS throughout gestation and lactation (Buttenoff et al. 2009b). Lee et al. (2015a) did find
a decrease in placental weight and placental capacity (ratio of fetal weight to placental weight) in mice
administered ≥0.5 mg/kg/day PFOS via gavage on GDs 11–16. Multigeneration studies with PFOS in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

333
2. HEALTH EFFECTS

Figure 2-31. Fecundability Relative to PFOS Levels (Presented as Adjusted Fecundability Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

334
2. HEALTH EFFECTS

Figure 2-32. Infertility Relative to PFOS Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

335
2. HEALTH EFFECTS

rats did not provide indications of reproductive toxicity. Exposure of male and female rats to up to
3.2 mg/kg/day PFOS by gavage before mating and continuing during gestation did not affect mating or
fertility parameters of the parental or F1 generation (Luebker et al. 2005a, 2005b). Dietary exposure of
rats to 1.3–1.8 mg/kg/day PFOS for 4 or 14 weeks did not induce gross or microscopic alterations in the
sex organs of males or females (Seacat et al. 2003). A similar study in Cynomolgus monkeys
administered up to 0.75 mg/kg/day PFOS administered via a capsule also reported no significant
morphological alterations in the sex organs, but serum estradiol was significantly decreased in males on
days 62, 91, and 182 of the study (Seacat et al. 2002). In addition, treatment with PFOS had no
significant effect on cell proliferation in the testes. Serum estradiol also was lower than in controls in one
male and one female monkey dosed with 2 mg/kg/day PFOS for 4 weeks, but little can be concluded from
results from just two animals (Thomford 2002a). In a 2-year dietary study in rats, administration of up to
1.04 mg/kg/day PFOS did not induce gross or microscopic alterations in the reproductive organs
(Butenhoff et al. 2012b; Thomford 2002b). Overall, the reproductive system does not seem to be a
sensitive target of PFOS toxicity, although some changes in testosterone and estradiol levels have been
observed.
PFHxS

Epidemiology Studies—Reproductive Hormone Levels. Two general population studies evaluated
possible effects of PFHxS on reproductive hormone levels. In young men, no associations between serum
PFHxS levels and testosterone, free androgen index, LH, estradiol, sex hormone binding globulin, or FSH
levels were found (Joensen et al. 2013). Similarly, no alterations in follicular estrogen or luteal
progesterone were observed in women (Barrett et al. 2015).

Epidemiology Studies—Effects on Sperm. With the exception of the finding of an inverse association
between serum PFHxS levels and percent normal sperm (Toft et al. 2012), general population studies
have not found associations between PFHxS and sperm parameters (Joensen et al. 2013; Toft et al. 2012);
it is noted that the Joensen et al. (2013) study of young men did not find alterations in sperm morphology.

Epidemiology Studies—Effects on Menstrual Cycle Length, Menopause Onset, Endometriosis, and
Breastfeeding Duration. Four general population studies have evaluated possible associations between
serum PFHxS levels and female reproductive outcomes. Taylor et al. (2014) reported increases in the risk
of earlier menopause in women with serum PFHxS levels in the 3rd tertile and the risk of hysterectomy in
women with serum PFHxS levels in the 2nd and 3rd tertiles. Other studies did not find associations with

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

336
2. HEALTH EFFECTS

the risk and severity of endometriosis (Buck Louis et al. 2012) or polycystic ovary syndrome (Vagi et al.
2014). Romano et al. (2016) did not find associations between maternal PFHxS levels and the risk of
breastfeeding ≤3 or 6 months.
Epidemiology Studies—Effects on Fertility. Four studies have evaluated possible effects on fertility
associated with female serum PFHxS levels. Vélez et al. (2015) found increases in time to pregnancy
(measured as a decreased fecundability OR) and risk of infertility, which were associated with serum
PFHxS levels in pregnant women. Vestergaard et al. (2012) reported an increase in the fecudability OR,
indicating a shorter time to pregnancy, when risk was calculated using continuous serum PFHxS;
however, when the subjects were divided into two groups based on serum PFHxS levels above and below
the median level, the fecundability ratio included unity in the above-median group (fecundability ratio
1.29, 95% CI 0.90–1.83), as compared to the below-median group. Studies by Bach et al. (2015a) and
Jørgensen et al. (2014a) did not find alterations in time to pregnancy or the risk of infertility.

Laboratory Animal Studies. Exposure to 10 mg/kg/day PFHxS did not result in alterations in
reproductive organ weights or histopathology in male rats exposed for a minimum of 42 days beginning
14 days prior to cohabitation and female rats sacrificed on lactation day 21 or GD 25 (rats that did not
deliver a litter) (exposure began 14 days prior to cohabitation) (Butenhoff et al. 2009a; Hoberman and
York 2003). Fertility was not affected by treatment with PFHxS and there were no significant effects on
sperm parameters. Also, estrous cycling was not affected by dosing with PFHxS.
PFNA

Epidemiology Studies—Reproductive Hormone Levels. Reproductive hormone alterations associated
with serum PFNA levels are limited to a finding for estradiol in young men (Joensen et al. 2013); no
associations with other reproductive hormones were found in this study. In another study of adolescent
and young adults, no associations between serum PFNA and sex hormone binding globulin, FSH, or
testosterone were found in males or females (subjects were segregated by sex and age range) (Tsai et al.
2015). A third study did not find alterations in follicular estradiol or luteal progesterone levels in women
(Barrett et al. 2015).

Epidemiology Studies—Effects on Sperm. Buck Louis et al. (2015) found associations between serum
PFNA and increases in the percentage of normal sperm and a decrease in the percentage of sperm with

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

337
2. HEALTH EFFECTS

coiled tails. No associations were found for other sperm parameters (Buck Louis et al. 2015; Joensen et
al. 2013; Toft et al. 2012).

Epidemiology Studies—Effects on Menstrual Cycle Length, Menopause Onset, Endometriosis, and
Breastfeeding Duration. Increases in the risk of earlier menopause and hysterectomy were found in
women with serum PFNA levels in the 3rd and ≥2nd serum PFNA tertiles (Taylor et al. 2014). The
investigators examined the possibility that these effects may be due to reverse causation and found that
serum PFNA levels increased post-menopause (Taylor et al. 2014). An increase in the risk of
endometriosis was associated with serum PFNA levels in a general population study (Buck Louis et al.
2012); however, adjustment for parity resulted in OR CIs that included unity. Vagi et al. (2014) did not
find an increased risk of polycystic syndrome that was associated with serum PFNA levels. No
associations between maternal PFNA levels and the risk of breastfeeding ≤3 or 6 months were found in a
general population study (Romano et al. 2016).

Epidemiology Studies—Effects on Fertility. Jørgensen et al. (2014a) found increases in time to
pregnancy (measured as a decrease in fecundability ratio) and an increase in infertility risk in a study of
pregnant women. In sensitivity analysis, the fecundability ratio for primiparous women was 0.99 and the
95% CI range included unity (0.88–1.22). Studies by Bach et al. (2015a) and Vestergaard et al. (2012)
did not find associations between serum PFNA levels and fecundability ratio; Bach et al. (2015a) also did
not find an increase in the risk of infertility.

Laboratory Animal Studies. Two acute-duration studies have evaluated the reproductive toxicity of
PFNA in male rats (Feng et al. 2009, 2010). Gavage administration of 5 mg/kg/day for 14 days resulted
in decreases in serum testosterone and increases in serum estradiol levels and atrophy of the seminiferous
tubules (Feng et al. 2009). Electron microscopic examination of the testes revealed large vacuoles
between the Sertoli cells and spermatogonia; these changes as well as increases in serum Mullerian
inhibiting substance and decreases in serum inhibin B cells were suggestive of damage to the secretory
function of the Sertoli cells (Feng et al. 2010).
PFDeA

Epidemiology Studies—Reproductive Hormone Levels. No associations were found between serum
PFDeA levels and testosterone, free androgen index, LH, estradiol, sex hormone binding globulin, or FSH

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

338
2. HEALTH EFFECTS

levels in young men (Joensen et al. 2013). Similarly, no alterations in follicular estradiol or luteal
progesterone levels were observed in women (Barrett et al. 2015).

Epidemiology Studies—Effects on Sperm. Two general population studies evaluated potential effects of
PFDeA exposure on sperm parameters. Buck Louis et al. (2015) found associations between serum
PFDeA levels and increases in sperm head length and decreases in the percentage of sperm with coiled
tails. No alterations were found for sperm viability, count, volume, motility, or other morphological
alterations (Buck Louis et al. 2015; Joensen et al. 2013).

Epidemiology Studies—Effects on Menstrual Cycle Length, Menopause Onset, and Endometriosis.
Only one study examined alterations in female reproductive outcomes associated with serum PFDeA
levels. In this study, no associations between serum PFDeA levels and the risk or severity of
endometriosis were found (Buck Louis et al. 2012).

Epidemiology Studies—Effects on Fertility. Two studies have examined the potential for PFDeA to alter
fertility. No alterations in time to pregnancy (measured as fecundability ratio) or risk of infertility were
observed in pregnant women (Bach et al. 2015a). Another study (Vestergaard et al. 2012) also found no
association between female serum PFDeA levels and time to pregnancy.
PFUA

Epidemiology Studies—Reproductive Hormone Levels. An inverse association between serum PFUA
levels and FSH levels was observed in adolescent girls (Tsai et al. 2015). The study did not find
alterations in sex hormone binding globulins or testosterone levels in adolescent and young adult males or
females. Another study of women did not find alterations in follicular estradiol or luteal progesterone
levels (Barrett et al. 2015).

Epidemiology Studies—Effects on Fertility. Only one study evaluated possible associations between
maternal serum PFUA levels and fertility; no alterations in time to pregnancy (measured as a
fecundability ratio) or infertility risk were observed in pregnant women (Bach et al. 2015a).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

339
2. HEALTH EFFECTS

PFBuS

Laboratory Animal Studies. Administration of up to 900 mg/kg/day PFBuS to rats by gavage for 28 days
did not cause any significant gross or microscopic alterations in primary or secondary sex organs from
males or females (3M 2001). A 2-generation study in which rats were exposed to gavage doses of
potassium PFBuS as high as 1,000 mg/kg/day did not result in alterations in fertility, sperm parameters,
estrus cycling, or histological alterations in reproductive tissues (Lieder et al. 2009b).
PFBA

Laboratory Animal Studies. No significant gross or microscopic alterations were reported in primary and
secondary reproductive organs from rats dosed with PFBA by gavage in doses of up to 184 mg/kg/day for
5 days, 150 mg/kg/day for 28 days, or 30 mg/kg/day for 90 days (3M 2007a; Butenhoff et al. 2012a; van
Otterdijk 2007a, 2007b).

PFDoA

Laboratory Animal Studies. Treatment of male rats with 1, 5, or 10 mg/kg/day PFDoA by gavage for
14 days induced a dose-related decrease in testes weight, which achieved statistical significance at
10 mg/kg/day (Shi et al. 2007). Measurement of serum hormone levels showed a significant decrease in
LH at 10 mg/kg/day and in testosterone at 5 and 10 mg/kg/day, no significant effect on FSH levels, and a
significant decrease in serum estradiol only at 5 mg/kg/day. Alterations in the ultrastructure of the testes
were seen in the 5 and 10 mg/kg/day groups and consisted of the presence of large clustered lipid droplets
and enlarged mitochondria in Sertoli cells, large vacuoles, and expanded mitochondria in Leydig and
spermatogenic cells. Morphological features of apoptosis were seen in cells in the 10 mg/kg/day group.
Assessment of messenger ribonucleic acid (mRNA) expression of genes involved in cholesterol transport
and steroidogenesis provided evidence of altered cholesterol transport and steroid hormone synthesis, but
no effects were noted for LH receptor and aromatase mRNA expression. Considering that serum total
cholesterol was unaffected at 5 mg/kg/day and increased at 10 mg/kg/day and that aromatase expression
was unaffected, the decrease in testosterone synthesis probably resulted from decreased steroidogenesis
gene expression.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

340
2. HEALTH EFFECTS

PFOSA

Epidemiology Studies. One study examined reproductive hormone levels and did not find an association
between serum PFOSA and follicular estradiol or luteal progesterone levels in women (Barrett et al.
2015). The only epidemiology study evaluating fertility did not find an increase in time to pregnancy, as
measured as a fecundability ratio, or decrease in the likelihood of becoming pregnant within the first six
menstrual cycles (Vestergaard et al. 2012).
Me-PFOSA-AcOH

Epidemiology Studies. In the only study identified that examined sperm parameters associated with
Me-PFOSA-AcOH, increases in the percentage of sperm with neck/midpiece abnormalities and increases
in the number of immature sperm were found (Buck Louis et al. 2015). No associations with sperm
viability, count, or motility were found. One study examining fertility did not find associations between
serum Me-PFOSA-AcOH levels in women and time to pregnancy (Vestergaard et al. 2012).
Et-PFOSA-AcOH

Epidemiology Studies. No associations between serum Et-PFOSA-AcOH levels in females and the time
to pregnancy were observed in a study of nulliparous couples (Vestergaard et al. 2012).
2.17 DEVELOPMENTAL
Overview. A large number of epidemiology studies have examined the potential of developmental
toxicity of perfluoroalkyls in the general population and in populations living in an area with high PFOA
drinking water contamination. The discussion of these developmental outcomes is divided into four
categories: pregnancy outcome, birth outcome, neurodevelopment, and sexual maturation. The
epidemiology studies examining pregnancy outcome are summarized in Table 2-22; the pregnancy
outcomes include miscarriage, stillbirth, preterm birth, and gestation age. Table 2-23 summarizes the
epidemiology studies examining birth outcomes, which include birth weight, birth size, low birth weight,
small for gestational age, birth defects, and sex ratio. Epidemiology studies examining
neurodevelopmental endpoints, particularly risks for ADHD, are summarized in Table 2-24. Studies
evaluating possible links between serum perfluoroalkyl levels and development of the reproductive
system are summarized in Table 2-25. Further details on these studies are presented in the Supporting

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

341
2. HEALTH EFFECTS

Table 2-22. Summary of Pregnancy Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

≥37.2 ng/mL (5th PFOA
quintile)

Preterm birth

OR 1.01 (0.55–1.86)

>39.4 ng/mL (5th PFOA
quintile)

Miscarriage risk
Parous subgroup
Nulliparous subgroup
Miscarriage
Stillbirth

OR 1.00 (0.63–1.58), 5th quintile
OR 1.06 (0.57–1.97), 5th quintile
OR 0.81 (0.38–1.71), 5th quintile
OR 0.9 (0.7–1.0)
OR 1.0 (0.5–1.8)

21.0–717.6 ng/mL
(5th maternal PFOA quintile)

Stillbirth
Preterm birth (<37 weeks)
Preterm birth (<32 weeks)

OR 0.8 (0.5–1.5)
OR 1.0 (0.9–1.2)
OR 1.0 (0.7–1.3)

83.3–921.3 ng/mL
(5th maternal PFOA quintile)

Preterm birth (<37 weeks)
Preterm birth (<32 weeks)

OR 1.2 (0.9–1.6)
OR 1.4 (0.5–3.6)

48.8 ng/mL (maternal mean
PFOA)

Miscarriage
Preterm birth

OR 0.9 (0.5–1.6), >90th percentile
OR 0.9 (0.6–1.5), >90th percentile

1.6 ng/mL (cord serum
median PFOA)

Gestational age

NS (p>0.05)

1.84 ng/mL (cord blood
geometric mean PFOA)

Gestational age
Preterm birth

NS (p>0.05)
OR 0.64 (0.40–1.02)

>2.1–18 ng/mL (maternal
3rd PFOA tertile)

Preterm birth

RR 1.31 (0.38–4.45), 3rd tertile

PFOA
Darrow et al. 2013
Community (C8) (n=1,330 women)
Darrow et al. 2014
Community (C8) (n=1,129 women)
Savitz et al. 2012a
Community (C8) (11,737 singleton infants)
Savitz et al. 2012b

63.1–934.3 ng/mL
(4th maternal PFOA quartile)

Community (13,243 cases stillbirth, preterm
birth, low birth weight or small for gestational
age)
Savitz et al. 2012b
Community (4,547 infants)
Stein et al. 2009
Community (C8) (n=1,845 pregnancies)
Apelberg et al. 2007b
General population (n=341 singleton births)
Chen et al. 2012a
General population (n=429 infants)
Hamm et al. 2010
General population (n=252 pregnant women)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

342
2. HEALTH EFFECTS

Table 2-22. Summary of Pregnancy Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Jensen et al. 2015

1.58 ng/mL (maternal median Miscarriage before gestation OR 0.64 (0.36–1.18)
PFOA)
week 12

General population (n=56 cases and
336 controls)
Whitworth et al. 2012a
General population (n=901 infants)
PFOS
Darrow et al. 2013
Community (C8) (n=1,330 women)
Darrow et al. 2014
Community (C8) (n=1,129 women)
Stein et al. 2009
Community (C8) (n=5,262 infants)
Chen et al. 2012a
General population (n=429 infants)
Fei et al. 2007, 2008a

Resultc

≥3.04 ng/mL (maternal
4th PFOA quartile)

Preterm birth

OR 0.1 (0.03–0.6)*, 4th quartile

15.6 ng/mL (mean PFOS)

Preterm birth

OR 1.07 (0.58–1.95)

>23.3 ng/mL (5th PFOS
quintile)

Miscarriage risk
Parous subgroup
Nulliparous subgroup
Miscarriage
Preterm birth

OR 1.41 (0.88–2.26), 5th quintile
OR 1.12 (0.58–2.17), 5th quintile
OR 2.02 (0.83–4.93), 5th quintile
OR 0.9 (0.7–1.3), >90th percentile
OR 1.4 (1.1–1.7)*, >90th percentile

Preterm birth

OR 2.45 (1.47–4.08)*

23.2–83.4 ng/mL
(>90th PFOS percentile)
5.94 ng/mL (cord blood
geometric mean PFOS)

35.3 ng/mL (maternal median Gestation length
PFOS)
Preterm birth

NS (p>0.01)
OR 1.43 (0.50–4.11), 4th quartile

13.0 and ≥16.59 ng/mL
(maternal median and
4th quartile PFOS)

OR 0.3 (0.1–1.0, p=0.03)*,
4th quartile

General population (n=1,400 pregnant women)
Hamm et al. 2010
>10–35 ng/mL (maternal
Preterm birth
RR 1.11 (0.36–3.38), 3rd tertile
rd
3 tertile PFOS)
General population (n=252 pregnant women)
Jensen et al. 2015
8.10 ng/mL (maternal median Miscarriage before gestation OR 1.16 (0.59–1.29)
PFOS)
week 12
General population (n=56 cases and
336 controls)
Whitworth et al. 2012a
General population (n=901 infants)

Preterm birth

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

343
2. HEALTH EFFECTS

Table 2-22. Summary of Pregnancy Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

>1.4–43 ng/mL (maternal
3rd tertile PFHxS)

Preterm birth

RR 0.31 (0.11–0.90)*, 3rd tertile

0.298 ng/mL (maternal
median PFHxS)

Miscarriage before gestation OR 1.53 (0.99–2.38)
week 12

2.36 ng/mL (cord blood
geometric mean PFNA)

Preterm birth

PFHxS
Hamm et al. 2010
General population (n=252 pregnant women)
Jensen et al. 2015
General population (n=56 cases and
336 controls)
PFNA
Chen et al. 2012a
General population (n=429 infants)
Jensen et al. 2015
General population (n=56 cases and
336 controls)
PFDeA
Jensen et al. 2015

OR 0.88 (0.71–1.11)

0.72 ng/mL (maternal median Miscarriage before gestation OR 16.46 (7.39–36.62)*
PFNA)
week 12

0.27 ng/mL (maternal median Miscarriage before gestation OR 2.30 (1.18–4.47)*
PFDeA)
week 12

General population (n=56 cases and
336 controls)
PFUA
Chen et al. 2012a

10.26 ng/mL (cord blood
geometric mean PFUA)

Preterm birth

OR 0.87 (0.64–1.16)

General population (n=429 infants)
aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 13 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

OR = odds ratio; NS = not significant; PFDeA = perfluorodecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFUA = perfluoroundecanoic acid; RR= risk ratio

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

344
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

≥37.2 ng/mL (5th PFOA
quintile)

Birth weight
Low birth weight

NS (p=0.70 for trend)
OR 0.92 (0.44–1.95)

NR

Birth weight

NS (p>0.05)

Low birth weight

OR 0.37 (0.16–0.86)*

NR

Congenital anomalies

OR 1.1 (0.34–3.3)

63.1–934.3 ng/mL
(4th maternal PFOA quartile)

Low birth weight
Birth defect

OR 0.37 (0.16–0.86)*
OR 1.0 (0.8–1.3)

7.7 ng/mL (estimated
maternal median PFOA)

Birth weight

Low birth weight

β -14.80 (-42.28–13.68), per
100 ng/mL increase in PFOA
OR 1.0 (0.86–1.15), per 100 ng/mL
increase in PFOA
OR 0.86 (0.67–1.11), per 100 ng/mL
increase in PFOA
OR 1.07 (0.96–1.18), per 100 ng/mL
increase in PFOA
OR 1.08 (1.01–1.16)*, per
100 ng/mL increase in PFOA
OR 0.8 (0.6–1.2), for serum PFOA
levels ≥80th percentile
OR -12.76 (-26.08–0.57), per
100 ng/mL increase in PFOA
OR 0.8 (0.3–1.9), 4th quartile

Birth defects

OR 1.7 (0.8–3.6), 4th quartile

PFOA
Darrow et al. 2013
Community (C8) (n=1,330 women)
Nolan et al. 2009
Community (n=1,555 singleton infants)
Nolan et al. 2010
Community (n=1,548 singleton infants)
Savitz et al. 2012a
Community (C8) (11,737 singleton infants)
Savitz et al. 2012b
Community (13,243 cases stillbirth, preterm
birth, low birth weight or small for gestational
age)
Savitz et al. 2012b

Low birth weight
Small for gestational age
13.4 ng/mL (estimated
maternal median PFOA)

Community (4,547 infants)

Low birth weight
Small for gestational age

Birth weight
Stein et al. 2009
Community (C8) (n=1,845 pregnancies)

50.0–<120.6 and 120.6–
894.4 ng/mL (3rd and 4th
maternal PFOA quartile)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

345
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Stein et al. 2014c

61.3 ng/mL (estimated in
utero mean PFOA)

Brain defects

OR 2.6 (1.3–5.1), interquartile range

Gastrointestinal defects
Kidney defects
Craniofacial defects
Eye defects
Limb defects
Genitourinary defects

OR 0.7 (0.3–1.4), interquartile range
OR 0.7 (0.3–1.8), interquartile range
OR 0.6 (0.3–1.3), interquartile range
OR 1.1 (0.6–2.1), interquartile range
OR 1.2 (0.7–2.0), interquartile range
OR 1.0 (0.6–1.7), interquartile range

Heart defects
Birth weight
Birth length
Head circumference
Ponderal index
Gestational weight gain

OR 1.2 (0.8–1.7), interquartile range
NS (p>0.05)
NS (p>0.05)
Inverse association (p>0.05)*
Inverse association (p>0.05)*
NS (p>0.1), serum PFOA
OR 1.04 (1.02–1.06)*, cord PFOA

Birth weight

NS, investigators noted no consistent
alterations across PFOA quartiles
NS, investigators noted no consistent
alterations across PFOA quartiles
NS, investigators noted no consistent
alterations across PFOA quartiles
OR 0.93 (0.68–1.26), maternal PFOA
OR 0.94 (0.72–1.23), paternal PFOA

Community (C8) (n=10,262 infants)

Apelberg et al. 2007b

1.6 ng/mL (cord serum
median PFOA)

General population (n=341 singleton births)

Ashley-Martin et al. 2016

1.70 and 0.39 ng/mL
(maternal and cord median
General population (n=1,723 pregnant women) PFOA)
Bach et al. 2016
2.0 ng/mL (median PFOA)
General population (n=1,507 nulliparous
women)

Birth length
Head circumference

Bae et al. 2015
General population (n=233 couples)

5.01 and 4.05 ng/mL and
Male birth
5.00 and 2.54 ng/mL
(geometric mean PFOA in
male and female nulliparous
parents and male and female
parous parents, respectively)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

346
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Chen et al. 2012a

1.84 ng/mL (cord blood
geometric mean PFOA)

Birth weight

NS (p>0.05)

Birth length
Head circumference
Ponderal index
Small for gestational age
Low birth weight
Birth weight

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
OR 1.24 (0.75–2.05)
OR 0.53 (0.18–1.55)
β -10.63 (-20.79 to -0.47)*

Birth length
Abdominal circumference
Head circumference
Low birth weight
Small for gestation age
Birth weight

β -0.069 (-0.113 to -0.024)*
β -0.059 (-0.106 to -0.012)*
β-0.030 (-0.064–0.004)
OR 2.44 (0.27–22.25), 4th quartile
OR 0.97 (0.55–1.70), 4th quartile
NS (p=0.473)

>2.1–18 ng/mL (maternal 3rd Birth weight
tertile PFOA)
Small for gestational age
1.46 ng/mL (maternal median Birth weight
PFOA)
Cord TSH

Change in weight 14.80 (-107.29–
136.89), 3rd tertile
RR 0.99 (0.25–3.92), 3rd tertile
NS (p>0.05)
Association (p<0.05)*

Cord T3
Cord T4
5.398 and 2.12 ng/mL (mean Thyroid stimulating
PFOA in cases and controls) immunoglobulin levels
TSH
T3
T4

NS (p>0.05)
NS (p>0.05)
Inverse association (p<0.05)*

General population (n=429 infants)

Fei et al. 2007, 2008a

5.6 ng/mL (maternal median
PFOA)

General population (n=1,400 pregnant women)

Govarts et al. 2016
General population (n=202 infants)
Hamm et al. 2010
General population (n=252 pregnant women)
Kim et al. 2011
General population (n=44 pregnant women)

Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

1.52 ng/mL (cord blood
geometric mean PFOA)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

347
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Lee et al. 2013

2.73 ng/mL (maternal mean
PFOA)

Birth weight

OR 0.54 (0.17–3.03)

1.84, 0.96, and 2.51 ng/mL
(maternal median PFOA for
General population (n=513 infants in
Greenland, Ukraine, and
Greenland subcohort, n=557 infants in Ukraine Poland subcohorts)
subcohort, and n=180 infants in Poland
subcohort
Maisonet et al. 2012
3.7 ng/mL (maternal median
PFOA)
General population (n=447 girls)

Birth weight

β -63.77 (-122.83 to -4.71,
p=0.035)*. 2 SD increase in lntransformed PFOA

Birth weight

Inverse association (p=0.0120 for
trend)*
NS (p=0.0978)
NS (p=0.5920)
NS (p=0.4147)

Monroy et al. 2008

Birth weight

NS (p>0.05), maternal serum and
cord blood PFOA

Birth weight

NS (p>0.05), maternal or paternal

General population (n=59 pregnant women)

Lee et al. 2016

Birth length
Ponderal index
Head circumference
1.11 ng/mL (cord blood mean Birth weight
PFOA)

General population (n=85 infants)
Lenters et al. 2016a, 2016b

General population (n=101 pregnant women)
Robledo et al. 2015a, 2015b
General population (n=234 couples)

Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)

1.81 and 1.58 ng/mL
(maternal and cord median
PFOA)
3.16 and 5.00 ng/mL
(maternal and paternal
geometric mean PFOA)

Birth length
Ponderal index
Body weight at 20 months

Birth length
Head circumference
Ponderal index
2.37 and 2.34 ng/mL (median Birth weight
maternal PFOA for boys and Birth length
girls)
Head circumference
Small for gestational age

***DRAFT FOR PUBLIC COMMENT***

OR 0.44 (0.12–1.58)
OR 0.56 (0.16–2.01)
OR 0.82 (0.24–13.65)
NS (p>0.05)

NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

348
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Washino et al. 2009

1.3 ng/mL (maternal median
PFOA)

Birth weight

NS (p=0.207)

Birth length
Chest circumference
Head circumference
Birth weight
Small for gestational age
Large for gestational age

NS (p=0.631)
NS (p=0.460)
NS (p=0.823)
NS (p=0.12)
NS (p=0.92)
NS (p=0.33)

General population (n=428 infants)

Whitworth et al. 2012a
General population (n=901 infants)
PFOS
Grice et al. 2007

2.2 and ≥3.04 ng/mL
(maternal median and
4th quartile PFOA)

1,300–1,970 ng/mL (range of Birth weight
PFOS)

NS (p=0.15)

15.6 ng/mL (mean PFOS)

NS (p=0.045 for trend), whole cohort
Association (p=0.006), women
(n=783) who conceived after blood
sample collection
OR 1.33 (0.60–2.96)
OR 1.6 (1.1–2.3)*, 75th-90th
percentile
OR 1.3 (0.8–2.1)
NS (p>0.05)

Occupational (n=263 females)
Darrow et al. 2013

Birth weight

Community (C8) (n=1,330 women)

Stein et al. 2009
Community (C8) (n=5,262 infants)
Apelberg et al. 2007b

17.7–<23.2 and 23.2–
83.4 ng/mL (75th–90th and
>90th PFOS percentile)
5 ng/mL (PFOS cord serum
median)

General population (n=341 singleton births)

Ashley-Martin et al. 2016
General population (1,723 pregnant women)

4.60 and 0.15 ng/mL
(maternal and cord PFOS
median)

Low birth weight
Low birth weight
Birth defects
Gestational age
Birth weight
Birth length
Head circumference
Ponderal index
Gestational weight gain

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
Inverse association (p>0.05)*
Inverse association (p>0.05)*
Association (p<0.1), serum PFOS in
underweight/normal weight subjects
OR 1.03 (1.00–1.05)*, cord PFOS

 PERFLUOROALKYLS

349
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Bach et al. 2016

8.3 ng/mL (PFOS median)

Inverse association reported by
investigators
NS, investigators noted no consistent
alterations across PFOS quartiles
NS, investigators noted no consistent
alterations across PFOS quartiles
OR 1.16 (0.88–1.53), maternal PFOS
OR 1.01 (0.78–1.33), paternal PFOS

General population (n=1,507 nulliparous
women)

Birth weight
Birth length
Head circumference

Bae et al. 2015
General population (233 couples)

Chen et al. 2012a
General population (n=429 infants)

21.7 and 14.5 ng/mL and
Male birth
21.5 and 10.8 ng/mL
(geometric mean PFOS in
male and female nulliparous
parents and male and female
parous parents, respectively)
5.94 ng/mL (cord blood
Gestational age
geometric mean PFOS)
Birth weight
Birth length
Head circumference

de Cock et al. 2014

1.611 ng/mL (cord blood
mean PFOS)

Ponderal index
Small for gestational age
Low birth weight
Weight

Height
BMI
Head circumference
Fei et al. 2007, 2008a
35.3 ng/mL (maternal median Birth weight
PFOS)
Birth length
General population (n=1,400 pregnant women)
Abdominal circumference
General population (n=89 infants)

Head circumference
Gestation length

***DRAFT FOR PUBLIC COMMENT***

Inverse association (p<0.001)*
β -110.2 g (-176.0 to -44.5,
p<0.001)*, per ln PFOS
NS (p>0.05)
β -0.25 cm (-0.46–0.05 cm,
p<0.05)*, per ln PFOS
NS (p>0.05)
OR 2.27 (1.25–4.15)*
OR 2.61 (0.185–8.03)
NS (p=0.802)
NS (p=0.975)
NS (p=0.586)
NS (p=0.649)
β -0.46 (-2.34–1.41)
β-0.002 (-0.011–0.006)
β -0.003 (-0.012–0.005)
β 0.000 (-0.006–0.007)
NS (p>0.01)

 PERFLUOROALKYLS

350
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Govarts et al. 2016

Serum perfluoroalkyl level Outcome evaluated

2.63 ng/mL (cord blood
geometric mean PFOS)

Resultc

Low birth weight

OR 4.82 (0.56–41.16), 4th quartile

Small for gestation age
Birth weight

OR 0.98 (0.58–1.65), 4th quartile
NS (p=0.798)

Birth weight

Change in weight 71.25 (54.97–
197.48), 3rd tertile

General population (n=202 infants)
Hamm et al. 2010

>10–35 ng/mL (maternal 3rd
tertile PFOS)

General population (n=252 pregnant women)

Small for gestational age
2.93 ng/mL (maternal median Birth weight
PFOS)
Cord TSH
Cord T3
Cord T4
5.326 and 4.05 ng/mL (mean Thyroid stimulating
PFOS in cases and controls) immunoglobulin levels
TSH
T3

RR 0.26 (0.10–0.70)*, 3rd tertile
NS (p>0.05)
NS (p>0.05)
Inverse association (p<0.05)*
NS (p>0.05)
NS (p>0.05)

T4
10.77 ng/mL (maternal mean Birth weight
PFOS)
Birth length
Ponderal index
Head circumference
0.87 ng/mL (cord blood mean Birth weight
PFOS)

NS (p>0.05)
OR 0.98 (0.32–3.03)
OR 0.97 (0.29–3.27)
OR 0.22 (0.05–0.90)*
OR 1.34 (0.20–8.90)
NS (p>0.05)

Kim et al. 2011
General population (n=44 pregnant women)

Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; N=13 controls)

Lee et al. 2013
General population (n=59 pregnant women)

Lee et al. 2016
General population (n=85 infants)
Lenters et al. 2016a, 2016b

20.09, 5.04, and 7.81 ng/mL
(maternal median PFOS for
General population (n=513 infants in
Greenland, Ukraine, and
Greenland subcohort, n=557 infants in Ukraine Poland subcohorts)
subcohort, and n=180 infants in Poland
subcohort)

Birth weight

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)

NS (p=0.109)

 PERFLUOROALKYLS

351
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

28.90 and 27.50 ng/mL
Congenital cerebral palsy
(maternal median PFOS in
Boys
General population (n=156 children diagnosed boy and girl cases)
Girls
27.60 and 26.20 ng/mL
with congenital cerebral palsy (cases) and
(maternal median PFOS in
550 controls)
boy and girl controls)
Maisonet et al. 2012
19.6 ng/mL (maternal median Birth weight
PFOS)
General population (n=447 girls)
Birth length

Resultc

Liew et al. 2014

Monroy et al. 2008
General population (n=101 pregnant women)
Robledo et al. 2015a, 2015b
General population (n=234 couples)

Washino et al. 2009

Ponderal index
14.54 mg/mL and 6.08 ng/mL Birth weight
(maternal and cord median
PFOS)
12.44 and 21.6 ng/mL
Birth weight
(maternal and paternal
Birth length
geometric mean PFOS)
Head circumference
Ponderal index
5.2 ng/mL (maternal median Birth weight
PFOS)
Males

Birth length
Chest circumference
Head circumference
Birth weight
Small for gestational age
Large for gestational age

NS (p=0.33)

Females

General population (n=901 infants)

13.0 and ≥16.59 ng/mL
(maternal median and 4th
quartile PFOS)

β -140.01 g (-238.14 to -41.89 g,
p=0.0053 for trend)*, 3rd tertile
β -0.63 cm (-1.11 to -0.15 cm,
p=0.103 for trend)* 3rd tertile
NS (p=0.1120)
NS (p>0.05), maternal serum and
cord blood PFOA
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
β -148.8 g (-297.0 to -0.5 g,
p=0.049)*, per log PFOS unit
NS (p=0.917)
β -269.4 g (-465.7 to -73.0 g,
p=0.007)*, per log PFOS unit
NS (p=0.167)
NS (p=0.718)
NS (p=0.488)
NS (p=0.10)
NS (p=0.51)

General population (n=428 infants)

Whitworth et al. 2012a

RR 1.7 (1.0–2.8)*
RR 0.7 (0.4–1.4)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

352
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

PFHxS
Ashley-Martin et al. 2016

1.00 and 0.10 ng/mL
(maternal and cord PFHxS
General population (n=1,723 pregnant women) median)
Bach et al. 2016
0.5 ng/mL (maternal PFHxS
median)
General population (n=1,507 nulliparous
women)

Gestational weight gain

NS (p>0.1), serum PFHxS
OR 1.01 (10.99–1.03), cord PFHxS

Birth weight

Inverse association reported by
investigators
NS, investigators noted no consistent
alterations across PFHxS quartiles
NS, investigators noted no consistent
alterations across PFHxS quartiles
Change in weight (25.99, 95%
CI -95.25–147.23), 3rd tertile
RR 2.35 (0.63–8.72), 3rd tertile

Birth length
Head circumference

Hamm et al. 2010

>1.4–43 ng/mL (maternal
3rd tertile PFHxS)

General population (n=252 pregnant women)
Kim et al. 2011
General population (n=44 pregnant women)

Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

Lee et al. 2013
General population (n=59 pregnant women)

Lee et al. 2016

Birth weight
Small for gestational age

0.55 ng/mL (maternal median Birth weight
PFHxS)
Cord TSH
Cord T3
Cord T4
1.228 and 1.17 ng/mL (mean Thyroid stimulating
PFHxS in cases and controls) immunoglobulin levels
TSH
T3
T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*

1.35 ng/mL (maternal mean
PFHxS)

OR 0.57 (0.19–1.75)
OR 0.44 (0.12–1.58)
OR 0.64 (0.19–2.23)
OR 0.90 (0.13–6.13)
NS (p>0.05)

Birth weight
Birth length
Ponderal index
Head circumference
0.60 ng/mL (cord blood mean Birth weight
PFHxS)

General population (n=85 infants)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

353
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Lenters et al. 2016a, 2016b

2.05, 1.56, and 2.28 ng/mL
Birth weight
(maternal median PFHxS for
General population (n=513 infants in
Greenland, Ukraine, and
Greenland subcohort, n=557 infants in Ukraine Poland subcohorts)
subcohort, and n=180 infants in Poland
subcohort)
Liew et al. 2014
0.96 and 0.90 ng/mL
Congenital cerebral palsy
(maternal median PFHxS in
Boys
General population (n=156 children diagnosed boy and girl cases)
Girls
0.92 and 0.92 ng/mL
with congenital cerebral palsy (cases) and
(maternal median PFHxS in
550 controls)
boy and girl controls)
Maisonet et al. 2012
1.6 ng/mL (maternal median Birth weight
PFHxS)
General population (n=447 girls)
Birth length

Monroy et al. 2008
General population (n=101 pregnant women)
PFNA
Bach et al. 2016
General population (n=1,507 nulliparous
women)

1.62 mg/mL (maternal
median PFHxS)

0.8 ng/mL (PFNA median)

Ponderal index
Birth weight

Birth weight
Birth length
Head circumference

Bae et al. 2015
General population (233 couples)

1.60 and 1.37 ng/mL and
Male birth
1.55 and 1.09 ng/mL
(geometric mean PFNA in
male and female nulliparous
parents and male and female
parous parents, respectively)

***DRAFT FOR PUBLIC COMMENT***

Resultc
NS (p=0.801)

RR 1.2 (0.9–1.7)
RR 1.1 (0.6–1.9)

Inverse association (p=0.0314 for
trend)*
Inverse association (p=0.0008 for
trend)*
NS (p=0.6802 for trend)
NS (p>0.05)

NS, investigators noted no consistent
alterations across PFNA quartiles
NS, investigators noted no consistent
alterations across PFNA quartiles
NS, investigators noted no consistent
alterations across PFNA quartiles
OR 0.94 (0.70–1.26), maternal PFNA
OR 0.94 (0.71–1.24), paternal PFNA

 PERFLUOROALKYLS

354
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Chen et al. 2012a

2.36 ng/mL (cord blood
geometric mean PFNA)

Gestational age

NS (p>0.05)

Birth weight
Birth length
Head circumference
Ponderal index
Small for gestational age
Low birth weight

NS (p>0.05)
Association (p<0.01)*
NS (p>0.05)
Inverse association (p<0.05)
OR 0.97 (0.74–1.26)
OR 0.76 (0.47–1.23)

1.931 and 0.633 ng/mL
(mean PFNA in cases and
controls)

Thyroid stimulating
immunoglobulin levels

NS (p>0.05)

General population (n=429 infants)

Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

Lee et al. 2016
General population (n=85 infants)
Lenters et al. 2016a, 2016b

TSH
T3
T4
0.36 ng/mL (cord blood mean Birth weight
PFNA)

0.69, 0.61, and 0.56 ng/mL
(maternal median PFNA for
General population (n=513 infants in
Greenland, Ukraine, and
Greenland subcohort, n=557 infants in Ukraine Poland subcohorts)
subcohort, and n=180 infants in Poland
subcohort)
Liew et al. 2014
0.46 and 0.39 ng/mL
(maternal median PFNA in
General population (n=156 children diagnosed boy and girl cases)
0.44 and 0.41 ng/mL
with congenital cerebral palsy (cases) and
(maternal median PFNA in
550 controls)
boy and girl controls)
Monroy et al. 2008
0.69 mg/mL (maternal
median PFNA)
General population (n=101 pregnant women)

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Birth weight

NS (p=0.065)

Congenital cerebral palsy
Boys
Girls

RR 1.2 (0.6–2.5)
RR 0.6 (0.3–1.2)

Birth weight

NS (p>0.05)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

355
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Robledo et al. 2015a, 2015b

1.211 and 1.566 ng/mL
(maternal and paternal
geometric mean PFNA)

NS (p>0.05), maternal or paternal

General population (n=234 couples)

Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)

PFDeA
Bach et al. 2016

Birth weight

Birth length
Head circumference
Ponderal index
1.55 and 1.58 ng/mL (median Birth weight
maternal PFNA for boys and
girls)
Birth length
Head circumference
Small for gestational age

0.3 ng/mL (PFDeA median)

General population (n=1,507 nulliparous
women)

Growth during childhood

NS (p>0.05)

Birth weight

NS, investigators noted no consistent
alterations across PFDeA quartiles
NS, investigators noted no consistent
alterations across PFDe quartiles
NS, investigators noted no consistent
alterations across PFDeA quartiles
OR 1.07 (0.81–1.42), maternal
PFDeA
OR 1.02 (0.78–1.34), paternal
PFDeA

Birth length
Head circumference

Bae et al. 2015
General population (233 couples)

Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

Lee et al. 2016

0.46 and 0.38 ng/mL and
0.49 and 0.46 ng/mL
(geometric mean PFDeA in
male and female nulliparous
parents and male and female
parous parents, respectively)
0.523 and 0.298 ng/mL
(mean PFDeA in cases and
controls)

NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
Inverse association (p>0.05)*, girls
only
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Male birth

Thyroid stimulating
immunoglobulin levels
TSH
T3
T4

0.14 ng/mL (cord blood mean Birth weight
PFDeA)

General population (n=85 infants)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

356
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Lenters et al. 2016a, 2016b

0.40, 0.16, and 0.22 ng/mL
Birth weight
(maternal median PFDeA for
General population (n=513 infants in
Greenland, Ukraine, and
Greenland subcohort, n=557 infants in Ukraine Poland subcohorts)
subcohort, and n=180 infants in Poland
subcohort)
Liew et al. 2014
0.18 and 0.16 ng/mL
Congenital cerebral palsy
(maternal median PFDeA in
Boys
General population (n=156 children diagnosed boy and girl cases)
Girls
0.17 and 0.16 ng/mL
with congenital cerebral palsy (cases) and
(maternal median PFDeA in
550 controls)
boy and girl controls)
Robledo et al. 2015a, 2015b
0.402 and 0.458 ng/mL
Birth weight
(maternal and paternal
Birth length
General population (n=234 couples)
geometric mean PFDeA)
Head circumference
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)

PFUA
Bach et al. 2016

Ponderal index
0.46 and 0.43 ng/mL (median Birth weight
maternal PFDeA for boys and
girls)
Birth length
Head circumference
Small for gestational age
0.3 ng/mL (PFUA median)

General population (n=1,507 nulliparous
women)

Birth weight

General population (n=429 infants)

RR 1.1 (0.7–1.7)
RR 0.6 (0.3–1.1)

NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
Inverse association (p>0.05)*, girls
only
NS (p>0.05)
NS (p>0.05)
OR 3.14 (1.07–9.19)*, girls only

Gestational age
Birth weight
Birth length

NS (p>0.05)

Birth length

10.26 ng/mL (cord blood
geometric mean PFUA)

NS (p=0.158)

NS, investigators noted no consistent
alterations across PFUA quartiles
NS, investigators noted no consistent
alterations across PFUA quartiles
NS, investigators noted no consistent
alterations across PFUA quartiles
NS (p>0.05)
NS (p>0.05)

Head circumference
Chen et al. 2012a

Resultc

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

357
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

Lee et al. 2016

Serum perfluoroalkyl level Outcome evaluated

0.982 and 0.438 ng/mL
(mean PFUA in cases and
controls)

0.70, 0.16, and 0.13 ng/mL
(maternal median PFUA for
General population (n=513 infants in
Greenland, Ukraine, and
Greenland subcohort, n=557 infants in Ukraine Poland subcohorts)
subcohort, and n=180 infants in Poland
subcohort)
Wang et al. 2016
General population (n=117 boys and 106
girls examined at age 2, 5, 8, and 11 years
of age)
PFHpA
Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

Head circumference

NS (p>0.05)

Ponderal index
Small for gestational age
Low birth weight
Thyroid stimulating
immunoglobulin levels
TSH
T3
T4

NS (p>0.05)
OR 0.93 (0.65–1.33)
OR 1.01 (0.53–1.91)
NS (p>0.05)

0.22 ng/mL (cord blood mean Birth weight
PFUA)

General population (n=85 infants)
Lenters et al. 2016a, 2016b

Resultc

Birth weight

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

NS (p=0.275)

3.52 and 3.31 ng/mL (median Birth weight
maternal PFUA for boys and
girls)
Birth length
Head circumference
Small for gestational age

Inverse association (p<0.05)*, girls
only

0.284 and 0.324 ng/mL
(mean PFHpA in cases and
controls)

NS (p>0.05)

Thyroid stimulating
immunoglobulin levels
TSH
T3
T4

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
OR 1.83 (1.01–3.32)*, girls only

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

358
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.464 and 0.220 ng/mL
(mean PFBA in cases and
controls)

NS (p>0.05)

PFBA
Kim et al. 2016a
General population (n=27 infants with
congenital hypothyroidism; n=13 controls)

PFDoA
Lee et al. 2016

Thyroid stimulating
immunoglobulin levels
TSH
T3
T4

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

0.14 ng/mL (cord blood mean Birth weight
PFDoA)

NS (p>0.05)

0.13, 0.04, and 0.05 ng/mL
Birth weight
(maternal median PFDoA for
Greenland, Ukraine, and
Poland subcohorts)

NS (p=0.440)

0.37 and 0.37 ng/mL (median Birth weight
maternal PFDoA for boys and
girls)
Birth length
Head circumference

Inverse association (p<0.05)*, girls
only
NS (p>0.05)
Inverse association (p<0.05)*, girls
only
NS (p>0.05)

General population (n=85 infants)
Lenters et al. 2016a, 2016b
General population (n=513 infants in
Greenland subcohort, n=557 infants in
Ukraine subcohort, and n=180 infants in
Poland subcohort)
Wang et al. 2016
General population (n=117 boys and 106 girls
examined at 2, 5, 8, and 11 years of age)

Small for gestational age
PFOSA
Bae et al. 2015
General population (233 couples)

0.11 and 0.10 ng/mL and
Male birth
0.10 and 0.12 ng/mL
(geometric mean PFOSA in
male and female nulliparous
parents and male and female
parous parents, respectively)

***DRAFT FOR PUBLIC COMMENT***

OR 1.07 (0.81–1.41), maternal
PFOSA
OR 1.14 (0.86–1.51), paternal
PFOSA

 PERFLUOROALKYLS

359
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Robledo et al. 2015a, 2015b

0.112 and 0.114 ng/mL
(maternal and paternal
geometric mean PFOSA)

General population (n=234 couples)

Birth weight
Boys
Girls
Birth length
Head circumference
Ponderal index

Me-PFOSA-AcOH
Bae et al. 2015
General population (233 couples)

Robledo et al. 2015a, 2015b
General population (n=234 couples)

Resultc
Inverse association (p<0.05)*,
maternal only
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal

0.29 and 0.28 ng/mL and
0.28 and 0.32 ng/mL
(geometric mean Me-PFOSAAcOH in male and female
nulliparous parents and male
and female parous parents,
respectively)
0.301 and 0.324 ng/mL
(maternal and paternal
geometric mean Me-PFOSAAcOH)

Male birth

Birth weight
Birth length
Head circumference
Ponderal index

NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal

0.11 and 0.10 ng/mL and
0.10 and 0.11 ng/mL
(geometric mean Et-PFOSAAcOH in male and female
nulliparous parents and male
and female parous parents,
respectively)

Male birth

OR 1.22 (0.92–1.60), maternal
Et-PFOSA-AcOH
OR 0.98 (0.75–1.29, paternal
Et-PFOSA-AcOH

Female birth

OR 0.98 (0.75–1.28), maternal
Me-PFOSA-AcOH
OR 0.80 (0.60–1.06), paternal
Me-PFOSA-AcOH
OR 0.34 (0.13–0.89)*, paternal
Me-PFOSA-AcOH 3rd tertile

Et-PFOSA-AcOH
Bae et al. 2015
General population (233 couples)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

360
2. HEALTH EFFECTS

Table 2-23. Summary of Birth Outcomes in Humansa
Reference and study populationb
Robledo et al. 2015a, 2015b
General population (n=234 couples)

Serum perfluoroalkyl level Outcome evaluated
0.109 and 0.105 ng/mL
(maternal and paternal
geometric mean Et-PFOSAAcOH)

Resultc

Birth weight

NS (p>0.05), maternal or paternal

Birth length
Head circumference
Ponderal index

NS (p>0.05), maternal or paternal
NS (p>0.05), maternal or paternal
β -0.09 (-0.16 to -0.02)*, maternal
NS (p>0.05), paternal

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 13 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

BMI = body mass index; Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; FSH = follicle stimulating hormone; HR = hazard ratio;
LH = luteinizing hormone; Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane sulfonamide) acetic acid; NS = not significant; NR = not reported; OR = odds ratio;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid; RR = relative
risk; T3 = triiodothyronine; T4 = thyroxine; TSH = thyroid stimulating hormone

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb
PFOA
Stein et al. 2013
Community (C8) (n=320 children 6–12 years
old)
Stein et al. 2014a, 2014b
Community (C8) (n=321 children 6–12 years
old)

Serum perfluoroalkyl level Outcome evaluated

Resultc

115.9 ng/mL (estimated in
utero mean PFOA)

β 4.61 (0.68–8.54)*, 4th quartile
NS
β -8.49 (-16.14 to -0.84)*, 4th
quartile
β -6.39 (-11.43 to -1.35)*, 4th
quartile boys
β -6.39 (-0.03–8.87), 4th quartile girls
β -6.42 (-13.29–0.45), 4th quartile
boys
β -1.92 (-10.39–6.55), 4th quartile
girls

Full scale IQ
Reading and math skills
Scores on tests of ADHD
(improvement)
94.1–838.6 ng/mL (4th PFOA Executive function scores
quartile measured 3–4 years (mother completed survey)
prior to behavioral
assessment)
Executive function scores
(teacher completed survey)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

361
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

β -3.82 (-8.96–1.31), 4th quartile boys
β 6.99 (2.47–11.51)*, 4th quartile
girls
β 2.30 (-1.18–5.77), 4th quartile boys
and girls
ADHD-like behaviors
β -9.25 (-18.78–0.27), 4th quartile
(teacher completed survey) boys
β -3.65 (-10.85–3.51) 4th quartile girls
β -6.03 (-11.40 to -0.66)*, 4th
quartile boys and girls
Behavioral problems and
β -1.55 (-5.91–2.82), 4th quartile boys
emotional disturbances
β 4.63 (0.72–8.53)*, 4th quartile girls
(mother completed survey)
Behavioral problems and
β -2.47 (-8.24–3.30), 4th quartile boys
emotional disturbances
β -0.91 (-6.19–4.37), 4th quartile girls
(teacher completed survey)
ADHD
OR 0.76 (0.64–0.90)*, 4th quartile
Learning problems
12–15 years old
OR 0.96 (0.73–1.26), 4th quartile
5–18 years old
OR 0.90 (0.76–1.06), 4th quartile
Social responsiveness scale β -2.0 (-4.4–0.4)
score (measure of autistic
behaviors)
ADHD-like behaviors
(mother completed survey)

Stein and Savitz 2011

65.3–2,070.6 ng/mL
(4th PFOA quartile)

Community (C8) (n=10,546 children aged 5–
18 year)
Braun et al. 2014
General population (n=175 children 4 and
5 years old)
Chen et al. 2013

5.5 ng/mL (maternal median
PFOA)

2.5 ng/mL (mean cord PFOA) Poor performance on tests

General population (239 children 2 years of
age)

***DRAFT FOR PUBLIC COMMENT***

NS OR 0.6 (0.08–4.8), whole test
OR1.3 (0.3–6.2), cognitive tests
OR 0.5 (0.06–4.0), language tests
OR 0.8 (0.1–4.7), gross motor tests
OR 2.8 (0.6–13.5), fine motor tests
OR 0.3 (0.02–2.7), social tests
OR 3.2 (0.7–14.3), self-help tests

 PERFLUOROALKYLS

362
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Donauer et al. 2015

5.49 ng/mL (maternal
geometric mean PFOA)

Social/easy going

NS (p>0.05)

Hypotonic

OR 3.79 (1.1–12.8)* per 10-fold
increase in PFOA

High arousal/difficult
Apgar scores <10
Motor and mental
development at 6 months
Neurobehavioral milestones
at 18 months
Behavioral problems
Motor coordination

NS (p=0.3533)
OR 1.14 (0.57–2.25)
NS (p>0.05)

General population (n=349 infants at 5 weeks
of age)
Fei et al. 2008b

5.6 ng/mL (maternal median
PFOA)

General population (n=1,400 infants)

Fei and Olsen 2011

5.4 ng/mL (maternal median
PFOA)

General population (n=526–787 7-year-old
children)
Forns et al. 2015

40 ng/L (median PFOA
breast milk level)

General population (n=843 infants)
Goudarzi et al. 2016

1.2 ng/mL (maternal median
PFOA)

General population (n=173 infants at 6 months
and 133 at 18 months)
Gump et al. 2011

3.23 ng/mL (mean PFOA)

General population (n=83 children aged 9–
11 years)
Hoffman et al. 2010

4.4 ng/mL (median PFOA)

NS (p>0.05)
NS (p>0.15 for trend)
NS (p=0.89 for trend)

Risk of an abnormal score
on neurobehavioral
assessment questionnaire

OR 1.05 (0.77-1.44) at 6 months of
age
OR 1.0 (0.78-1.28) at 24 months of
age
MDI/PDI at 6 months of age NS (p>0.05), boys and girls
Inverse association (p<0.05)*,
females
MDI/PDI at 18 months of
NS (p>0.05)
age
Performance on task
NS (p>0.05)
requiring behavioral
inhibition
ADHD (parent reported)

General population (NHANES) (n=571 children
12–15 years)

***DRAFT FOR PUBLIC COMMENT***

OR 1.12 (1.01–1.23)*, per 1 ng/mL
PFOA

 PERFLUOROALKYLS

363
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Høyer et al. 2015a

1.4 and 1.9–9.8 ng/mL
(maternal median and
General population (n=1,106 children aged 5– 3rd tertile PFOA)
9 years)
Lien et al. 2016
1.55 ng/mL (cord blood
weighted average PFOA)
General population (n=282 children aged
7 years)

Liew et al. 2015
General population (n=215, ADHD cases,
213 autism cases, 545 controls)

4.06, 3.88, and 4.00 ng/mL
(maternal median PFOA for
ADHD, autism, and controls)

Motor skills

β -0.2 (-1.2–0.9)

Abnormal behavior
Hyperactivity

OR 2.7 (1.2–6.3)*, 3rd tertile
OR 3.1 (1.3–7.2)*, 3rd tertile

Inattention
Hyperactivity/impulsivity
Emotional symptoms

NS (p=0.7758)
NS (p=0.2997)
NS (p=0.691)

Conduct problems
Hyperactivity/inattention
ADHD
Autism

NS (p=0.2664)
NS (p=0.774)
RR 0.98 (0.82–1.16)
RR 0.98 (0.73–1.31)

Ode et al. 2014

1.80 and 1.83 ng/mL (cord
ADHD
blood median PFOA in cases
General population (n=206 children with ADHD and controls)
and 206 controls; children were 5–17 years old
at time of diagnosis)
Quaak et al. 2016
0.9056 ng/mL (cord mean
Score on test evaluating
PFOA)
ADHD
Males
General population (n=76 infants 18 months of
Females
age)
Scores on test evaluating
externalizing problem
Males

Strøm et al. 2014
General population (n=876 adults age
20 years)

3.7 ng/mL (median maternal
PFOA)

Resultc

Females
ADHD
Depression
Scholastic achievement

***DRAFT FOR PUBLIC COMMENT***

OR 0.98 (0.91–1.02), per 1 ng/mL
increase in PFOA

NS (p=0.72), 3rd tertile
NS (p=0.22), 3rd tertile
NS (p=0.31), 3rd tertile
NS (p=0.31), 3rd tertile
Association (p=0.05 and 0.09)*,
2nd and 3rd tertiles
NS (p=0.74), 3rd tertile
NS (p=0.45 for trend of 3rd tertile)
NS (p=0.28 for trend of 3rd tertile)
NS (p=0.21 for trend of 3rd tertile)

 PERFLUOROALKYLS

364
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Vuong et al. 2016

5.4 ng/mL (maternal median
PFOA)

Behavioral regulation

β 1.11(-1.22–3.44)

2.50 and 2.50 ng/mL
(maternal median PFOA for
5 and 8 year olds)

IQ score
Age 5 years
Age 8 years

14.8–<20.2, 20.22–<27.9,
and 27.9–202.1 ng/mL (2nd,
3rd, and 4th PFOS quartiles)

ADHD
Learning problems
5–18 years old
12–15 years old
Social responsiveness scale
score (test of autism)

OR 0.99 (0.76–1.30), 4th quartile

Neurobehavioral outcomes

NS (p>0.05)

General population (n=256 children aged 5 or
8 years)
Wang et al. 2015b
General population (n=120 children age
5 years and 120 children aged 8 years)
PFOS
Stein and Savitz 2011
Community (C8) (n=10,546 children aged 5–
18 years)
Braun et al. 2014
General population (n=175 children 4 and
5 years old)
Donauer et al. 2015
General population (n=349 infants)
Fei et al. 2008b
General population (n=1,400 infants)

13 ng/mL (maternal PFOS
median)

13.25 ng/mL (maternal
geometric mean PFOS)

Metacognition
β 0.58 (-1.77–2.93)
Global executive functioning β 1.06 (-1.33–3.45)

Hypotonic
High arousal/difficult
35.3 ng/mL (maternal median Apgar scores <10
PFOS)
Neurobehavioral milestones
Delay in age of sitting
Earlier use of word-like
sounds
Delays in using 2-word
sentences
Other milestones
Motor and mental
development at 6 months

***DRAFT FOR PUBLIC COMMENT***

NS
NS

OR 0.83 (0.70–0.98)*, 2nd quartile
OR 0.68 (0.52–0.89)*, 3rd quartile
No association

NS (p=0.3996)
NS (p=0.4678)
OR 1.20 (0.67–2.14)
Association (p=0.041 for trend)*
Association (p=0.039 for trend)*
Association (p=0.050 for trend)*
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

365
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Fei and Olsen 2011

34.4 ng/mL (maternal median Behavioral health
PFOS)
Motor coordination

NS (p>0.39 for trend)

General population (n=526–787 children)
Forns et al. 2015

110 ng/L (median PFOS
breast milk level)

General population (n=843 infants)
Goudarzi et al. 2016

5.7 ng/mL (maternal median
PFOS)

General population (n=173 at 6 months and
133 at 18 months)

NS (p=0.41 for trend)

Risk of an abnormal score
on neurobehavioral
assessment questionnaire

OR 0.96 (0.76-1.20) at 6 months of
age
OR 0.93 (0.74-1.17) at 24 months of
age
MDI/PDI at 6 months of age NS (p>0.05)
MDI/PDI at 18 months of
NS (p>0.05)
age

Gump et al. 2011

9.90 ng/mL (mean PFOS)

Inverse association (p<0.05)*

General population (n=83 children aged 9–
11 years)
Hoffman et al. 2010

Performance on task
requiring behavioral
inhibition

22.6 ng/mL (median PFOS)

ADHD (parent reported)

OR 1.03 (1.01–1.05)*, per 1 ng/mL
PFOS

Motor skills
Abnormal behavior
Hyperactivity
Inattention
Hyperactivity/impulsivity

β -0.1 (-1.2–1.1)
OR 1.5 (0.5–4.8), 3rd tertile
OR 1.4 (0.4–4.9), 3rd tertile
NS (p=0.8508)
NS (p=0.6857)

Emotional symptoms
Conduct problems
Hyperactivity/inattention
ADHD

NS (p=0.9431)
NS (p=0.4938)
NS (p=0.5226)
RR 0.87 (0.74–1.02)
RR 0.79 (0.64–0.98)*, 4th quartile
RR 0.92 (0.69–1.22)

General population (NHANES) (n=571 children
12–15 years)
Høyer et al. 2015a
10.0 and 16.6–87.3 ng/mL
(maternal median and 3rd
General population (n=1,106 children)
tertile PFOS)
Lien et al. 2016

4.79 ng/mL (cord blood
weighted average PFOS)

General population (n=282 children aged
7 years)

Liew et al. 2015
General population (n=215, ADHD cases,
213 autism cases, 545 controls)

26.80, 25.40, and
27.40 ng/mL (maternal
median PFOS for ADHD,
autism, and controls)

Autism

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

366
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Ode et al. 2014

6.92 and 6.77 ng/mL (cord
ADHD
blood median PFOS in cases
General population (n=206 children with ADHD and controls)
and 206 controls)
Quaak et al. 2016
1.5836 ng/mL (cord mean
Score on test evaluating
PFOS)
ADHD
Males
General population (n=76 infants 18 months of
Females
age)
Scores on test evaluating
externalizing problem
Males
Females
Strøm et al. 2014
21.4 ng/mL (median maternal ADHD
PFOS)
Depression
General population (n=876 adults age
Scholastic achievement
20 years)
Vuong et al. 2016
13.2 ng/mL (maternal median Behavioral regulation
PFOS)
Metacognition
General population (n=256 children 5 or
Global executive functioning
8 years of age)
Global executive functioning
composite score >60
Wang et al. 2015b
13.25 and 12.28 ng/mL
IQ score
(maternal median PFOS for
Age 5 years
General population (n=120 children age
5 and 8 year olds)
Age 8 years
5 years and 120 children aged 8 years)
PFHxS
Stein and Savitz 2011
2.9–<5.2 and 10.1–
ADHD
276.4 ng/mL (2nd and
5–18 years
12–15 years
Community (C8) (n=10,546 children aged 5–
4th PFHxS quartiles)
18 years)
Learning problems
5–18 years old
12–15 years old

***DRAFT FOR PUBLIC COMMENT***

Resultc
OR 0.98 (0.92–1.04), per 1 ng/mL
increase in PFOS

NS (p=0.19), 3rd tertile
NS (p=0.35), 3rd tertile
NS (p=0.43), 3rd tertile
NS (p=0.31), 3rd tertile
NS (p=0.74), 3rd tertile
NS (p=0.31), 3rd tertile
NS (p=0.38 for trend of 3rd tertile)
NS (p=0.14 for trend of 3rd tertile)
NS (p=0.59 for trend of 3rd tertile)
β 3.14 (0.68–5.61)*
β 3.10 (0.62–5.58)*
β 3.38 (0.86–5.90)*
OR 2.19 (1.03–4.66)*

NS (p>0.05)
NS (p>0.05)

OR 1.27 (1.06–1.52)*, 2nd quartile
OR 1.46 (1.10–1.93)*, 2nd quartile
OR 1.19 (1.00–1.41), 4th quartile
OR 1.05 (0.79–1.40), 4th quartile

 PERFLUOROALKYLS

367
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Braun et al. 2014

1.6 ng/mL (maternal PFHxS
median)

Social responsiveness scale No association
score (test for autism)

General population (n=175 children 4 and
5 years old)
Gump et al. 2011

6.06 ng/mL (mean PFHxS)

Inverse association (p<0.01)*

General population (n=83 children aged 9–
11 years)
Hoffman et al. 2010

Performance on task
requiring behavioral
inhibition

2.2 ng/mL (median PFHxS)

ADHD (parent reported)

OR 1.06 (1.02–1.11)*, per 1 ng/mL
PFOS

ADHD

RR 0.67 (0.54–0.83)*, 4th quartile

Autism

RR 1.10 (0.92–1.33)

General population (NHANES) (n=571 children
12–15 years)
Liew et al. 2015
0.84. 0.92, and 0.92 ng/mL
(maternal median PFHxS for
General population (n=215, ADHD cases,
ADHD, autism, and controls)
213 autism cases, 545 controls)
Vuong et al. 2016
1.5 ng/mL (maternal median
PFHxS)
General population (n=256 children 5 or
8 years of age)

Wang et al. 2015b
General population (n=120 children age
5 years and 120 children aged 8 years)
PFNA
Stein and Savitz 2011
Community (C8) (n=10,546 children aged 5–
18 years)

Behavioral regulation
Metacognition
Global executive functioning
Global executive functioning
composite score >60
0.69 and 0.69 ng/mL
IQ score
(maternal median PFHxS for
Age 5 years
5 and 8 year olds)
Age 8 years

1.2–<1.5, 1.5–<2.0, and 2.0– ADHD
24.1 ng/mL (2nd, 3rd, and
5–18 years
12–15 years
4th PFNA quartiles)
Learning problems
5–18 years old
12–15 years old

***DRAFT FOR PUBLIC COMMENT***

Resultc

β 1.19 (-0.54–5.40)
β 1.31 (-0.43–3.04)
β 1.36 (-0.41–3.12)
OR 1.71 (1.05–2.77)*

NS (p>0.05)
NS (p>0.05)

OR 0.99 (0.84–1.18), 4th quartile
OR 1.00 (0.75–1.32), 4th quartile
OR 0.81 (0.69–0.95)*, 3rd quartile
OR 0.73 (0.55–0.98)*, 4th quartile

 PERFLUOROALKYLS

368
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Braun et al. 2014

0.9 ng/mL (maternal PFNA
median)

Social responsiveness scale No association
score (tests for autism)

General population (n=175 children 4 and
5 years old)
Gump et al. 2011

0.82 ng/mL (mean PFNA)

Inverse association (p<0.05)*

General population (n=83 children aged 9–
11 years)
Hoffman et al. 2010

Performance on task
requiring behavioral
inhibition

0.6 ng/mL (median PFNA)

ADHD (parent reported)

OR 1.32 (0.86–2.02), per 1 ng/mL
PFNA

Inattention

Inverse association (p=0.0129)*

Hyperactivity/impulsivity
Emotional symptoms
Conduct problems
Hyperactivity/inattention
ADHD
Autism

NS (p=0.0588)
NS (p=0.1902)
NS (p=0.6931)
Inverse association (p=0.0484)*
RR 0.80 (0.62–1.03)
RR 0.80 (0.58–1.11)

General population (NHANES) (n=571 children
12–15 years)
Lien et al. 2016
4.49 ng/mL (cord blood
weighted average PFNA)
General population (n=282 children aged
7 years)

Liew et al. 2015
General population (n=215, ADHD cases,
213 autism cases, 545 controls)
Vuong et al. 2016
General population (n=256 children 5 or
8 years of age)
Wang et al. 2015b
General population (n=120 children age
5 years and 120 children aged 8 years)

0.42, 0.41, and 0.43 ng/mL
(maternal median PFNA for
ADHD, autism, and controls)

Resultc

0.9 ng/mL (maternal median
PFNA)

Behavioral regulation
β 2.57 (-0.26–5.40)
Metacognition
β 1.37 (-1.49–4.23)
Global executive functioning β 2.01 (-0.89–4.92)

1.59 and 1.44 ng/mL
(maternal median PFNA for
5 and 8 year olds)

IQ score
IQ scores, age 8 years
Full scale IQ
Visual IQ
Performance IQ

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*
NS (p>0.05)

 PERFLUOROALKYLS

369
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.26 ng/mL (mean PFDeA)

Inverse association (p<0.05)*

PFDeA
Gump et al. 2011
General population (n=83 children aged 9–
11 years)
Liew et al. 2015
General population (n=215, ADHD cases,
213 autism cases, 545 controls)
Vuong et al. 2016
General population (n=256 children 5 or
8 years of age)
Wang et al. 2015b
General population (n=120 children age
5 years and 120 children aged 8 years)
PFUA
Lien et al. 2016

0.15, 0.15, and 0.17 ng/mL
ADHD
(maternal median PFDeA for
ADHD, autism, and controls) Autism
0.2 ng/mL (maternal median
PFDeA)

General population (n=120 children age
5 years and 120 children aged 8 years)

Behavioral regulation
Metacognition

RR 0.76 (0.64–0.91)*
RR 0.53 (0.43–0.66)*, 4th quartile
RR 0.79 (0.63–1.01)
RR 0.52 (0.35–0.77)*, 4th quartile
β 0.70 (-3.31–1.92)
β 1.24 (-3.87–1.39)

Global executive functioning β -1.13 (-3.79–1.54)
0.44 and 0.44 ng/mL
IQ score
(maternal median PFDeA for
Age 5 years
5 and 8 year olds)
Age 8 years

7.96 ng/mL (cord blood
weighted average PFUA)

General population (n=282 children aged
7 years)

Wang et al. 2015b

Performance on task
requiring behavioral
inhibition

3.42 and 3.13 ng/mL
(maternal median PFUA for
5 and 8 year olds)

NS (p>0.05)
NS (p>0.05)

Inattention
Hyperactivity/impulsivity
Emotional symptoms
Conduct problems
Hyperactivity/inattention
IQ score

NS (p=0.6177)
NS (p=0.3642)
NS (p=0.0517)
NS (p=0.1207)
NS (p=0.9991)
NS (p>0.05)

IQ scores, age 8 years
Full scale IQ
Visual IQ
Performance IQ

NS (p>0.05)
NS (p>0.05)
Inverse association (p<0.05)*

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

370
2. HEALTH EFFECTS

Table 2-24. Summary of Neurodevelopmental Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.38 and 0.37 ng/mL
IQ score
(maternal median PFDoA for
Age 5 years
5 and 8 year olds)
Age 8 years

NS (p>0.05)
NS (p>0.05)

0.75 ng/mL (mean PFOSA)

Inverse association (p<0.05)*

PFDoA
Wang et al. 2015b
General population (n=120 children age
5 years and 120 children aged 8 years)
PFOSA
Gump et al. 2011
General population (n=83 children aged 9–
11 years)

Performance on task
requiring behavioral
inhibition

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 13 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

ADHD = attention deficit hyperactivity disorder; MDI/PDI = mental and psychomotor development indices; NHANES = National Health and Nutrition Examination
Survey; NS = not significant; OR = odds ratio; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid;
PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide;
PFUA = perfluoroundecanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

371
2. HEALTH EFFECTS

Table 2-25. Summary of Effects on the Development of the Reproductive System in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

PFOA
Lopez-Espinosa et al. 2011

26 and 20 ng/mL (median
PFOA in boys and girls)

Age of puberty
Boys
Girls

Community (C8) (n=3,076 boys and 2,931 girls
aged 8–18 years)
Lopez-Espinosa et al. 2016
34.8 and 30.1 ng/mL (median Estradiol
PFOA in boys and girls)
Community (C8) (n=1,169 boys and 1,123 girls
aged 6–9 years)
Total testosterone

3.7 ng/mL (maternal median
PFOA)

Earlier age of menarche

NS (interquartile difference of 4.3,
95% CI -0.4–9.1), boys
NS (4.2, 95% CI -0.7–9.4), girls
Inverse association (-4.9, 95%
CI -8.7 to -0.8)*, boys
NS (-2.5, 95% CI -6.7–1.8), girls
NS (-0.4, 95% CI -3.4–2.7), boys
Inverse association (-3.6, 95%
CI -6.6 to -0.5)*, girls
OR 1.01 (0.61–1.68)

1.4 ng/mL (maternal median
PFOA)

Cord estradiol

NS (p>0.05)

Cord testosterone
Cord testosterone: estradiol
ratio
Cord progesterone
Cord prolactin
Cord LH
Cord FSH
Cord SHGB

NS (p>0.05)
NS (p>0.05)

Cord insulin-like factor 3
Cord inhibin
Males
Females

NS (p>0.05)

Insulin-like growth factor-1

Christensen et al. 2011
General population (n=448 girls)
Itoh et al. 2016
General population (n=189 infants)

OR 0.95 (0.84–0.07)
OR 0.54 (0.35–0.84)*, 2nd quartile

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Association (p=0.040)*
NS (p>0.05)

 PERFLUOROALKYLS

372
2. HEALTH EFFECTS

Table 2-25. Summary of Effects on the Development of the Reproductive System in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Kristensen et al. 2013

3.6 and 4.4–19.8 ng/mL
Age of menarche
(maternal median PFOA and Menstrual cycle length
3rd PFOA tertile)
Total testosterone
SHBG
Free androgen index
Dehydroepiandrosterone
sulphate
Anti-Müllerian hormone
Number of follicles/ovary

Association (p=0.01)*

>4.1 ng/mL (maternal 3rd
tertile PFOA)

Testosterone
SHBG

β 0.24 (0.05–0.43)*, 3rd tertile
β 5.02 (-13.07–11.00), 3rd tertile

2.6 and 2.1 ng/mL (median
cord blood PFOA Denmark
and Finland cohorts)

Cryptorchidism

OR 0.51 (0.21–1.20), whole cohort
OR 0.35 (0.12–0.99, p=0.04 for
trend)*, Finland cohort 3rd tertile

20 and 18 ng/mL (PFOS
median in boys and girls)

Age of puberty
Boys
Girls

General population (n=343 females
approximately 20 years of age)

Maisonet et al. 2015
General population (n=72 girls aged 15 years)
Vesterholm Jensen et al. 2014
General population (n=107 cases
cryptorchidism [29 from Denmark and 78 from
Finland] and 108 matched controls from
Denmark and Finland)
PFOS
Lopez-Espinosa et al. 2011

Community (C8) (n=3,076 boys and 2,931 girls
aged 8–18)
Lopez-Espinosa et al. 2016
22.4 and 20.9 ng/mL (PFOS
median in boys and girls)
Community (C8) (n=1,169 boys and 1,123 girls
aged 6–9 years)

Estradiol

Total testosterone

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

OR 0.58 (0.37–0.90)*, 3rd quartile
OR 0.55 (0.35–0.86)*, 3rd quartile
Inverse association (interquartile
difference of -4.0, 95% CI -7.7
to -0.1)*, boys
NS (-0.3, 95% CI -4.6–4.2), girls
Inverse association (-5.8, 95%
CI -9.4 to -2.0)*, boys
Inverse association (-6.6, 95%
CI -10.1 to -2.8)*, girls

 PERFLUOROALKYLS

373
2. HEALTH EFFECTS

Table 2-25. Summary of Effects on the Development of the Reproductive System in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated
Insulin-like growth factor-1

Christensen et al. 2011

19.8 ng/mL (maternal median Earlier age of menarche
PFOS)

Resultc
Inverse association (-5.9, 95%
CI -8.3 to -3.3)*, boys
Inverse association (-5.6, 95%
CI -8.2 to -2.9)*, girls
OR 0.68 (0.40–1.13)

General population (n=448 girls)
Itoh et al. 2016
General population (n=189 infants)

5.2 ng/mL (maternal median
PFOS)

Cord estradiol
Males
Females
Cord testosterone
Cord testosterone: estradiol
ratio
Males
Females
Cord progesterone
Males
Females
Cord prolactin
Males
Females
Cord LH
Cord FSH
Cord SHGB
Cord insulin-like factor 3
Cord inhibin
Males
Females

***DRAFT FOR PUBLIC COMMENT***

Association (p=0.021)*
NS (p>0.05)
NS (p>0.05)

Inverse association (p=0.008)*
NS (p>0.05)
Association (p=0.043)*
Association (p=0.002)*
NS (p>0.05)
Association (p=0.001)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.001)*
NS (p>0.05)

 PERFLUOROALKYLS

374
2. HEALTH EFFECTS

Table 2-25. Summary of Effects on the Development of the Reproductive System in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

Kristensen et al. 2013

21.1 ng/mL (maternal median Age of menarche
PFOS)
Menstrual cycle length
Total testosterone
SHBG
Free androgen index
Dehydroepiandrosterone
sulphate
Anti-Müllerian hormone
Number of follicles/ovary

NS (p=0.28)

>1.4 ng/mL (amniotic fluid
3rd tertile PFOS)

Cryptorchidism
Hypospadias
Testosterone
Androstendione
Progesterone
Cortisol
DHEAS

OR 1.01 (0.66–1.53), 3rd tertile
OR 0.69 (0.35–1.38), 3rd tertile
Association (p=0.002)*
Association (p=0.001)*
Association (p=0.001)*
Association (p<0.001)*
NS (p=0.93)

Insulin-like factor 3
Cryptorchidism

Inverse association (p<0.001)*
OR 0.83 (0.44–1.58), whole cohort

General population (n=343 young women
approximately 20 years of age)

Toft et al. 2016
General population (270 cases cryptorchidism,
75 cases hypospadias, and 300 controls)

Vesterholm Jensen et al. 2014
General population (n=107 cases
cryptorchidism [29 from Denmark and 78 from
Finland] and 108 matched controls from
Denmark and Finland)

9.1 and 5.2 ng/mL (median
cord blood PFOS Denmark
and Finland cohorts)

***DRAFT FOR PUBLIC COMMENT***

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

 PERFLUOROALKYLS

375
2. HEALTH EFFECTS

Table 2-25. Summary of Effects on the Development of the Reproductive System in Humansa
Reference and study populationb

Serum perfluoroalkyl level Outcome evaluated

Resultc

8.1 and 7.0 ng/mL (PFHxS
median in boys and girls)

PFHxS
Lopez-Espinosa et al. 2016

1.6 ng/mL (maternal median
PFHxS)

Earlier age of menarche

NS (interquartile difference of -1.3,
95% CI -5.5–3.1), boys
NS (2.1, 95% CI -2.2–6.5), girls
NS (-2.7, 95% CI -6.4–1.2), boys
NS (0.2, 95% CI -3.5–4.0), girls
NS (-2.5, 95% CI -5.2–0.3), boys
NS (-2.1, 95% CI -4.8–0.7), girls
OR 0.89 (0.65–1.22)

>1.9 ng/mL (maternal 3rd
tertile PFHxS)

Testosterone
SHBG

β 0.18 (0.00–0.35), 3rd tertile
β 5.31 (-21.61–11.00), 3rd tertile

1.7 and 1.7 ng/mL (PFNA
median in boys and girls)

Estradiol

NS (interquartile difference of -2.5,
95% CI -6.2–1.4), boys
NS (-2.4, 95% CI -6.3–1.7), girls
NS (-2.1, 95% CI -5.5–1.3), boys
NS (-1.9, 95% CI -5.5–1.9), girls
Inverse association (-3.5, 95%
CI -6.0 to -1.0)*, boys
Inverse association (-3.8, 95%
CI -6.4 to -1.2)*, girls

Community (C8) (n=1,169 boys and 1,123 girls
aged 6–9 years)

Estradiol

Total testosterone
Insulin-like growth factor-1

Christensen et al. 2011
General population (n=448 girls)
Maisonet et al. 2015
General population (n=72 girls aged 15 years)
PFNA
Lopez-Espinosa et al. 2016
Community (C8) (n=1,169 boys and 1,123 girls
aged 6–9 years)

Total testosterone
Insulin-like growth factor-1

Maisonet et al. 2015

>0.6 ng/mL (maternal
3rd tertile PFNA)

Testosterone
SHBG

β 0.05 (-0.14–0.24), 3rd tertile
β 7.91 (-8.69–24.52), 3rd tertile

0.2 ng/mL (maternal median
PFOSA)

Earlier age of menarche

OR 0.91 (0.67–1.24)

General population (n=72 girls aged 15 years)
PFOSA
Christensen et al. 2011
General population (n=448 girls)
Me-PFOSA-AcOH

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

376
2. HEALTH EFFECTS

Table 2-25. Summary of Effects on the Development of the Reproductive System in Humansa
Reference and study populationb
Christensen et al. 2011
General population (n=448 girls)
Et-PFOSA-AcOH
Christensen et al. 2011

Serum perfluoroalkyl level Outcome evaluated

Resultc

0.4 ng/mL (maternal median
Me-PFOSA-AcOH)

Earlier age of menarche

OR 0.86 (0.66–1.12)

0.6 ng/mL (maternal median
Et-PFOSA-AcOH)

Earlier age of menarche

OR 1.03 (0.75–1.43)

General population (n=448 girls)
aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 13 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

CI = confidence interval; DHEAS = dihydroepiandrosterone sulfate; Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; FSH = follicle
stimulating hormone; LH = luteinizing hormone; Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane sulfonamide) acetic acid; NS = not significant; OR = odds ratio;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; SHBG = sex
hormone binding globulin

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

377
2. HEALTH EFFECTS

Document for Epidemiological Studies for Perfluoroalkyls, Table 13. Studies examining childhood
growth and examining the possible relationship between maternal serum perfluoroalkyl levels and body
weight and BMI in children and adults are discussed in Section 2.3, Body Weight. In general, the
epidemiology studies did not find associations between perfluoroalkyl exposure and adverse pregnancy
outcomes (miscarriage, preterm birth, or gestational age) for PFOA, PFOS, PFHxS, PFNA, or PFUA.
Mixed results have been found for birth outcomes, particularly birth weight. Some epidemiology studies
have found associations between maternal PFOA or PFOS exposure and decreases in birth weight, and
meta-analyses of these data have found that increases in maternal PFOA or PFOS were associated with
15–19 g or 5 g decreases in birth weight, respectively; accounting for maternal glomerular filtration rates
attenuated these results by about 50%. No consistent associations for alterations in birth weight were
found for other perfluoroalkyls (PFHxS, PFNA, PFDeA, PFUA, PFDoA, Me-PFOSA-AcOH, Et-PFOSAAcOH). Overall, no associations were found between serum PFOA, PFOS, PFHxS, PFNA, or PFUA and
increases in the risk of low birth weight or small for gestational age infants. No consistent results for
risks of birth defects have been found; these potential endpoints were only examined for a few
perfluoroalkyls. The available epidemiology data do not suggest associations between perfluoroalkyls
and IQ or scholastic achievement for PFOA, PFOS, PFHxS, PFNA, PFDeA, PFUA, or PFDoA.
Similarly, no associations were found between PFOA, PFOS, PFHxS, PFNA, or PFDeA and increased
risk of ADHD; several studies have found decreased risk of ADHD. Inconsistent results have been found
between PFOA and PFOS and delays in puberty or age of puberty, especially in girls.

Summaries of laboratory animal studies are presented in Tables 2-1, 2-3, 2-4, and 2-5 and the NOAEL
and LOAEL values are presented in Figures 2-4, 2-6, 2-7, and 2-8. Laboratory animal studies provide
strong evidence of the developmental toxicity of a number of perfluoroalkyl compounds. Prenatal losses
and decreases in pup survival were observed following exposure to PFOA, PFOS, PFNA, and PFDeA; no
deaths were observed in a single study of PFBuS. Decreases in fetal weights, birth weight, and pup
weight were observed in studies of PFOA, PFOS, PFNA, PFDeA, and PFUA; no effects on weight were
observed in studies on PFHxS or PFBuS. In PFOA studies, delays in mammary gland development were
observed at fairly low doses. Several studies have demonstrated biphasic alterations in motor activity in
rodents exposed to PFOA, PFOS, and PFHxS; no effects on locomotor activity were observed in a study
of PFDeA. Studies in laboratory animals have examined a number of developmental endpoints, including
pup survival, malformations, birth weight, mammary gland development, and neurodevelopment.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

378
2. HEALTH EFFECTS

PFOA

Epidemiology Studies—Pregnancy Outcomes. The results of available epidemiology studies of women
living near a PFOA facility and the general population do not suggest an association between serum
PFOA levels and adverse pregnancy outcomes. No increases in risk of miscarriage (Darrow et al. 2014;
Jensen et al. 2015; Savitz et al. 2012b; Stein et al. 2009), stillbirths (Savitz et al. 2012b), or pre-term birth
(Chen et al. 2012a; Darrow et al. 2013; Hamm et al. 2010; Stein et al. 2009; Whitworth et al. 2012a) were
found; the Whitworth et al. (2012a) general population studies reported a decrease in the risk of preterm
births among women with serum PFOA levels in the 4th quartile. Additionally, Apelberg et al. (2007b)
and Chen et al. (2012a) did not find associations between maternal serum PFOA levels and gestational
age.

Epidemiology Studies—Birth Outcomes. Community and general population exposure studies have
evaluated a number of birth outcomes including birth weight; risk of low birth weight; risk of small for
gestational age; birth length; head, chest, and abdominal circumferences; ponderal index; sex ratio; and
birth defects. In highly exposed populations, no association between maternal serum PFOA levels and
birth weight were found (Darrow et al. 2013; Nolan et al. 2009; Savitz et al. 2012b). Several general
population studies have found associations between maternal serum PFOA and birth weight. Fei et al.
(2007, 2008a), Lenters et al. (2016a), and Maisonet et al. (2012) found an inverse association between
maternal serum PFOA and birth weight. However, 12 other general population studies did not find
associations (Bach et al. 2016; Chen et al. 2012a; Govarts et al. 2016; Hamm et al. 2010; Kim et al. 2011;
Lee et al. 2013, 2016; Monroy et al. 2008; Robledo et al. 2015a; Wang et al. 2016; Washino et al. 2009;
Whitworth et al. 2012a). As illustrated in Figure 2-33, most studies found no association between
maternal serum PFOA levels and the risk of low birth weight infants (typically defined as <2,500 g)
(Chen et al. 2012a; Darrow et al. 2013; Fei et al. 2007, 2008a; Savitz et al. 2012b; Stein et al. 2009) or
found a decreased risk of low birth weight infants (Nolan et al. 2009; Savitz et al. 2012a). Similarly,
there were no increases in the risk for small for gestational age (Chen et al. 2012a; Fei et al. 2007, 2008a;
Hamm et al. 2010; Savitz et al. 2012b; Wang et al. 2016; Whitworth et al. 2012a); these data are
presented in Figure 2-34. One study (Savitz et al. 2012b) of C8 participants did find an increase in the
risk of small for gestational age; however, when the maternal serum PFOA levels were categorized into
percentiles, the risk was not increased in infants whose maternal serum PFOA levels were ≥80th percentile
(21.0–717.6 ng/mL). Two general population studies found inverse associations between maternal serum
PFOA levels and birth length, abdominal circumference, and/or ponderal index (ratio of birth weight to
birth length) (Apelberg et al. 2007b; Fei et al. 2007, 2008a). However, most studies did not find

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

379
2. HEALTH EFFECTS

Figure 2-33. Risk of Low Birth Weight Infant Relative to PFOA Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

380
2. HEALTH EFFECTS

Figure 2-34. Risk of Small For Gestational Age Infant Relative to PFOA Levels (Presented as Adjusted Odds
Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

381
2. HEALTH EFFECTS

associations between maternal serum PFOA levels and birth length; head, chest, or abdominal
circumference; or ponderal index (Bach et al. 2016; Chen et al. 2012a; Lee et al. 2013; Maisonet et al.
2012; Robledo et al. 2015a; Wang et al. 2016).

In a systematic review of 19 epidemiology studies discussed above, Johnson et al. (2014) evaluated the
possible association between PFOA exposure and fetal growth and concluded that there was sufficient
evidence that PFOA reduces fetal growth based on a moderate rating of the human evidence. A metaanalysis of the Apelberg et al. (2007b), Chen et al. (2012a), Fei et al. (2007, 2008a), Fromme et al.
(2010), Hamm et al. 2009), Kim et al. (2011), and Maisonet et al. (2012) studies showed an association
between PFOA and birth weight; a 1 ng/mL increase in serum or plasma PFOA was associated with
a -18.9 g (95% CI -29.8 to -7.9) change in birth weight. A second meta-analysis (Verner et al. 2015) of
the Apelberg et al. (2007b), Chen et al. (2012a), Fei et al. (2007), Hamm et al. (2010), Maisonet et al.
(2012), Washino et al. (2009), and Whitworth et al. (2012a) studies found a similar result, a 1 ng/mL
increase in PFOA levels was associated with a 14.72 g (95% CI -21.66 to -7.78) decrease in birth weight.
Verner et al. (2015) also utilized a PBPK model to evaluate the influence of glomerular filtration rate on
the association between maternal PFOA and birth weight. Utilizing simulated maternal PFOA levels at
delivery, a 1 ng/mL increase in PFOA was associated with a 7.13 g (95% CI -8.46 to -5.80) decrease in
birth weight; suggesting that glomerular filtration rate may be a confounding factor.

A small number of studies have examined the potential associations between PFOA exposure and risks of
birth defects. In a study of C8 Health Study participants, no increases in the risk of brain, gastrointestinal,
kidney, craniofacial, eye, limb, genitourinary, or heart defects were found (Stein et al. 2014c).

Epidemiology Studies—Neurodevelopmental Outcomes. A number of epidemiology studies have
evaluated neurodevelopment at various ages using maternal serum PFOA or cord blood PFOA as a
biometric of exposure. Fei et al. (2008b) did not find an increased risk of Apgar scores of <10 in
newborns. Utilizing the Neonatal Intensive Care Unit Network Neurobehavioral Scale (NNNS) in
5-week-old infants, Donauer et al. (2015) found an increased risk of reduced muscle tone (hypotonia),
which was associated with maternal serum PFOA levels, but found no associations on tests of social/easy
going or high arousal/difficult. Goudarzi et al. (2016) reported lower scores on tests of mental and
psychomotor development in female 6-month-old infants; no association was found when male and
female infants were grouped together. When the infants were tested at 18 months of age, no association
between maternal PFOA levels and mental and psychomotor indices were found. Fei et al. (2008b) did
not find associations between maternal PFOA levels and the risk of delays in motor, cognitive, or

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

382
2. HEALTH EFFECTS

language development in 6- and 18-month-old infants. It is noted that in the Fei et al. (2008b) study, the
mothers were asked to recall at what age the infants reached a developmental milestone, whereas
standardized tests of development were used in the other two studies. Although the Donauer et al. (2015)
and Goudarzi et al. (2016) studies suggest some delays in neurodevelopment in young infants, more
research is needed before establishing a possible relationship with PFOA.

Studies in children have examined possible links between PFOA and IQ, motor skills, behavior, and
ADHD. An association between estimated in utero PFOA levels and IQ was found in 6–12-year-old
children participating in the C8 Health Studies (Stein et al. 2013); higher IQ scores were found in children
with the highest estimated PFOA exposure levels. The study did not find an association with reading or
math skills. A general population study (Wang et al. 2015b) did not find an association between maternal
serum PFOA levels and IQ scores in children 5 or 8 years of age. In a study of adults (20 years of age),
Strøm et al. (2014) did not find an association between maternal PFOA levels and scholastic achievement.
A community study of children and adolescents did not find an association between serum PFOA levels
and learning problems in 12–15 or 5–18 year olds (Stein and Savitz 2011). Two studies (Fei and Olsen
2011; Høyer et al. 2015a) did not find associations between maternal PFOA levels and motor
coordination in 7 year olds or motor skills in 5–9 year olds.

Several studies have examined possible associations between maternal or child PFOA levels and scores
on tests/surveys that assess behavioral problems. No associations between maternal PFOA levels and
behavioral problems in 7 year olds (Fei and Olsen 2011) or behavioral regulation problems in 5 or 8 year
olds (Vuong et al. 2016). Similarly, no associations between serum PFOA levels and performance on
tasks requiring behavioral inhibition were observed in 9–11-year-old children (Gump et al. 2011). No
associations between breast milk PFOA levels and behavioral development in 6- and 24-month-old
infants were observed (Forns et al. 2015). In contrast, Høyer et al. (2015a) found an association between
maternal PFOA levels and behavioral problems in 5–9-year-old children; the risk was increased in
children with maternal PFOA levels in the 3rd tertile. Stein et al. (2014a) found an association between
the children’s serum PFOA levels and survey results on behavioral problems and emotional disturbances
in girls aged 6–12 years of age; this association was not found in boys or in boys and girls combined.
Additionally, the association was only found when the survey was completed by mothers, but not when
completed by the child’s teacher.

Ten studies have looked for a possible link between PFOA and ADHD in children. Two studies of
participants of the C8 Health Study found lower scores on tests for ADHD (Stein et al. 2013) or lower

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

383
2. HEALTH EFFECTS

risks of ADHD (Stein and Savitz 2011) associated with estimated in utero PFOA or child PFOA levels,
respectively. In a third community study in which parents and teachers completed surveys regarding
ADHD-like behaviors (Stein et al. 2014a), no association between the child’s serum PFOA (measured 3–
4 years before the surveys were completed) and ADHD-like behaviors were found when the mothers
completed the survey and an inverse association was found when the teachers completed the survey.
Segregating the children by sex resulted in an association in girls (mother-completed survey only) and no
associations in boys. Two general population studies have found associations between the risk of ADHD
or increases in ADHD behavior in children. An increase in the risk of parent-reported ADHD diagnosis
was observed in a study of 12–15-year-old NHANES participants (Hoffman et al. 2010). The second
study (Høyer et al. 2015a) found increases in hyperactivity among 5–9-year-old children with maternal
serum PFOA levels in the 3rd tertile. When this multinational cohort was segregated by country, the
association was only found in group of children from Greenland, but not in the Ukrainian cohort. Median
serum PFOA levels were slightly higher in the Greenland cohort; it is also noted that the median maternal
PFOS levels were 4 times higher in the Greenland cohort than in the Ukraine cohort. Other general
population studies have not found associations. Two case-control studies of children did not find
increased risks of being diagnosed with ADHD associated with maternal PFOA levels (Liew et al. 2015)
or cord blood PFOA levels (Ode et al. 2014). Two studies did not find associations between cord blood
PFOA levels and performance on tests evaluating for ADHD symptoms in 7-year-old children (Lien et al.
2016) or 18-month-old infants (Quaak et al. 2016). A third study found no association between maternal
PFOA levels and ADHD in 20 year olds (Strøm et al. 2014). In addition to looking at possible
relationships between PFOA and ADHD, two studies did not find associations between maternal PFOA
levels and autism behaviors (Braun et al. 2014) or the risk of autism diagnosis (Liew et al. 2015).

Epidemiology Studies—Development of the Reproductive System. Studies exploring possible links
between PFOA and alterations in the development of the reproductive system have examined several
outcomes including hormone levels in cord blood, hormone levels in children and adolescents, congenital
malformations of reproductive organs, and age of puberty in boys and girls.

A multinational case-control general population study (Vesterholm Jensen et al. 2014) found a decrease in
the risk of cryptorchidism in the Finnish cohort, but not in the Danish cohort or in the combined cohort.
With the exception of inhibin levels, no associations between maternal serum PFOA levels and cord
blood levels of reproductive hormones were found (Itoh et al. 2016). Cord inhibin was associated with
maternal serum PFOA levels in male infants, but not in female infants (Itoh et al. 2016). Some alterations
in reproductive hormone levels were found in 6–9-year-old boys and girls participating in the C8 Health

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

384
2. HEALTH EFFECTS

Study (Lopez-Espinosa et al. 2016). In boys, an inverse association between serum PFOA levels and total
testosterone levels were observed; no associations were found for estradiol levels or insulin-like growth
factor 1. In girls, an inverse association was found for insulin-like growth factor 1 levels and no
associations were found for estradiol or testosterone levels. In adolescent girls, an association between
maternal PFOA levels and testosterone levels was found (Maisonet et al. 2015a). This association was
not found in young adult females (Kristensen et al. 2013). Other reproductive hormones were not shown
to be associated with maternal PFOA levels (Kristensen et al. 2013; Maisonet et al. 2015a).

In a community exposure study (Lopez-Espinosa et al. 2011), increasing levels of serum PFOA were
associated with delays in menarche in girls aged 8–18 years. Serum PFOA levels in the 2nd, 3rd, and
4th quartiles were associated with 142-, 163-, and 130-day delays in the onset of menarche, respectively.
Using PBPK modeling, Wu et al. (2015) examined whether the association between serum PFOA and
delays in the onset of menarche observed in the Lopez-Espinosa et al. (2011) study were due to reverse
causality using a Monte Carlo PBPK model. They found that rapid growth around the time of
menstruation onset may contribute to the apparent association between PFOA and delay of menarche. In
the PBPK simulated study, the delay in the onset of menarche was 48 days for the 4th quartile (OR 0.82,
95% CI 0.76–0.88). A delay in menarche was also observed in a general population study; a 162-day
delay was estimated in the daughters of women with maternal serum PFOA levels in the 3rd tertile
(Kristensen et al. 2013). A second general population study did not find an association between maternal
serum PFOA levels and an earlier age of menarche (Christensen et al. 2011).

The only study available on age of puberty in males (Lopez-Espinosa et al. 2011) did not find an
association with serum PFOA levels.
Laboratory Animal Exposure Studies. Exposure of pregnant Sprague-Dawley rats to 25 mg/m3 APFO
on GDs 6–15 resulted in a statistically significant reduction (10.3%) in neonatal body weight on PND 1,
but the difference over controls was no longer significant on PND 4 (Staples et al. 1984). Exposure
concentrations ≤10 mg/m3 did not affect neonatal body weight. The incidence of malformations and
variations among the exposed groups and controls was comparable.

In utero exposure to PFOA resulted in prenatal losses and decreases in pup survival. An increase in
resorptions was observed in mice administered ≥5 mg/kg/day throughout gestation (Lau et al. 2006) or
2 mg/kg/day on GDs 11–16 (Suh et al. 2011). Prenatal losses were also observed in PFOA mouse studies
administering ≥6 mg/kg/day (Abbott et al. 2007), 5 mg/kg/day (White et al. 2011), or 20 mg/kg/day (Lau

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

385
2. HEALTH EFFECTS

et al. 2006) throughout gestation; an increase in the percentage of dams with total litter loss was also
observed at 5 mg/kg/day (Wolf et al. 2007). Administration of 20 mg/kg/day PFOA on GDs 7–17 or 10–
17 did not result in litter loss (Wolf et al. 2007); no effect on litter size was observed as a result of
administration of 5 mg/kg/day on GDs 8–17 (White et al. 2009). Gestational exposure (GDs 1–17) to
PFOA also resulted in perinatal losses in mice administered 3 mg/kg/day PFOA (Ngo et al. 2014) and
decreases in pup survival in mice exposed to ≥0.6 mg/kg/day (Abbott et al. 2007), 3 mg/kg/day (Albrecht
et al. 2013), or 5 mg/kg/day (Lau et al. 2006; Yahia et al. 2010; White et al. 2011, Wolf et al. 2007);
100% pup mortality was observed in the offspring of mice exposed to 10 mg/kg/day throughout gestation
(Yahia et al. 2010). Decreased pup survival was also observed in mice exposed to 5 mg/kg/day PFOA on
GDs 15–17 (Wolf et al. 2007). No alterations in fetuses/litter or survival were observed at 1 mg/kg/day
PFOA (Lau et al. 2006; White et al. 2011). Butenhoff et al. (2004b) also reported increases in pup
mortality on PNDs 6–8 in the offspring of rats administered 30 mg/kg/day PFOA throughout gestation
and during lactation.

Decreases in birth weight have not been consistently found in mouse studies with PFOA. No significant
alterations in birth weight were observed in mice exposed to 3 mg/kg/day (Albrecht et al. 2013), 5 or
10 mg/kg/day (Lau et al. 2006), or 20 mg/kg/day (Abbott et al. 2007); decreases in birth or fetal weight
were observed at 5 mg/kg/day (Hines et al. 2009; Yahia et al. 2010), 10 mg/kg/day (Suh et al. 2011), and
20 mg/kg/day (Lau et al. 2006; Li et al. 2016). A decrease in mean litter weight on PNDs 2–14 was
observed in mice administered ≥0.5 mg/kg/day PFOA on GDs 6–17 (Hu et al. 2010) and a decrease in
pup body weight on PND 20 was observed in mice exposed to 5 mg/kg/day on GDs 8–17 or 12–17
(White et al. 2007). In utero exposure of mice to PFOA throughout gestation resulted in decreases in pup
body weight in mice exposed to 1 mg/kg/day (Abbott et al. 2007; Hines et al. 2009), ≥3 mg/kg/day (Lau
et al. 2006; Wolf et al. 2007), and 5 mg/kg/day (Yahia et al. 2010; White et al. 2007, 2011). In a crossfostering study, lactation-only exposure (maternal dose of 5 mg/kg/day PFOA) resulted in decreased body
weight in female pups on some PNDs (2, 3, 4, and 22, but not on PNDs 7, 10, 15, or 17) (Wolf et al.
2007). Hines et al. (2009) monitored body weights from birth to 18 months of age in female mice
exposed in utero to PFOA on GDs 1–17. At weaning, decreases in body weight were observed at 1 and
5 mg/kg/day; by 10 weeks of age, there were no differences in body weight between the controls and
mice exposed to ≥1 mg/kg/day. Significant increases in body weight were observed in mice exposed to
0.1 and 0.3 mg/kg/day, and by 20–29 weeks of age, the increases in body weight were observed in mice
exposed to 0.01, 0.1, or 0.3 mg/kg/day. The largest increase in body weight gain (9.6%) was observed at
0.1 mg/kg/day; because the weight increase was less than 10%, the 0.1 mg/kg/day was considered a
NOAEL. At 40 weeks of age, the increased body weight was observed in the 0.1 and 0.3 mg/kg/day

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

386
2. HEALTH EFFECTS

groups. At termination (18 months of age), there were no differences in body weight between the
controls and mice exposed to 0.01–3 mg/kg/day; a decrease in body weight was observed at 5 mg/kg/day.
During the period of increased body weight in the lower-dose animals, there were no changes in serum
glucose levels or the response to a glucose challenge, but there were significant increases in insulin and
leptin levels at 0.01 and 0.1 mg/kg/day. Although there were no changes in the percentage of body fat to
body weight measurements in mice at 42 weeks of age, at 18 months of age, significant decreases in
abdominal body fat and increases in intrascapular brown fat was observed at ≥1 mg/kg/day PFOA (Hines
et al. 2009). Based on systematic review of pup body weight data from the Abbott et al. (2007), Hines et
al. (2009), Lau et al. (2006), White et al. (2007, 2009, 2011), and Wolf et al. (2007) mouse studies,
Koustas et al. (2014) concluded that there was sufficient evidence that exposure to PFOA adversely
affected fetal growth in animals. A meta-analysis estimate was a decrease of 0.023 g pup body weight per
1 mg/kg/day increase in PFOA dose.

A few studies have examined the potential of PFOA to induce malformations/variations. Lau et al. (2006)
reported reductions in ossification of the proximal phalanges at ≥1 mg/kg/day and supraoccipital at 10 or
20 mg/kg/day. This study also reported enlarged fontanels in pups exposed to ≥1 mg/kg/day and tail and
limb defects at ≥5 mg/kg/day; however, there was no clear dose-response for these effects. Koskela et al.
(2016) found altered femur and tibial bone morphology and decreased tibial mineral density in the
offspring of mice exposed to 0.3 mg/kg/day in the diet on GDs 1–21. An increased percentage of litters
with microcardia was also observed in the offspring of mice exposed to 10 or 20 mg/kg/day (Lau et al.
2006). No increases in the occurrence of malformations/variations were observed in the offspring of rats
administered 100 mg/kg/day on GDs 6–15 (Staples et al. 1984) or in a 2-generation study at doses as high
as 30 mg/kg/day (Butenhoff et al. 2004b).
Delayed eye opening was observed in the offspring of mice administered ≥1 mg/kg/day PFOA on GDs 1–
17 (Abbott et al. 2007) and in mice administered 5 mg/kg/day throughout gestation (Lau et al. 2006; Wolf
et al. 2007). Neither Albrecht et al. (2013) nor Lau et al. (2006) found alterations in eye opening in mice
exposed to 3 mg/kg/day PFOA on GDs 1–17. Lau et al. (2006) also reported advanced preputial
separation at ≥1 mg/kg/day and delayed vaginal opening at 20 mg/kg/day. The effect in the male
offspring is in contrast to the Butenhoff et al. (2004b) study, which found delays in preputial separation in
rats exposed to 30 mg/kg/day PFOA; a delay in vaginal patency was also observed at this dose.

A series of studies conducted by White and associates found significant delays in mammary gland
development in the offspring of mice administered 1 mg/kg/day PFOA via gavage on GDs 8–17 (White et

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

387
2. HEALTH EFFECTS

al. 2011) or 5 mg/kg/day PFOA on GDs 1–17, 8–17, 12–17, 10–17, 13–17, or 15–17 (White et al. 2007,
2009, 2011). The delay was characterized as reduced ductal elongation and branching and delays in
timing and density of terminal end buds and was observed at all observational periods (PNDs 10, 20, 22,
and 42, and 63 and 18 months of age). Decreases in mammary epithelial growth, as assessed by
developmental scoring, were observed in the offspring of mice exposed to 0.01 mg/kg/day on GDs 1–17
(Tucker et al. 2015), 0.3 mg/kg/day on GDs 1–17 (Macon et al. 2011), or 0.01 mg/kg/day on GDs 10–17
(Macon et al. 2011). Tucker et al. (2015) noted that the delays in mammary gland development began at
puberty and continued during young adulthood. Albrecht et al. (2013) did not find any alterations in
mammary gland development on PND 20 in mouse offspring following in utero exposure to PFOA on
GDs 1–17. Delayed mammary gland development was also observed in offspring only exposed via
lactation (maternal dose of 3 mg/kg/day PFOA on GDs 1–17); the effects were observed on PNDs 42 and
63, but not on PND 22 (White et al. 2009). In a multigeneration study conducted by White et al. (2011),
delays in mammary gland development were not consistently observed in the F2 offspring of F1 females
that were exposed in utero to 1 or 5 mg/kg/day PFOA. However, delays in mammary gland development
were observed in the F1 and F2 offspring exposed to 0.001 mg/kg/day in utero (GDs 7–17) and
postnatally. The investigators (White et al. 2011) noted that the delay in mammary gland development
did not appear to affect lactational support based on normal survival and growth of the F2 pups. Tucker
et al. (2015) noted dose-related strain differences on the effect of PFOA on mammary gland differences;
effects were observed in CD-1 mice at ≥0.01 mg/kg/day and in C57BL/6 mice at ≥0.3 mg/kg/day (the
highest NOAEL for this strain was 0.1 mg/kg/day); it is noted that the serum PFOA concentrations at a
given dose were lower in the C57Bl/6 mice than in the CD-1 mice.

A consistent finding in the four mouse studies evaluating the neurodevelopmental toxicity of PFOA is an
increase in motor activity. Increases in horizontal and ambulatory locomotor activity (tested on PND 60)
were observed in the offspring of mice exposed to 0.1 mg/kg/day in the diet on GD 7 through PND 21
(Sobolewski et al. 2014); a decrease in resting time was also observed in the males. Significant increases
in open field activity were observed at PND 36 in the offspring of mice exposed to 1.6 mg/kg/day
throughout gestation and lactation (Cheng et al. 2013). Johansson et al. (2008) and Onishchenko et al.
(2011) demonstrated a biphasic alteration in motor activity: an initial period of decreased activity
followed by increased activity. Johansson et al. (2008) administered a single dose of 8.7 mg/kg/day
PFOA to mice on PND 10 and monitored spontaneous activity for a 1-hour period when the mice were
2 or 4 months of age. In the first 20-minute period, there was a decrease in spontaneous activity, followed
by a 20-minute period with an activity level similar to controls, and a 20-minute period with significantly
increased spontaneous activity. Similarly, Onishchenko et al. (2011) reported an increase in activity in a

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

388
2. HEALTH EFFECTS

48-hour period in the adult offspring of mice exposed to 0.3 mg/kg/day PFOA throughout gestation;
however, there was a decrease in activity during the initial 3 hours of testing. Johansson et al. (2008) also
found an increased susceptibility of the cholinergic system in mice exposed to 0.58 or 8.7 mg/kg/day
PFOA on PND 10. In control mice, an injection of nicotine resulted in increases in activity; mice
exposed to 0.58 mg/kg/day also responded with an increase in activity, although the increase was less
than that observed in the controls. In contrast, nicotine resulted in a decrease in activity in mice exposed
to 8.7 mg/kg/day. Exposure to PFOA did not alter learning or memory, as evidenced by the lack of effect
on maze tests (Cheng et al. 2013; Johansson et al. 2008). Tests of neurobehavioral development found
altered motor coordination and impaired negative geotaxis reflex, but no effect on righting reflex or cliff
avoidance, in the offspring of mice exposed to 1.6 mg/kg/day throughout gestation and lactation (Cheng
et al. 2013). Decreases in initial novel object exploratory behavior were also observed at 0.1 mg/kg/day,
but there were no alterations in recognition time for novel objects (Sobolewski et al. 2014).

Although not investigated in mammalian species, studies in chicken embryos and hatchlings demonstrate
the developmental cardiotoxicity of PFOA (Jiang et al. 2012, 2013, 2016). The effects following in ovo
exposure include thinning of the right ventricular wall in chick embryos and alterations in left ventricular
posterior wall dimension, volume, heart rate, stroke volume, and ejection fraction in the hatchlings (Jiang
et al. 2012). Tests with WY 14,643, a PPARα agonist, and PFOA provide evidence that the
cardiotoxicity involves both PPARα and bone morphorgenic protein 2 (BMP2) pathways (Jiang et al.
2013). Comparisons of results following in ovo exposure and in vitro exposure suggest that the
cardiotoxicity was not likely due to cytotoxicity, but rather an alteration in early cardio morphology and
function processes (Jiang et al. 2016).

Summary. Epidemiology studies have examined a number of potential developmental outcomes in
communities living near a PFOA facility and in the general populations. Although not consistently
reported, the available general population studies suggest an inverse association between maternal serum
PFOA levels and birth weight; a number of studies have not found this association. Several systematic
reviews of these data have concluded that there was sufficient evidence that maternal PFOA levels are
associated with reductions in fetal growth. After correcting for glomerular filtration rate, a small decrease
in birth weight was associated with increases in maternal serum PFOA. Two of the three studies
evaluating possible effects of sexual maturation found small delays in the start of menarche associated
with maternal serum PFOA levels. Overall, the data do not suggest associations between serum PFOA
levels and adverse pregnancy outcomes such as miscarriages or stillbirths, most birth outcomes (e.g., risk
of low birth weight, risk of small for gestational age, birth length, ponderal index, sex ratio, or birth

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

389
2. HEALTH EFFECTS

defects), or neurodevelopmental outcomes (IQ or scholastic achievement, motor skills, and risk of
ADHD). Animal studies provide strong evidence that developmental toxicity is a sensitive target of
PFOA toxicity. Observed effects include prenatal losses and decreases in pup survival, decreases in birth
weight, developmental delays such as delayed eye opening, delays in mammary gland development, and
increased motor activity.

PFOS

Epidemiology Studies—Pregnancy Outcomes. No associations between maternal PFOS levels and the
risk of miscarriages were observed in several studies (Darrow et al. 2014; Jensen et al. 2015; Stein et al.
2009). Two studies reported increases in the risk of preterm birth associated with maternal serum PFOS
levels in the >90th percentile (>23.2 ng/mL) (Stein et al. 2009) or cord blood PFOS levels in the 3rd and
4th quartiles (≥5.68 ng/mL) (Chen et al. 2012a), and one study reported a decrease risk in preterm birth
(Whitworth et al. 2012a). Two other studies did not find associations for preterm birth (Fei et al. 2007,
2008a; Hamm et al. 2010).

Epidemiology Studies—Birth Outcomes. Occupational, community, and general population exposure
studies have examined the possible links between maternal PFOS levels and a number of birth outcomes
including birth weight; risk of low birth weight; risk of small for gestational age; birth length; head, chest,
and abdominal circumferences; ponderal index; sex ratio; and birth defects. Most studies did not find
associations between maternal serum PFOS levels and birth weight (Apelberg et al. 2007b; AshleyMartin et al. 2016; Darrow et al. 2013; Bach et al. 2016; Fei et al. 2007, 2008a; Govarts et al. 2016;
Hamm et al. 2010; Kim et al. 2011; Lee et al. 2013, 2016; Lenters et al. 2016a; Maisonet et al. 2012;
Monroy et al. 2008; Robledo et al. 2015a; Whitworth et al. 2012a), including an occupational exposure
study (Grice et al. 2007) in which female workers were exposed to very high levels of PFOS (serum
levels ranged from 1,300 to 1,970 ng/mL). Three studies did find inverse associations between birth
weight and maternal serum PFOS levels. In the Washino et al. (2009) study, an inverse association was
found between maternal serum PFOS levels and birth weight; segregating by sex resulted in an inverse
association in girls, but not in boys. The magnitude of the change was small, 148.8 g decrease in birth
weight per log unit increase in maternal PFOS for combined. Maisonet et al. (2012) also reported small
decreases in birth weight (140.1 g) in infants whose mother’s serum PFOS levels were in the 3rd tertile.
Similarly, Chen et al. (2012a) reported an inverse association between cord blood PFOS and birth weight,
but the magnitude was small (110.2 g decrease per ln unit increase in cord PFOS levels). Although these
studies found decreases in birth weight associated with PFOS levels, no studies found increases in the risk

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

390
2. HEALTH EFFECTS

of low birth weight infants (Chen et al. 2012a; Darrow et al. 2013; Fei et al. 2007, 2008a; Stein et al.
2009) or small for gestational age infants (Chen et al. 2012a; Fei et al. 2007, 2008a; Hamm et al. 2010;
Whitworth et al. 2012a). The ORs for low birth weight and small for gestational age risks are presented
in Figures 2-35 and 2-36. Maternal PFOS was not associated with birth length (Apelberg et al. 2007b;
Bach et al. 2016; Chen et al. 2012a; Fei et al. 2007, 2008a; Lee et al. 2013; Robledo et al. 2015a;
Washino et al. 2009) with the exception of the finding of a small decrease in birth length (-0.63 cm) that
was associated with 3rd tertile maternal serum PFOS levels. Two studies reported inverse associations
between ponderal index and cord blood PFOS levels (Apelberg et al. 2007b) and maternal serum PFOS
levels (Lee et al. 2013); other studies did not find this effect (Chen et al. 2012a; Maisonet et al. 2012;
Robledo et al. 2015a). Two studies reported small decreases in head circumference, which were
associated with maternal serum PFOS levels (Apelberg et al. 2007b) and cord blood PFOS (Chen et al.
2012a); other studies have not found associations (Bach et al. 2016; de Cock et al. 2014; Fei et al. 2007,
2008a; Lee et al. 2013; Robledo et al. 2015a; Washino et al. 2009). Verner et al. 2015) conducted a metaanalysis of the Apelberg et al. (2007b), Chen et al. (2012a), Fei et al. (2007), Hamm et al. (2010),
Maisonet et al. (2012), Washino et al. (2009), and Whitworth et al. (2012a) studies and found that a
1 ng/mL increase in maternal PFOS levels was associated with a 5.00 g (95% CI -8.92 to -1.09) decrease
in birth weight. When the data were re-analyzed utilizing a PBPK model to account for glomerular
filtration rate, the magnitude of the effect of PFOS on birth weight decreased (Verner et al. 2015). A
1 ng/mL increase in PFOS was associated with a 2.72 g (95% CI -3.40 to -2.04) decrease in birth weight.

One study reported no increases in the risk of birth defects associated with maternal serum PFOS levels
(Stein et al. 2009); a second study found an increased risk of congenital cerebral palsy in girls, but not in
boys (Liew et al. 2014). Bae et al. (2015) did not find associations between the odds of having a boy and
paternal or maternal serum PFOS levels.

Epidemiology Studies—Neurodevelopmental Outcomes. Epidemiology studies examined several
aspects of neurodevelopment, including age of reaching neurobehavioral milestones, IQ, motor
development, behavior, ADHD, and autism. Fei et al. (2008b) did not find associations between maternal
PFOS levels and the risk of having an Apgar score of <10 or in motor and mental development at
6 months. However, the study did find that some neurobehavioral milestones (delay in sitting, early use
of word-like sounds, and delays in using two- word sentences) were associated with maternal PFOS
levels. Goudarzi et al. (2016) did not find alterations on mental and psychomotor development in 6- and
18-month-old infants that were associated with maternal serum PFOS levels. A third study of infants did
not find alterations in neurobehavioral or muscle coordination tests (Donauer et al. 2015).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

391
2. HEALTH EFFECTS

Figure 2-35. Risk of Low Birth Weight Infant Relative to PFOS Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

392
2. HEALTH EFFECTS

Figure 2-36. Risk of Small for Gestational Age Infant Relative to PFOS Levels (Presented as Adjusted Odds
Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

393
2. HEALTH EFFECTS

In the only study evaluating IQ, Wang et al. (2015b) did not find associations between maternal PFOS
levels and IQ score in children 5 or 8 years of age. Strøm et al. (2014) found no associations between
scholastic achievement in 20 year olds and maternal PFOS levels. In a study of children living in a
community with high PFOA contamination, Stein and Savitz (2011) found decreases in the risk of
learning problems in children 5–18 or 12–15 years of age. In contrast, Vuong et al. (2016) found
increased risks of global executive functioning and metacognition problems that were associated with
maternal PFOS levels. Three studies have not found links between maternal PFOS levels and behavioral
health and motor coordination/skills in children (Fei and Olsen 2011; Høyer et al. 2015a) or between
breast milk PFOS levels and behavioral development in 6- and 24-month-old infants (Forns et al. 2015).
A third study (Vuong et al. 2016) found an increased risk for problems with behavioral regulation. The
available data do not suggest an association between maternal PFOS levels or cord blood PFOS levels
and the risk of ADHD or ADHD behaviors (Hoffman et al. 2010; Liew et al. 2015; Ode et al. 2014;
Quaak et al. 2016; Stein and Savitz 2011; Strøm et al. 2014), although Liew et al. (2015) found a
decreased risk of ADHD diagnosis in children whose mothers had serum PFOS levels in the 4th quartile.
Similarly, Høyer et al. (2015a) did not find increases in the risk of hyperactivity in children and Gump et
al. (2011) found a decrease in impulsivity. Braun et al. (2014) and Liew et al. (2015) did not find
associations between maternal PFOS and autism risk.

Epidemiology Studies—Development of Reproductive System. Several epidemiology studies have
examined the possible associations between PFOS and the development of the reproductive system,
including the risk of congenital defects to reproductive organs, alterations in reproductive hormone levels,
and age of puberty; the results of these studies are summarized in Table 2-25. No alterations in the risk of
cryptorchidism (Toft et al. 2016; Vesterholm Jensen et al. 2014) or hypospadias (Toft et al. 2016) were
found in two studies. Itoh et al. (2016) reported associations between maternal PFOS levels and
alterations in cord blood hormone levels, in particular estradiol in males, testosterone:estradiol ratio in
males (inverse association), progesterone levels in males and females, prolactin levels in females, and
inhibin levels in males. Similarly, Toft et al. (2016) found associations between amniotic fluid PFOS
levels and levels of testosterone, androstenedione, progesterone, and insulin-like factor 3 (inverse
association) in amniotic fluid. Lopez-Espinosa et al. (2016) also found a number of alterations in
reproductive hormone levels in 6–9-year-old boys and girls. In the boys, inverse associations between
serum PFOS levels and estradiol, total testosterone, and insulin-like growth factor 1 were observed.
Inverse associations between total testosterone and insulin-like growth factor 1 and serum PFOS levels
were also observed in the girls. A study of young adult women found no associations between
reproductive hormone levels and maternal PFOS levels (Kristensen et al. 2013).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

394
2. HEALTH EFFECTS

A study of 8–18-year-old children found delays in the age of puberty in boys and girls (Lopez-Espinosa et
al. 2011) that were associated with serum PFOS levels. In the children with serum PFOS levels in the
3rd and 4th quartiles, the respectively delays were 131 and 190 days in boys and 141 and 138 days in girls.
In contrast, two other studies have not found alterations in either the age of menarche or an earlier age of
menarche that were associated with maternal PFOS levels (Christensen et al. 2011; Kristensen et al.
2013). The differences in the biomarker of exposure and the potential exposure to high levels of PFOA in
the Lopez-Espinosa et al. (2011) community study make it difficult to compare the results of these three
studies. As discussed in the PFOA section, Wu et al. (2009) reanalyzed the Lopez-Espinosa et al. (2011)
data using a Monte Carlo PBPK model, which accounted for rapid growth occurring around puberty, and
found much shorter delays in the age of menarche than found in the Lopez-Espinosa et al. (2011) study.
In the girls with simulated serum PPFOS levels in the 4th quartile, the delay was 72 days (OR 0.75, 95%
CI 0.70–0.81).

Laboratory Animal Exposure Studies. Increases in fetal mortality and decreases in pup survival have
also been observed in rats and mice exposed to PFOS in utero (Abbott et al. 2009; Chen et al. 2012b;
Fuentes et al. 2006; Grasty et al. 2003, 2005; Lau et al. 2003; Lee et al. 2015a; Luebker et al. 2005a,
2005b; Ngo et al. 2014; Thibodeaux et al. 2003; Xia et al. 2011; Yahia et al. 2008). Increases in the
number of resorptions and dead fetuses were observed in mice administered ≥0.5 mg/kg/day (Lee et al.
2015a); increases in abortions between GD 22 and 28 were observed in rabbits treated with
3.75 mg/kg/day PFOS by gavage on GDs 6–20 (Case et al. 2001). Decreases in the number of live
fetuses were observed in mice exposed to ≥2.0 mg/kg/day on GDs 11–16 and 20 mg/kg/day on GDs 1–17
(Thibodeaux et al. 2003) or GDs 0–17 (Yahia et al. 2008). Increases in perinatal losses were observed in
the litters of mice administered ≥0.1 mg/kg/day PFOS on GDs 1–17 (Ngo et al. 2014). Pup survival is
affected at lower maternal doses. Significant decreases in pup survival were observed in rats at
1.6 mg/kg/day (dams were exposed for 6 weeks prior to mating and during gestation through lactation
days 4 or 21) (Luebker et al. 2005a, 2005b) and in mice exposed to 4.5 mg/kg/day on GDs 15–18 (Abbott
et al. 2009); no alterations in pup survival were observed in rats or mice exposed to 1 mg/kg/day (Luebker
et al. 2005b; Yahia et al. 2008). A series of studies by Grasty et al. (2003) in rats that were exposed for
4 days during different gestational periods showed that the pup was more susceptible if exposure occurred
later in gestation. On PND 4, pup survival was 70, 50, 60, 20, or 5% for exposures on GDs 2–5, 6–9, 10–
13, 14–17, or 17–20, respectively. Grasty et al. (2003) and others (Abbott et al. 2009; Chen et al. 2012b;
Lau et al. 2003) also noted that most deaths occurred within the first 4 PNDs, with the highest rates
occurring on PND 1. Lau et al. (2003) and Luebker et al. (2005a) found that cross fostering did not

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

395
2. HEALTH EFFECTS

significantly improve pup survival; deaths were observed in the in utero only exposure group. However,
Luebker et al. (2005a) showed that rats exposed in utero and during lactation had the highest pup
mortality, as compared to other cross-fostered groups. The mechanism involved in the early pup
mortality has not been identified, but there is some indication that pulmonary deficits may be a
contributing factor. At high doses (50 mg/kg/day administered on GDs 19–20), pups demonstrated
difficulty breathing within minutes of birth (Grasty et al. 2003). Histological examination of the lungs of
pups exposed to 25 or 50 mg/kg/day on GDs 19–20 showed evidence of delayed lung maturation (Grasty
et al. 2003, 2005), specifically, an increase in the proportion of solid lung tissue and a decrease in the
proportion of small airway tissue. A comparison of the lungs of PFOS-exposed neonates to control
fetuses (GD 21) showed that 17 and 50% of the lung tissue in the neonates exposed to 25 or
50 mg/kg/day, respectively, on GDs 19–20 was not histologically different from the control fetuses
(Grasty et al. 2005). Administration of therapeutic agents known to enhance terminal lung maturation
and accelerate surfactant production did not improve pup survival (Grasty et al. 2005). Histological
damage has also been reported in pups exposed to lower PFOS levels. Lung atelectasis was observed in
pups exposed to 10 mg/kg/day on GDs 0–18 (Yahia et al. 2008). No lung effects were observed in pups
exposed to 1 mg/kg/day or in fetuses exposed to 20 mg/kg/day on GDs 0–17 (Yahia et al. 2008).
Alveolar hemorrhage, thickened epithelial walls of the pulmonary alveolus, focal lung consolidation, and
focal infiltration of inflammation cells were observed in pups exposed to 2 mg/kg/day on GDs 0–21; no
lung effects were observed at 0.1 mg/kg/day (Chen et al. 2012b).

Decreases in fetal body weight, birth weight, and pup body weight have been observed in rats, mice, and
rabbits exposed to PFOS (Case et al. 2001; Chen et al. 2012b; Era et al. 2009; Fuentes et al. 2006, 2007b;
Grasty et al. 2003; Lau et al. 2003; Lee et al. 2015a; Luebker et al. 2005a, 2005b; Rogers et al. 2014; Xia
et al. 2011; Yahia et al. 2008). In rats, the lowest-adverse-effect level for decrease in fetal body weight
was 10 mg/kg/day following administration on GDs 2–20 (Thibodeaux et al. 2003) and the highest noeffect level was 5 mg/kg/day, also identified in the Thibodeaux et al. (2003) study. Decreases in rat pup
birth weight and body weight on PND 4 were observed in the offspring of rats exposed to 0.4 mg/kg/day
for 42 days prior to mating and gestation through lactation day 4 (Luebker et al. 2005b). Mice appear to
be less sensitive to the effect of PFOS on pup body weight than rats (Lau et al. 2003). Exposure of rats to
2 mg/kg/day PFOS on GDs 2–21 resulted in significant decreases in birth weight and pup body weight on
PNDs 1–3; exposure to 5 mg/kg/day resulted in decreases in pup body weight through PND 19. In
contrast, no alterations in birth weight or pup body weight were observed in mice exposed to doses as
high as 5 mg/kg/day on GDs 1–18. Fuentes et al. (2007b) reported the lowest LOAEL of 6 mg/kg/day for
decreases in pup weight in mice exposed on GDs 12–18. Decreases in fetal body weight were observed

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

396
2. HEALTH EFFECTS

in mice exposed to 10 mg/kg/day on GDs 0–17 (Yahia et al. 2008). Fuentes et al. (2006) did not find
decreases in fetal body weight following exposure to 6 mg/kg/day on GDs 6–18. In rabbits, a decrease in
fetal body weight was observed following exposure to 2.5 mg/kg/day on GDs 6–20, but not at
1 mg/kg/day (Case et al. 2001).

Several studies also reported delays in developmental milestones. Delays in eye opening were observed
in rats exposed to 2 mg/kg/day on GDs 2–21 (Lau et al. 2003) or 0.4 mg/kg/day for 42 days prior to
mating and throughout the gestation and lactation periods (Luebker et al. 2005a) and in mice exposed to
8.5 mg/kg/day on GDs 15–18 (Abbott et al. 2009). Fuentes et al. (2007b) did not find a delay in eye
opening in mouse pups exposed to 6 mg/kg/day on GDs 12–18, but did find a delay in pinna detachment
at this dose level. A decrease in neuromuscular development, as evidenced by a delay in tail pull reflex,
climbing ability, and forelimb grip strength, was observed in mice exposed to 6 mg/kg/day on GDs 12–18
(Fuentes et al. 2007b).

Prenatal exposure to PFOS has resulted in malformations/anomalies/variations in rats, mice, and rabbits
(Case et al. 2001; Era et al. 2009; Thibodeaux et al. 2003; Yahia et al. 2008). An increased incidence of
cleft palate was observed in rats exposed to 10 mg/kg/day on GDs 2–20 (Thibodeaux et al. 2003) and in
mice exposed to 10 mg/kg/day on GDs 0–17 (Yahia et al. 2008), 15 mg/kg/day on GDs 1–17
(Thibodeaux et al. 2003), 20 mg/kg/day on GDs 1–17 (Era et al. 2009), and 50 mg/kg/day on GDs 11–15
(Era et al. 2009). Other skeletal and external alterations included sternal defects in rats exposed to
10 mg/kg/day on GDs 2–20 (Thibodeaux et al. 2003) and mice exposed to 1 mg/kg/day on GDs 0–17
(Yahia et al. 2008), delayed skeletal ossification in rabbits exposed to 2.5 mg/kg/day on GDs 6–20 (Case
et al. 2001), wavy ribs and spina bifida occulta in mice exposed to 10 mg/kg/day on GDs 1–17 (Yahia et
al. 2008), and tail abnormalities and delayed ossification of phalanges at 20 mg/kg/day (Yahia et al.
2008). Visceral abnormalities, consisting of enlarged right atrium at 10 mg/kg/day, and ventricular septal
defects at 20 mg/kg/day were observed in mice exposed on GDs 1–17 (Thibodeaux et al. 2003). No
malformations/anomalies/variations were found by Thibodeaux et al. (2003) in mice exposed to
1 mg/kg/day on GDs 1–17 or by Fuentes et al. (2006) in mice exposed to 6 mg/kg/day on GDs 6–18. In
addition to the previously discussed histological alterations observed in the pups exposed to lethal doses,
mild to severe intracranial dilatation of blood vessels was observed in fetuses exposed to 20 mg/kg/day on
GDs 0–17 and in pups exposed to 10 mg/kg/day on GDs 0–18 (Yahia et al. 2008). No histological
alterations were observed in the heart of rat pups exposed to 2 mg/kg/day on GDs 2–21 (Xia et al. 2011);
the study also found no alterations in heart rate or blood pressure. Lee et al. (2015b) found increases in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

397
2. HEALTH EFFECTS

cholesterol levels in fetal livers of mice exposed to PFOA on GDs 1–17 and Wan et al. (2014b) found
increases in relative liver weights in pups on PND 21.

Neurodevelopmental studies have shown that prenatal and/or postnatal exposure to PFOS can affect
motor activity, but does not appear to affect learning or memory. A significant decrease in locomotion
was observed in male mice aged 5–8 weeks exposed to 0.3 mg/kg/day on GDs 1–17 when they were
placed in a novel environment (Onishchenko et al. 2011). Hallgren et al. (2015) reported biphasic
alterations in spontaneous activity in 2-month-old mice administered a single dose of 11.3 mg/kg on
PND 10; locomotor activity was reduced during the first 20-minute period, was unchanged in the second
period, and increased during the third period. Decreases in circadian activity were noted in males and
increases in the number of inactive periods were noted in males and females when they were observed
over a 48-hour period. The study also found increased inactivity in an elevated plus maze test. In an
open field test of 70-day-old mice exposed to 6 mg/kg/day on GDs 12–18, an increase in the amount of
time spent in the center of the field was found; no changes in vertical movement were found (Ribes et al.
2010). In 3-month-old mice exposed to 6 mg/kg/day on GDs 12–18, a decrease in the distance traveled
was observed after 20–25 minutes in an open field apparatus; activity was not affected during the first
5 minutes of the test (Fuentes et al. 2007a). In a 15-minute open field test, prenatal exposure to
6 mg/kg/day PFOS on GDs 12–18 did not alter motor activity in 3-month-old mice (Fuentes et al. 2007b).
In contrast, Butenhoff et al. (2009b) found a significant increase in locomotion in male rats exposed to
0.3 or 1.0 mg/kg/day PFOS throughout gestation and lactation. However, this effect was only observed in
male rats on PND 17; no significant alterations were observed on PNDs 13, 21, or 61. An increase in
locomotion was observed in female rats on PND 21 exposed to 1.0 mg/kg/day, but not at other time
points. To evaluate the biological relevance of the increased activity, activity was analyzed by 1-minute
sequential time periods. The investigators concluded that the increased activity observed in the
0.3 mg/kg/day males at PND 17 and 1.0 mg/kg/day females at PND 21 was not treatment-related due to
the lack of significant changes in total or ambulatory activity and the similarity in habituation pattern
between the treated groups and controls. In the 1.0 mg/kg/day PND 17 males, the pattern of habituation
differed from controls and there was an increase in ambulatory activity; this increase in locomotor activity
was considered to be related to PFOS exposure. The increased activity was observed in the last three time
periods. Postnatal exposure (PND 10) to 11.3 mg/kg/day resulted in an initial decrease in motor activity
followed by an increase in activity in 2- and 4-month-old mice (Johansson et al. 2008). In 2-month-old
mice exposed to 0.75 mg/kg/day, there was a decrease in total activity during the first 20 minutes of
testing, but not during the remaining 40 minutes of the test; no changes in activity were observed in the
4-month-old mice exposed to 0.75 mg/kg/day. Johansson et al. (2009) also found an altered response to

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

398
2. HEALTH EFFECTS

nicotine exposure. Exposure to 11.3 mg/kg/day PFOS resulted in a decrease in motor activity in response
to nicotine exposure, as compared to the increased activity observed in controls; no significant alteration
was observed at 0.75 mg/kg/day. Two studies testing muscle coordination did not find alterations in the
offspring of rats exposed to 3.2 mg/kg/day for 6 weeks prior to mating and throughout gestation and
lactation (Luebker et al. 2005a) or mice exposed to 6 mg/kg/day on GDs 12–18 (Fuentes et al. 2007b). A
decrease in muscle coordination was observed in mice exposed to 0.3 mg/kg/day on GDs 1–17
(Onishchenko et al. 2011). Perinatal exposure to PFOS did not significantly alter learning or memory in
rats exposed to 2 mg/kg/day on GDs 2–21 and tested on PND 21 (Lau et al. 2003), the offspring of rats
exposed to 3.2 mg/kg/day for 6 weeks prior to mating and throughout gestation and lactation and tested
on PNDs 21 and 70 (Luebker et al. 2005a), or mice exposed to 6 mg/kg/day on GDs 12–18 and tested at
3 months of age (Fuentes et al. 2007a). In contrast, decreases in spatial learning ability were observed in
the offspring of mice exposed to 0.8 mg/kg/day on GD 1 through PND 1 or on PNDs 1–7 (Wang et al.
2015b).

The effect of pre- and/or postnatal exposure to PFOS on serum lipid levels, thyroid function, and immune
function has also been evaluated by a small number of studies. In the offspring of rats exposed to
1.6 mg/kg/day for 6 weeks prior to mating through GD 20, a significant decrease in fetal serum
cholesterol levels and an increase in LDL cholesterol levels were observed (Luebker et al. 2005b). In rats
exposed through PND 4, there was a decrease in serum triglyceride levels in the pups exposed to
1 mg/kg/day (Luebker et al. 2005b). No alterations in thyroid histology or follicular morphology were
observed in rats exposed to 1 mg/kg/day on GD 0–PND 20 (Chang et al. 2009), and no alterations in TSH
levels were observed in the Chang et al. (2009) study or in rats exposed to 2 mg/kg/day on GDs 2–21
(Lau et al. 2003). Decreases in total and free T4 levels were observed in rats exposed to 1 mg/kg/day on
GDs 2–21 (Lau et al. 2003); free T4 levels remained low through PND 35. Similarly, a cross-fostering
study found decreases in T4 levels in rats exposed to 3.2 mg/kg/day in utero, during lactation only, and
throughout gestation and lactation (Yu et al. 2009b). Altered immune function was observed in mice
exposed to PFOS on GDs 1–17 (Keil et al. 2008). At 5 mg/kg/day, an altered IgM antibody response to
sRBCs was observed in 8-week-old males; decreases in CD3+ and CD4+ lymphocytes were also
observed. At 1 mg/kg/kg/day, there was decreased in NK cell activity in males; no effects were observed
at 0.1 mg/kg/day.

Summary. A number of epidemiology studies have evaluated developmental outcomes in occupational,
community (living near a PFOA facility), and general exposure populations. Overall, these studies have
not found associations between serum PFOS and adverse pregnancy outcomes (miscarriage, preterm

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

399
2. HEALTH EFFECTS

birth), most birth outcomes (risks of low birth weight or small for gestational age, birth length, head, chest
or abdominal circumferences, ponderal index, sex ratio, or birth defects), or neurodevelopmental
outcomes (IQ, motor development, behavior, ADHD, or autism). It is noted that some studies have found
associations for these effects and for some effects, only a couple of studies examined the endpoint.
Although most studies did not find associations between maternal PFOS and birth weight, a meta-analysis
did find a small decrease in birth weight was associated with increasing maternal PFOS levels, after
adjustment for glomerular filtration rate. There is also some suggestive evidence that PFOS levels may
be associated with small delays in the age of puberty in boys and girls. Studies in laboratory animals
clearly indicate that developmental toxicity is a sensitive outcome of PFOS exposure. Oral exposure
studies have reported increases in fetal mortality and decreases in pup survival; decreases in fetal body
weight, birth weight, and pup body weight; delays in developmental milestones such as eye opening;
increases in skeletal malformations/anomalies/variations such as cleft palate and delayed skeletal
ossification; and decreases in offspring motor activity.

PFHxS

Epidemiology Studies—Pregnancy Outcomes. Two studies, summarized in Table 2-22 have evaluated
possible links between pregnancy outcomes and maternal PFHxS levels. Jensen et al. (2015) did not find
an association between maternal PFHxS levels and the risk of miscarriage. Hamm et al. (2010) found a
decreased risk of preterm births among women with serum PFHxS levels in the 3rd tertile.

Epidemiology Studies—Birth Outcomes. General population studies have evaluated possible
associations between maternal PFHxS levels and birth outcomes including birth weight, length, small for
gestation age, and birth defects; studies are summarized in Table 2-23. Bach et al. (2016) and Maisonet et
al. (2012) reported inverse associations between maternal PFHxS levels and birth weight; however, other
studies have not found associations (Hamm et al. 2010; Kim et al. 2011; Lee et al. 2013, 2016; Lenters et
al. 2016a; Monroy et al. 2008). The risk of low birth weight infants was not examined in the available
epidemiology studies; Hamm et al. (2010) did not find an association between maternal PFHxS level and
the relative risk of small for gestational age. Several studies did not find associations between maternal
PFHxS levels and birth length, head circumference, or ponderal index (Bach et al. 2016; Lee et al. 2013).
Maisonet et al. (2012) found an inverse association for birth length, but no association for ponderal index.
Only one study examined possible birth defects; Liew et al. (2014) did not find an association between
maternal PFHxS levels and the risk of congenital cerebral palsy in a case-control study.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

400
2. HEALTH EFFECTS

Epidemiology Studies—Neurodevelopmental Outcomes. Epidemiology studies, summarized in
Table 2-24, have examined PFHxS-related alterations in risks of ADHD, autism, IQ, and behavior. Wang
et al. (2015b) did not find associations between maternal PFHxS levels and IQ in 5- or 8-year-old
children. However, Vuong et al. (2016) found a higher risk of performing poorly on tests of global
executive function with increasing maternal PFHxS levels. No association between serum PFHxS levels
and the risk of learning problems was found in children living in a community with high PFOA levels
(Stein and Savitz 2011). Gump et al. (2011) found an inverse association between serum PFHxS levels
and performance on tasks requiring behavioral inhibition; Vuong et al. (2016) did not find alterations in
behavioral regulation. Two studies evaluated the risk of ADHD and reported conflicting findings. Stein
and Savitz (2011) reported increases in risk of ADHD in 5–18 and 12–15 year olds with serum PFHxS
levels in the 2nd, 3rd, or 4th quartile, whereas Liew et al. (2015) reported an inverse association between
maternal PFHxS levels and risk of ADHD. This study also did not find an increase in the risk of autism;
Braun et al. (2014) also found no association between maternal PFHxS levels and performance on tests
assessing autism.

Epidemiology Studies—Development of the Reproductive System. No associations between
reproductive hormone levels and serum PFHxS levels (Lopez-Espinosa et al. 2016) or maternal serum
PFHxS levels (Maisonet et al. 2015a) were found in boys and girls 6–9 years of age or in girls 15 years of
age. Christensen et al. (2011) did not find an association between maternal PFHxS levels and risk of an
earlier menarche. Summaries of these epidemiology studies are presented in Table 2-25.

Laboratory Animal Studies. Administration of 9.2 mg/kg/day PFHxS on PND 10 resulted in a decrease
in spontaneous motor activity during the first 20 minutes of the test and an increase in activity in the last
20 minutes of the test (Viberg et al. 2013). The study also assessed the influence of PFHxS on nicotineinduced behavior. In the 9.2 mg/kg/day PFHxS group, exposure to nicotine did not significantly affect
spontaneous motor activity, which was in contrast to the nicotine-induced increases in spontaneous motor
activity observed in the controls and lower PFHxS groups. Another study evaluating the developmental
toxicity of PFHxS did not find alterations in litter size, pup survival, or pup body weight in rats exposed
to 10 mg/kg/day PFHxS for 14 days prior to mating and throughout gestation and lactation (Butenhoff et
al. 2009a; Hoberman and York 2003).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

401
2. HEALTH EFFECTS

PFNA

Epidemiology Studies—Pregnancy Outcomes. Two studies (summarized in Table 2-22) have examined
pregnancy outcome. Jensen et al. (2015) found an increase in the risk of having a miscarriage before
gestation week 12, which was associated with maternal serum PFNA levels. No alterations in the risk of
preterm birth was found in a study conducted by Chen et al. (2012a).

Epidemiology Studies—Birth Outcomes. Several studies have examined the possible link between birth
outcomes and maternal PFNA levels, these studies are summarized in Table 2-23. Most studies did not
find an association between birth weight and maternal PFNA levels (Bach et al. 2016; Chen et al. 2012a;
Lee et al. 2016; Lenters et al. 2016a; Monroy et al. 2008; Robledo et al. 2015a). No alterations in the risk
of low birth weight or small for gestational age were found in a study conducted by Chen et al. (2012a).
Wang et al. (2016) did find an inverse association between maternal PFNA levels and birth weight in girls
only. Chen et al. (2012a) found an association between maternal PFNA levels and birth length, but other
studies have not found alterations (Bach et al. 2016; Robledo et al. 2015a; Wang et al. 2016). Most
studies did not find alterations in ponderal index or head circumference (Bach et al. 2016; Chen et al.
2012a; Robledo et al. 2015a; Wang et al. 2016); Chen et al. (2012a) reported an inverse association
between cord PFNA levels on ponderal index. No associations between maternal PFNA or paternal
PFNA levels and the odds of a male birth were observed in a general population study (Bae et al. 2015).
Liew et al. (2014) did not find alterations in the risk of congenital cerebral palsy that were associated with
maternal PFNA levels.

Epidemiology Studies—Neurodevelopmental Outcomes. Several potential neurodevelopmental
outcomes have been examined in epidemiology studies; these studies are summarized in Table 2-24. No
association between maternal PFNA levels and full-scale IQ scores were observed in 8-year-old children
(Wang et al. 2015b); however, an association was found for visual IQ. Maternal PFNA levels were not
associated with IQ scores in 5 year olds (Wang et al. 2015b). Stein and Savitz (2011) found a decrease in
the risk of learning problems in 5–18 or 12–15 year olds with serum PFNA levels in the two highest
quartiles or in the 4th quartile, respectively. Vuong et al. (2016) did not find an association between
maternal PFNA levels and metacognition or global executive functioning in 5 or 8 year olds. Mixed
results have been found in studies on behavior. Gump et al. (2011) found a decrease in behavioral
response inhibition that was associated with serum PFNA levels in children aged 9–11 years, and Lien et
al. (2016) reported inverse associations between cord blood PFNA levels in inattention and hyperactivity/
inattention in 7-year-old children, but no effect on hyperactivity/impulsivity. Vuong et al. (2016) did not

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

402
2. HEALTH EFFECTS

find an association between maternal PFNA levels and behavior regulation. Three studies have not found
associations between PFNA levels and ADHD risk (Hoffman et al. 2010; Liew et al. 2015; Stein and
Savitz 2011). Similarly, maternal PFNA levels do not appear to be associated with autism (Braun et al.
2014; Liew et al. 2015).

Epidemiology Studies—Development of the Reproductive System. An inverse association between
PFNA levels and insulin-like growth factor 1 was found in boys and girls aged 6–9 years (LopezEspinosa et al. 2016). No associations were found between PFNA and estradiol or total testosterone in 6–
9 years olds (Lopez-Espinosa et al. 2016) or between maternal PFNA and testosterone or sex hormone
binding globulin levels in 15-year-old girls (Maisonet et al. 2015a). Additionally, no association between
maternal serum PFHxS levels and risk of earlier age of menarche were observed in girls (Christensen et
al. 2011). Summaries of these three studies are presented in Table 2-25.

Laboratory Animal Studies. Three studies were identified that examined the developmental toxicity of
PFNA in laboratory animals. Full litter resorptions were observed in mice administered 10 mg/kg/day on
GDs 1–17; maternal weight loss was also observed at this dose level (Das et al. 2015). At
≤1.5 mg/kg/day, decreases in postnatal survival were observed (Das et al. 2015; Wolf et al. 2010).
Decreases in birth weight were observed in female offspring of rats administered 5 mg/kg/day PFNA on
GDs 1–20 (Rogers et al. 2014) and in male offspring of mice administered ≥3 mg/kg/day PFNA on
GDs 1–17 (Das et al. 2015). No skeletal or visceral abnormalities were observed in mouse pups (Das et
al. 2015). Reductions in nephron endowment (number of functioning nephrons at birth) were observed in
male rat pups on PND 22 (Rogers et al. 2014). This study also found increases in systolic blood pressure
in pups at 10 weeks of age; no alterations were observed at 26 or 52 weeks of age. Delays in eye opening
and decreased in pup body weight gain were observed in offspring of mice administered 2.0 mg/kg/day
on GDs 1–18 (Wolf et al. 2010). Studies in PPARα knockout mice did not find alterations in pup
survival, birth weight, pup body weight gain, or day of eye opening at maternal doses as high as
2.0 mg/kg/day (Wolf et al. 2010). Comparison between the results in tests using wild-type mice and
knockout mice suggests that PPARα plays a role in PFNA developmental toxicity (Wolf et al. 2010).
PFDeA

Epidemiology Studies—Pregnancy Outcomes. In the only identified epidemiology study examining
pregnancy outcomes, Jensen et al. (2015) found an increased risk of miscarriage that was associated with
maternal PFDeA levels.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

403
2. HEALTH EFFECTS

Epidemiology Studies—Birth Outcomes. A small number of epidemiology studies examined risks of
adverse birth outcomes associated with maternal PFDeA exposure; these studies are summarized in
Table 2-23. Wang et al. (2016) found an inverse association between maternal PFDeA levels and birth
weight in female infants only. This study also found an increased risk for small for gestational age among
female infants. Other studies have not found associations (Bach et al. 2016; Lee et al. 2016; Lenters et al.
2016a; Robledo et al. 2015a). Epidemiology studies have not found associations between birth length,
ponderal index, and/or head circumference and maternal PFDeA levels (Bach et al. 2016; Robledo et al.
2015a; Wang et al. 2016). Liew et al. (2014) did not find alterations in the risk of congenital cerebral
palsy in boys or girls and Bae et al. (2015) did not find alterations in odds of a male birth associated with
maternal or paternal PFDeA levels. Additionally, Kim et al. (2016b) did not find associations between
serum PFDeA levels and thyroid parameters.

Epidemiology Studies—Neurodevelopmental Outcomes. Several studies have evaluated the potential of
PFDeA to adversely affect neurodevelopment; see Table 2-24 for a summary of the studies. Wang et al.
(2015b) did not find associations between maternal PFDeA levels and IQ in 5- and 8-year-old children.
Similarly, Vuong et al. (2016) did not find alterations in scores on tests of global executive functioning
and metacognition in 5- or 8-year-old children. This study also found no alteration in behavioral
regulation. In contrast, Gump et al. (2011) found increases in impulsivity. Liew et al. (2015) found
decreases in the risk of ADHD and autism in children.

Laboratory Animal Studies. An increase in fetal mortality was observed in mice exposed to
12.8 mg/kg/day PFDeA on GDs 6–15 (Harris and Birnbaum 1989); this dose level was also associated
with a marked decrease in fetal weight/litter (50% lower than controls), 100% incidence of variations in
ossification of the braincase, decreases in maternal body weight, and maternal mortality. Decreases in
fetal body weight/litter were observed at ≥1 mg/kg/day. The study did not find alterations in the
occurrence of cleft palate, soft tissue malformations, or skeletal malformations. In mice exposed to
10.8 mg/kg/day PFDeA on PND 10, there was no effect on spontaneous activity, habituation,
performance on an elevated maze test, or response to a nicotine injection (Johansson et al. 2008). These
results differ from the Johansson et al. (2008) findings when mice were exposed to PFOA or PFOS and
the findings of Viberg et al. (2013) in mice exposed to PFHxS.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

404
2. HEALTH EFFECTS

PFUA

Epidemiology Studies—Pregnancy Outcomes. Limited epidemiology studies evaluated pregnancy
outcomes. Jensen et al. (2015) did not find an alteration in the risk of miscarriage before gestation
week 12.

Epidemiology Studies—Birth Outcomes. The results from a study conducted by Wang et al. (2016)
found an inverse association between maternal PFUA levels and birth weight and an increased risk of
small for gestation age among female infants. The remaining epidemiology studies have not found
alterations in infant size (birth weight, birth length, ponderal index, head circumference) (Bach et al.
2016; Chen et al. 2012a; Lee et al. 2016; Lenters et al. 2016a) or the risks of low birth weight (Chen et al.
2012a) or small for gestational age (Chen et al. 2012a). No association between serum PFUA levels and
thyroid parameters were observed in infants (Kim et al. 2016a). The results of the epidemiology studies
examining associations between birth outcome and PFUA are presented in Table 2-23.

Epidemiology Studies—Neurodevelopmental Outcomes. The results of two studies examining possible
associations between neurodevelopmental outcome and PFUA are summarized in Table 2-24. Wang et
al. (2015b) found no association between maternal PFUA levels and IQ score in 5 and 8 year olds; the
study did find an inverse association with scores on tests assessing performance IQ. Lien et al. (2016)
found no associations between cord blood PFUA levels and performance on behavioral tests.

Laboratory Animal Studies. One study was identified that examined the potential developmental toxicity
of PFUA (Takahashi et al. 2014); the study found decreases in pup body weight at birth and on PND 4 in
the offspring of rats administered via gavage 1.0 mg/kg/day PFUA.
PFBuS

Laboratory Animal Studies. No alterations in pup survival, body weight, or development were observed
at doses as high as 1,000 mg/kg/day in a 2-generation rat study of potassium PFBuS (Lieder et al. 2009b).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

405
2. HEALTH EFFECTS

PFBA

Epidemiology Studies—Birth Outcomes. In the only available epidemiology studies examining birth
outcomes, no associations were found between serum PFBA levels and thyroid parameters in infants
(Kim et al. 2016a).

Laboratory Animal Studies. A delay (approximately 1 day) in eye opening was observed in the offspring
of mice administered via gavage 35 mg/kg/day PFBA on GDs 1–17 (Das et al. 2008).
PFDoA

Epidemiology Studies—Birth Outcomes. General population studies conducted by Lee et al. (2016) and
Lenters et al. (2016a) did not find associations between cord blood PFDoA or maternal PFDoA levels and
birth weight. Wang et al. (2016) found an inverse association between maternal PFDoA levels and birth
weight and head circumference in female infants; no alteration in the risk of small for gestation age was
found. The results of these three studies are summarized in Table 2-23.

Epidemiology Studies—Neurodevelopmental Outcomes. As summarized in Table 2-24, only one study
examined neurodevelopmental outcomes. In this study, maternal PFDoA levels were not associated with
IQ scores in 5- or 8-year-old children (Wang et al. 2015b).
PFOSA

Epidemiology Studies. Robledo et al. (2015a) found an inverse association between maternal PFOSA
levels and birth weight in boys, but not in girls; paternal PFOSA levels were not associated with birth
weight. The study did not find alterations in birth length, head circumference, or ponderal index (see
Table 2-23). Bae et al. (2015) did not find alterations in the odds of a male birth that was associated with
maternal or paternal PFOSA levels. As summarized in Table 2-24, only one study evaluated possible
associations between PFOSA and neurodevelopmental outcomes. Gump et al. (2011) reported an inverse
association between serum PFOSA levels and performance on tasks requiring behavioral inhibition. In
the only study examining development of the reproductive system, Christensen et al. (2011) did not find
an association between maternal serum PFOSA levels and the risk of an earlier age of menarche in girls
(see Table 2-25).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

406
2. HEALTH EFFECTS

Me-PFOSA-AcOH

Epidemiology Studies. Bae et al. (2015) found decreased odds of a female birth that was associated with
paternal Me-PFOSA-AcOH levels; there were no alterations in the odds of male births associated with
maternal or paternal Me-PFOSA-AcOH levels. Robledo et al. (2015a) found no associations between
maternal or paternal Me-PFOSA-AcOH levels and birth weight, birth length, head circumference, or
ponderal index. The results of these studies are presented in Table 2-23. Christensen et al. (2011) did not
find an association between maternal Me-PFOSA-AcOH levels and an earlier age of menarche; the results
of the study are summarized in Table 2-25.
Et-PFOSA-AcOH

Epidemiology Studies. Two epidemiology studies (summarized in Table 2-23) evaluated birth outcomes.
No associations were found between paternal or maternal Et-PFOSA-ACOH levels and the odds of male
birth (Bae et al. 2015) or birth weight, birth length, head circumference, or ponderal index (Robledo et al.
(2015a). An earlier age of menarche was not associated with maternal Et-PFOSA-AcOH levels were
found in a study conducted by Christensen et al. (2011).
2.18 OTHER NONCANCER
Overview. A number of epidemiology studies have examined the possible associations between
perfluoroalkyls and outcomes related to diabetes; the results of these studies are summarized in
Table 2-26, with additional study details presented in the Supporting Document for Epidemiological
Studies for Perfluoroalkyls, Table 14. Overall, the epidemiology studies do not provide support for an
association between serum perfluoroalkyl levels and increases in the risk of diabetes or related outcomes
(e.g., increases in blood glucose, glucose tolerance) for PFOA, PFOS, PFHxS, PFNA, PFDeA, PFUA,
PFHpA, or PFOSA. Additionally, results of studies on PFOA, PFOS, PFHxS, and Me-PFOSA-AcOH do
not suggest a link between perfluoroalkyls and gestational diabetes. Only one laboratory animal study
examined other noncancer endpoints (Table 2-5).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

407
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Leonard 2006

>250–≤750 ng/mL PFOA

Diabetes deaths

SMR 183 (112–283)* (males only)

Occupational (n=6,027)
Leonard et al. 2008

NR

Diabetes deaths

SMR 197 (123–298)*

Occupational (n=6,027)
Lundin et al. 2009

Probable exposure

Diabetes deaths

SMR 2.0 (1.2–3.2)*

Occupational (n=3,992)
Raleigh et al. 2014

NR

Diabetes deaths

SMR 0.76 (0.50–1.11)

Occupational (n=9,027)
Steenland et al. 2015

Cumulative

Risk of diabetes

RR 1.10 (0.77–1.57) no lag
RR1.12 (0.76–1.66) 10-year lag

Occupational (n=3,713)
Steenland and Woskie 2012

580 ng/mL (median PFOA)

Diabetes deaths

Occupational (n=1,088)
Anderson-Mahoney et al. 2008

NR

Self-reported diabetes

SMR 1.90 (1.35–2.61)* no lag
SMR 1.90 (0.98–3.33) 10-year lag
SMR 1.73 (0.83–3.18) 20-year lag
SPR 1.54 (1.16–2.05)*

Community (n=566)
Karnes et al. 2014

Cumulative

Self-reported diabetes

122.7 ng/mL (mean PFOA)

Fasting blood glucose
Validated diabetes

PFOA

Community (n=32,254 C8 participants)

MacNeil et al. 2009
Community (n=13,922 C8 participants)

***DRAFT FOR PUBLIC COMMENT***

HR 1.00 (0.99–1.00, p=0.60),
retrospective analysis
HR 1.00 (1.00–1.01, p=0.31),
prospective analysis
NS (p>0.05)
OR 0.72 (0.52–1.00) (10th decile)

 PERFLUOROALKYLS

408
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Fisher et al. 2013

2.46 ng/mL (geometric mean
PFOA)

Insulin

NS (p=0.12)

Blood glucose
HOMA-IR
Insulin
β-cell function
Fasting blood glucose
HOMA-IR

NS (p=0.17)
NS (p=0.10)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

Insulin
β-cell function
Fasting blood glucose
HOMA-IR
Diabetes
HOMA-IR

Association (p<0.05)
Association (p<0.05)
NS (p>0.05)
NS (p>0.05)
OR 0.97 (0.61–1.53, p=0.88)
NS (p=0.20)

Self-reported diabetes

OR 0.69 (0.41–1.16, p=0.158)

General population (NHANES) (n=3,966)

10.39 ng/mL (M)
9.47 ng/mL (F)
(4th PFOA quartile)

Nelson et al. 2010

4.6 ng/mL (mean PFOA)

HOMA (adolescent)
HOMA (adult)

NS (p=0.16) (M), NS (p=0.11) (F)
NS (p>0.05)

1.68 ng/mL (geometric mean
PFOA)

Gestational diabetes
Impaired glucose
tolerance
Diabetes

NS (p=0.86 for trend)
NS (p=0.36 for trend)

General population (n=2,700)
Lin et al. 2009
General population (NHANES)
(n=474 adolescents);

1.51 ng/mL (mean log PFOA)

1.48 ng/mL (mean log PFOA)
General population (NHANES) (n=969 adults)

Lind et al. 2014
General population (n=1,016)
Melzer et al. 2010

General population (NHANES)
(n=306 adolescent and 524 adults)
Shapiro et al. 2016

3.3 ng/mL (median PFOA)

General population (1,274 pregnant women)
Su et al. 2016
General population (n=571)

5.8–8.0 ng/mL
(2nd PFOA quartile)

Fasting blood glucose
Glucose tolerance

***DRAFT FOR PUBLIC COMMENT***

OR 0.39 (0.16–0.96)* (inverse
association)
Inverse association (p<0.01 for
trend)*
Inverse association (p<0.01 for
trend)*

 PERFLUOROALKYLS

409
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Glycated hemoglobin
3.07 and 3.94 ng/mL
(geometric mean PFOA in
women with or without
gestational diabetes)

Gestational diabetes

Inverse association (p=0.04 for
trend)*
OR 1.86 (1.14–3.02)*

8.04 ng/mL
(geometric mean PFOS)

Insulin
Blood glucose

NS (p=0.88)
NS (p=0.96)

HOMA-IR
Insulin
HOMA-IR
β-cell function
Blood glucose
Insulin
HOMA-IR

NS (p=0.25
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
Association (p<0.05)*
Association (p<0.05)*

β-cell function
Blood glucose
Diabetes
HOMA

Association (p<0.05)*
NS (p>0.05)
OR 1.43 (0.94–22.16, p=0.09)
NS (p=0.51)

Self-reported diabetes

OR 0.87 (0.57–1.31, p=0.491)

General population (NHANES) (n=3,966)

57.73 ng/mL (M)
50.96 ng/mL (F)
(4th quartile mean PFOS)

Nelson et al. 2010

25.3 ng/mL (mean PFOS)

HOMA (adolescent)
HOMA (adult)

NS (p=0.18) (M), NS (p=0.22) (F)
NS (p>0.05)

4.58 ng/mL (geometric mean
PFOS)

Gestational diabetes
Impaired glucose
tolerance

NS (p=0.70 for trend)
NS (p=0.74 for trend)

Zhang et al. 2015a
General population (n=258)
PFOS
Fisher et al. 2013
General population (n=2,700)
Lin et al. 2009

3.11 ng/mL (log mean PFOS)

General population (NHANES)
(n=474 adolescents)
General population (NHANES) (969 adults)

Lind et al. 2014
General population (n=1,016)
Melzer et al. 2010

General population (NHANES)
(n=306 adolescent and 524 adults)
Shapiro et al. 2016
General population (1,274 pregnant women)

3.19 ng/mL (log mean PFOS)

13.2 ng/mL (median PFOS)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

410
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Su et al. 2016

>4.8 ng/mL
(4th PFOS quartile)

Diabetes

OR 3.37 (1.18–9.65)*

Fasting blood glucose
Glucose tolerance test
Glycated hemoglobin
Gestational diabetes

Association (p<0.01 for trend)*
Association (p≤0.01 for trend)*
Association (p=0.04 for trend)*
OR 1.13 (0.75–1.72)

Insulin
Blood glucose
HOMA-IR

NS (p=0.89)
NS (p=0.98)
NS (p=0.20)

Insulin
HOMA-IR
Β-cell function
Blood glucose
0.60 ng/mL (log mean PFHxS) Insulin
HOMA-IR

NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

General population (n=571)

Zhang et al. 2015a
General population (n=258)
PFHxS
Fisher et al. 2013

13.10 and 12.04 ng/mL
(geometric mean PFOS in
women with or without
gestational diabetes)
2.18 ng/mL (geometric mean
PFHxS)

General population (n=2,700)
Lin et al. 2009
General population (NHANES)
(n=474 adolescents);

General population (NHANES) (n=969 adults)

0.95 ng/mL (log mean)

Lind et al. 2014

2.1 ng/mL (median PFHxS)

General population (n=1,016)
Nelson et al. 2010

2.6 ng/mL (mean PFHxS)

General population (NHANES)
(n=306 adolescent and 524 adults)
Shapiro et al. 2016
General population (n=1,274 pregnant women)

1.02 ng/mL
(geometric mean PFHxS)

Β-cell function
Blood glucose
Diabetes
HOMA

NS (p>0.05)
NS (p>0.05)
OR 1.00 (0.74–1.35, p=0.98)
NS (p=0.29)

HOMA (adolescent)
HOMA (adult)

NS (p=0.20) (M),
Inverse association (p=0.001)* (F)
NS (p>0.05)

Gestational diabetes

NS (p=0.73 for trend)

Impaired glucose
tolerance

NS (p=0.44 for trend)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

411
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

0.35 ng/mL (log mean PFNA)

Insulin
HOMA-IR
β-cell function
Blood glucose
Insulin
HOMA-IR

Inverse association (p<0.05)*
NS (p>0.05)
Inverse association (p<0.05)*
NS (p>0.05)
NS (p>0.05)
NS (p>0.05)

β-cell function
Blood glucose
Diabetes
HOMA

NS (p>0.05)
NS (p>0.05)
OR 1.30 (0.85–1.97, p=0.22)
NS (p=0.90)

PFNA
Lin et al. 2009
General population (NHANES)
(n=474 adolescents)
General population (NHANES) (n=969 adults)

0.21 ng/mL (log mean PFNA)

Lind et al. 2014

0.7 ng/mL (median PFNA)

General population (n=1,016)
Nelson et al. 2010

1.3 ng/mL (mean PFNA)

HOMA (adolescent)
HOMA (adult)

NS (p=0.83) (M), (p=0.20) (F)
NS (p>0.05)

>5.1 ng/mL
(4th PFNA quartile)

Diabetes

Glycated hemoglobin

OR 0.31 (0.11–0.88)* 4th quartile
(inverse association)
NS (p=0.10 for trend)
Inverse association (p<0.01 for
trend)*
NS (p=0.11 for trend)

Gestational diabetes

OR 1.06 (0.70–1.60)

General population (NHANES)
(n=306 adolescent and 524 adults)
Su et al. 2016
General population (n=571)

Zhang et al. 2015a
General population (n=258)

Fasting blood glucose
Glucose tolerance test

1.23 and 1.20 ng/mL
(geometric mean) PFNA in
women with or without
gestational diabetes)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

412
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

0.41 and 0.40 ng/mL ng/mL
(geometric mean PFDeA in
women with or without
gestational diabetes)

Gestational diabetes

OR 1.04 (0.70–1.53)

0.3 ng/mL (median PFUA)

Diabetes

OR 0.95 (0.59–1.52; p=0.81)

HOMA

NS (p=0.32)

Diabetes

Glycated hemoglobin

OR 0.24 (0.08–0.78)* 3rd quartile
(inverse association)
Inverse association (p<0.01 for
trend)*
Inverse association (p<0.01 for
trend)*
NS (p=0.17 for trend)

Diabetes

OR 1.02 (0.77–1.34, p=0.90)

HOMA

NS (p=0.56)

Diabetes

OR 1.07 (0.75–1.53, p=0.71)

HOMA

NS (p=0.070)

PFDeA
Zhang et al. 2015a
General population (n=258)
PFUA
Lind et al. 2014
General population (n=1,016)
Su et al. 2016

6.4–9.2 ng/mL
(3rd PFUA quartile)

General population (n=571)

Fasting blood glucose
Glucose tolerance test

PFHpA
Lind et al. 2014
General population (n=571)
PFOSA
Lind et al. 2014
General population (n=571)

0.05 ng/mL (median PFHpA)

0.11 ng/mL (median PFOSA)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

413
2. HEALTH EFFECTS

Table 2-26. Summary of Outcomes Related to Diabetes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Me-PFOSA-AcOH
Zhang et al. 2015a
General population (n=258)

0.30 and 0.29 ng/mL ng/mL
Gestational diabetes
(geometric mean Me-PFOSAAcOH in women with or without
gestational diabetes)

OR 1.05 (0.71–1.54)

aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 14 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

(F) = females; HOMA = homeostatic model assessment; HR = hazard ratio; IR = insulin resistance; (M) = males; Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane
sulfonamide) acetic acid; NHANES = National Health and Nutrition Examination Survey; NR = not reported; NS = not significant; OR = odds ratio;
PFDeA = perfluorodecanoic acid; PFHpA = perfluoroheptanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid;
SMR = standardized mortality ratio; SPR = standardized prevalence ratio

1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

414
2. HEALTH EFFECTS

1

PFOA

2
3

Epidemiology Studies. A cohort mortality study conducted by Leonard et al. (2008; Leonard 2006) of

4

workers at the Washington Works facility found a significant increase in deaths from diabetes, as

5

compared to workers at other DuPont facilities in the region. In an update of the Leonard et al. (2008)

6

study, Steenland and Woskie (2012) found an increased risk of diabetes deaths when compared to other

7

regional DuPont employees, but not when compared to the U.S. population. However, when the workers

8

were categorized by estimated cumulative exposure levels, the exposure-response trend was not

9

statistically significant. Lundin et al. (2009) also found an increase in deaths from diabetes in workers

10

exposed to APFO at the 3M Cottage Grove facility in Minnesota, as compared to Minnesota death rates.

11

The increase was only found in workers with probable exposure to APFO, but not with definite exposure;

12

no deaths from diabetes were observed in the workers with definite exposure to APFO. As noted by

13

Steenland and Woskie (2012), diabetes mortality may not be a good surrogate for the underlying diabetes

14

incidence data. Raleigh et al. (2014) did not find an increase in diabetes deaths at the Cottage Grove

15

facility and Steenland et al. (2015) did not find an increased risk of diabetes associated with cumulative

16

PFOA exposure at the Washington Works facility.

17
18

In community exposure studies, Anderson-Mahoney et al. (2008) found an increased prevalence of self-

19

reported diabetes in residents living near the Washington Works facility, as compared to expected rates

20

taken from NHANES. In contrast, Karnes et al. (2014) did not find an increased risk of self-reported

21

diabetes associated with cumulative PFOA levels and MacNeil et al. (2009) did not find an increased risk

22

of validated diabetes in C8 Health Study participants. General population studies found either an inverse

23

association between serum PFOA and risk of diabetes (Su et al. 2016) or no association (Lind et al. 2014;

24

Melzer et al. 2010). Additionally, general population studies have not found associations between serum

25

PFOA levels and insulin (Fisher et al. 2013; Lin et al. 2009), blood glucose levels (Fisher et al. 2013; Lin

26

et al. 2009; Su et al. 2016), homeostatic model assessment for insulin resistance (HOMA-IR) (Fisher et al.

27

2013; Lin et al. 2009; Lind et al. 2014; Nelson et al. 2010), or glucose tolerance (Su et al. 2016). Two

28

studies evaluated the risk of gestational diabetes and found mixed results. In a case-control study, Zhang

29

et al. (2015a) found an increased risk of gestational diabetes associated with serum PFOA, whereas

30

Shapiro et al. (2016) did not find associations between serum PFOA and gestational diabetes or impaired

31

glucose tolerance. The ORs for the risk of diabetes and gestational diabetes are graphically presented in

32

Figure 2-37.

33

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

415
2. HEALTH EFFECTS

Figure 2-37. Diabetes Risk Relative to Serum PFOA Levels (Presented as Adjusted Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

416
2. HEALTH EFFECTS

PFOS

Epidemiology Studies. In a general population study conducted by Su et al. (2016), an increased risk of
diabetes was noted, as well as associations between serum PFOS levels and fasting blood glucose,
response to glucose tolerance test, and glycated hemoglobin levels. Two other general population studies
did not find increased risks of diabetes (Lind et al. 2014; Melzer et al. 2010). Several studies have not
found associations between serum PFOS levels and insulin, blood glucose, or HOMA-IR levels (Fisher et
al. 2013; Lin et al. 2009; Lind et al. 2014; Nelson et al. 2010). Lin et al. (2009) found associations
between serum PFOS and insulin and HOMA-IR in a study of NHANES adult participants; these
associations were not found in adolescent participants. No alterations in the risk of gestational diabetes
were observed in two general population studies (Shapiro et al. 2016; Zhang et al. 2015a); the Shapiro et
al. (2016) study also found no association between serum PFOS and glucose tolerance. The ORs for the
risk of diabetes and gestational diabetes are graphically presented in Figure 2-38.

PFHxS

Epidemiology Studies. Five general population studies have examined diabetes-related outcomes and
have not found associations between serum PFHxS levels and diabetes risk (Lind et al. 2014), gestational
diabetes (Shapiro et al. 2016) or insulin, blood glucose, or HOMA-IR levels (Fisher et al. 2013; Lin et al.
2009; Lind et al. 2014; Nelson et al. 2010).
PFNA

Epidemiology Studies. An inverse association between serum PFNA levels and the risk of diabetes was
observed in a general population study (Su et al. 2016). Two other studies did not find associations for
diabetes (Lind et al. 2014) or gestational diabetes (Zhang et al. 2015a). The Su et al. (2016) study also
reported an inverse association between PFNA levels and response on a glucose tolerance test. A study of
adolescent NHANES participants found decreasing levels of insulin with increasing serum PFNA levels
(Lin et al. 2009); this association was not found in adult NHANES participants (Lin et al. 2009). Several
studies did not find associations between serum PFNA levels and HOMA-IR (Lin et al. 2009; Lind et al.
2014; Nelson et al. 2010).

Laboratory Animal Studies. An increase in serum glucose levels was observed in rats administered via
gavage 1 mg/kg/day PFNA for 14 days (Fang et al. 2012a).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

417
2. HEALTH EFFECTS

Figure 2-38. Diabetes Risk Relative to Serum PFOS Levels (Presented as Adjusted Odds Ratios)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

418
2. HEALTH EFFECTS

PFDeA

Epidemiology Studies. In the only study identified, Zhang et al. (2015a) did not find an association
between serum PFDeA levels and the risk of gestational diabetes.

PFUA

Epidemiology Studies. The two epidemiology studies evaluating associations between PFUA and
diabetes-related outcomes have found conflicting results. Su et al. (2016) found inverse associations
between serum PFUA levels and diabetes risk, fasting blood glucose levels, and glucose tolerance test
results, whereas Lind et al. (2014) found no alterations in the risk of diabetes or HOMA.
PFHpA

Epidemiology Studies. Lind et al. (2014) did not find associations between serum PFHpA levels and the
risk of diabetes or HOMA alterations.
PFOSA

Epidemiology Studies. In the one epidemiology study identified, no associations between serum PFOSA
levels and the risk of diabetes or HOMA were found (Lind et al. 2014).
Me-PFOSA-AcOH

Epidemiology Studies. In a case-control study, Zhang et al. (2015a) did not find alterations in the risk of
gestational diabetes associated with serum Me-PFOSA-AcOH levels.
2.19 CANCER
Overview. A number of occupational exposure, community, and general population studies have
examined possible associations between perfluoroalkyls and cancer risk; these studies are summarized in
Table 2-27 and the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 15.
Occupational and community exposure studies have found increases in the risk of testicular and kidney
cancer associated with PFOA. No consistent epidemiologic evidence for other cancer types were found

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

419
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

NR

All cancer deaths

SMR 0.86 (0.72–1.01), males
SMR 0.75 (0.56–0.99), females

Prostate cancer

SMR 2.03 (0.55–4.59)
RR 1.13 (1.01–1.27) for a 1-year
increase in employment length
RR 3.3 (1.02–10.6)* for a 10-year
employment length
SMR 100 (88–114), males
SMR 149 (77–260), females
SMR 185 (95–323), males
SMR 133 (53–274)

PFOA
Gilliland and Mandel 1993
Occupational (n=389 deaths)
Reference population: Minnesota general
population

Leonard 2006; Leonard et al. 2008
Occupational (n=6,027)
Reference population: DuPont workers at
other regional facilities

Lundin et al. 2009
Occupational (n=3,993)
Reference population: Minnesota general
population; for HR analysis comparisons with
workers with low exposure or <1 year of
exposure

5–9,550 ng/mL (estimated
range of PFOA)

All cancer deaths
Kidney cancer deaths
Biliary passages and
liver cancer deaths
Pancreatic cancer
deaths
Bladder or other urinary
organ cancer deaths
Prostate cancer deaths
Bronchus, trachea, lung
cancer deaths
All cancer deaths

2,600–5,200 and 300–
1,500 ng/mL (range of PFOA in
subset of current workers with Pancreas cancer
definite exposure jobs and
deaths
probable exposure jobs)
Trachea, bronchus, and
lung cancer deaths
Prostate cancer deaths

***DRAFT FOR PUBLIC COMMENT***

SMR 100 (50–180)
SMR 131 (53–269)
SMR 65 (34–114)
SMR 81 (63–104)
SMR 0.9 (0.5–1.4), definite exposure
SMR 0.9 (0.8–1.1), probable exposure
SMR 0.9 (0.0–4.7), definite exposure
SMR 1.0 (0.4–2.1), probable exposure
SMR 1.2 (0.5–2.3), definite exposure
SMR 1.0 (0.7–1.4), probable exposure
SMR 2.1 (0.4–6.1), definite exposure
SMR 0.9 (0.4–1.8), probable exposure
HR 6.6 (1.1–37.7), high exposure
≥6 months
HR 3.7 (1.3–10.4), definite exposure
for ≥5 years

 PERFLUOROALKYLS

420
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Raleigh et al. 2014

>7.9x10-4 μg/m3 (cumulative
exposure, 4th PFOA quartile)

All cancer deaths

SMR 0.87 (0.78–0.97)

Occupational (n=9,207)
Reference population: Minnesota general
population

Raleigh et al. 2014
Occupational (n=9,207)
Reference population: non-APFO exposed
workers at a St. Paul facility

Steenland et al. 2015

Pancreatic cancer
SMR 0.85 (0.50–1.34)
deaths
Prostate cancer deaths SMR 0.83 (0.53–1.23)

≤7.9x10-4 and >7.9x10-4 μg/m3
(cumulative exposure,3rd and
4th PFOA quartile)

Cumulative exposure

Kidney cancer deaths
Liver cancer deaths
Breast cancer deaths
Bladder cancer deaths
Pancreatic cancer
deaths
Pancreatic cancer

SMR 0.53 (0.20–1.16)
SMR 0.81 (0.35–1.59)
SMR 0.82 (0.41–1.47)
SMR 0.89 (0.38–1.76)
HR 1.23 (0.50–3.00), 3rd and 4th
quartiles combined
HR 1.36 (0.59–3.11), 3rd and 4th
quartiles combined
Prostate cancer deaths HR 1.32 (0.61–2.84), 4th quartile
Prostate cancer
HR 1.11 (0.82–1.49), 4th quartile
Kidney cancer deaths HR 0.39 (0.11–1.32), 3rd and 4th
quartiles combined
Kidney cancer
HR 0.73 (0.21–2.48), 4th quartile
Liver cancer deaths

HR 0.67 (0.14–3.27), 3rd and 4th
quartiles combined

Breast cancer deaths

HR 0.54 (0.15–1.94), 3rd and 4th
quartiles combined

Breast cancer
Bladder cancer deaths

HR 1.27 (0.70–2.31), 4th quartile
HR 1.96 (0.63–6.15), 3rd and 4th
quartiles combined
HR 1.66 (0.86–3.18), 4th quartile
Inverse association (p=0.04 or
p=0.06 for trend) with no lag or
10-year lag
RR 0.23 (0.05–0.93), 4th quartile with
no lag

Bladder cancer
Bladder cancer

Occupational (n=3,713)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

421
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Colorectal cancer

NS (p=0.91 and 0.86 for trend), with no
lag or 10-year lag
NS (p=0.83 and 0.91 for trend), no lag
or 10-year lag),
NS (p=0.16 and 0.55 for trend), no lag
or 10-year lag)
SMR 0.94 (0.76–1.16), 4th quartile

Prostate cancer
Melanoma
Steenland and Woskie 2012
Occupational (n=1,084 deceased workers)
Reference population: DuPont workers at other
regional facilities

≥2,700,000 ng/mL-years
All cancer deaths
(cumulative 4th PFOA quartile) Pancreatic cancer
deaths
Lung cancer deaths
Prostate cancer deaths
Bladder cancer deaths
Kidney cancer deaths

Barry et al. 2013

Cumulative exposure

Community and occupational (n=32,254)

24.2 and 112.7 ng/mL (median
PFOA)

Testicular cancer

Kidney cancer

Breast cancer

Colorectal cancer

***DRAFT FOR PUBLIC COMMENT***

SMR 0.92 (0.30–2.16), 4th quartile
SMR 0.75 (0.48–1.11), 4th quartile
SMR 0.57 (0.16–1.46), 4th quartile
SMR 0.36 (0.10–2.01), 4th quartile
SMR 2.66 (1.15–5.24)*, 4th quartile
SMR 2.82 (1.13–5.81)*, 10-year lag
SMR 3.67 (1.48–7.57)*, 20-year lag
HR 1.34 (1.00–1.79, p=0.05)* no lag
HR 1.28 (0.95–1.73, p=0.10) 10-year
lag
HR 3.17 (0.75–13.45, p=0.04 for
trend)*, 4th quartile
HR 1.10 (0.98–1.24, p=0.10), no lag
(continuous)
HR 1.58 (0.88–2.84, p=0.18 for trend),
4th quartile
HR 0.94 (0.89–1.00, p=0.05)*, no lag
HR 0.93 (0.88–0.99, p=0.03)*, 10-year
lag
HR 0.99 (0.92–1.07, p=0.84), no lag

 PERFLUOROALKYLS

422
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Ducatman et al. 2015a, 2015b

86.6 ng/mL (mean PFOA)

Prostate specific
antigen

NS (p<0.05)

13.5–27.8 ng/mL (2nd PFOA
quartile)

Colorectal cancer

OR 0.47 (0.31–0.74)*, 2nd quartile

Community (C8) (n=25,412 men)
Innes et al. 2014
Community (n=208 cases of colorectal cancer
and 47,359 cancer-free adults)
Vieira et al. 2013
Community (n=25,107)

Bonefeld-Jorgensen et al. 2011
General population (n=31 breast cancer cases
and115 matched controls)
Bonefeld-Jorgensen et al. 2014

30.8–109 and 110–655 ng/mL Kidney cancer
(estimated PFOA in high and
very high exposure groups)

AOR 2.0 (1.3–3.2)*, high exposure
group
AOR 2.0 (1.0–3.9), very high exposure
group

Testes cancer

AOR 2.8 (0.8–9.2), very high exposure
group

Prostate cancer

AOR 1.5 (0.9–2.5), very high exposure
group

Breast cancer
2.5 and 1.6 ng/mL (median
PFOA in cases and controls)

Breast cancer

AOR 1.4 0.9–2.3) very high exposure
group, females only
AOR 1.20 (0.77–1.88, p=0.43)

5.2 ng/mL (mean PFOA)

Breast cancer

RR 1.00 (0.90–1.11).

Prostate cancer
Bladder cancer
Pancreas cancer
Liver cancer

IRR 1.18 (0.84–1.65)
IRR 0.81 (0.53–1.24)
IRR 1.55 (0.85–2.80)
IRR 0.60 (0.26–1.37)

General population (n=250 breast cancer cases
and 233 matched controls)
Eriksen et al. 2009
6.8 and 6.0 ng/mL (median
PFOA in male and female
General population (n=713 for prostate cancer, cancer patients)
n=332 for bladder cancer, n=128 for pancreatic 6.9 and 5.4 ng/mL (median
PFOA in male and female
cancer, n=67 for liver cancer, and
n=772 controls)
controls)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

423
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Hardell et al. 2014

2.3 and 1.9 ng/mL (mean
PFOA in cases and controls)

Prostate cancer

OR 1.1 (0.7–1.7)
OR 2.6 (1.2–6.0)*, among subjects
with a heredity risk and serum
PFOA above the median

NR

All cancer deaths

SMR 0.84 (0.50–1.32), high potential
exposure group
SMR 12.77 (2.63–37.35)*, high
potential exposure group
SMR 16.12 (3.32–47.14)*, high
exposure group ≥1 year exposure
SIR 1.74 (0.64–3.79), high potential
exposure group
SIR 1.43 (0.16–5.15) ≥10-year
exposure group

General population (n=201 cases and
186 controls)
PFOS
Alexander et al. 2003
Occupational (n=2,083; 145 deaths)
Reference population: Alabama general
population
Alexander and Olsen 2007
Occupational (n=1,895; 1,488 deaths)
Reference population: NIOSH SEER referent
data
Grice et al. 2007
Occupational (n=1,400 current, retired, or
former workers)
Olsen et al. 2004a

Bladder and other
urinary organs cancer

NR

Bladder cancer

1,300–1,970 ng/mL (PFOS
levels in high potential
exposure group)

Colon cancer
Melanoma

OR 1.69 (0.68–4.17)
OR 1.01 (0.25–4.11)

Prostate cancer

OR 1.08 (0.44–2.69)

NR

Malignant melanoma of RREpC 12 (1.0–>100)
the skin
RREpC 10 (0.7–>100), >10 years
employment
Malignant neoplasm of RREpC 5.4 (0.5–>100)
the colon
RREpC 12 (0.8–>100), >10 years
employment

22.18 ng/mL (mean PFOS)

Prostate specific
antigen

NS (p<0.05)

13.6–20.1 ng/mL (2nd PFOS
quartile)

Colorectal cancer

OR 0.35 (0.24–0.53)*

Occupational (current and retired workers)

Ducatman et al. 2015a, 2015b
Community (C8) (n=25,412 men)
Innes et al. 2014
Community (n=208 cases of colorectal cancer
and 47,359 cancer-free adults)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

424
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Bonefeld-Jorgensen et al. 2011

45.6 and 21.9 ng/mL (median
PFOS in cases and controls)

Breast cancer

OR 1.03 (1.001–1.07, p=0.05)*

30.6 ng/mL (mean PFOS)

Breast cancer

RR 0.99 (0.98–1.01)

Prostate cancer
Bladder cancer
Pancreas cancer
Liver cancer

IRR 1.05 (0.97–1.14)
IRR 0.93 (0.83–1.03)
IRR 0.99 (0.86–1.14)
IRR 0.59 (0.27–1.27)

Prostate cancer

OR 1.0 (0.6–1.5)
OR 2.7 (1.04–6.8)*, among subjects
with a heredity risk and serum
PFOS above the median

3.58 ng/mL (mean PFHxS)

Prostate specific
antigen

NS (p<0.05)

1.2 ng/mL (mean PFHxS)

Breast cancer

RR 0.66 (0.47–0.94)*

Prostate cancer

OR 1.3 (0.8–1.9)
OR 4.4 (1.7–12)*, among subjects
with a heredity risk and serum
PFHxS above the median

General population (n=31 breast cancer cases
and 115 matched controls)
Bonefeld-Jorgensen et al. 2014

General population (n=250 breast cancer cases
and 115 matched controls)
Eriksen et al. 2009
35.1 and 32.1 ng/mL and
35.0 and 29.3 ng/mL (median
General population (n=713 for prostate cancer, PFOS in male and female
n=332 for bladder cancer, n=128 for pancreatic cancer patients and males and
cancer, n=67 for liver cancer, and
females in the comparison
n=772 controls)
group)
Hardell et al. 2014
11 and 10 ng/mL (mean PFOS
in cases and controls)
General population (n=201 cases of prostate
cancer and 186 controls)
PFHxS
Ducatman et al. 2015a, 2015b
Community (C8) (n=25,412 men)
Bonefeld-Jorgensen et al. 2014

General population (n=250 breast cancer cases
and 115 matched controls)
Hardell et al. 2014
1.1 and 0.940 ng/mL (mean
PFHxS in cases and controls)
General population (n=201 cases of prostate
cancer and 186 controls)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

425
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

Ducatman et al. 2015a, 2015b

1.47 ng/mL (mean PFNA)

Prostate specific
antigen

NS (p<0.05)

Community (C8) (n=25,412 men)
Bonefeld-Jorgensen et al. 2014

0.5 ng/mL (mean PFNA)

Breast cancer

RR 0.76 (0.30–1.94)

PFNA

General population (n=250 breast cancer cases
and 115 matched controls)
Hardell et al. 2014
0.679 and 0.631 ng/mL (mean Prostate cancer
PFNA in cases and controls)
General population (n=201 cases of prostate
cancer and 186 controls)
PFDeA
Hardell et al. 2014
0.338 and 0.291 ng/mL (mean Prostate cancer
PFDeA in cases and controls)
General population (n=201 cases of prostate
cancer and 186 controls)
PFUA
Hardell et al. 2014

0.308 and 0.285 ng/mL (mean Prostate cancer
PFUA in cases and controls)

General population (n=201 cases of prostate
cancer and 186 controls)

***DRAFT FOR PUBLIC COMMENT***

OR 1.2 (0.8–1.8)
OR 2.1 (0.9–4.8), among subjects with
a heredity risk and serum PFNA above
the median
OR 1.4 (0.9–2.1)
OR 2.6 (1.1–6.1)*, among subjects
with a heredity risk and serum
PFDeA above the median
OR 1.2 (0.8–1.9)
OR 2.6 (1.1–5.9)*, among subjects
with a heredity risk and serum
PFUA above the median

 PERFLUOROALKYLS

426
2. HEALTH EFFECTS

Table 2-27. Summary of Cancer Outcomes in Humansa
Reference and study populationb

Serum perfluoroalkyl level

Outcome evaluated

Resultc

3.5 ng/mL (mean PFOSA)

Breast cancer

RR 1.89 (1.01–3.54)*, among women
with serum PFOSA >5.75 ng/mL

PFOSA
Bonefeld-Jorgensen et al. 2014
General population (n=250 breast cancer cases
and 115 matched controls)
aSee

the Supporting Document for Epidemiological Studies for Perfluoroalkyls, Table 15 for more detailed descriptions of studies.
in occupational exposure may have also lived near the site and received residential exposure; community exposure studies involved subjects living
near PFOA facilities with known exposure to high levels of PFOA.
cAsterisk indicates association with perfluoroalkyl compound; unless otherwise specified, values in parenthesis are 95% confidence intervals.
bParticipants

AOR = adjusted odds ratio; CI = confidence interval; HR = hazard ratio; IRR = incidence rate ratio; NIOSH = National Institute for Occupational Safety and Health;
NR = not reported; NS = not significant; OR = odds ratio; PFDeA = perfluorodecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic
acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid;
RR = relative risk; RREpC = risk ratio episodes of care; SEER = Surveillance Epidemiology and End Results; SIR = standardized incidence ratio; SMR =
standardized mortality ratio

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

427
2. HEALTH EFFECTS

for PFOA. For PFOS, one occupational exposure study reported an increase in bladder cancer, but this
was not supported by subsequent occupational studies. General population studies have not consistently
reported increases in malignant tumors for PFOS. Epidemiology studies examining other perfluoroalkyl
compounds consisted of two case-control studies. No increases in breast cancer risk were observed for
PFHxS or PFNA; an increased breast cancer risk was observed for PFOSA. Another case-control study
did not find increases in prostate cancer for PFOA, PFOS, PFHxS, PFNA, PFDeA, or PFUA. However,
among men with a first-degree relative with prostate cancer, associations were found for PFOA, PFOS,
PFHxS, PFDeA, and PFUA, but not for PFNA.

Laboratory animal studies have evaluated the carcinogenicity of PFOA and PFOS; the results of these
studies are summarized in Tables 2-3 and 2-4. In laboratory animals, there is some evidence for increases
in Leydig cell adenomas and pancreatic acinar cell adenomas in male rats exposed to PFOA in the diet.

An increase in hepatocellular adenomas was observed in male rats exposed to dietary PFOS for 2 years;
thyroid follicular cell adenomas were observed in rats exposed to PFOS for 1 year and allowed to recover
for an additional year.

EPA (2016e, 2016f) has concluded that there is suggestive evidence of the carcinogenic potential of
PFOA and PFOS in humans. IARC (2017) concluded that PFOA is possibly carcinogenic to humans
(Group 2B).
PFOA

Epidemiology Studies. Several studies have examined the possible association between occupational
exposure to PFOA and increased cancer risk in workers at two U.S. facilities—3M facility in Cottage
Grove, Minnesota (Gilliland and Mandel 1993; Lundin et al. 2009; Raleigh et al. 2014) and DuPont
Washington Works facility in West Virginia (Leonard 2006; Leonard et al. 2008; Steenland and Woskie
2012; Steenland et al. 2015). In addition, the potential carcinogenicity of PFOA has been assessed in the
community near the Washington Works facility (Barry et al. 2013; Innes et al. 2014; Vieira et al. 2013)
and in the general population (Bonefeld-Jorgensen 2011, 2014; Eriksen et al. 2009; Hardell et al. 2014).

Occupational exposure studies have not found increases in the risk of all cancer deaths (Gilliland and
Mandel 1993; Leonard 2006; Leonard et al. 2008; Lundin et al. 2009; Raleigh et al. 2014; Steenland and
Woskie 2012). The occupational exposure studies have consistently found no increases in the risk of

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

428
2. HEALTH EFFECTS

pancreatic, liver, or respiratory tract cancers or deaths from these cancers (Leonard 2006; Leonard et al.
2008; Lundin et al. 2009; Raleigh et al. 2014; Steenland and Woskie 2012); a general population casecontrol study also found no associations between serum PFOA and pancreas or liver cancer (Eriksen et al.
2009). Additionally, two case-control studies did not find associations between serum PFOA levels and
risk of breast cancer (Bonefeld-Jorgensen et al. 2011, 2014). Steenland et al. (2015) found an inverse
association between cumulative PFOA exposure and bladder cancer in workers; other studies have not
found associations (Eriksen et al. 2009; Gilliland and Mandel 1993; Leonard 2006; Leonard et al. 2008;
Raleigh et al. 2014; Steenland and Woskie 2012).

Some associations between PFOA and cancer effects have been observed, including prostate, kidney, and
testicular cancers. Ten years of employment in the Chemical Division of the 3M Cottage Grove facility
was associated with a 3.3-fold increase in the relative risk of prostate cancer mortality, as compared to no
employment in PFOA production areas (Gilliland and Mandel 1993); no increase in prostate cancer risk
was observed when all workers in the Chemical Division were analyzed. The investigators noted that the
prostate cancer findings are based on a small number of cases and could have resulted from chance or
unrecognized confounding from exposure to other factors. An update of this study conducted by Lundin
et al. (2009) did not find an increase in prostate cancer deaths in workers with definite PFOA exposure.
When the cohort was divided into the three exposure categories and duration of definite exposure,
increased risks for prostate cancer were found in the high-exposure category and in workers with definite
exposure for at least 5 years, as compared with workers in the low-exposure category and with the
shortest cumulative exposure duration, respectively. Interpretation of the Gilliland and Mandel (1993)
and Lundin et al. (2009) studies is limited by the qualitative assessment of potential exposure and the fact
that workers in the low exposure categories were likely research-and-development professionals rather
than production workers (Raleigh et al. 2014). In the most recent evaluation of the Cottage Grove
facility, which involved extensive exposure assessment, Raleigh et al. (2014) did not find increases in
prostate cancer deaths when compared to the general population or to workers at another facility and did
not find an increase in the incidence of prostate cancer when the workers were categorized by cumulative
exposure levels. Studies of the Washington Works facility workers did not find increases in prostate
cancer deaths (Leonard et al. 2008; Steenland and Woskie 2012) or incidence (Steenland et al. 2015). A
case-control general population study by Hardell et al. (2014) did find an increase in prostate risk only
among subjects with a heredity risk (first-degree relative with prostate cancer) and serum PFOA levels
above the median. In a study of community members, Ducatman et al. (2015b) did not find an
association between prostate-specific antigen (PSA) levels and serum PFOA levels in men 20–49 or 50–
69 years of age.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

429
2. HEALTH EFFECTS

In the earliest cancer assessment of workers at the Washington Works facility (Leonard 2006; Leonard et
al. 2008), an increase in the number of deaths from kidney cancer relative to workers at other regional
DuPont facilities was observed; however, the CI included unity. In a follow-up study that used serum
PFOA levels collected in current workers to assess job title exposure (Steenland and Woskie 2012), an
increase in kidney cancer deaths was observed in workers with the highest exposures when analyzed with
no lag, a 10-year lag, or a 20-year lag. Steenland and Woskie (2012) also found an increase in deaths
from mesothelioma; the investigators noted that this was likely due to asbestos exposure. Steenland and
Woskie (2012) noted that tetrafluoroethylene, a rodent kidney carcinogen, is used in the manufacture of a
variety of fluoropolymers and noted that the tetrafluoroethylene is well controlled due to its volatile and
explosive properties. It is noted that in a multisite study of tetrafluoroethylene workers, which included
workers at the Washington Works facility (Consonni et al. 2013), an increased risk of renal cancer (SMR
1.44, 95% CI 0.69–2.65) was found, although the CI included unity. Consonni et al. (2013) noted that
88% of the workers were also exposed to PFOA. When PFOA exposure was used as an exposure
variable, the findings were the similar as when tetrafluoroethylene was used as the exposure variable, and
thus, it was difficult for the investigators to evaluate separate associations for each compound. It is noted
that increases in kidney deaths were not observed in the Cottage Grove facility (Raleigh et al. 2014),
which did not use tetrafluoroethylene (Chang et al. 2014).

Three studies have examined the community living near the Washington Works facilities; some of these
studies also included workers at the facility. Barry et al. (2013) reported an increased risk of testicular
cancer that was associated with cumulative PFOA exposure. Vieira et al. (2013) also reported an increase
in testicular cancer, but the CIs of the adjusted odds ratio (AOR) included unity. When the participants
were grouped by water district, an increased risk of testicular cancer (AOR 5.1, 95% CI 1.6–15.6) was
observed in the Little Hocking water district, which had the highest PFOA levels in the water. The Vieira
et al. (2013) study also found increased risks of kidney cancer among participants with high or very high
exposure to PFOA; Barry et al. (2013) also concluded that there was an association between cumulative
PFOA exposure and kidney cancer, although the CIs for the hazard ratio included unity. The third study
of the Washington Works community found an inverse association between serum PFOA and risk of
colorectal cancer (Innes et al. 2014).

In their review of the available epidemiology data, IARC (2017) concluded that the evidence for testicular
cancer was “considered credible and unlikely to be explained by bias and confounding, however, the
estimate was based on small numbers.” Similarly, IARC (2017) concluded that the evidence for kidney

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

430
2. HEALTH EFFECTS

cancer was also credible but noted that chance, bias, and confounding could not be ruled out with
reasonable confidence. They considered that there was limited evidence in humans for the
carcinogenicity of PFOA.

Laboratory Animal Exposure Studies. Two studies have examined the carcinogenic potential of PFOA
in rats. In the first study of male and female Sprague-Dawley rats exposed to PFOA in the diet for 2 years
(3M 1983), significant increases in the incidence of fibroadenoma of the mammary gland in females and
Leydig cell adenoma were found in males exposed to 15 mg/kg/day. A high incidence of pituitary
adenoma occurred among all groups, including controls. The incidence of hepatocellular carcinoma was
not significantly increased. The investigators noted that the incidence of fibroadenoma in the mammary
gland in the 15 mg/kg/day group was similar to the incidence found in untreated aging rats and that the
incidence of Leydig cell adenoma was similar to the spontaneous incidence of this tumor in aged rats.
The mammary gland pathology slides from the 3M (1983) study were re-examined in 2005 by a
Pathology Working Group (PWG) using current diagnostic criteria (Hardisty et al. 2010). The incidences
of fibroadenoma found by the PWG were 36, 44, and 46% in the 0, 1, and 15 mg/kg/day groups,
respectively; there were no statistically significant differences between the groups (Hardisty et al. 2010).
Additionally, there were no significant differences in the incidence of adenocarcinoma, total benign
neoplasms, or total malignant neoplasms between the groups. In the second study of male SpragueDawley rats exposed to PFOA in the diet for 2 years (Biegel et al. 2001), an increase in the incidence of
hepatocellular adenomas was found, but there were no hepatocellular carcinomas in the treated group.
PFOA also increased the incidence of Leydig cell adenomas. In addition, PFOA increased the incidence
of pancreatic acinar cell adenomas; a pancreatic carcinoma was observed in one treated rat. Hepatic
peroxisome proliferation was increased significantly at all interim evaluation time points (1, 3, 6, 9, 12,
15, 18, and 21 months), but there was no increase in cell proliferation. In Leydig cells, neither
peroxisome proliferation nor cell proliferation were increased.

PFOA was a positive modulator of hepatocarcinogenesis in male Wistar rats in a biphasic (initiation with
diethylnitrosamine followed by oral treatment with PFOA) or triphasic (initiation with diethylnitrosamine
[DEN] followed by dosing with 2-acetylaminofluorene and then PFOA) promotion protocol (Abdellatif et
al. 1991, 2004). PFOA induced a marked increase in acylCoA oxidase activity and only a slight increase
in catalase activity (Abdellatif et al. 2004). Since PFOA did not significantly increase 8-hydroxydeoxyguanosine (a marker of oxidative DNA damage in vivo) in isolated liver DNA, it appeared that
PFOA did not require extensive DNA damage for its promoting activity (Abdellatif et al. 2004). PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

431
2. HEALTH EFFECTS

was also found to act as a promoter in male Wistar rats in an initiation-selection-promotion protocol
(Nilsson et al. 1991).

IARC (2017) concluded that there was limited evidence in experimental animals for the carcinogenicity
of PFOA.

PFOS

Epidemiology Studies. Four studies have evaluated the carcinogenic potential in workers at a Decatur,
Alabama perfluorooctanesulphonyl fluoride (PFOSF) based fluorochemical production facility. In the
earliest study, no increase in all cancer deaths was found, as compared to the Alabama general population
(Alexander et al. 2003). An increased risk of bladder cancer was observed in workers with high potential
exposure and in workers with a high potential exposure for ≥1 year; the mortality ratio was based on three
cases in the high exposure group. In a reanalysis of workers at this facility conducted by Alexander and
Olsen (2007), 11 cases of bladder cancer were identified from worker surveys (n=6) and death certificates
(n=5). Only two of the six self-reported bladder cancer diagnosis were confirmed via medical records;
the other four subjects declined to give consent for medical verification. When compared to incidence
data from the National Institute for Occupational Safety and Health (NIOSH) Surveillance Epidemiology
and End Results (SEER) referent data, the standardized incidence ratios for the high potential exposure
group were elevated, but the CIs included unity. When compared with workers with <1 year of high
exposure, workers with 5–<10 and ≥10 years of high exposure had relative risks of 1.92 (95% CI 0.30–
12.06) or 1.52 (95% CI 0.21–10.99). Although the study did not adjust for smoking, the investigators
noted that 83% of the living bladder cancer cases (five of the six subjects) reported cigarette use, as
compared to 56% reported in the noncases. An additional limitation of the study is inclusion of four cases
of bladder cancer that were not verified by medical records. The results of this study do not appear to
confirm the findings of increased bladder cancer in the mortality study (Alexander et al. 2003). In a
subsequent study of this facility, treatment for bladder cancer was not reported among current workers
(Olsen et al. 2004a). The study did find increases in the number of episodes of care for malignant
neoplasm of the prostate or malignant neoplasms of the colon, as compared to long-term workers in
another division, but the CIs included unity. No increases in the risk ratio episodes of care were found for
liver, rectum, or respiratory tract (Olsen et al. 2004a). A fourth study of this facility (Grice et al. 2007)
examined possible associations between colon cancer, melanoma, and prostate cancer and PFOS
exposure. The risks of these cancers were not associated with any of the PFOS-exposure categories for
analyses that included all self-reported or only validated cancers.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

432
2. HEALTH EFFECTS

General population case-control studies have evaluated several cancer types. Innes et al. (2014) reported
an inverse association between PFOS and colorectal cancer. A small-scale study of 31 cases by
Bonefeld-Jorgensen et al. (2011) found a slight increase in breast cancer risk, a finding not replicated in
another larger study of a different population (Bonefeld-Jorgensen et al. 2014). Eriksen et al. (2009) and
Hardell et al. (2014) did not find increases in the risk of prostate cancer associated with serum PFOS.
However, an increased risk of prostate cancer was found among subjects with a first-degree relative with
prostate cancer and PFOS levels above the median level (Hardell et al. 2014). Eriksen et al. (2009) also
found no associations between serum PFOS and the risk of bladder cancer, pancreatic cancer, or liver
cancer. Ducatman et al. (2015b) did not find an association between serum PFOS levels and PSA levels
in men participating in the C8 studies.

Laboratory Animal Studies. In a 2-year PFOS dietary exposure study bioassay in male and female
Sprague-Dawley rats (Butenhoff et al. 2012b; unpublished study by Thomford 2002b), a significant
positive trend of hepatocellular adenoma was observed in males; the incidence was significantly higher
than controls at 1.04 mg/kg/day. No hepatocellular adenomas were seen in a group of rats exposed to
1.17 mg/kg/day for 1 year and allowed to recover for the second year. High-dose males from the
recovery group showed a significant increase in thyroid follicular cell adenoma relative to controls. No
significant increase in this type of tumor was observed in rats exposed for 2 years. In females, there was a
significant positive trend for incidences of hepatocellular adenoma, which was associated with a
significant increase in the 1.04 mg/kg/day group. In females, there were also significant negative trends
for mammary adenoma and fibroadenoma carcinoma combined.

PFHxS

Epidemiology Studies. Two case-control studies have examined the possible association between serum
PFHxS and cancer. Bonefeld-Jorgensen et al. (2014) found in inverse association between PFHxS levels
and breast cancer risk. No association between PFHxS and prostate cancer was observed (Hardell et al.
2014), with the exception of increased risk in men with a first-degree relative with prostate cancer and
above-median serum PFHxS levels. No associations between serum PFHxS and PSA levels were
observed in men 20–49 or 50–69 years of age participating in the C8 Health Studies (Ducatman et al.
2015b).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

433
2. HEALTH EFFECTS

PFNA

Epidemiology Studies. The carcinogenic potential of PFNA has been examined in two case-control
studies. No associations between serum PFNA levels and breast cancer (Bonefeld-Jorgensen et al. 2014)
or prostate cancer (Hardell et al. 2014) were found. Serum PSA levels were not associated with serum
PFNA levels in men participating in the C8 Health Study (Ducatman et al. 2015b).
PFDeA

Epidemiology Studies. Hardell et al. (2014) examined the possible association between the serum
PFDeA level and risk of prostate cancer and only found an association in men with a heredity risk factor
and PFDeA levels above the median.
PFUA

Epidemiology Studies. An increased risk of prostate cancer was found in men with first-degree relatives
with prostate cancer and serum PFUA levels above the median (Hardell et al. 2014).
PFOSA

Epidemiology Studies. Bonefeld-Jorgensen et al. (2014) reported an increased risk of breast cancer
among women with serum PFOSA levels >5.75 ng/mL.

2.20 MECHANISM OF TOXICITY
The primary effects observed in laboratory animals exposed to perfluoroalkyl compounds are liver
toxicity, developmental toxicity, and immune toxicity. The cellular mechanisms by which hepatic effects
are induced have been extensively studied, while more limited data are available on mechanisms for other
effects. The available data indicate that perfluoroalkyl compounds produce a number of adverse effects
through activation of the PPARα, a member of the nuclear receptor superfamily that mediates a broad
range of biological responses (Issemann and Green 1990). However, some adverse effects of
perfluoroalkyl compounds occur through PPARα-independent mechanisms, which may include activation
of other nuclear receptors, increased oxidative stress, dysregulation of mitochondrial function, and
inhibition of gap junction intercellular communication (GJIC). In the sections below, cellular

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

434
2. HEALTH EFFECTS

mechanisms of action that are mediated by PPARα and independent of PPARα are discussed, followed by
discussions of mechanisms specific to the hepatic, developmental, immunotoxic, and hormone effects of
perfluoroalkyl compounds.
2.20.1 Cellular Mechanisms of Toxicity
PPARα-Dependent Mechanisms
Activation of the PPARα receptor in rodents initiates a characteristic sequence of morphological and
biochemical events, principally, but not exclusively, in the liver. These events include marked
hepatocellular hypertrophy due to an increase in number and size of peroxisomes, a large increase in
peroxisomal fatty acid β oxidation, increased cytochrome 450-mediated ω hydroxylation of lauric acid,
and alterations in lipid metabolism. Proliferation of peroxisomes in laboratory animals exposed to
perfluoroalkyl compounds is discussed in Section 2.9 (Hepatic); as discussed in that section, hepatic
peroxisome proliferation has been shown in rats exposed to PFOA and in mice exposed to PFDeA.

Many, but not all, of the adverse effects induced by perfluoroalkyl compounds are mediated through
activation of the PPARα. Ligands, including perfluoroalkyl compounds, bind to and activate PPARα,
causing a conformational change in the receptor that leads to dissociation of co-repressors and enables
heterodimerization with the retinoid X receptor (Corton et al. 2014). The activated receptor complex
binds to a DNA direct repeat motif (the peroxisome proliferator response element or PPRE) located in the
promoters of peroxisome proliferator responsive genes. The binding of the receptor complex leads to
recruitment of co-activators that acetylate histones and remodel chromatin, enabling RNA polymerase to
transcribe the target gene(s). PPARα regulates lipid homeostasis by modulating the expression of genes
involved in fatty acid uptake, activation, and oxidation. Activation of nuclear receptors including PPARα
is a complex, dynamic process that depends on levels of expression of the receptors in different tissues,
competition among receptors for endogenous and exogenous ligands and for binding sites on chromatin,
and availability and abundance of co-activators and/or co-repressors (Corton et al. 2014).
PPARα Receptor Activation. Many perfluoroalkyl compounds, including PFOA, PFOS, PFUA, PFHpA,
PFDoA, and perfluoropentanoic acid have been shown to activate PPARα in mammalian cells in vitro
(Bjork and Wallace 2009; Bjork et al. 2011; Shipley et al. 2004; Takacs and Abbott 2007; Vanden Heuvel
et al. 2006; Wolf et al. 2008b, 2012). Cell systems used in these studies include COS-1 cells expressing
mouse, rat, or human PPARα, and cultured rat, mouse, and human hepatocytes. In these studies,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

435
2. HEALTH EFFECTS

perfluoroalkyl sulfonate compounds were less potent than perfluoroalkyl carboxylate compounds in
activating PPARα-induced gene expression, and the potency of stimulation within each class increased
with carbon chain length (Bjork and Wallace 2009; Wolf et al. 2008b, 2012). In comparison with
naturally occurring long-chain fatty acids such as linoleic and α linoleic acids, PFOA and PFOS are
relatively weak ligands for PPARα (Vanden Heuvel et al. 2006)
PPARα-Dependent Gene Expression Changes. Perfluoroalkyl compounds have been shown to induce
changes in the expression of genes under the regulation of PPARα. The expression of PPARα target
genes involved in fatty acid metabolism, cell cycle control, peroxisome biogenesis, and proteasome
structure and organization were upregulated, while inflammatory response genes were downregulated, in
wild-type mice exposed orally to PFOA or the PPARα agonist WY-14,643 (Rosen et al. 2008a).
Furthermore, PFOA and PFDA have been shown to downregulate, via PPARα activation, genes involved
in bile transport in the livers of mice exposed by intraperitoneal administration (Cheng and Klaassen
2008a). Both compounds decreased expression of organic anion transporting polypeptides [OATP1a1,
1a4, and 1b2], and PFDA also downregulated sodium-taurocholate cotransporting polypeptide [Nctp], via
activation of PPARα. Many of these expression changes may play roles in the hepatic effects of
perfluoroalkyl compounds, and the diminished expression of inflammatory response genes may be
involved in immunosuppression induced by perfluoroalkyl compounds.

Gene expression changes induced by perfluoroalkyl compounds have been extensively studied in
experiments aimed at determining the extent to which the adverse effects of these compounds are
dependent on activation of PPARα or interaction with other nuclear receptors (Foreman et al. 2009;
Rosen et al. 2008a, 2008b, 2010, 2017). These studies, comparing gene expression changes in wild-type
and PPARα-null mice exposed to perfluoroalkyl compounds, demonstrate the following:
•

A majority of the gene expression changes induced in rodents by perfluoroalkyl compounds
tested to date, especially PFOA and PFNA, are dependent on activation of PPARα.

•

Perfluoroalkyl compounds also induce gene expression changes that are independent of PPARα.

•

The extent to which gene expression changes induced by perfluoroalkyls are dependent on
activation of PPARα varies by compound.

•

PPARα-independent gene expression changes induced by perfluoroalkyl compounds are similar
to those induced by compounds that activate PPARγ, constitutive androstane receptor (CAR), and
estrogen receptor alpha (ERα) (discussed further under PPARα-independent mechanisms).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

436
2. HEALTH EFFECTS

Species Differences in PPARα Activation. Species differences in response to PPARα activators have
been reviewed by Corton et al. (2014). Studies of PPARα activation by structurally diverse ligands in
various species have shown that rats and mice are the most sensitive species to PPARα agonists, whereas
guinea pigs, hamsters, nonhuman primates, and humans are less responsive (Corton et al. 2014).
However, the differences do not appear to result from differences in the PPARα gene itself: PPARα
cDNA from humans is indistinguishable from the rodent PPARα. Species differences in response to
exogenous PPARα activators may stem from any or all of the following: (1) differences in the expression
of PPARα in a given tissue; (2) differences in the gene product structure; and (3) differences in the
ligand-mediated transactivation of PPARα. Experiments quantifying mRNA and/or protein levels of
PPARα show ~10-fold higher expression of PPARα in the livers of mice and rats compared with humans
and guinea pigs, but available data are limited and require further study to validate these differences
(Corton et al. 2014). In humans, variants of PPARα that may affect its transactivation potential have been
identified. For example, humans produce higher levels of a truncated PPARα (that lacks a ligand binding
domain) compared with mice and rats (Corton et al. 2014). The truncated form appears to inhibit the
activity of the full-length receptor, possibly via sequestering critical co-activators. Other, non-truncated
variants of PPARα have been identified in humans, but the sensitivity of these variants to PPARα
activators does not differ markedly from that of the wild-type receptor.
Species and compound-related differences in PPARα transactivation by perfluoroalkyl compounds have
been demonstrated in vitro (Shipley et al. 2004; Takacs and Abbott 2007; Vanden Heuvel et al. 2006;
Wolf et al. 2008b, 2012). In a comparison of human and mouse PPARα transactivation by different
perfluoroalkyl compounds in transfected COS-1 cells, Wolf et al. (2008b, 2012; see Table 2-28) found
that some perfluoroalkyls exhibited marked species differences in transactivation potency (for example,
PFUA, PFDeA, PFDoA), while other compounds showed similar transactivation potency for both human
and mouse PPARα (for example, PFNA, PFOA, perfluoropentanoic acid [PFPeA]).

Table 2-28. Transactivation of Human and Mouse PPARα in Transfected Cos-1
Cells Exposed to Perfluoroalkyl Compounds (In Order of Decreasing C20max
in the Mouse)

Perfluoroalkyl

Carbon
number

NOEC
(µM)

PFNA

9

1

PFOAb

8

5, 0.5

Mouse
NOEC
LOEC
LOEC (µM) (µM)
(µM)

C20max (µM)a

Human

5
10,1

Human

Mouse

1

5

11

5

0.5, 1

1, 3

7

6

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

437
2. HEALTH EFFECTS

Table 2-28. Transactivation of Human and Mouse PPARα in Transfected Cos-1
Cells Exposed to Perfluoroalkyl Compounds (In Order of Decreasing C20max
in the Mouse)

Perfluoroalkyl

Carbon
number

NOEC
(µM)
50

Mouse
NOEC
LOEC
LOEC (µM) (µM)
(µM)

C20max (µM)a

Human

PFUA

11

PFHpA

7

PFDeA

10

100

>100c

PFDoA

12

75

PFPeA

5

0.5

PFHxA

6

PFBA

Mouse

5

10

86

8

3

5

15

11

<5

5

–

20

90

3

5

NA

33

1

1

5

52

45

5

10

10

20

86

45

4

30

40

30

40

75

51

PFHxS

6

5

10

10

20

81

76

PFOS

9

20

30

60

90

262

94

PFBuS

4

20

30

120

150

206

317

<0.5

75

Human

0.5

aPerfluoroalkyl

concentration yielding 20% of maximum response given by the most active compound (PFNA).
from two separate experiments.
cSlope for human PPARα dose-response line was not significant.
bResults

– = not active; LOEC = lowest-observed-effect concentration; NA = not available; NOEC = no-observed-effect
concentration; PFBA = perfluorobutyric acid; PFBuS = perfluorobutane sulfonic acid; PFDeA = perfluorodecanoic
acid; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxA = perfluorohexanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFPeA = perfluoropentanoic acid;
PFUA = perfluoroundecanoic acid; PPAR = peroxisome proliferator activated receptor
Source: Wolf et al. 2008b, 2012

In addition to differences in transactivation of PPARα, Corton et al. (2014) noted that there are species
differences in the transcripts controlled by PPARα. While PPARα activation leads to hypolipidemic
changes in both humans and laboratory rodents, the gene sets responsible for these changes may differ. In
a comparison between human and mouse hepatocytes exposed to the prototypical PPARα ligand
WY-14,643, some genes (ACOX1, ECH1, PEX11A, and ACAA1) were induced in both species, while
some (Ehhadh, Pxmp4, Acot4, and Peci) were induced only in mouse hepatocytes (Corton et al. 2014).
Importantly, PPARα activators induce large increases in the expression of fatty acyl-CoA oxidase (ACO,
which is believed to play a role in oxidative stress-induced liver cancer) in rodent hepatocytes, but
relatively weak increases in human hepatocytes (Corton et al. 2014). Other hypothesized explanations for
the species difference in response to exogenous PPARα ligands include variations in the structure of the
PPRE that alter the response of the human genes compared with rodents; differences between humans and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

438
2. HEALTH EFFECTS

rodents in the functions of genes under the regulation of PPARα; and differences in the ability of ligandbound human and mouse PPARα receptor complex to recruit or interact with co-activators (Corton et al.
2014).
PPARα-Independent or Associative Mechanisms
Experiments using PPARα-null mice have demonstrated that perfluoroalkyl compounds exert some
adverse effects, including developmental and hepatic effects, through mechanisms other than activation of
PPARα. These may include activation of other nuclear receptors, increased oxidative stress,
dysregulation of mitochondrial function, and inhibition of GJIC. While some of these effects have been
seen after exposure to PPARα activators (Corton et al. 2014), these mechanisms may also occur
independent of PPARα activation.
Activation of Other Nuclear Receptors. Examination of gene expression changes, as well as studies
using other knock-out mice, have shown that some of the PPARα-independent effects induced by
perfluoroalkyl compounds may be mediated by activation of other nuclear receptors, especially PPARγ,
CAR, and ERα. In a series of experiments, Rosen et al. (2008b, 2010, 2017) compared the gene
expression changes induced by perfluoroalkyl compounds in wild-type and PPARα -null mice with gene
expression changes induced by known agonists of PPARγ, CAR, and ERα. Using these data, the study
authors estimated the percentage of gene expression changes that were independent of activation of
PPARα, and identified other nuclear receptors potentially involved in the changes induced by the
perfluoroalkyls. The results, summarized in Table 2-29, show that between 10 and 24% of gene
expression changes induced by perfluoroalkyl compounds are independent of PPARα. All four
compounds tested (PFOA, PFOS, PFNA, and PFHxS) were shown to alter the expression of PPARγ- and
CAR-regulated genes in PPARα-null mice, and PFNA and PFHxS also altered the expression of ERαregulated genes in the knock-out mice. In contrast, none of the compounds altered the expression of
genes commonly affected by an agonist of LXR in either wild-type or null mice.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

439
2. HEALTH EFFECTS

Table 2-29. Gene Expression Changes Induced by Perfluoroalkyl Compounds
Gene expression changes similar to those induced by
prototypical agonist
% PPARαPPARγ
CAR
ERα
LXR
Dose
independent
(mg/kg/day gene
PPARαPPARαPPARαPPARαfor 7 days) changes
WT null
WT null
WT null
WT null
PFOA
PFOS
PFNA
PFHxS
WY14,643

3
10
1
3
3
10
0.1% in diet

~14
~16
~10
~17
24
22

+
+
+
+
+
+

+
+
+/–
+
–
+

+
+
+
+
+
+

+
+
–
+
–
+

+
+
+
+
+
+

–
–
–
+
+
+

–
–
–
–
–
–

–
–
–
–
–
–

2

+

–

–

–

+

–

–

–

+ = significant (p<0.0001) similarity to gene expression changes induced by prototypical receptor agonist as
assessed by running Fisher test; CAR = constitutive androstane receptor; ER = estrogen receptor; LXR = liver X
receptor; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; PPAR = peroxisome proliferator activated receptor; WT = wild-type
Sources: Rosen et al. 2017, 2008b, 2013; Wolf et al. 2008b

Gene expression changes typical of CAR and PXR activators (phenobarbital and pregnenolone
16 α-carbonitrile [PCN]) were also observed in rat liver after oral exposure to PFOA and PFOS (Ren et al.
2009). In addition, PFDA was shown to activate CAR-dependent genes in a study comparing wild-type
and CAR-null mice exposed by intraperitoneal injection (Cheng and Klaassen 2008b).

These data suggest that perfluoroalkyl compounds induce gene expression changes through activation of
other nuclear receptors including PPARγ, CAR, and ERα. Support for these findings are available from
in vitro studies demonstrating binding and/or transactivation of PPARγ, CAR, and ERα by perfluoroalkyl
compounds. Both PFOA and PFOS activated PPARγ in cultured human, mouse, and rat hepatocytes,
albeit with much lower potency than the known agonist rosiglitazone; neither LXRβ nor RXRα was
activated in this system (Vanden Heuvel et al. 2006). Zhang et al. (2014) observed binding of PFOA and
PFOS to human PPARγ in transfected Escherichia coli. However, in experiments conducted by Takacs
and Abbott (2007), neither PFOA nor PFOS activated the mouse or human PPARγ.
Oxidative Stress. Perfluoroalkyl compounds increase oxidative stress in the liver, kidney, and brain.
Increases in oxidative stress may be mediated in part via PPARα activation, but may also result from
activation of the Nrf2 receptor (Xu et al. 2016).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

440
2. HEALTH EFFECTS

Oxidative stress may contribute to oxidative DNA damage, tumor promotion, perturbation of lipid
homeostasis, and stimulation of inflammation, among other changes; thus, increases in oxidative stress
can have diverse physiological effects. Evidence that perfluoroalkyl compounds increase oxidative stress
is available from in vivo and in vitro studies. For example, oxidative DNA damage (measured as 8-OHdG levels) was significantly increased in the liver, but not the kidneys, of male rats exposed to PFOA via
feed for 2 weeks (Takagi et al. 1991). In HepG2 cells cultured with PFOA or PFOS, significant increases
in ROS (measured as 2’7’-dichlorofluorescein diacetate fluorescence) were observed, but there was no
evidence of DNA damage measured with the comet assay (Eriksen et al. 2010). In this system, PFNA,
PFBS, and PFHxA did not induce ROS production, but a significant increase in DNA damage was seen in
cells exposed to PFNA (Eriksen et al. 2010).

In male, but not female, KM mouse pups administered a single subcutaneous injection of PFOS at 1, 2, 3,
4, or 5 weeks of age, brain total antioxidant capacity (T-AOC) was lower than controls at most time
points, and significantly decreased after treatment on PND 21 (Liu et al. 2009). In the liver, T-AOC was
decreased in male pups treated on PNDs 7 and 14, and in females treated on PND 21. Significant
decreases in superoxide dismutase (SOD) activity were noted in the brain of males treated on PNDs 7 and
21, and in the liver of females treated on PND 14. Thus, these data demonstrated increased sensitivity of
younger male mouse pups, compared with older pups or female pups, to oxidative damage by PFOS.
Increases in oxidative stress can lead to NFκB activation (Corton et al. 2014). NFκB activation plays a
role in tumorigenesis, and NFκB transcription factors coordinate immune responses. Few studies have
examined NFκB activation after exposure to perfluoroalkyls. An increase in NFκB mRNA level was seen
in the hippocampus of neonatal rats exposed to PFOS in utero (Zeng et al. 2011). In addition, NFκB
nuclear translocation was accelerated, and NFκB was activated, in breast cancer cells exposed to PFOA
(Zhang et al. 2014). The activation of NFκB was associated with increased invasiveness of the breast
cancer cells, as coexposure to an inhibitor of NFκB reduced the invasiveness induced by PFOA.
Gap Junction Intercellular Communication (GJIC) Inhibition. Perfluoroalkyl compounds also have
been shown to inhibit GJIC both in vivo and in vitro (Corton et al. 2014). GJIC plays an important role in
maintenance of tissue homeostasis, intercellular transmission of regulatory signals, and metabolic
cooperation. Disruption of GJIC is thought to be involved in neurological, reproductive, and endocrine
abnormalities, as well as in carcinogenesis (Corton et al. 2014; EPA 2016h). There are limited data
examining the effects of perfluoroalkyls on GJIC. The available studies showed that both PFOA and
PFOS inhibited GJIC in the livers of rats exposed via diet for 1 week or 3 or 21 days, respectively (Hu et

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

441
2. HEALTH EFFECTS

al. 2002; Upham et al. 1998, 2009). In vitro studies in WB-344 rat liver epithelial cells also showed
inhibition of GJIC after exposure to PFOS (Hu et al. 2002) and to perfluorinated fatty acids with 7–
10 carbons (Upham et al. 1998, 2009). In this system, PFOA activated extracellular receptor kinase,
which may play a role in the inhibition of GJIC. In addition, inhibition of phosphatidylcholine-specific
phospholipase C partially mitigated the GJIC inhibition, suggesting that PFOA-induced activation of this
enzyme may also be involved in GJIC inhibition (Upham et al. 1998, 2009).

PFOS was also shown to inhibit GJIC in dolphin kidney epithelial cells and rat Sertoli cells in vitro (Hu et
al. 2002; Wan et al. 2014a). In Sertoli cells, GJIC plays an important role in maintenance of the
blood:testes barrier and in intercellular communication during spermatogenesis (EPA 2016i).

Impaired Mitochondrial Function. Mitochondrial function, including cellular respiration as well as
mitochondrial membrane potential, has been shown to be perturbed by perfluoroalkyl compounds.
Available data suggest that PFOA and PFOS are relatively weak mitochondrial toxicants (EPA 2016h,
2016i). Mitochondrial proliferation was observed in rats exposed orally to PFOA for 28 days and in mice
exposed to PFOA during gestation and lactation (Quist et al. 2015a, 2015b; Waters et al. 2009). In
isolated rat liver mitochondria, higher concentrations of either PFOA or PFOS were noted to slightly
increase resting respiration rate and decrease membrane potential, possibly due to these compounds’
effects on membrane fluidity (Starkov and Wallace 2002). Testing of other perfluoroalkyls for effects on
mitochondrial respiration rate and oxidative phosphorylation showed a wide range of inhibitory activities,
with PFOS demonstrating the highest potency (3-fold higher than PFOA and 20–30-fold higher than
PFBuS and PFHxA) (Wallace et al. 2013).

2.20.2 Hepatic Toxicity Mechanisms
Hepatic effects of perfluoroalkyl compounds in rodents likely result from a combination of PPARαdependent and independent changes; see Table 2-30. For example, increased liver weight has been
observed in both wild-type and PPARα-null mice orally exposed to PFOA or APFO (Nakagawa et al.
2012; Rosen et al. 2008a), PFOS (Qazi et al. 2009b; Rosen et al. 2010), PFNA (Das et al. 2017; Rosen et
al. 2017), or PFHxS (Das et al. 2017; Rosen et al. 2017), but not in null mice exposed to PFBA by
intraperitoneal injection (Foreman et al. 2009). Similarly, both wild-type and PPARα-null mice exposed
to APFO exhibited increased hepatocyte vacuolation and proliferation, while exposure to WY-14,643 did
not induce such changes in the null mice (Wolf et al. 2008b). Das et al. (2017) showed that PFOA,
PFNA, and PFHxS also increased hepatocyte cell size, percent lipid, and hepatic triglyceride levels, and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

442
2. HEALTH EFFECTS

decreased hepatic DNA content, in both wild-type and PPARα-null mice, while WY-14,643 did not,
indicating that these effects were not dependent on PPARα activation. Similarly, Nakagawa et al. (2012)
showed that at a lower APFO dose (1.0 mg/kg/day for 6 weeks), increases in hepatic triglyceride levels
were observed in wild-type, PPARα-null, and humanized PPAR (hPPAR) mouse strains; however, at a
higher dose (5 mg/kg/day), hepatic triglyceride levels were still increased in PPARα-null and hPPAR
mice, but decreased in wild-type mice.

Table 2-30. Hepatic Effects of Perfluoroalkyl Compounds in Wild-Type and
PPARα-Null Mice Exposed Orally
↑ Relative liver
weight
Dose
PPARα(mg/kg/day) WT null
PFOA
3
+++ +++
PFOS
10
++
++
1
++
+
PFNA
3
+++ ++
10
+++ +++
3
+
+
PFHxS
10
+++ +++
WY-14,643 50
+++ –

↑ % Lipid by
↑ Hepatic
↑ Hepatocyte ↓ Hepatic DNA
cell area
triglycerides
cell size
content
PPARαPPARαPPARαPPARαWT
null
WT null
WT null
WT null
+++ +++ +++ +++
+
+
ND
ND
+++
ND
+++
+++

+++

+++ +++

+++ ++

+

+

+++
–

+++ –
–
–

+++ +++
+++ –

+
+

+
–

+ = statistically significant change from control (the number of plus signs indicates degree of change from controls);
– = not statistically significantly different from control; DNA = deoxyribonucleic acid; ND = no data;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; PPAR = peroxisome proliferator activated receptor; WT = wild type
Sources: Das et al. 2017; Rosen et al. 2008a, 2010, 2017

Lipid homeostasis is maintained through a balance between fatty acid synthesis or accumulation and fatty
acid oxidation. Available data indicate that perfluoroalkyl compounds affect both sides of this balance,
but a growing body of evidence indicates that fatty acid accumulation induced by perfluoroalkyl
compounds tips the balance in favor of hepatic steatosis (Das et al. 2017). As discussed above,
perfluoroalkyl compounds alter lipid homeostasis via PPARα activation, which upregulates genes
involved in fatty acid oxidation and reduces lipid levels. However, as noted above, Das et al. (2017)
indicate that perfluoroalkyl compounds also perturb lipid homeostasis via PPARα-independent
mechanisms. In addition to the effects noted in Table 2-30, increased incidences of hepatic steatosis were
seen in PPARα-null mice exposed to perfluoroalkyl compounds (Das et al. 2017; Minata et al. 2010;
Nakagawa et al. 2012), but not in those exposed to WY-14,643 (Das et al. 2017). Additionally,
microvesicular steatosis was observed in hPPAR mice (Nakagawa et al. 2012). The findings are
consistent with earlier studies showing triglyceride accumulation in rodent livers after exposure to

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

443
2. HEALTH EFFECTS

perfluoroalkyl compounds (Kudo and Kawashima 1997, 2003; Kudo et al. 1999); hepatic steatosis and
glucose intolerance in adult rats exposed to PFOS during the prenatal and postnatal periods (Lv et al.
2013); and inhibited hepatic secretion of VLDL, resulting in steatosis, in APOE3-Leiden mice (a rodent
model with lipoprotein metabolism similar to humans) exposed to PFOS or PFHxS (Bijland et al. 2011).

Das et al. (2017) investigated whether the steatosis induced by PFOA, PFNA, and PFHxS was mediated
by increased fatty acid or triglyceride synthesis or by inhibition of mitochondrial fatty acid transport or
β-oxidation. Microarray analysis of mouse liver after exposure to these compounds showed upregulation
of genes involved in fatty acid and triglyceride synthesis in both wild-type and PPARα-null mice. In
contrast, in vitro experiments demonstrated that these perfluoroalkyl compounds did not affect
mitochondrial fatty acid oxidation in isolated rat liver mitochondria, and neither PFOA nor PFOS altered
fatty acid oxidation in HepG2/C3A human liver cells. The authors suggested that perfluoroalkyl
compounds induce hepatic steatosis by perturbing lipid homeostasis in favor of the accumulation of fatty
acids and triglycerides in the liver.

2.20.3 Developmental Toxicity Mechanisms
Developmental effects observed in laboratory rodents exposed to perfluoroalkyl compounds include
prenatal loss, reduced neonate weight and viability, neurodevelopment toxicity, and delays in mammary
gland differentiation, eye opening, vaginal opening, and first estrus (see Section 2.17 Developmental).
During development, PPARα, PPARβ, and PPARγ mRNA and protein are expressed in the embryo of
rodents and humans (Abbott 2009; Abbott et al. 2010). In humans, the fetal expression levels are
equivalent to levels in adult tissues (Abbott et al. 2010). PPARα activation also appears to be involved in
some, but not all, of the developmental effects of perfluoroalkyl compounds in mice, and the role of
PPARα in mediating developmental toxicity differs among the various compounds. For example, a
gestational exposure study of PFOA resulted in decreases in postnatal survival in wild-type mice, but not
in PPARα-null mice, while the occurrence of full-litter resorptions was similar between the two genotypes
(Abbott 2009; Abbott et al. 2007). In contrast, gestational exposure to PFOS resulted in decreased pup
survival in both wild-type and PPARα-null mice (Abbot et al. 2009). The developmental effects of
PFNA, including reduced pup survival and body weight and delayed eye opening, were seen only in wildtype, and not in PPARα-null mice; however, maternal pregnancy rate was affected only in the null mice
(Wolf et al. 2010).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

444
2. HEALTH EFFECTS

Abbott et al. (2012) showed that PFOA altered expression of genes that are involved in homeostatic
control of lipids and glucose, and postulated that decreased neonatal survival and body weights may be, in
part, due to metabolic disruption. It has been suggested that PFOS interacts with pulmonary surfactants,
and that this effect is responsible for neonatal mortality seen in rats. Grasty et al. (2003, 2005) showed
that neonatal mortality in PFOS-exposed rats was highest when exposure occurred during the gestational
period of lung maturation (GDs 17–20), and that the morphometry of the lungs in exposed neonates was
consistent with immaturity. However, treatment of neonates with rescue agents that hasten lung
maturation did not prevent neonatal mortality induced by PFOS, and examination of the pulmonary
surfactant profile in exposed animals showed no difference from controls, leading Grasty et al. (2005) to
conclude that neonatal mortality in neonatal rats exposed to PFOS was not due to immaturity. Other
hypotheses pertaining to the mechanisms of developmental toxicity of perfluoroalkyl compounds were
not located. However, other molecular- and cellular-level effects of perfluoroalkyl compounds, including
increased oxidative stress, dysregulation of mitochondrial function, and receptor-mediated events, may be
involved in the observed developmental effects of these compounds.

2.20.4 Immunotoxicity Mechanisms
NTP (2016b) conducted a systematic review of the human, animal, and in vitro data examining
immunotoxic effects of PFOA and PFOS. The conclusion of the systematic review was that both PFOA
and PFOS are “presumed to be immune hazards to humans.” Evidence was considered strong that both
compounds were associated with suppression of the antibody response, while there was weaker evidence
for PFOA-induced impairment of infectious disease resistance, increased hypersensitivity-related
outcomes, and increased autoimmune disease incidence, and for PFOS-induced suppression of natural
killer cell activity. A recent study comparing the T-cell dependent antibody response (TDAR) in female
wild-type and PPARα knock-out mice after exposure to PFOA with or without antigen exposure showed
that PFOA suppressed TDAR in both wild-type and knock-out mice, indicating that the mechanism for
antibody response suppression is independent of PPARα activation (DeWitt et al. 2016). These
investigators observed no treatment-related changes in splenic lymphocyte subpopulations in exposed
mice of either genotype, suggesting that PFOA suppressed TDAR via impairment of B-cell/plasma cell
function rather than by altering lymphocyte numbers. DeWitt et al. (2012) and Corsini et al. (2014)
reviewed mechanistic data for perfluoroalkyl-induced suppression of antibody response, and postulated
that perfluoroalkyl compounds may modulate cell-signaling responses critical to antibody production,
including c-Jun, NF-κB, and IL-6.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

445
2. HEALTH EFFECTS

2.20.5 Endocrine Mechanisms
Perfluoroalkyl compounds have been shown to induce alterations in thyroid hormone levels in rats, and
associations between serum perfluoroalkyl concentrations and thyroid hormone levels have been reported
in human epidemiological studies (see Section 2.13). Few data examining mechanisms of thyroid
hormone disruption are available, but suggest that effects of perfluoroalkyl compounds on thyroid
function may be mediated by binding to the thyroid hormone receptor, and/or by altering expression of
genes involved in thyroid function or thyroid hormone regulation. Several perfluoroalkyl compounds
were shown to bind to the human thyroid hormone receptor in cultured GH2 cancer cells and in molecular
docking experiments (Ren et al. 2015). In the in vitro tests, all 16 of the tested compounds exhibited
lower affinity for the receptor than T3 (Ren et al. 2015). Among the tested compounds, PFOS exhibited
the strongest agonist activity (Ren et al. 2015). Alterations in the mRNA or protein levels of thyroidregulating genes have been observed after oral exposure of male Sprague-Dawley rats to PFOS. PFOS
exposure for 5 or 90 days resulted in decreased hepatic levels of mRNA type 1 deiodinase (DIO1, which
bioactivates T3 by deiodination of T4) (Martin et al. 2007; Yu et al. 2009a); after 5 days of exposure,
hepatic mRNA for type 3 deiodinase (DIO3, which inactivates T3) was increased relative to controls
(Martin et al. 2007). After 90 days, hepatic levels of uridine diphosphoglucuronosyl transferase 1A1
(UGT1A1, which plays a role in T4 turnover) mRNA and thyroid levels of DOI1 protein were increased,
while there were no changes in thyroid levels of the sodium iodide symporter, thyrotropin (THS) receptor,
or activity of thyroid peroxidase (Yu et al. 2009a).

Limited data from in vitro studies suggest the possibility that perfluoroalkyl compounds may interact with
the estrogen and androgen receptors. PFOA, PFOS, PFHxS, PFNA, and PFDeA were all shown to be
antagonists of the androgen receptor, while PFOA, PFOS, and PFHxS induced transactivation of the
estrogen receptor (Kjeldsen and Bonefeld-Jorgensen 2013). Recently, analysis of gene expression data
from the livers of wild-type and PPARα-null mice exposed to PFOA, PFOS, PFHxS, and PFNA by
gavage for 7 days indicated similarities to gene expression changes induced by known ERα agonists
(Rosen et al. 2017), providing indirect evidence for perfluoroalkyl changes in the liver mediated via ER
activation. However, at oral doses up to 1 mg/kg, PFOA failed to induce treatment-related alterations in
uterine weight, ER-dependent gene expression, or morphology of reproductive organs in uterotrophic
assays using immature CD-1 mice (Dixon et al. 2012; Yao et al. 2014), suggesting that PFOA is either
inactive in vivo or of very low estrogenic potency.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

446
2. HEALTH EFFECTS

2.20.6 Cancer Mechanisms
Like many other PPARα agonists, PFOA induced hepatocellular adenomas, Leydig cell adenomas, and
pancreatic acinar cell adenomas in rats (Biegel et al. 2001). Liver tumors induced by PFOA are believed
to be mediated largely through PPARα activation, and considered to be of limited or no relevance to
humans (EPA 2016h), based on species differences in response to PPARα (see details above under
PPARα activation). An expert panel convened by EPA’s Science Advisory Board in 2006 to review
issues related to the toxicity of PFOA agreed that the weight of evidence supports the hypothesis that
induction of liver tumors in rats by PFOA is mediated by a PPARα agonism mode of action (EPA 2006);
this conclusion is also reflected in the EPA Health Effects Support Document for PFOA (EPA 2016h). A
recent review by a panel of experts from academia, government, industry, and consulting groups updated
the Klaunig et al. (2003) assessment of PPARα agonism as a liver cancer mode of action, and drew the
same conclusion: while the PPARα mode of action for liver tumors is biologically plausible, species
differences in response to PPARα activation indicate that liver tumors are unlikely to be induced by
PPARα induction in humans (Corton et al. 2014).
Studies conducted in rainbow trout, an animal model that is similar to humans in terms of insensitivity to
peroxisome proliferators, suggest that some perfluoroalkyls may induce liver cancer by alternate
mechanisms (Benninghoff et al. 2011, 2012). The investigators (Benninghoff et al. 2011) found that
PFOA, PFNA, PFDeA, and PFUA were potent inducers of vitellogenin, an estrogen-responsive
biomarker protein. In vitro, PFOA, PFOS, PFNA, and PFDeA also had weak to very weak affinities for
estrogen receptors (ERα) for several species including humans, mice, and rats (Benninghoff et al. 2011).
In vivo studies demonstrated that PFOA, PFOS, PFNA, and PFDeA enhanced liver carcinogenesis via a
mechanism that likely involves interactions with hepatic estrogen receptors (Benninghoff et al. 2012).

Although Leydig cell tumors are commonly induced by peroxisome proliferating agents such as
perfluoroalkyl compounds, the mode of action by which these tumors are induced, and thus their
relevance to humans, is much less clear (Corton et al. 2014; EPA 2016h; Klaunig et al. 2003). One mode
of action proposed for the induction of Leydig cell tumors involves PFOA-induced inhibition of
testosterone biosynthesis, leading to increased production of gonadotropin releasing hormone and
circulating LH, which promotes Leydig cell proliferation. Activation of PPARα may be involved in the
decreased serum testosterone levels; PPARα-null mice did not exhibit the reduction in testosterone
concentration seen in wild-type mice exposed to PFOA (Li et al. 2011). Evidence of decreased serum
testosterone and increased serum estradiol was seen in studies of male rats exposed orally to PFOA for

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

447
2. HEALTH EFFECTS

14 days (Biegel et al. 1995; Cook et al. 1992; Liu et al. 1996). Reduced testosterone levels may occur
through the conversion of testosterone to estradiol via the enzyme aromatase. Hepatic aromatase activity
was shown to be markedly increased in male rats exposed to APFO by gavage for 14 days, and aromatase
activity was positively correlated with serum estradiol levels in these animals (Liu et al. 1996). The
relevance of Leydig cell tumors induced by PFOA to human risk assessment is uncertain. For example,
an intermediate-duration study in Cynomolgus monkeys exposed to PFOA did not find treatment-related
alterations in serum estradiol, estrone, estriol, or testosterone (Butenhoff et al. 2002). Studies of humans
occupationally exposed to PFOA have not consistently reported alterations in estradiol or testosterone
levels (Klaunig et al. 2012). In addition, humans are less sensitive than rats to LH stimulation, and the
average number of LH receptors per Leydig cell is 13-fold higher in rats than humans (Klaunig et al.
2012). In summary, the induction of Leydig cell tumors by PFOA may be mediated by effects on
aromatase activity or testosterone biosynthesis, both of which may be related to PPARα activation (EPA
2016h). While the relevance of the PPARα mode of action to humans is uncertain, the data supporting
this mode of action for Leydig cell tumors is not sufficient to rule out human relevance (EPA 2016h).

The mechanism of PFOA-induced pancreatic acinar cell tumors has not been elucidated, and relevant data
are limited. A proposed mode of action involves stimulation of PPARα leading to reduced bile flow
and/or changes in bile acid composition with subsequent increase in cholecystokinin (CCK), which
stimulates pancreatic cell proliferation and tumor formation (EPA 2016h). Effects on bile acid
composition induced by PFOA may be mediated by effects on bile acid transporters. PFOA exposure has
been shown to decrease expression of OATPs and increase expression of MRP3 and MRP4 (Cheng and
Klassen 2008a; Maher et al. 2008). In a study using wild-type and PPARα-null mice, increased biliary
excretion of PFOA was seen in wild-type mice compared with null mice, and biliary excretion of bile
acids was highest in the null mice (Minata et al. 2010). These observations suggest the possibility that
increased excretion of PFOA could diminish the excretion of bile acids that require the same transporters.
However, given the limitations in available data, information is insufficient to fully characterize the mode
of action for PFOA-induced pancreatic tumors (EPA 2016e).
2.21 GENOTOXICITY
No studies of genotoxicity in humans exposed to perfluoroalkyl compounds were located.

Administration of a single intraperitoneal injection of 100 mg/kg PFOA to male Fischer-344 rats resulted
in a significant increase in the levels of 8-hydroxydeoxyguanosine (a marker of oxidative DNA damage)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

448
2. HEALTH EFFECTS

in liver DNA, but not in kidney DNA; the same dose of PFBA had no effect on liver or kidney DNA
(Takagi et al. 1991). Oral administration of approximately 20 mg/kg/day PFOA or 10 mg/kg/day PFDeA
in the diet for 2 weeks to male Fischer-344 rats induced hepatomegaly and also increased the levels of
8-hydroxydeoxyguanosine in liver DNA but not in kidney DNA (Takagi et al. 1991). These findings led
the authors to conclude that induction of peroxisome proliferation also leads to organ-specific oxidative
DNA damage. Unpublished studies summarized by Butenhoff et al. (2014) and Kennedy et al. (2004) did
not find increases in the occurrence of polychromatic erythrocytes in the bone marrow of mice orally
exposed to PFOA. PFOS induced micronuclei frequency and decreased the ratio of polychromatic
erythrocytes to normochromatic erythrocytes in bone marrow of rats following oral exposure to 0.6–
2.5 mg/kg for 30 days (Celik et al. 2013; Eke and Celik 2016).

In vitro studies provide evidence that PFOA and PFOS are not mutagenic at noncytotoxic concentrations.
Butenhoff et al. (2014) and Kennedy et al. (2004) summarized the results of various unpublished
mutagenicity studies with PFOA. Negative results were found in reverse mutation assays using
Salmonella typhimurium (strains TA98, TA100, TA1535, TA1537, and TA1538), Saccharomyces
cerevisiae, and Escherichia coli (WP2uvrA strain) with or without metabolic activation. In mammalian
cells, PFOA was negative for forward mutations using Chinese hamster ovary cells, for chromosomal
aberrations in Chinese hamster ovary cells and human lymphocytes, and for cell transformation in C3H
10T1/2 cells. Similarly, two in vivo assays found no alterations in the occurrence of micronuclei in mice.
PFOA also was not mutagenic in S. typhimurium TA1535/pSK1002 (hisG46, rfa, uvrB) with or without
metabolic activation using the umu test (Oda et al. 2007) or in S. typhimurium TA98, TA100, TA102, and
TA104 strains with or without metabolic activation using an Ames assay (Fernández Freire et al. 2008).

Incubation of human hepatoma HepG2 cells with 50–400 µM PFOA caused DNA strand breaks and 100–
400 µM increased the incidence of micronuclei, in a dose-related manner in both cases (Yao and Zhong
2005). These effects were accompanied by a significant increase in reactive oxygen species (ROS),
which the investigators suggested caused the DNA damage. Bjork and Wallace (2009) measured mRNA
expression for DNA damage inducible Ddit3 to assess DNA damage in primary rat and human hepatocyte
cultures and in HepG2/C3a hepatoma cells. Significant increases in mRNA transcription for Ddit3 were
found in primary rat hepatocytes at 100 µM PFOA and in primary human hepatocytes and HepG2/C3a
hepatoma cells at 200 µM PFOA. Although both studies provide evidence of DNA damage, the tested
concentrations were very high as compared to what could be expected to occur in the environment. A
significant increase in mutation frequencies was observed in hamster-human hybrid cells exposed to
200 µM PFOA for 1–16 days; a 79% decrease in cell viability was also observed at this concentration

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

449
2. HEALTH EFFECTS

(Zhao et al. 2011). Concurrent treatment with a ROS inhibitor significantly decreased the mutagenic
potential, indicating that ROS may play an important role in mediating the genotoxic effects of PFOA.
PFOA induced DNA damage in Paramecium caudatum following exposure to 100 µM for 12 and
24 hours (Kawamoto et al. 2010). Intracellular ROS was significantly increased and DNA damage was
not reversed by the application of glutathione, a ROS-inhibitor, indicating that intracellular ROS may not
be the cause of PFOA-induced DNA damage. PFOS did not induce DNA damage in this study. In
contrast, no increases in DNA damage or micronuclei formation were found in human hepatoma HepG2
cells following a 24-hour exposure to PFOA concentrations as high as 800 µM (Florentin et al. 2011);
cytotoxicity was observed at ≥200 µM. Eriksen et al. (2010) also found no evidence of DNA damage in
HepG2 cells incubated with 100 or 400 µM PFOA for 24 hours.

OECD (2002) summarized unpublished mutagenicity studies conducted with PFOS. PFOS was negative
in all tested assays. It did not induce reverse mutations in S. typhimurium or E. coli with or without
metabolic activation. A study published after this review also found that PFOS was not mutagenic in
S. typhimurium TA1535/pSK1002 (hisG46, rfa, uvrB) with or without metabolic activation using the umu
test (Oda et al. 2007). As summarized by OECD (2002), PFOS did not induce chromosomal aberrations
in human lymphocytes with or without metabolic activation and did not induce unscheduled DNA
synthesis in primary cultures of rat hepatocytes. In addition, PFOS did not induce micronuclei in the
bone marrow of CD-1 mice in an in vivo assay. PFOS did not result in DNA damage in Syrian hamster
embryo cells at concentrations up to 50 µg/mL, but did induce cell transformation at noncytotoxic
concentrations (0.2 and 2 µg/mL) following 5 and 24 hours of exposure (Jacquet et al. 2012). Similarly,
PFOS did not induce DNA damage or micronuclei formation in human hepatoma HepG2 cells following
a 24-hour exposure to PFOA concentrations as high as 600 µM; cytotoxicity was observed at ≥300 µM
(Florentin et al. 2011). Another study of with HepG2 cells did not find evidence of DNA damage at
concentrations of 100 and 400 µM PFOS (Eriksen et al. 2010).

Limited in vitro data on the genotoxicity of other perfluoroalkyls were located. No DNA damage was
found in HepG2 cells incubated with 100 or 400 µM PFHxS or PFBuS for 24 hours (Eriksen et al. 2010).
A “modest” increase in DNA damage was observed at 400 µM PFNA, a cytotoxic concentration (Eriksen
et al. 2010).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

450

CHAPTER 3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS,
BIOMARKERS, CHEMICAL INTERACTIONS
3.1

TOXICOKINETICS

Toxicokinetic data on perfluoroalkyls examined in this profile are available from studies in humans and
animals. Most studies in animals administered perfluoroalkyls by the oral route. These data are briefly
summarized below.
•

Absorption
-

Perfluoroalkyls are absorbed following oral, inhalation, and dermal exposure.

-

Quantitative estimates of the fractional absorption of orally administered perfluoroalkyls in
animals range from >50% for PFHxS to >95% for PFOA, PFBA, PFNA, PFDeA, PFUA, and
PFDoA.

-

No quantitative estimates of the fractional absorption of perfluoroalkyls following inhalation
or dermal exposure were identified.

•

Distribution
-

Perfluoroalkyls are widely distributed in the body, with the highest concentrations in the
liver, kidneys, and blood.

•

-

In the blood, perfluoroalkyls bind to albumin and other proteins.

-

Perfluoroalkyls can be transferred to the fetus during pregnancy and to nursing infants.

Metabolism
-

Results of available oral and in vitro studies suggest that perfluoroalkyls are not metabolized
and do not undergo chemical reactions in the body.

-

Although no studies examining metabolism of perfluoroalkyls following inhalation or dermal
exposure were identified, metabolism by these exposure routes is not expected.

•

Excretion
-

Studies of elimination rates (i.e., half-lives) of perfluoroalkyls show that elimination t1/2
values are similar following intravenous, intraperitoneal, and oral exposures. Findings
suggest that route of absorption has no substantial effect on rates of elimination of absorbed
perfluoroalkyls.

-

Perfluoroalkyls are primarily eliminated in the urine, with smaller amounts eliminated in
feces and breast milk.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

451

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

-

Perfluoroalkyls undergo biliary excretion, but substantial reabsorption occurs; therefore,
biliary excretion is not a major elimination pathway.

-

Rates of elimination of perfluoroalkyls vary substantially across chemical species and animal
species, and show sex differences and age-dependencies within certain species.

-

In general, perfluoroalkyl sulfonates are eliminated more slowly than perfluoroalkyl
carboxylates; elimination rate decreases with increasing chain length, and increases with
increased branching.

-

In humans, estimates for elimination t1/2 range from hours (PFBA: 72–81 hours) to several
years (PFOA: 2.1–8.5 years; PFOS: 3.1–7.4 years; PFHxS: 4.7-15.5 years).

-

Evidence for sex differences in elimination of perfluoroalkyls in humans is not as strong as in
rats. Menstruation may contribute to faster elimination of PFOS in younger women
(≤50 years) women compared to men and older women.

3.1.1

Absorption

Inhalation Exposure. Studies of the absorption of perfluoroalkyls in humans following inhalation
exposure were not located; elevated serum concentrations of perfluoroalkyls in workers in fluorochemical
production industry have been reported (see Table 5-22), indicating that perfluoroalkyls are absorbed
following inhalation exposure. Occupational exposures in these workers are likely to have included
inhalation of aerosols of perfluoroalkyls complexed with airborne dusts. Higher serum levels in workers
compared to the general population (see Table 5-20) probably reflects a predominant contribution from
inhaled perfluoroalkyls.

Studies conducted in rodents provide direct evidence for absorption of inhaled perfluoroalkyls. PFOA
was detected in plasma of rats within 30 minutes of initiating nose-only exposures to aerosols (mass
median aerodynamic diameter [MMAD]=1.9–2.1 µm) of 1–25 mg ammonium PFOA/m3. Plasma
concentrations increased during the 6-hour exposure, with the highest concentrations observed at 9 hours
(3 hours after cessation of exposure) in male rats and at 7 hours (1 hour after cessation of exposure) in
females (Hinderliter et al. 2006a). Assuming an elimination t1/2 of absorbed PFOA of approximately
160 hours in male rats, a peak plasma concentration at 9 hours would correspond to an absorption t1/2 of
approximately 1.3 hours (see discussion below, Equations 3-1 and 3-2). The earlier time of highest
plasma concentration observed in female rats appears to be associated with faster elimination of absorbed
PFOA in female rats, compared to male rats (see Section 3.1.4).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

452

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Nose-only exposure of male rats to dusts of ammonium perfluorononanoate induced significant increases
in absolute and relative liver weight, assessed 5 and 12 days after exposure, providing indirect evidence of
absorption of this compound through the respiratory airways (Kinney et al. 1989).

Oral Exposure. Studies of absorption of perfluoroalkyls through the gastrointestinal tract in humans are
not available. A study of the general population of Europe and North America estimated that the greatest
portion of the chronic exposure to PFOS and PFOA results from the intake of contaminated food,
including drinking water (Trudel et al. 2008). Direct evidence of oral absorption of perfluoroalkyl
compounds was provided in studies that found associations between environmental levels (e.g., drinking
water) and perfluoroalkyl concentrations in human serum (Emmett et al. 2006a; Hoffman et al. 2011;
Hölzer et al. 2008; Seals et al. 2011; Wilhelm et al. 2008) and by reductions in serum levels after
exposures from water were eliminated or reduced (Bartell et al. 2010; Emmett et al. 2009).

Animal data provide quantitative estimates of the fractional absorption of orally administered PFOA,
PFOS, PFBA, and PFHxS, PFHpA, PFNA, PFDeA, PFUA, and PFDoA, with estimates ranging from
>50% for PFHxS to >95% for PFOA, PFBA, PFNA, PFDeA, PFUA, and PFDoA. Greater than 95% of
an oral dose of ammonium [14C]PFOA was absorbed in rats that received single gavage doses ranging
from 0.1 to 25 mg/kg (Kemper 2003). In male and female mice, comparison of the 24-hour area under
the curve (AUC) for oral and intravenous administration showed that 90–100% of the oral dose was
absorbed for PFOA (females), PFNA (males and female), PFDeA (males and females), PFUA (males and
females), and PFDoA (males and female); however, absorption of PFOA in males was 80%, compared to
100% in females (Fujii et al. 2015a, 2015b). Based on comparison of the AUC for oral and intravenous
administration, the estimated oral absorption fractions were 50% in female rats administered a single
10 mg/kg dose of potassium [18O3- ]PFHxS (Sundström et al. 2012) and 79 and 55% in male and female
rats administered a single dose of 4 mg/kg sodium [18O3- ]PFHxS (Kim et al. 2016b). Sundström et al.
(2012) stated that this estimate may not be reliable due to the short (24 hours) observation period. A
comparison of 14C disposition in rats, mice, hamsters, and rabbits following an oral dose of 10 mg
ammonium [14C]PFOA/kg showed that similar fractions of the dose were absorbed (Hundley et al. 2006).
The estimated absorbed fractions (i.e., 14C in tissues, urine, and exhaled air measured 120–168 hours after
the dose) in males were 89% in rats, 82% in mice, 92% in hamsters, and 88% in rabbits. Corresponding
values for females were 76% in rats, 61%, in mice, 75% in hamsters, and 88% in rabbits. These estimates
exclude 14C excreted in feces, which may have been absorbed and secreted in bile before excretion (see
Section 3.1.4). Fasting appears to increase absorption of PFOA. Plasma PFOA concentrations in rats,
24 hours following a gavage dose of 10 mg ammonium PFOA/kg, were 2–3 times higher when

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

453

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

administered to fasted rats, compared to fed rats (Hinderliter et al. 2006b). The estimated absorption
fractions of ingested ammonium [14C]PFOA or potassium [14C]PFOS (administered as a 4.2 mg/kg oral
dose) were >93 and >95% in rats, respectively (Johnson and Ober 1979, 1999a). Based on combined
urinary excretion and retention in the carcass (excluding the gastrointestinal tract and its contents), the
estimated oral absorption fraction of [14C]PFOS (administered as a single 4.2 mg/kg dose of potassium
[14C]PFOS) in male rats was >95% (Chang et al. 2012). The estimated absorption fraction of PFBA
(administered as 30 mg/kg oral dose of PFBA) was >95% in rats (Chang et al. 2008a). Cumulative
excretion of PFBA 24 hours after an oral dose (administered as 10, 30, or 100 mg/kg ammonium PFBA)
was approximately 35% in urine and 4–11% in feces in male mice; and 65–69% in urine, and 5–7% in
feces in female mice (Chang et al. 2008a).

Studies examining the rate of absorption of PFOA, PFBA, and PFBuS show rapid absorption from the
gastrointestinal tract, with values for absorption t1/2 of <2 hours. For PFOA, the highest observed
concentrations of 14C in plasma occurred in male rats at approximately 10 hours (range 7.5–15 hours)
following single oral doses ranging from 0.1 to 25 mg ammonium PFOA/kg (Kemper 2003). The
elimination t1/2 of 14C in plasma estimated in these same animals was approximately 170 hours (range
138–202 hours), corresponding to an elimination rate constant (ke) of 0.0044 hour-1 (range 0.004–
0.005 hour-1). The corresponding absorption t1/2 of approximately 1.5 hours (ka=0.45 hour-1) can be
calculated from these observations (Equations 3-1 and 3-2):

t max = ln
t1 / 2 =

ka
1
⋅
k e (k a − k e )

ln(2)
ke

Eq. (3-1)

Eq. (3-2)

Where tmax = time of maximum concentration of 14C; ke = elimination rate constant; and ka = absorption
constant. The absorption rate of PFOA appears to be greater in female rats compared to male rats. The
time to peak concentrations of 14C in plasma occurred at approximately 1.1 hour (range 0.6–1.5 hours) in
female rats and 10 hours (range 7–15 hours) in male rats following single oral doses ranging from 0.1 to
25 mg ammonium PFOA/kg (Kemper 2003). The elimination t1/2 of 14C in plasma estimated in these
same animals varied with dose and ranged from 3.2 hours at the lowest dose (ke=0.23 hour-1) to
16.2 hours at the highest dose (ke=0.059 hour-1). The estimated absorption t1/2 from the observations
made at all doses (0.1, 1, 5, and 25 mg/kg), based on Equations 3-1 and 3-2, was approximately
0.25 hours (range 0.12–0.38 hours). The absorption t1/2 of PFBA in male and female rats following
administration of a single oral dose (30 mg/kg ammonium PFBA) was 0.23 hours (3.04 hour-1) in males
and 0.17 hours (4.15 hour-1) in females (Chang et al. 2008a). In male and female mice administered 10–

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

454

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

30 mg/kg ammonium PFBA, the absorption t1/2 was <1 hour, although the absorption rate may be dosedependent in males, with higher absorption t1/2 at doses >30 mg/kg (Chang et al. 2008a). Similar results
for were reported by Olsen et al. (2009) based on estimated compartmental pharmacokinetic parameters
for PFBuS in serum of male and female rats following a single intravenous or gavage dose of 30 mg
potassium PFBuS. Plasma concentration-time profiles were fit to a two-compartment elimination model.
The absorption t1/2 can be approximated from these data using Equation 3-1, with the elimination rate
constant represented by the fast-phase elimination rate constant estimated for either the oral or
intravenous dose. Using the oral or intravenous parameters yield similar values for the absorption t1/2
(0.12–0.16 hours). The estimated tmax values following the gavage dose were 0.42 hours in males and
0.33 hours in females. The fast-phase elimination rate constants following the gavage dose were
0.892 hours-1 (t1/2=0.79 hours) in males and 1.308 hours-1 (t1/2=0.53 hours) in females. The corresponding
values for absorption t1/2 were 0.14 hours (ka=5.0 hours-1) in males and 0.12 hours (ka=5.8 hours-1) in
females. Use of the fast-phase elimination rate constants estimated following intravenous administration
(male: 1.143 hours-1; female: 1.956 hours-1) yielded values for the absorption t1/2 of 0.16 hours in males
(ka=4.30 hours-1) and females (ka=4.45 hours-1).

Mechanisms of oral absorption of perfluoroalkyls have not been elucidated.

Dermal Exposure. Dermal exposures of rats to ammonium PFOA have been shown to produce systemic
(e.g., liver, immunotoxicity) toxicity in animals (see Chapter 2). Estimates of the amount or rates of
dermal absorption in humans or animals have not been reported. PFOA was detected in serum of mice
following dermal application of PFOA dissolved in acetone (Franko et al. 2012). The investigators noted
PFOA ingestion may have occurred during grooming and may have contributed to the body burden.
Dermal absorption of PFOS was assessed following application of single doses of potassium PFOS (doses
up to 0.30 mg/kg) and the diethanolamine salt of PFOS (doses up to 20 µg/kg) to clipped, intact skin of
rabbits (Johnson 1995a, 1995b). Analysis of the liver 28 days after application showed no increase in
content of total organic fluoride compared to controls, indicating that dermal absorption was not
detectable at low dose levels using this methodology. Dermal penetration of PFOA has been studied in
preparations of isolated rat, mouse, and human epidermis (Fasano et al. 2005; Franko et al. 2012). These
studies indicate that the rat and mouse skin may be more permeable to PFOA than human skin.
Approximately 24% of a dermal dose of PFOA (0.5 mg in 1% acetone) was absorbed across isolated full
thickness human skin in 24 hours and 45% of the dose was retained in skin (Franko et al. 2012); it is
noted that the acetone, as well as the glycerol used to pretreat the skin may have enhanced PFOA
absorption. Permeability was sensitive to pH and was higher when the skin was buffered at pH 2.5

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

455

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

(5.5x10-2 cm/hour) compared to pH 5.5 (4.4x10-5 cm/hour), well above the pKa for the terminal
carboxylic acid of PFOA (Franko et al. 2012). This suggests that permeability of the unionized acid is
greater than that of the dissociated anion. Lower permeability of ionized PFOA is also suggested by
relatively low permeability of the ammonium salt of PFOA in isolated preparations of rat and human skin.
Following application of the ammonium salt of PFOA to isolated human or rat epidermis (150 µL/cm2 of
a 20% aqueous solution of ammonium PFOA; approximately 30 mg ammonium PFOA/cm2),
approximately 0.048% of the dose was absorbed across human epidermis and 1.44% was absorbed across
rat epidermis in 40 hours. The estimated dermal penetration coefficients were 9.49x10-7 cm/hour in the
isolated human epidermis and 3.25x10-5 cm/hour in the isolated rat epidermis.
3.1.2

Distribution

Available information on the distribution of perfluoroalkyls is obtained from oral exposure studies in
laboratory animals and occupational exposure studies in which exposure is predominantly by inhalation.
Studies specifically examining the distribution of perfluoroalkyls by inhalation or dermal exposure were
not identified. As discussed in Section 3.1.3 (Metabolism), perfluoroalkyls do not undergo metabolism.
Therefore, distribution is expected to be the same regardless of the route of administration.

Distribution in Blood. In a study of perfluoroalkyl workers, serum:plasma ratios for PFHxS, PFOS, and
PFOA were 1:1, and this ratio was independent of the concentrations measured (Ehresman et al. 2007).
The ratio of whole blood:plasma (or serum) was approximately one-half, which corresponded to volume
displacement by red blood cells, suggesting that these perfluoroalkyls do not enter cellular components of
blood. In studies conducted in animals, most of the PFOA in blood is in the plasma fraction. In rats,
24 or 48 hours following an oral dose of 11.4 mg ammonium [14C]PFOA/kg, the red blood cell:plasma
PFOA concentration ratio ranged from 0.2 to 0.3, suggesting that there was no selective retention of
PFOA by red blood cells (Johnson and Ober 1999a). Blood:plasma (or serum) ratios of approximately
0.5 have also been observed in rats following intravenous injection of PFOA (Kudo et al. 2007).

Perfluoroalkyls in plasma bind to serum albumin. The dissociation constant for binding of PFOA to
serum albumin is approximately 0.4 mM (0.38 mM, ±0.04 standard deviation [SD] for human serum
albumin; 0.36 nM, ±0.08 SD for rat serum albumin) and involves 6–9 binding sites (Han et al. 2003).
Given a dissociation constant (KD) of 0.4 mM and an albumin concentration of approximately 0.6 mM,
>90% of PFOA in serum would be expected to be bound to albumin when the serum concentration of
PFOA is <1 mM (<440 mg/L). This is consistent with observations of the bound fraction of

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

456

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

perfluoroalkyls in plasma of rats that received a gavage dose of 25 mg PFOA/kg (Han et al. 2003, 2005;
Ylinen and Auriola 1990), and in human, rat, and monkey plasma incubated in vitro with perfluoroalkyls
(e.g., PFHxA, PFOA, PFOS, PFNA, PFDeA) (Ohmori et al. 2003; Kerstner-Wood et al. 2003).
Comparison of dissociation constants for binding of PFOA and PFOS to human serum albumin indicates
that PFOS (KD: 8x10-8) has a higher binding affinity than PFOA (KD: 1x10-4) for albumin, consistent with
the longer t1/2 of PFOS versus PFOA in humans (Beesoon and Martin 2015; see section 3.1.4 for
additional information). PFOS has also been shown to bind to human hemoglobin in vitro (Wang et al.
2016). PFBuS was found to bind only to albumin, whereas PFOS, PFOA, and PFHxS were found to have
the potential to bind to other human serum binding proteins, including plasma gamma-globulin, alphaglobulin, alpha-2-macroglobulin, transferrin, and beta-lipoproteins (Kerstner-Wood et al. 2003).

Distribution to Extravascular Tissues. Absorbed perfluoroalkyls distribute from plasma to soft tissues,
with the highest extravascular concentrations achieved in liver. An analysis of samples from human
cadavers attempted to quantify PFOA, PFOS, PFOSA, and PFHxA concentrations in serum and liver
(Olsen et al. 2003c). The route of exposure was unknown. Mean serum PFOS concentration was
17.7 ng/mL (95% CI 13.0–22.5, range of <6.9 [limit of quantification] to 57 ng/mL, n=24) and was not
different in males (18.2 ng/mL, n=13) and females (17.2 ng/mL, n=11). The mean liver concentration
was 18.8 ng/g (95% CI 14.1–23.5; range <7.3–53.8 ng/g, n=30). The mean liver:serum concentration
ratio was 1.3 (95% CI 0.9–1.7, n=23) and was not different in males (1.3, n=13) and females (1.3, n=10).
Most liver and serum concentrations for PFOA, PFOSA, and PFHxA were below the limit of
quantification; these limits were <17.9–<35.9 ng/mL for PFOA, <7.5–<19.6 ng/g for PFOSA, and <3.4–
<18.5 ng/mL for PFHxA.

Studies conducted in nonhuman primates and rodents have provided additional information on the
distribution of absorbed perfluoroalkyls to extravascular tissues. Distribution, as assessed from tissue
perfluoroalkyl concentrations and tissue:serum ratios, exhibits profound species and sex differences as
well as dose-dependencies (e.g., tissue levels that change disproportionately with dose). These
differences have been attributed, in part, to species and sex differences in elimination kinetics of absorbed
perfluoroalkyls and dose-dependence of elimination kinetics (see Section 3.1.4). In general, a consistent
finding across species is that the liver receives a relatively high fraction of the absorbed dose and may
also experience relatively high tissue concentrations compared with other tissues, with blood
(i.e., plasma) and kidney also showing relatively high concentrations. The most extensive investigations
of tissue distribution have been conducted in rodents.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

457

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Bogdanska et al. (2011) examined distribution of 35S following dietary exposure to adult male C57/BL6
mice to low (environmentally relevant; 0.031 mg/kg/day) and high (experimentally relevant;
23 mg/kg/day) doses of [35S]PFOS for 1–5 days. For both low and high doses after 1, 3, and 5 days of
exposure, 35S was distributed to the following tissues: blood, liver, lung, kidney, skin, whole bone,
pancreas, spleen, thymus, heart, testes, epididymal fat, fat pads, brain, and muscle; 35S was also detected
in tissues throughout the gastrointestinal tract. Similar tissue:blood ratios were observed in both dose
groups. In low-dose animals after 5 days of treatment, the highest tissue concentrations (excluding the
gastrointestinal tract) were liver (tissue:blood=5.8), followed by lung (tissue:blood=1.4), whole bone,
including marrow (tissue:blood=1.1), blood, and kidney (tissue:blood=0.94). In high-dose animals, the
highest tissue concentrations were liver (tissue:blood=3.6), followed by lung (tissue:blood=1.6), blood,
kidney (tissue:blood=0.81), and whole bone, including marrow (tissue:blood=0.72). A similar pattern of
distribution was observed following intravenous administration of [14C]potassium PFOS (4.2 mg/kg) to
male rats (Johnson and Ober 1980). For both dose groups, the tissue:blood ratios for all other tissues
were <1. In male and female CD-1 mice administered a single oral dose (4.2 mg/kg) of [14C]PFOS, the
highest concentrations of 14C was observed in the liver, followed by serum, and then kidney, with similar
tissue levels observed in males and females (Chang et al. 2012). In male and female rats fed diets
containing 0, 2, 20, 50, or 100 mg/kg [13C]sodium PFOS (equivalent to 0, 0.14, 1.33, 3.21, and 6.34 and
0, 0.15, 1.43, 3.73, and 7.58 mg/kg/day in males and females, respectively) for 28 days, PFOS levels were
highest in liver, followed by spleen, heart, and serum. Liver:serum ratios for the 2, 20, 50, and
100 mg/kg/day diets were approximately 52, 42, 41, and 35, respectively, in males and 30, 47, 20, and 23,
respectively, in females (Curran et al. 2008). Except for rats fed diets containing 20 mg/kg, the
liver:serum ratio in males was higher than in females. No additional data were reported to determine if
PFOS distribution differed between male and female rats.
Kemper (2003) determined the distribution of 14C in male and female rats at the approximate time of
maximum plasma concentration in both sexes, following single gavage doses of [14C]PFOA (as
ammonium PFOA, 0.1–25 mg/kg). This design allows a more direct comparison of patterns of tissue
distribution in male and female rats at similar plasma concentrations, even though the elimination kinetics
in the female rat are substantially faster than in male rats (see Section 3.1.4). The highest concentrations
of 14C were observed in blood, liver, and kidney (Figure 3-1). Liver, blood, and kidney accounted for
approximately 22, 22, and 2% of the administered dose of 1 mg/kg in male rats; and 6, 7, and, 3% in
female rats (the sex difference reflected more rapid excretory elimination in females). Although blood,
liver, and kidney concentrations appeared to increase proportionately with increasing dose in male rats, in
female rats, a disproportionately higher concentration in kidney was observed following the 25 mg/kg dose

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

458

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

(Figure 3-1). Concentrations in other tissues ranged from 0.1 to 0.25 of that in liver or kidney;
concentrations in bone and fat were <0.1 of that in liver or kidney. Profound sex differences and dosedependencies in tissue concentrations of PFOA were also observed in rats that received oral doses of
PFOA for 28 days at doses of 3, 10, or 30 mg PFOA/kg/day (Ylinen et al. 1990; Figure 3-2). Mean serum,
kidney, or liver concentrations did not increase proportionally with dose in either sex. Kidney
concentrations exhibited a disproportionate increase as the dose increased from 3 to 10 mg/kg/day, with
little further increase at the 30 mg/kg/day dose. Sex differences in tissue distribution of PFOA in rats are
not explained by sex differences in bioavailability since the differences persist in animals that received
parenteral doses of PFOA (Johnson and Ober 1999b; Vanden Heuvel et al. 1991b, 1991c). The differences
have been attributed to more rapid elimination of PFOA in female rats, compared to male rats (see
Section 3.1.4).

Figure 3-1. Tissue Concentrations of 14C in Male and Female Rats Following a
Single Gavage Dose of [14C]PFOA at 1, 5, or 25 mg/kg*

180
Blood, M

Tissue (ug/g)

160

Liver, M

140

Kidney, M

120

Blood, F

100

Liver, F
Kidney, F

80
60
40
20
0
0

5

10

15

20

Dose (mg/kg)
*Tissue levels are measured at time of maximum concentration in each tissue.
Source: Kemper 2003

***DRAFT FOR PUBLIC COMMENT***

25

30

 PERFLUOROALKYLS

459

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-2. Tissue Concentrations of 14C in Male (Upper Panel) and Female
(Lower Panel) Rats Following Oral Doses of PFOA for 28 Days
at Doses of 3, 10, or 30 mg/kg/day

100

Male rat

Tissue (ug/g)

90
80

Serum
Liver

70
60

Kidney

50
40
30
20
10
0
0

5

10

15

20

25

30

35

Dose (mg/kg/day)

16

Female rat

14

Tissue (ug/g)

12
10
8
6
Serum

4

Liver

2

Kidney

0
0

5

10

15

20

Dose (mg/kg/day)
Source: Ylinen et al. 1990

***DRAFT FOR PUBLIC COMMENT***

25

30

35

 PERFLUOROALKYLS

460

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

A comparison of PFOA disposition in rats, mice, hamsters, and rabbits showed pronounced species and
sex differences (Hundley et al. 2006; Table 3-1). In this study, rats, mice, hamsters, or rabbits received an
oral dose of 10 mg ammonium [14C]PFOA/kg and 14C in tissues was measured at 120 or 168 hours
(rabbits) hours post-dosing. In male rats, the highest concentrations of 14C occurred in blood, liver and
kidney, and all tissues combined accounted for approximately 60% of the dose. However, in female rats,
concentrations of 14C in all tissues were below limits of quantification. In mice, liver concentrations were
similar in males and females, and liver showed the highest concentrations; 14C levels in all tissues
combined were lower in females compared to males. The opposite pattern was evident in hamsters and
rabbits, with males having lower tissue levels than females, although, in common with rats and mice,
blood, liver and kidney had the highest concentrations. Male rats that received a single oral dose of 5 mg
PFOSA/kg had liver PFOSA concentrations that were 3–5 times higher than serum concentrations 1 day
post-dosing (Seacat and Luebker 2000).

Sex differences in elimination that give rise to sex differences in tissue levels following oral exposure to
perfluoroalkyls in rats are not evident in studies conducted with nonhuman primates. Rhesus monkeys
that received 3 or 10 mg ammonium PFOA/kg/day for 90 days had liver concentrations of 48 µg/g (one
male) or 50 µg/g (one female) at the low dose and 45 µg/g (one male) and 72 µg/g (one female) at the
higher dose, with corresponding serum concentrations of 3 and 7 µg/mL, and 9 and 10 µg/mL,
respectively (Griffith and Long 1980). Although limited to only one animal per sex, these results suggest
that liver levels did not increase proportionately with increasing dose. A similar observation was made in
a study of male Cynomolgus monkeys (Butenhoff et al. 2004c). In male monkeys that received daily oral
doses of 3 or 10 mg ammonium PFOA kg/day for 27 weeks, liver PFOA concentrations ranged from
11 to 18 µg/g at the low dose and from 6 to 22 µg/g at the higher dose. Mean serum concentrations
measured after 6 weeks of exposure (which may have represented steady-state concentrations) were
77,000 ng/mL in the low-dose group and 86,000 ng/mL in the higher dose group. In this same study, an
analysis of serum PFOA kinetics following an intravenous dose of PFOA revealed similar elimination
kinetics in males and females (Butenhoff et al. 2004c; see Section 3.1.4). In Cynomolgus monkeys that
received daily oral doses of PFOS (0, 0.03, 0.15, or 0.75 mg PFOS/kg/day) for 26 weeks, liver
concentrations of PFOS and serum concentration were similar in males and females (liver:serum ratios
ranged from 1 to 2) and increased in approximate proportion to the administered dose (Seacat et al. 2002).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

461

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-1. Tissue Distribution and Excretion of 14C-Radioactivity from Both
Sexes of Rats, Mice, Hamsters, and Rabbits Dosed with 14C-Labeled APFOa

Sample

Male

Blood
Liver
Kidneys
Lungs
Heart
Skin
Testes
Muscle
Fat
Brain

23.5
40.0
24.0
8.7
6.4
4.8
3.2
1.9
1.7
0.6

μg Equivalent per g (mL) wet weightb
Rat
Mouse
Hamster
Female Male
Female
Male
Female
<0.1
<0.1
<0.1
<0.1
<0.1
<0.01
–
<0.1
<0.1
<0.1

13.8
43.2
2.9c
1.4c
1.2c
3.5
0.9c
1.1
1.6
0.2c

10.1
45.3
2.2c
1.3c
0.6c
0.2
–
0.5
1.3
0.8c

Rabbit
Male
Female

0.1
0.3
0.2
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1

8.8
7.3
7.1
3.8
2.9
3.4
–
0.9
1.5
0.3

<0.1
0.1
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1

0.1
1.5
0.4
0.1
<0.1
<0.1
–
<0.1
<0.1
<0.1

0.7
90.3
8.2
1.3
0.6
101.1

26.5
45.3
9.3
2.9
2.1
86.1

<0.1
76.8
4.2
No data
0.5
81.6

0.3
87.9
4.6
No data
4.8
97.6

Percent of dose
Tissues
Urine
Feces
Expiration
Cage wash
Percent recovered

59.6
25.6
9.2
3.6
0.6
98.5

0.6
73.9
27.8
1.5
0.8
104.6

73.6
3.4
8.3
5.2
4.9
95.4

50.0
6.7
5.4
4.4
4.9
71.4

aThe

rabbits were sacrificed 168 hours after dosing; all other animals were sacrificed 120 hours after dosing.
μg equivalent calculations were based on the specific activity of 14C-labeled APFO, which was
1.1x106 DPM/mg. The μg equivalent per g wet weight could not accurately be determined below 0.1 μg/g.
cRepresents the μg equivalents for the entire organ.
bThe

APFO = ammonium perfluorooctanoate
Source: Hundley et al. 2006

Bogdanska et al. (2014) examined distribution of 35S in 20 tissues following dietary exposure of adult
male C57/BL6 mice to PFBuS (16 mg/kg/day) for 1–5 days.

35

S was detected in all tissues and

concentrations reached plateau levels after 3 days of exposure. After 5 days, tissue:blood ratios
(excluding stomach and small intestine) were >1 for liver (tissue:blood=1.6), kidney (tissue:blood=1.3),
whole bone (tissue:blood=1.1), and cartilage (tissue:blood=1.1). At all-time points, approximately 90%
of the ingested 35S was recovered in combined blood, liver, bone, skin, and muscle.

Subcellular Distribution. The subcellular distribution of perfluoroalkyls has been examined in rats (Han
et al. 2004, 2005; Kudo et al. 2007; Vanden Heuvel et al. 1992b). Two hours following an oral dose of
25 mg ammonium [14C]PFOA/kg, sex differences were noted in the subcellular distribution of 14C in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

462

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

liver; females had approximately 50% of total 14C in the cytosolic fraction compared to 26% in males
(Han et al. 2005). The distributions to other cell fractions were: nuclear/cell debris fraction, 30%
females, 40% males; lysosomes, 12% females, 14% males; mitochondria, 8% females, 16% males; and
ribosomes, <3% males and females. In kidney, 80 and 70% of the 14C was associated with the cytosolic
fraction in males and females, respectively, 16–22% in the nuclear/cell debris fraction, and the remainder
in lysosome/mitochondria/ribosome fractions. In liver, approximately 55% of cytosolic 14C was bound to
proteins (>6,000 Da) in both males and females, whereas in kidney, 42% of the cytosolic fraction was
bound to protein in males and 17% in females. The subcellular distribution of PFOA is dose-dependent.
In rats, 2 hours following an intravenous dose of 0.041 mg [14C]PFOA/kg, approximately 5% 14C in the
liver was associated with the cytosolic fraction, whereas approximately 45% was in the cytosolic fraction
following a dose of 16.6 mg/kg (Kudo et al. 2007). A small component of tissue-associated PFOA and
PFDeA appeared to be bound covalently to protein. Following an intraperitoneal dose of 9.4 µmol/kg
[14C]PFDeA or [14C]PFOA (4.2 mg/kg), approximately 0.1–0.5% of liver 14C was bound covalently
(i.e., was not removed by repeated extraction with a methanol/ether and ethyl acetate; Vanden Heuvel et
al. 1992b). Covalent binding was detected when cytosolic or microsomal fractions of rat liver were
incubated in vitro with [14C]PFDeA (Vanden Heuvel et al. 1992b).
PFOA binds to rat kidney and urine α2u-globulin; dissociation constants were estimated to be
approximately 1.5 and >2 mM (for a single binding site) for the proteins isolated from rat kidney and
urine, respectively. These values suggest relatively low affinity for the protein, compared to other ligands
that are known to induce hyaline droplet nephropathy (10-4–10-7 M; Han et al. 2004).

Maternal-fetal Transfer. Perfluoroalkyls can be transferred to the fetus during pregnancy (Cariou et al.
2015; Fei et al. 2007; Fisher et al. 2016; Fromme et al. 2010; Glynn et al. 2012; Gützkow et al. 2012;
Hanssen et al. 2010, 2013; Inoue et al. 2004; Kato et al. 2014; Kim et al. 2011, Lee et al. 2013; Lien et al.
2013; Liu et al. 2011; Manzano-Salgado et al. 2015; Midasch et al. 2007; Monroy et al. 2008; Needham
et al. 2011; Ode et al. 2013; Porpora et al. 2013; Yang et al. 2016a, 2016b). Studies that measured
perfluoroalkyls in maternal and fetal cord blood of matched mother-infant pairs found relatively strong
correlations (r>0.8) between maternal and fetal serum (or plasma); however, fetal/maternal serum ratios
vary depending on the structure of the perfluoroalkyl (Table 3-2). With some exceptions, longer
fluoroalkyl chain length and a terminal sulfonate group are associated with lower fetal/maternal ratios
(Glynn et al. 2012; Gützkow et al. 2012; Hanssen et al. 2013; Kim et al. 2011; Liu et al. 2011; Needham
et al. 2011). PFOS was detected in amniotic fluid obtained from amniocentesis (Jensen et al. 2012). The

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

463

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

median concentration in amniotic fluid samples from 300 pregnancies (from the Danish amniotic fluid
sample back) was 1.1 ng/mL.

Table 3-2. Serum (or Plasma) Concentrations in Matched Human Maternal-Infant
Pairs

Study
Glynn et al. 2012

Cariou et al. 2015

Fisher et al. 2016

Fromme et al. 2010

Gützkow et al. 2012

Hanssen et al. 2013

Inoue et al. 2004
Kim et al. 2011

Perfluoro- Perfluoroalkyl
alkyl
chain length
N

Maternal
(ng/mL)

Cord
(ng/mL)

Ratioa r

PFOA
PFOS
PFNA
PFHxS
PFOA
PFOS
PFNA
PFHxS
PFOA
PFOS
PFHxS
PFOA
PFOS
PFNA

7
8
8
6
7
8
8
6
7
8
6
7
8
8

413
413
413
59
89
94
22
315
865
648
53
53
53
53

4
29
0.6
0.62
1.05
3.07
0.43
NR
NR
NR
0.60
2.60
3.50
0.60

1
5
0.1
0.34
0.86
1.11
0.27
NR
NR
NR
0.30
1.70
1.10
<0.4

NR
NR
NR
0.56
0.78
0.38
0.51
0.23
0.28
0.14
0.50
0.65
0.31
ND

0.89
0.86
0.53
0.99
0.83
0.88
0.92
NR
NR
NR
0.89
0.94
0.89
ND

PFHxS
PFOA
PFOS
PFNA
PFDeA
PFUA
PFHxS
PFOA
PFOS
PFNA
PFOSA
PFUA
PFOA
PFHxS
PFOA
PFOS
PFNA
PFDeA
PFUA

6
7
8
8
9
10
6
7
8
8
8
10
7
6
7
8
8
9
10

123
123
123
123
123
123
7
7
7
7
7
7
15
20
20
20
20
20
20

0.34
1.25
5.37
0.40
0.10
0.19
0.26
1.50
10.70
0.89
0.41
0.33
8.90
0.89
1.60
5.60
0.79
0.36
1.60

0.23
1.03
1.78
0.16
0.04
0.06
0.17
1.26
3.93
0.50
0.45
0.16
2.90
0.58
1.10
2.00
0.37
0.01
0.46

0.68
0.82
0.33
0.40
ND
0.32
0.65
0.84
0.37
0.56
1.10
0.48
0.32
0.65
0.69
0.36
0.47
0.03
0.29

0.70
0.82
0.74
0.64
ND
0.67
ND
ND
ND
ND
ND
ND
0.94
ND
ND
ND
ND
ND
ND

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

464

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-2. Serum (or Plasma) Concentrations in Matched Human Maternal-Infant
Pairs

Study
Kato et al. 2014

Lee et al. 2013

Liu et al. 2011

Manzano-Salgado et
al. 2015

Midasch et al. 2007
Monroy et al. 2008

Needham et al. 2011

Ode et al. 2013

Porpora et al. 2013

Perfluoro- Perfluoroalkyl
alkyl
chain length
N
PFHxS
PFOA
PFOS
MePFOSAAcOH
PFNA
PFDeA
PFHS
PFOA
PFOS

6
7
8
8

PFHxS
PFOA
PFOS
PFNA
PFDeA
PFUA
PFDoA
PFHxS
PFOA
PFOS
PFNA
PFOA
PFOS
PFHxS
PFOA
PFOS
PFNA
PFHxS
PFOA
PFOS
PFNA
PFDeA
PFOA
PFOS
PFNA
PFOA
PFOS

Maternal
(ng/mL)

Cord
(ng/mL)

Ratioa r

78
78
78
78

1.20
3.30
8.50
0.20

0.60
3.10
3.50
0.30

0.50
0.89
0.31
1.50

0.89
0.88
0.82
0.92

8
9
6
7
8

78
78
70
70
70

0.66
0.20
1.35
2.73
10.77

0.41
ND
0.67
2.09
3.44

0.62
ND
0.57
0.84
0.35

0.79
ND
ND
ND
ND

6
7
8
8
9
10
11
6
7
8
8
7
8
6
7
8
8
6
7
8
8
9
7
8
8
7
8

50
50
50
50
50
50
50
66
66
66
66
11
11
101
101
101
101
12
12
12
12
12
263
263
263
38
38

0.08
1.66
3.18
0.55
0.58
0.56
0.08
0.84
2.97
6.99
0.85
2.70
12.10
4.05
2.24
16.19
0.80
12.30
4.20
19.70
0.76
0.34
2.30
17.00
0.31
2.90
3.20

0.06
1.50
1.69
0.33
0.24
0.30
ND
0.40
1.90
1.86
0.32
3.40
7.20
5.05
1.94
7.19
0.94
9.10
3.10
6.60
0.37
0.10
2.80
7.40
0.26
1.60
1.40

0.79
0.91
0.53
0.61
0.41
0.53
ND
0.446
0.746
0.299
0.4
1.30
0.60
1.25
0.87
0.44
1.18
0.74
0.72
0.34
0.50
0.29
1.30
0.45
0.93
0.55
0.44

0.59
0.91
0.75
0.82
0.82
0.70
ND
NR
NR
NR
NR
0.42
0.72
ND
0.94
0.91
ND
0.05
0.91
0.82
0.84
0.91
0.74
0.76
0.51
0.70
0.72

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

465

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-2. Serum (or Plasma) Concentrations in Matched Human Maternal-Infant
Pairs

Study
Yang et al. 2016a

Yang et al. 2016b

aRatio

Perfluoro- Perfluoroalkyl
alkyl
chain length
N
PFHxS
PFOA
PFOS
PFNA
PFDeA
PFUA
PFDoA
PFHxS
PFOA
PFOS
PFNA
PFDeA
PFUA
PFDoA

6
7
8
8
9
10
11
6
7
8
8
9
10
11

50
50
50
50
50
50
50
157
157
157
157
157
157
157

Maternal
(ng/mL)
0.064
1.24
2.98
0.55
0.56
0.55
0.085
0.53
1.74
4.23
0.46
0.37
0.38
0.040

Cord
(ng/mL)

Ratioa r

0.033
1.03
1.23
0.35
0.22
0.23
0.058
0.26
1.32
1.52
0.23
0.13
0.14
0.026

0.52
0.83
0.41
0.64
0.39
0.42
0.68
0.43
0.71
0.36
0.49
0.35
0.36
0.61

0.80
0.93
0.88
0.89
0.92
0.88
0.76
0.68
0.81
0.63
0.70
0.65
0.63
0.52

of cord:maternal perfluoroalkyl level

ND = no data (detected but below limit of quantification); NR = not reported; PFDeA = perfluorodecanoic acid;
PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide;
PFTrDA = perfluorotridecanoic acid; PFUA = perfluoroundecanoic acid;

Studies in rats and mice provide further support for maternal-fetal transfer of perfluoroalkyls. Following
gavage administration of 0.1–10 mg/kg/day PFOS to rats during gestation, PFOS was distributed to fetal
serum, liver, and brain, with fetal concentrations increasing with maternal dose (Chang et al. 2009; Lau et
al. 2003; Luebker et al. 2005a, 2005b; Thibodeaux et al. 2003). Levels in fetal serum and liver generally
were similar and higher than in brain. Studies did not report on concentrations of PFOS in other fetal
tissues. Paired fetal-maternal levels of PFOS were examined in rats following exposure (gavage) to
potassium PFOS at doses of 0.1, 0.4, 1.6, or 3.2 mg/kg/day on GDs 0–20 (Luebker et al. 2005b). On
GD 21, fetal:maternal serum ratios were 2.1, 1.7, 1.6, and 1.1 at doses of 0.1, 0.4, 1.6, and 3.2 mg/kg/day,
respectively; these results suggest that fetal:maternal serum ratios varied inversely with dose.
Fetal:maternal liver ratios (0.37–0.44) were similar across the dose range. In mice administered a single
gavage dose of 12.5 mg/kg [35S]PFOS on GD 16, fetal organ:maternal blood ratios of 35S on GD 18 were
2.8 for kidneys, 2.6 for liver, 2.3 for blood, 2.1 for lungs, and 1.2 for brain (Borg et al. 2010).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

466

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Maternal-fetal transfer of PFOA has also been studied in rats and mice (Das et al. 2008; Hinderliter et al.
2005). In rats, PFOA concentrations in amniotic fluid, placenta, and fetus (measured on days 10, 15, or
21 of gestation) increased with increasing maternal oral dose (3, 10, or 10 mg/kg/day, administered daily
beginning on GD 4) (Hinderliter et al. 2005). Fetal plasma concentrations of PFOA measured on GD 21
were approximately 40% of maternal plasma concentration. Following gavage administration of 0.01, 1,
or 5 mg/kg ammonium PFOA on GD 17 in mice, PFOA was detected in amniotic fluid and pup serum,
with dose-dependent increases (Fenton et al. 2009). On PND 1, pup serum PFOA concentrations were
approximately 1.7–2.0-fold greater than levels in maternal serum.

Following administration of ammonium PFBA (35, 175, or 350 mg/kg) to pregnant mice on GDs 0–17,
fetal serum and liver levels of PFBA were determined on PND 1 (Das et al. 2008). The fetal:maternal
serum ratio of PFBA was approximately 0.15 and did not vary with maternal dose. Fetal liver:serum
ratios were 0.44, 0.75, and 0.78 at maternal doses of 35, 175, and 350 mg/kg, respectively. PFHxS was
detected in fetal blood and in the liver of neonates following exposure of dams to potassium PFHxS (0.3,
1, 3, and 10 mg/kg) throughout gestation (Butenhoff et al. 2009a); concentrations in serum and liver
increased with dose.

Maternal-infant Transfer. Perfluoroalkyls can be transferred to nursing infants (Barbarossa et al. 2013;
Cariou et al. 2015; Fromme et al. 2010; Kärrman et al. 2007; Kim et al. 2011; Kuklenyik et al. 2004; Liu
et al. 2011; Tao et al. 2008a, 2008b). Studies that measured perfluoroalkyls in maternal serum (or
plasma) and breast milk in matched mother-infant pairs found highly variable correlations (Table 3-3).
Relatively high correlations have been reported for PFOA (Kärrman et al. 2007; Liu et al. 2011).
Transfer to breast milk appears to be a significant route of elimination of perfluoroalkyls during
breastfeeding. Comparisons of serum concentrations of women who did or did not breastfeed their
infants showed that breastfeeding significantly decreases maternal serum concentrations of PFOA, PFOS,
PFHxS, and PFNA (Bjermo et al. 2013; Brantsaeter et al. 2013; Mondal et al. 2012, 2014; von Ehrenstein
et al. 2009). The decrease was estimated to be 2–3% decrease per month of breastfeeding (Brantsaeter et
al. 2013; Mondal et al. 2012, 2014). Concentrations of perfluoroalkyls in breast milk also decrease with
breastfeeding duration (Tao et al. 2008b; Thomsen et al. 2010). Numerous perfluoroalkyl compounds
(including PFOS, PFOA, PFBuS, PFHxS, PFNA, PFDeA, PFDoA, PFUA, and PFOSA) have been
detected in breast milk samples in women in China, Korea, Japan, Malaysia, Cambodia, India, Korea,
Vietnam, Indonesia, Norway, Philippines, Sweden, and the United States (Forns et al. 2015; Fujii et al.
2012; Kang et al. 2016; Kärrman et al. 2007; Kim et al. 2011; Liu et al. 2010, 2011; Mondal et al. 2014;
So et al. 2006b; Tao et al. 2008a). The mean concentrations for perfluoroalkyls in breast milk collected

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

467

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

from 45 women in Massachusetts were 0.131 ng/mL (range of <0.032.0–617 ng/mL) for PFOS,
0.043.8 ng/mL (<0.0301–0.161 ng/mL) for PFOA, and 0.0145 ng/mL (<0.0120–0.0638 ng/mL) for
PFHxS (Tao et al. 2008b). PFHpA, PFDeA, PFUA, PFDoA, and PFBuS were also detected in the breast
milk; however, ≤4 samples had levels that exceeded the limit of quantitation. Serum concentrations in
breastfed infants can be higher than maternal levels (Fromme et al. 2010; Mondal et al. 2014; Post et al.
2012). Mogensen et al. (2015b) reported that following weaning, significant (<0.0001) decreases were
observed in infant serum concentrations of PFOS, PFOA, and PFHxS.

Table 3-3. Matched Serum (or Plasma) and Breast Milk Concentrations in
Humans

Study

Perfluoroalkyl
Perfluoroalkyl chain length
N

Cariou et al. 2015

PFHxS

6

9

2.28

0.026

0.011

0.36

PFOA

7

10

1.22

0.041

0.034

0.72

PFOS

8

19

3.62

0.040

0.011

0.85

6

12

4.7

0.085

0.020

PFOA

7

12

3.8

0.49

0.120

0.88

PFOS

8

12

20.7

0.20

0.010

0.83

PFOSA

8

12

0.24

0.013

0.070

ND

PFNA

8

12

0.80

0.017

0.010

ND

PFHxS

6

20

0.89

0.007

0.008

NS

PFOA

7

20

1.60

0.041

0.026

NS

PFOS

8

20

5.60

0.061

0.011

PFNA

8

20

0.79

<0.0088 ND

ND

PFDeA

9

20

0.36

<0.018

ND

ND

PFUA

10

20

1.60

<0.024

ND

ND

PFHxS

6

50

0.08

ND

ND

PFOA

7

50

1.66

0.181

0.109

0.77

PFOS

8

50

3.18

0.056

0.018

0.57

PFNA

8

50

0.55

0.026

0.048

0.62

PFDeA

9

50

0.58

0.02

0.034

0.54

PFUA

10

50

0.56

0.026

0.046

0.44

PFDoA

11

50

0.08

ND

ND

ND

PFTrDA

12

50

0.08

ND

ND

ND

Kärrman et al. 2007a PFHxS

Kim et al. 2011

Liu et al. 2011

Serum Milk
(ng/mL) (ng/mL) Ratioa

aMilk

ND

r

ND

0.60

to serum ratio
ND = no data (detected but below limit of quantification); NS = not significantly correlated;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHxS = perfluorohexane sulfonic acid;
PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid;
PFOSA = perfluorooctane sulfonamide; PFTrDA = perfluorotridecanoic acid; PFUA = perfluoroundecanoic acid;

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

468

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Studies conducted in rats and mice provide further support for maternal-infant transfer of perfluoroalkyls
through breast milk (Fenton et al. 2009; Hinderliter et al. 2005; Lau et al. 2003; Luebker et al. 2005a; Yu
et al. 2009b). PFOA concentrations in breast milk of nursing rats increased with increasing maternal oral
dose (3, 10, or 10 mg/kg/day, administered daily beginning on GD 4). Milk concentrations of PFOA
measured on postpartum days 3, 7, 14, or 21 in rats were approximately 0.1 of maternal plasma
concentration. In dams exposed to 0.1, 1, or 5 mg/kg PFOA by gavage on GD 17, a dose-dependent
increase in PFOA concentrations in breast milk was observed on PND 2, with breast milk:serum ratios of
approximately 0.15, 0.38, and 0.25 at 0.1, 1, and 5 mg/kg doses, respectively; milk/serum concentration
ratios for PFOA ranged from 0.15 to 0.56 (Fenton et al. 2009). Following lactational exposure of control
rat pups to PFOS in breast milk of dams treated with dietary PFOS (3.2 mg/kg diet; approximately
equivalent to 0.33 mg/kg/day), pup serum and liver concentrations increased throughout the 35-day
lactation period (Yu et al. 2009b). At PND 35, the pup liver:serum PFOS ratios were 2.55 and 2.43 in
male and female pups, respectively. Results of a cross-foster study show that pups are exposed to PFOS
through breast milk (Luebker et al. 2005a). Postnatal toxicity observed in cross-fostered pups that nursed
from exposed dams provides additional evidence of maternal-infant transfer of PFOS in rats and mice
(see Section 2.17).

Mechanisms of Distribution. Perfluoroalkyls in plasma bind to serum albumin and various other plasma
proteins including gamma-globulin, alpha-globulin, alpha-2-macroglobulin, transferrin, and betalipoproteins (Bischel et al. 2011; Butenhoff et al. 2012c; Chen and Guo 2009; Han et al. 2003, 2005;
Kerstner-Wood et al. 2003; Luo et al. 2012; Ohmori et al. 2003; Salvalaglio et al. 2010; Vanden Heuvel et
al. 1992b; Wu et al. 2009; Ylinen and Auriola 1990; Zhang et al. 2009). The dissociation constant for
albumin-bound PFOA in serum is approximately 0.4 mM (0.38 mM, ±0.04 SD for human serum albumin;
0.36 nM, ±0.08 SD for rat serum albumin) and involves 6–9 biding sites (Han et al. 2003). Noncovalent
binding appears to be at the same sites as fatty acids (Chen and Guo 2009). Interactions between PFOS
and human serum albumin include interaction of PFOS polar sulfonyl groups with albumin hydrophilic
sites and interaction of perfluorinated groups with albumin hydrophobic sites (Luo et al. 2012).

Absorbed perfluoroalkyls distribute from plasma to soft tissues, with the highest extravascular
concentrations achieved in liver. Mechanisms by which perfluoroalkyls enter the liver have not been
elucidated and may involve interactions with organic anion transporters that function in the distribution of
fatty acids or other organic anions (Andersen et al. 2008). PFOA appears to be a substrate for organic
anion transporters in the luminal and basolateral membranes of renal tubular epithelial cells, which

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

469

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

facilitates entry of PFOA into renal tubular cells (Kudo et al. 2002; Nakagawa et al. 2008; Vanden Heuvel
et al. 1992b; Weaver et al. 2010). The subcellular distribution of PFOA is sex- and dose-dependent in
rats (Han et al. 2005; Kudo et al. 2007) and the association with the membrane fraction of liver cells
decreases with increasing dose (Kudo et al. 2007), consistent with limited capacity of membrane proteins
that bind PFOA (e.g., membrane transport proteins). Intracellular PFOA binds to proteins; protein
complexes formed have not been fully characterized. PFOA exhibits a low affinity for binding to rat
kidney and urine alpha-2µ-globulin (dissociation constants 1.5 and >2 mM, respectively) (Han et al.
2004).
3.1.3

Metabolism

Results of available intraperitoneal and in vitro studies suggest that the perfluoroalkyls discussed in this
profile are not metabolized and do not undergo chemical reactions in the body. Studies conducted in male
and female rats did not detect fluorine metabolites in the urine, plasma, or liver following a single
injection of 4–150 mg/kg PFOA or 5–50 mg/kg PFDeA (Goecke et al. 1992; Vanden Heuvel et al. 1991b,
1991c; Ylinen and Auriola 1990). Following a single intraperitoneal dose of approximately 4 mg/kg of
14

C-PFOA, only parent compound was excreted in the urine and bile (Vanden Heuvel et al. 1991c).

PFOA was not metabolized when incubated with microsomal fractions of human or rat intestine, kidney,
or liver homogenates (Kemper and Nabb 2005). Although no studies examining metabolism of
perfluoroalkyls following inhalation, oral, or dermal exposure were identified, metabolism by these
exposure routes is not anticipated.
3.1.4

Excretion

As noted in Section 3.1.3 (Metabolism), there is presently no evidence that perfluoroalkyls undergo
metabolism. Therefore, route-specific differences in excretion patterns are not expected. Selected studies
in which elimination half-lives rates (i.e., t1/2) of perfluoroalkyls have been determined (see summaries in
Table 3-5) show that, in general, elimination t1/2 values are similar following intravenous, intraperitoneal,
and oral exposures. Findings suggest that the route of absorption has no substantial effect of rates of
elimination of absorbed perfluoroalkyls (Butenhoff et al. 2004c; Chang et al. 2008a; Kemper 2003; Kudo
et al. 2002; Ohmori et al. 2003; Vanden Heuvel et al. 1991b; Ylinen et al. 1990). As discussed in this
section, perfluoroalkyls are primarily eliminated in the urine, with smaller amounts eliminated in the feces
and breast milk (see Section 3.1.2; Distribution, Maternal-fetal Transfer). Perfluoroalkyls undergo biliary
excretion, but substantial reabsorption occurs; therefore, biliary excretion does not represent a major
elimination pathway. Perfluoroalkyls do not appear to be eliminated in sweat, as induction of perspiration

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

470

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

by exercise or sauna does not alter clearance of PFOA, PFOA, PFHxA, or PFNA (Genuis et al. 2013).
Perfluoroalkyls are eliminated in menstrual fluid, which appears to contribute to sex differences in serum
elimination rates (Wong et al. 2014, 2015; Zhang et al. 2013).

In humans, absorbed perfluoroalkyls are excreted in urine. Estimates of renal clearance of PFOA and
PFOS from serum in humans ranged from 0.8 to 3.3 mL/day for PFOA (serum concentration range: 5–
16 ng/mL) and 0.1–1.5 mL/day for PFOS (serum concentration range 9–49 ng/mL). These clearance
values were <0.001% of glomerular filtration rate (Harada et al. 2005a). Assuming that 99% of the serum
PFOA and PFOS was bound to albumin (see Section 3.1.2), <0.1% of filtered perfluoroalkyls were
excreted in urine, suggesting extensive reabsorption of filtered PFOA and PFOS in the renal tubule.
Renal clearance was not different in males and females. Mean renal clearances for PFOA were
2.12 mL/day (±0.80 SD, n=5) in males and 1.15 (±0.33 SD, n=5) in five females (mean age 22 and
23 years, respectively). Mean renal clearances for PFOS were 0.66 mL/day (±0.48 SD, n=5) in males and
0.91 (±0.56 SD, n=5) for females. Fujii et al. (2015a) reported renal clearances (mL/day/kg; mean±SD)
for several perfluoroalkyls in humans (three males and five females), including PFOA (0.044±0.01),
PFNA (0.038±0.01), PFDeA (0.015±0.01), PFUA (0.005±0.00), and PFDoA (0.005±0.00). Zhang et al.
(2013) reported renal clearances for several perfluoroalkyls (mL/day/kg) and found that clearance of
PFOS was similar in younger females (≤50 years, 0.050 mL/day/kg, 95% CI 0.037–0.064) and a
combined group of males and older females (grouped together since there were no significant differences
in serum concentrations) (0.037 mL/day/kg, 95% CI 0.026–0.049). However, there appeared to be
differences in renal clearance for PFOA; clearance rates were 0.30 mL/day/kg (95% CI 0.11–0.49) in
young females and 0.77 mL/day/kg (95% CI 0.47–1.1) in the combined older women and all males group.
Urinary excretion of perfluoroalkyls may show sex and age differences (Zhang et al. 2015b). Urinary
excretion of PFOA as a fraction of estimated intake in male adults (n=29) was 31% (p=0.002) higher than
in nonpregnant female adults (n=25). In addition, urinary excretion of PFOS was inversely correlated
with age (r=0.334; p=0.015).

Absorbed PFOA and PFOS are also secreted into bile in humans, but the biliary pathway is not a major
excretory pathway because PFOA and PFOS are reabsorbed after biliary secretion. Estimates of total
body clearance, serum-to-urine clearance, and serum-to-bile clearance of PFOA and PFOS in humans are
presented in Table 3-4 (Harada et al. 2007). Biliary clearances of PFOA and PFOS were 1.06 and
2.98 mL/kg body weight/day, respectively, and greatly exceeded total body clearance (0.150 and
0.106 mL/kg/day) and urinary clearance (0.030 and 0.015 mL/kg/day). Based on these estimates,
approximately 89% of the PFOA secreted into bile and 97% of secreted PFOS was estimated to have been

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

471

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

reabsorbed from the gastrointestinal tract. Fujii et al. (2015a) also reported that biliary clearances of
several perfluoroalkyls (PFOA, PFNA, PFDeA, PFUA, PFDoA) were much higher than total body
clearance in humans, further supporting that perfluoroalkyls excreted in bile undergo extensive
reabsorption.

Table 3-4. Excretory Clearance of PFOA and PFOS in Humans
Parameter

Units

PFOA

PFOS

Serum t1/2a
Total clearanceb
Urinary clearancec
Biliary clearanced
Reabsorbed from bilee

day
mL/kg/day
mL/kg/day
mL/kg/day
%

1,387
0.150
0.030
1.06
89

1,971
0.106
0.150
2.98
97

aEstimates

from Olsen et al. (2005).
where Vd is the volume of distribution (300 mL/kg).
cEstimates from Harada et al. (2005a).
dEstimates from Harada et al. (2007).
e1-(Total-Urinary)/Biliary.
bln(t

1/2)xVd,

PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid
Source: Harada et al. (2007)

Studies conducted in nonhuman primates and rodents provide further evidence that urine is the major
route of excretion of perfluoroalkyls, accounting for >93% of absorbed PFOA and PFOS (Benskin et al.
2009; Butenhoff et al. 2004c; Chang et al. 2008a, 2012; Chengelis et al. 2009; Hanhijarvi et al. 1982,
1987; Hundley et al. 2006; Johnson and Ober 1979, 1980, 1999a, 1999b; Kemper 2003; Kudo et al. 2001;
Olsen et al. 2009; Sundström et al. 2012; Vanden Heuvel et al. 1991b, 1991c). Studies conducted in rats
have shown that PFDeA, PFNA, PFOA, and PFHxA are secreted in bile and undergo extensive
reabsorption from the gastrointestinal tract (Kudo et al. 2001; Vanden Heuvel et al. 1991b, 1991c).
PFOS, PFHxS, and PFBuS are excreted in feces following intravenous dosing of rats, suggesting that
these perfluoroalkyls may also be secreted into bile (Chang et al. 2012; Johnson et al. 1984; Olsen et al.
2009; Sundström et al. 2012). The percentage of the dose excreted in the feces appears to vary with
compound, 8–13% for PFOS, <0.5% for PFHxS, and 0.13–0.36% for PFBuS. Renal clearances of PFOA
from plasma in rats were approximately 0.032 mL/minute/kg body weight in male rats and
0.73 mL/minute/kg in female rats; plasma concentrations of PFOA during these measurements ranged
from approximately 0.8 to 80 µg/mL (Kudo et al. 2002). In the latter study, approximately >95% of
plasma PFOA was bound to high molecular weight protein and the glomerular filtration rate was
approximately 10 mL/minute/kg; therefore, urinary excretion of PFOA was approximately 6% of the rate

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

472

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

of glomerular filtration of PFOA in males and 146% in females. These estimates indicate that net renal
tubular reabsorption of filtered PFOA occurred in male rats, whereas net renal tubular secretion of PFOA
occurred in female rats (i.e., clearance of free PFOA in plasma > glomerular filtration rate). The
pronounced sex difference in renal clearance of PFOA has been attributed to modulation of renal
excretory transport of PFOA by testosterone and estradiol (Kudo et al. 2002; Vanden Heuvel et al. 1992a;
see Section 3.1.5).

Rates of elimination of perfluoroalkyls vary substantially across chemical species and animal species, and
show sex differences and age-dependencies within certain species. Table 3-5 summarizes estimates of the
elimination t1/2 for perfluoroalkyls in humans and experimental animals. In compiling the estimates
presented in Table 3-5, preference was given to the terminal t1/2 when multiple t1/2 values were reported.
The significance of the terminal t1/2 is that it determines the time required for complete elimination of the
perfluoroalkyl as well as the exposure duration required to achieve a steady state. Most of the t1/2 values
in Table 3-5 were estimated from analyses of data on declining serum concentrations of perfluoroalkyls
after a single dose or following cessation of a period of repeated dosing. Estimates of the terminal t1/2
based on serum concentrations can vary with the length of the observation period following the last dose
and with the modeling approach used to estimate the t1/2. Longer observation times are required to
estimate the slowest phases of elimination. As a result, estimates of t1/2 based on observation periods of
1–2 days can be much shorter than estimates for the same perfluoroalkyl based on observation periods of
several weeks. Direct comparisons of t1/2 values should be made with consideration of whether or not the
observation periods were comparable. Differences in estimation methodology can also contribute to
differences in t1/2 values. Values reported in Table 3-5 are based on fitting data to single or multicompartment models, or noncompartmental modeling of the data. While the terminal t1/2 provides a
metric for comparing times required for complete elimination and steady state, it does not always provide
a measure of how rapidly the perfluoroalkyl is cleared from the body. A more useful metric for this is the
systemic clearance (ClS), typically estimated from the absorbed dose (AD) and the area under the serum
concentration curve (AUCS):

ClS =

AD
AUCS

Eq. (3-3)

Equation 3-3 will provide an accurate estimate of systemic clearance following an oral dose if the oral
dose is completely absorbed. Accurate estimation of AUCS also depends on fitness of the underlying

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

473

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

model used to predict serum concentrations. Estimates of systemic clearance based on pharmacokinetics
analyses of serum data from animal studies are presented in Table 3-6.

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Route Dose

Exposure Elimination halfdurationa lifeb

Reference

NA

NA

NA

Olsen et al. 2007a

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

3.8 years (95% CI
3.1–4.4, GM 3.5)
2.8 years (95 %CI
2.4–3.4)
2.4 years (95% CI
2.0–3.0)
2.6 years (SE 0.4,
GM 1.2)
2.1 years (SE 0.3,
GM 1.5)
3.9 years

Human (n=5),
22±0.9, M
Human (n=5), 68±5,
M
Human (n=5), 23±3,
F
Human (n=5), 69±5,
F
Human (n=200)
54±15, M, F
Human (n=643)
Adults, M, F

NA

NA

NA

2.3 years

Harada et al. 2005a

NA

NA

NA

2.6 years

Harada et al. 2005a

NA

NA

NA

3.5 years

Harada et al. 2005a

NA

NA

NA

2.9 years

Harada et al. 2005a

Oral

NA

NA

Bartell et al. 2010

Oral

NA

NA

Human (n=1,029)
Adults, M, F

Oral

NA

NA

Humans (n=17)
Oral
Adults, M, F
PFOS—Human
Human (n=26), adult, NA
M (24) F (2)
NA
Human (n≅1,000),
>12–>80 years, M
NA
Human (n≅1,000),
>12–>80 years, F

NA

NA

2.3 years (95% CI
2.1–2.4)
2.9 years
(<4 years)
(95% CI 2.3–3.8)
10.1 years
(>4 years)
8.5 years
(<9 years)
(95% CI 7.1–10.1)
5.1 years (SD 1.7,
GM 4.8)

NA

NA

Olsen et al. 2007a

NA

NA

5.4 years (95% CI
3.9–6.9, GM 4.8)
4.7 years

NA

NA

4.3 years (95% CI
4.1–4.5)

Wong et al. 2015

PFOA—Human
Human (n=26), adult,
M (24) F (2)
Human (n=20) 15–
50 years M
Human (n=30) 15–
50 years F
Human (n=66),
>50 years, M, F
Human (n=20),
≤50 years, F
Human (n=45) M, F

***DRAFT FOR PUBLIC COMMENT***

Li et al. 2018
Li et al. 2018
Zhang et al. 2013
Zhang et al. 2013
Worley et al. 2017a

Seals et al. 2011

Seals et al. 2011

Costa et al. 2009

Wong et al. 2014

 PERFLUOROALKYLS

474

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Route Dose

Exposure Elimination halfdurationa lifeb

Reference

Human (n=66),
>50 years, M, F
Human (n=20),
≤50 years, F
Human (n=45) M, F

NA

NA

NA

Zhang et al. 2013

NA

NA

NA

NA

NA

NA

27 years (SE 3.1,
GM 18)
6.2 years (SE 0.5,
GM 5.8)
3.3 years

Human (n=20) 15–
50 years M
Human (n=30) 15–
50 years F
Human (n=5),
22±0.9, M
Human (n=5), 68±5,
M
Human (n=5), 23±3,
F
Human (n=5), 69±5,
F
PFHxS—Human
Human (n=26), adult,
M (24), F (2)
Human (n=20),
≤50 years, F
Human (n=20) 15–
50 years M
Human (n=30) 15–
50 years F
Human (n=45) M, F

NA

NA

NA

Li et al. 2018

NA

NA

NA

NA

NA

NA

4.6 years (95% CI
3.7–6.1)
3.1 years (95% CI
2.7–3.7)
4.9 years

NA

NA

NA

7.4 years

Harada et al. 2005a

NA

NA

NA

4.5 years

Harada et al. 2005a

NA

NA

NA

4.6 years

Harada et al. 2005a

NA

NA

NA

Olsen et al. 2007a

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

8.5 years (95% CI
6.4–10.6, GM 7.3)
7.7 years (SE 0.6,
GM 7.1)
7.4 years (95% CI
6.0–9.7)
4.7 years (95% CI
3.9–5.9)
15.5 years

Human (n=66),
>50 years, M, F
PFBA—Human
Human (n=3), adult,
M
Human (n=9), adult,
M (7), F (2)
PFBuS—Human
Human (n=6), adult
M (5), F(1)
PFNA—Human

NA

NA

NA

35 years (SE 3.9,
GM 25)

Zhang et al. 2013

NA

NA

NA

81 hours (SD 41)

Chang et al. 2008b

NA

NA

NA

72 hours (SD 38)

Chang et al. 2008b

665 hours (SD
266)

Olsen et al. 2009

4.3 years (SE 0.5,
GM 3.2)
2.5 years (SE 0.6,
GM 1.7)

Zhang et al. 2013

Human (n=66),
>50 years, M, F
Human (n=20),
≤50 years, F

NA

NA

NA

NA

NA

NA

***DRAFT FOR PUBLIC COMMENT***

Zhang et al. 2013
Worley et al. 2017a

Li et al. 2018
Harada et al. 2005a

Zhang et al. 2013
Li et al. 2018
Li et al. 2018
Worley et al. 2017a

Zhang et al. 2013

 PERFLUOROALKYLS

475

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Route Dose

Exposure Elimination halfdurationa lifeb

Reference

NA

NA

NA

Zhang et al. 2013

NA

NA

NA

12 years (SE 1.5,
GM 7.1)
4.5 years (SE 0.4,
GM 4.0)

NA

NA

NA

Zhang et al. 2013

NA

NA

NA

12 years (SE 2.0,
GM 7.4)
4.5 years (SE 0.5,
GM 4.0)

NA

NA

Zhang et al. 2013

NA

NA

1.2 years (SE 0.2,
GM 0.82)
1.5 years (SE 0.3,
GM 1.0)

10 mg/kg/day

6 months 20.1 days

10 mg/kg

1 day

10 mg/kg

1 day

0.15 mg/kg/day

6 months 170 days

Seacat et al. 2002

0.75 mg/kg/day

6 months 170 days

Seacat et al. 2002

0.15 mg/kg/day

6 months 170 days

Seacat et al. 2002

0.75 mg/kg/day

6 months 170 days

Seacat et al. 2002

2 mg/kg

1 day

132 days (SE 7)

Chang et al. 2012

2 mg/kg

1 day

110 days (SE 15)

Chang et al. 2012

10 mg/kg

1 day

5.3 days (SD 2.5)

Chengelis et al. 2009

10 mg/kg

1 day

2.4 days (SD 1.7)

Chengelis et al. 2009

PFDeA—Human
Human (n=66),
>50 years, M, F
Human (n=20),
≤50 years, F

Zhang et al. 2013

PFUA—Human
Human (n=66),
>50 years, M, F
Human (n=20),
≤50 years, F
PFHpA—Human

Human (n=66),
NA
>50 years, M, F
Human (n=20),
NA
≤50 years, F
PFOA—Nonhuman primate
Cynomolgus
Oral
monkey, adult, M
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
PFOS—Nonhuman primate
Cynomolgus
Oral
monkey, adult, M
Cynomolgus
Oral
monkey, adult, M
Cynomolgus
Oral
monkey, adult, F
Cynomolgus
Oral
monkey, adult, F
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
PFHxA—Nonhuman primate
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F

Zhang et al. 2013

Zhang et al. 2013

Butenhoff et al. 2004c

20.9 days (SD
Butenhoff et al. 2004c
12.5)
32.6 days (SD 8.0) Butenhoff et al. 2004c

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

476

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Exposure Elimination halfdurationa lifeb

Reference

10 mg/kg

1 day

141 days (SE 30.)

Sundström et al. 2012

10 mg/kg

1 day

87 days (SE 27)

Sundström et al. 2012

10 mg/kg

1 day

Chang et al. 2008b

10 mg/kg

1 day

40.3 hours (SD
2.4)
41.0 hours (SD
4.7)

10 mg/kg

1 day

10 mg/kg

1 day

10 mg/kg

1 day

10 mg/kg

1 day

11.4 mg/kg

1 day

Oral

0.1 mg/kg

1 day

Johnson and Ober
1980
202 hours (SD 38) Kemper 2003

Oral

1 mg/kg

1 day

138 hours (SD 32) Kemper 2003

Oral

1 mg/kg

1 day

44 hours

Oral

5 mg/kg

1 day

174 hours (SD 29) Kemper 2003

Oral

25 mg/kg

1 day

157 hours (SD 38) Kemper 2003

IV

1 mg/kg

1 day

185 hours (SD 19) Kemper 2003

IV

1 mg/kg

1 day

39 hours

Oral

0,4 mg/kg

1 day

322 hours (SD 38) Benskin et al. 2009

Oral

0.022 mg/kg/day 12 weeks 218 hours (95% CL De Silva et al. 2009
127–792)
21.5 mg/kg
1 day
136 hours (SD 24) Kudo et al. 2002
20.1 mg/kg
1 day
135 hours (SD 29) Ohmori et al. 2003

Route Dose

PFHxS—Nonhuman primate
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
PFBA—Nonhuman primate
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
PFBuS—Nonhuman primate
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
PFOA—Rat
Rat (CR), adult, M
Oral
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (Wistar), adult, M
Rat (Wistar), adult, M

IV
IV

Chang et al. 2008b

15.0 hours (SD
Chengelis et al. 2009
9.8)
8.0 hours (SD 2.0) Chengelis et al. 2009
95.2 hours
(SE 27.1)
83.2 hours
(SE 41.9)

Olsen et al. 2009
Olsen et al. 2009

115 hours

***DRAFT FOR PUBLIC COMMENT***

Kim et al. 2016b

Kim et al. 2016b

 PERFLUOROALKYLS

477

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Route Dose

Exposure Elimination halfdurationa lifeb

IP

3.9 mg/kg

1 day

IP
Oral

50 mg/kg
0.1 mg/kg

1 day
1 day

216 hours (SE
30.9)
105 hours
3.2 hours (SD 0.9)

Oral

1 mg/kg

1 day

3.5 hours (SD 1.1) Kemper 2003

Oral

1 mg/kg

1 day

3.6 hours

Oral

5 mg/kg

1 day

4.6 hours (SD 0.6) Kemper 2003

Oral

25 mg/kg

1 day

IV

1 mg/kg

1 day

16.2 hours (SD
Kemper 2003
9.9)
2.8 hours (SD 0.5) Kemper 2003

IV

1 mg/kg

1 day

4.6 hours

IV

21.5 mg/kg

1 day

1.9 hours (SD 0.7) Kudo et al. 2002

Rat (Wistar), adult, F IV

20.1 mg/kg

1 day

1.9 hours (SD 0.7) Ohmori et al. 2003

Rat (SpragueIP
Dawley), adult, F
Rat (Wistar), adult, F IP

3.9 mg/kg

1 day

50 mg/kg

1 day

2.9 hours (SE 0.2) Vanden Heuvel et al.
1991c
24 hours
Ylinen et al. 1990

Oral

4.2 mg/kg

1 day

179 hours

Oral

0.27 mg/kg

1 day

809 hours

Oral

Oral

0.023 mg/kg/day 12 weeks 1,968 hours (95%
CL 1.584–2.568)
2 mg/kg
1 day
1,495 hours (SE
50)
2 mg/kg
1 day
635 hours

Oral

15 mg/kg

1 day

Oral

1,707 hours
Chang et al. 2012
(SE 270)
0.023 mg/kg/day 12 weeks 1,992 hours (95% De Silva et al. 2009
CL 1,752–2,280)
2 mg/kg
1 day
919 hours (SE 56) Chang et al. 2012

Oral

2 mg/kg

1 day

564 hours

Oral

15 mg/kg

1 day

989 hours (SE 48) Chang et al. 2012

Rat (SpragueDawley), adult, M
Rat (Wistar), adult, M
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (Wistar), adult, F

PFOS—Rat
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F

Oral

Oral

***DRAFT FOR PUBLIC COMMENT***

Reference
Vanden Heuvel et al.
1991c
Ylinen et al. 1990
Kemper 2003

Kim et al. 2016b

Kim et al. 2016b

Johnson and Ober
1979
Benskin et al. 2009
De Silva et al. 2009
Chang et al. 2012
Kim et al. 2016b

Kim et al. 2016b

 PERFLUOROALKYLS

478

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Route Dose

Exposure Elimination halfdurationa lifeb

Reference

IV

2 mg/kg

1 day

689 hours

Kim et al. 2016b

IV

2 mg/kg

1 day

595 hours

Kim et al. 2016b

Oral

5.0 mg/kg

1 day

125 hours

Seacat and Luebker
2000

IP

4.8 mg/kg

1 day

1,008 hours

IV

25 mg/kg

1 day

Rat (Wistar), adult, F IV

25 mg/kg

1 day

IP

4.8 mg/kg

1 day

958 hours (SD
207)
1,406 hours
(SD 140)
552 hours

Vanden Heuvel et al.
1991b
Ohmori et al. 2003

Oral

0.2 mg/kg

1 day

974 hours

Oral
Oral

0.029 mg/kg/day 12 weeks 1,128 hours (95%
CL 935–1,416)
1, 3, or 10 mg/kg 1 day
734.4 hours

IV

22.6 mg/kg

Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F
PFOSA—Rat
Rat (SpragueDawley), adult, M
PFDeA—Rat
Rat (SpragueDawley), adult, M
Rat (Wistar), adult, M

Rat (SpragueDawley), adult, F
PFNA—Rat
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult M
Rat (Wistar), adult, M

Rat (SpragueOral
Dawley), adult F
Rat (Wistar), adult, F IV

1 day

Ohmori et al. 2003
Vanden Heuvel et al.
1991b
Benskin et al. 2009
De Silva et al. 2009

Tatum-Gibbs et al.
2011
710 hours (SD 55) Ohmori et al. 2003

1, 3, or 10 mg/kg 1 day

33.6 hours

22.6 mg/kg

1 day

58.6 hours (SD
9.8)

PFHpA—Rat
Rat (Wistar), adult, M IV

17.7 mg/kg

1 day

2.4 hours (SD 1.2) Ohmori et al. 2003

Rat (Wistar), adult, F IV

17.7 mg/kg

1 day

1.2 hours (SD 0.2) Ohmori et al. 2003

IV

10 mg/kg

1 day

1.0 hour

Chengelis et al. 2009

Oral

50 mg/kg

1 day

2.2 hours

Chengelis et al. 2009

Oral

150 mg/kg

1 day

2.4 hours

Chengelis et al. 2009

Oral

300 mg/kg

1 day

2.5 hours

Chengelis et al. 2009

IV

10 mg/kg

1 day

0.42 hour

Chengelis et al. 2009

PFHxA—Rat
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F

***DRAFT FOR PUBLIC COMMENT***

Tatum-Gibbs et al.
2011
Ohmori et al. 2003

 PERFLUOROALKYLS

479

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
PFHxS—Rat
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
PFBA—Rat
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F
PFBuS—Rat
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), Rat (SD),
adult, M
Rat (SpragueDawley), adult, M
Rat (SpragueDawley), adult, F
Rat (SpragueDawley), adult, F

Route Dose

Exposure Elimination halfdurationa lifeb

Reference

Oral

50 mg/kg

1 day

2.6 hours

Chengelis et al. 2009

Oral

150 mg/kg

1 day

2.2 hours

Chengelis et al. 2009

Oral

300 mg/kg

1 day

2.1 hours

Chengelis et al. 2009

Oral

0.030 mg/kg

1 day

382 hours

Benskin et al. 2009

Oral

4 mg/kg

1 day

645.6 hours

Kim et al. 2016b

Oral

4 mg/kg

1 day

41.28 hours

Kim et al. 2016b

IV

4 mg/kg

1 day

496.8 hours

Kim et al. 2016b

IV

10 mg/kg

1 day

Sundström et al. 2012

IV

4 mg/kg

1 day

688 hours (SE
14.4)
21.12 hours

IV

10 mg/kg

1 day

39 hours (SE 1.9)

Sundström et al. 2012

Oral

30 mg/kg

1 day

Chang et al. 2008b

IV

30 mg/kg

1 day

Oral

30 mg/kg

1 day

IV

30 mg/kg

1 day

9.22 hours
(SE 0.75)
6.38 hours
(SE 0.53)
1.76 hours
(SE 0.26)
1.03 hours
(SE 0.03)

IV

10 mg/kg

1 day

2.1 hours

Chengelis et al. 2009

IV

30 mg/kg

1 day

4.51 hours
(SE 2,22)

Olsen et al. 2009

Oral

30 mg/kg

1 day

Olsen et al. 2009

IV

10 mg/kg

1 day

4.68 hours
(SE 0.07)
0.64 hours

IV

30 mg/kg

1 day

3.96 hours
(SE 0.21)

Olsen et al. 2009

***DRAFT FOR PUBLIC COMMENT***

Kim et al. 2016b

Chang et al. 2008b
Chang et al. 2008b
Chang et al. 2008b

Chengelis et al. 2009

 PERFLUOROALKYLS

480

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex

Route Dose

Exposure Elimination halfdurationa lifeb

Reference

Rat (SpragueOral
Dawley), adult, F
PFOS—Mouse
Mouse (CD), adult, M Oral

30 mg/kg

1 day

7.42 hours
(SE 0.79)

Olsen et al. 2009

1 mg/kg

1 day

1,027 hours

Chang et al. 2012

Mouse (CD), adult, M Oral

20 mg/kg

1 day

874 hours

Chang et al. 2012

Mouse (CD), adult, F Oral

1 mg/kg

1 day

907 hours

Chang et al. 2012

Mouse (CD), adult, F Oral

20 mg/kg

1 day

731 hours

Chang et al. 2012

1 mg/kg

1 day

732 hours

Sundström et al. 2012

Mouse (CD), adult, M Oral

20 mg/kg

1 day

671 hours

Sundström et al. 2012

Mouse (CD), adult, F Oral

1 mg/kg

1 day

597 hours

Sundström et al. 2012

Mouse (CD), adult, F Oral

20 mg/kg

1 day

643 hours

Sundström et al. 2012

Oral

1 or 10 mg/kg

1 day

Oral

1 or 10 mg/kg

1 day

823.2–
1,653.6 hours
619.2–
1,641.6 hours

Tatum-Gibbs et al.
2011
Tatum-Gibbs et al.
2011

Oral

10 mg/kg

1 day

Oral

30 mg/kg

1 day

Oral

100 mg/kg

1 day

Oral

10 mg/kg

1 day

Oral

30 mg/kg

1 day

Oral

100 mg/kg

1 day

PFHxS—Mouse
Mouse (CD), adult, M Oral

PFNA—Mouse
Mouse (CD-1), adult
M
Mouse (CD-1), adult
F
PFBA—Mouse
Mouse (CD1), adult,
M
Mouse (CD1), adult,
M
Mouse (CD1), adult,
M
Mouse (CD1), adult,
F
Mouse (CD1), adult,
F
Mouse (CD1), adult,
F

13.34 hours
Chang et al. 2008b
(SE 4.55)
16.3 hours (SE 7.2) Chang et al. 2008b
5.22 hours
(SE 2.27)
2.87 hours
(SE 0.30)
3.08 hours
(SE 0.26)
2.79 hours (SE 0.3)

***DRAFT FOR PUBLIC COMMENT***

Chang et al. 2008b
Chang et al. 2008b
Chang et al. 2008b
Chang et al. 2008b

 PERFLUOROALKYLS

481

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-5. Summary Elimination Half-Lives for Perfluoroalkyls
Estimated in Humans and Experimental Animals
Species, age, and
sex
PFOS—Rabbit
Rabbit (New
Zealand), adult, F
aExposure
bReported

Exposure Elimination halfdurationa lifeb

Route Dose
Oral

0.085 mg/kg/day 102 days 87 days (SD 31)

Reference
Tarazona et al. 2016

durations of 1 day indicate that a single dose was administered.
half-lives are arithmetic means for the terminal elimination phase if multiple elimination phases were

observed.
CI = confidence interval; CL = confidence limit; F = female; GM = geometric mean; IP = intraperitoneal;
IV = intravenous; M = male; NA = not applicable; PFBA = perfluorobutyric acid; PFDeA = perfluorodecanoic acid;
PFHpA = perfluoroheptanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide;
SD = standard deviation; SE = standard error

Table 3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in
Experimental Animals

Species, age, and sex

Dose Exposure Systemic clearance
Route (mg/kg) duration (mL/day/kg)a
Reference

PFOA—Nonhuman primate
Cynomolgus monkey,
IV
adult, M
Cynomolgus monkey,
IV
adult, F
PFOS—Nonhuman primate
Cynomolgus monkey,
IV
adult, M
Cynomolgus
IV
monkey, adult, F
PFHxA—Nonhuman primate
Cynomolgus
IV
monkey, adult, M
Cynomolgus
IV
monkey, adult, F
PFHxS—Nonhuman primate
Cynomolgus monkey,
IV
adult, M
Cynomolgus monkey,
IV
adult, F

10

1 day

12.4 (SD 7.4)

Butenhoff et al. 2004c

10

1 day

5.3 (SD 3.3)

Butenhoff et al. 2004c

2

1 day

1.10 (SE 0.06)

Chang et al. 2012

2

1 day

1.65 (SE 0.04)

Chang et al. 2012

10

1 day

569

Chengelis et al. 2009

10

1 day

535

Chengelis et al. 2009

10

1 day

1.3 (SE 0.1)

Sundström et al. 2012

10

1 day

1.9 (SE 0.4)

Sundström et al. 2012

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

482

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in
Experimental Animals

Species, age, and sex

Dose Exposure Systemic clearance
Route (mg/kg) duration (mL/day/kg)a
Reference

PFBA—Nonhuman primate
Cynomolgus monkey,
IV
adult, M
Cynomolgus monkey,
IV
adult, F
PFBuS—Nonhuman primate
Cynomolgus monkey,
IV
adult, M
Cynomolgus monkey,
IV
adult, F
Cynomolgus monkey,
IV
adult, M
Cynomolgus monkey,
IV
adult, F
PFOA—Rat
Rat (Sprague-Dawley),
Oral
adult, M
Rat (Sprague-Dawley),
Oral
adult, M
Rat (Sprague-Dawley),
Oral
adult, M
Rat (Sprague-Dawley),
Oral
adult, M
Rat (Sprague-Dawley),
Oral
adult, M
Rat (Sprague-Dawley),
IV
adult, M
Rat (Sprague-Dawley),
IV
adult, M
Rat (Sprague-Dawley),
Oral
adult, F
Rat (Sprague-Dawley),
Oral
adult, F
Rat (Sprague-Dawley),
Oral
adult, F
Rat (Sprague-Dawley),
Oral
adult, F
Rat (Sprague-Dawley),
Oral
adult, F
Rat (Sprague-Dawley),
IV
adult, F
Rat (Sprague-Dawley),
IV
adult, F

10

1 day

2,371 (SE 293)

Chang et al. 2008a

10

1 day

1,075 (SE 91)

Chang et al. 2008a

10

1 day

159

Chengelis et al. 2009

10

1 day

238

Chengelis et al. 2009

10

1 day

12,264 (SE 3384)

Olsen et al. 2009

10

1 day

8,832 (SE 2880)

Olsen et al. 2009

0.1

1 day

23.1 (SD 5.8)

Kemper 2003

1

1 day

20.9 (SD 3.8)

Kemper 2003

1

1 day

40.40 (SD 2.29)

Kim et al. 2016b

5

1 day

20.4 (SD 5.0)

Kemper 2003

25

1 day

27.1 (SD 7.4)

Kemper 2003

1

1 day

21.5 (SD 2.0)

Kemper 2003

1

1 day

47.39 (SD 3.40)

Kim et al. 2016b

0.1

1 day

778 (SD 144)

Kemper 2003

1

1 day

655 (SD 173)

Kemper 2003

1

1 day

645.12 (SD 43.44)

Kim et al. 2016b

5

1 day

1,164 (SD 118)

Kemper 2003

25

1 day

842 (SD 166)

Kemper 2003

1

1 day

816 (SD 221)

Kemper 2003

1

1 day

612.84 (SD 32.54)

Kim et al. 2016b

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

483

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in
Experimental Animals

Species, age, and sex

Dose Exposure Systemic clearance
Route (mg/kg) duration (mL/day/kg)a
Reference

Rat (Wistar), adult, M

IV

21.5

1 day

50.4 (SD 14.4)

Kudo et al. 2002

Rat (Wistar), adult, F

IV

21.5

1 day

2,233 (SD 805)

Kudo et al. 2002

Rat (Wistar), adult, M

IV

20.1

1 day

135 (SD 29)

Ohmori et al. 2003

Rat (Wistar), adult, F

IV

20.1

1 day

2,233 (SD 805)

Ohmori et al. 2003

Oral

2

1 day

7.33 (SD 0.55)

Kim et al. 2016b

Oral

2

1 day

11.3 (SE 0.56)

Chang et al. 2012

Oral

15

1 day

4.9 (SE 0.52)

Chang et al. 2012

IV

2

1 day

9.24 (SD 0.37)

Kim et al. 2016b

Oral

2

1 day

8.52 (SD 0.37)

Kim et al. 2016b

Oral

2

1 day

22.2 (SE 0.28)

Chang et al. 2012

Oral

15

1 day

5.4 (SE 20)

Chang et al. 2012

IV

2

1 day

9.82 (SD 0.21)

Kim et al. 2016b

IV

25

1 day

207 (SD 0.054)

Ohmori et al. 2003

Rat (Wistar), adult, F

IV

25

1 day

140 (SD 0.008)

Ohmori et al. 2003

PFNA—Rat
Rat (Wistar), adult, M

IV

22.6

1 day

6.9 (SD 0.6)

Ohmori et al. 2003

Rat (Wistar), adult, F

IV

22.6

1 day

106 (SD 31)

Ohmori et al. 2003

PFHpA—Rat
Rat (Wistar), adult, M

IV

17.7

1 day

1,604 (SD 558)

Ohmori et al. 2003

Rat (Wistar), adult, F

IV

17.7

1 day

3,071 (SD 781)

Ohmori et al. 2003

IV

10

1 day

2,784

Chengelis et al. 2009

IV

10

1 day

18,600

Chengelis et al. 2009

Oral

4

1 day

7.15 (SD 0.06)

Kim et al. 2016b

IV

4

1 day

9.01 (SD 0.05)

Kim et al. 2016b

PFOS—Rat
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, F
PFDeA—Rat
Rat (Wistar), adult, M

PFHxA—Rat
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
PFHxS—Rat
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, M

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

484

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in
Experimental Animals

Species, age, and sex
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, F
PFBA—Rat
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
PFBuS—Rat
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
Rat (Sprague-Dawley),
adult, M
Rat (Sprague-Dawley),
adult, F
PFOA—Mouse
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
PFOS—Mouse
Mouse (CD), adult, M
Mouse (CD), adult, M
Mouse (CD), adult, F
Mouse (CD), adult, F
PFHxS—Mouse
Mouse (CD), adult, M

Dose Exposure Systemic clearance
Route (mg/kg) duration (mL/day/kg)a
Reference
IV

10

1 day

6.7 (SE 0.06)

Sundström et al. 2012

Oral

4

1 day

124.83 (SD 3.40)

Kim et al. 2016b

IV

4

1 day

227.93 (SD 6.73)

Kim et al. 2016b

IV

10

1 day

53.4 (SE 4.38)

Sundström et al. 2012

IV

30

1 day

851 (SE 61)

Chang et al. 2008a

IV

30

1 day

2,949 (SE 59)

Chang et al. 2008a

Oral

30

1 day

494 (SE 29)

Chang et al. 2008a

Oral

30

1 day

1,527 (SE 145)

Chang et al. 2008a

IV

10

1 day

946

Chengelis et al. 2009

IV

10

1 day

7,464

Chengelis et al. 2009

IV

30

1 day

2,856 (SE 816)

Olsen et al. 2009

IV

30

1 day

11,265 (SE 960)

Olsen et al. 2009

Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b

IV

0.13

1 day

14.2 (SD 8.4)

IV

0.13

1 day

11.8 (SD 6.1)

Oral

1.3

1 day

13.1 (SD 7.4)

Oral

1.3

1 day

9.0 (SD 1.9)

Oral
Oral
Oral
Oral

1
20
1
20

1 day
1 day
1 day
1 day

4.7
4.7
5.0
6.0

Chang et al. 2012
Chang et al. 2012
Chang et al. 2012
Chang et al. 2012

Oral

1

1 day

2.9

Sundström et al. 2012

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

485

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in
Experimental Animals

Species, age, and sex
Mouse (CD), adult, M
Mouse (CD), adult, F
Mouse (CD), adult, F
PFNA—Mouse
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
PFDeA—Mouse
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
PFUA—Mouse
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
PFDoA—Mouse
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F
Mouse (FVB/NJcl),
adult, M
Mouse (FVB/NJcl),
adult, F

Dose Exposure Systemic clearance
Route (mg/kg) duration (mL/day/kg)a
Reference
Oral
Oral
Oral

20
1
20

1 day
1 day
1 day

4.8
2.7
3.8

Sundström et al. 2012
Sundström et al. 2012
Sundström et al. 2012
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b

IV

0.14

1 day

3.9 (SD 1.9)

IV

0.14

1 day

5.1 (SD 2.3)

Oral

1.4

1 day

4.0 (SD 1.7)

Oral

1.4

1 day

2.4 (SD 1.0)

IV

0.16

1 day

2.2 (SD 0.9)

IV

0.16

1 day

2.8 (SD 1.2)

Oral

1.6

1 day

3.9 (SD 1.8)

Oral

1.6

1 day

2.2 (SD 1.1)

IV

0.17

1 day

2.8 (SD 1.0)

IV

0.17

1 day

3.4 (SD 1.5)

Oral

1.7

1 day

5.7 (SD 2.6)

Oral

1.7

1 day

3.1 (SD 1.7)

IV

0.19

1 day

4.4 (SD 1.6)

IV

0.19

1 day

4.8 (SD 2.4)

Oral

1.9

1 day

9.4 (SD 4.1)

Oral

1.9

1 day

5.2 (SD 3.2)

***DRAFT FOR PUBLIC COMMENT***

Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b
Fujii et al. 2015a,
2015b

 PERFLUOROALKYLS

486

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-6. Summary Systemic Clearance for Perfluoroalkyls Estimated in
Experimental Animals

Species, age, and sex

Dose Exposure Systemic clearance
Route (mg/kg) duration (mL/day/kg)a
Reference

PFBA—Mouse
Mouse (CD1), adult, M

Oral

10

1 day

280 (SE 72)

Chang et al. 2008b

Mouse (CD1), adult, M

Oral

30

1 day

296 (SE 640)

Chang et al. 2008b

Mouse (CD1), adult, M

Oral

100

1 day

784 (SE 112)

Chang et al. 2008b

Mouse (CD1), adult, F

Oral

10

1 day

564 (SE 24)

Chang et al. 2008b

Mouse (CD1), adult, F

Oral

30

1 day

696 (SE 32)

Chang et al. 2008b

Mouse (CD1), adult, F

Oral

100

1 day

1,336 (SE 64)

Chang et al. 2008b

aAs

reported in units of mL/day/kg or converted from mL/hour (x24), mL/hour (x24/body weight) or mL/minute
(x60x24).

CI = confidence interval; F = female; IV = intravenous; M = male; PFBA = perfluorobutyric acid;
PFBuS = perfluorobutane sulfonic acid; PFDeA = perfluorodecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxA = perfluorohexanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; SD = standard deviation; SE = standard error

Elimination of Perfluoroalkyls in Humans. Elimination t1/2 values for PFOA, PFOS, PFHxS, PFBA,
and PFBuS have been estimated in humans (Bartell et al. 2010; Costa et al. 2009; Chang et al. 2008a;
Glynn et al. 2012; Harada et al. 2005a; Li et al. 2018; Olsen et al. 2007a, 2009; Seals et al. 2011;
Spliethoff et al. 2008; Yeung et al. 2013; Wong et al. 2014, 2015; Worley et al. 2017a; Zhang et al. 2013).
Estimates in humans are based on measurements of the decline in serum perfluoroalkyl concentrations
following cessation or an abrupt decrease in exposure, or on measurements of renal plasma clearance
from serum in a general population sample from Japan (Harada et al. 2005a). The latter clearance
estimates were converted to t1/2 values, for display in Table 3-5 as follows (Equations 3-4 and 3-5):

ke =

Cl
V

t1 / 2 =

ln(2)
ke

Eq. (3-4)
Eq. (3-5)

where ke is the elimination rate constant (e.g., day-1), Cl is the renal plasma clearance (e.g., mL
plasma/day/kg), and V is the plasma volume (L/kg), which is assumed to be 4.3% of body weight (ICRP
1981). In general, these studies show that longer chain length is associated with slower elimination rates.
For example, the elimination t1/2 for PFBA was estimated to be 70–80 hours (Chang et al. 2008a),
whereas the t1/2 values for PFHxS, PFOS, and PFOA range from 2 to 35 years (Bartell et al. 2010; Harada
et al. 2005a; Li et al. 2018; Olsen et al. 2007a; Seals et al. 2011; Worley et al. 2017a; Zhang et al. 2013).
Longer t1/2 values for PFOA have been reported with longer monitoring follow-up times, which allow the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

487

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

detection of slower elimination phases of multiphasic elimination kinetics (Seals et al. 2011).
Perfluoroalkyl sulfonates are eliminated more slowly in humans than corresponding carboxylates of the
same chain length (Zhang et al. 2013). Analytical methods typically used to measure serum
perfluoroalkyls do not discriminate between linear and branched isomers and, as a result, these studies
estimate elimination rates for the isomer mixture. A study that compared elimination rates of isomers of
PFOA found that linear isomers tend to be eliminated more slowly than branched isomers (Zhang et al.
2013), consistent with results of studies conducted in rats (Benskin et al. 2009; De Silva et al. 2009).

An analysis of serum PFOS data from NHANES indicated that t1/2 in females may be shorter (4.3 years)
compared to males (4.7 years; Wong et al. 2014, 2015). The NHANES data are cross-sectional and,
therefore, the estimates of t1/2 required fitting the data to age patterns of PFOS intake. An improved fit to
the data for females was achieved when estimated losses of PFOS in menstrual fluids were considered,
suggesting that menstrual loss of PFOS may account for some, but not all, of the sex difference in the
elimination rate (Verner and Longnecker 2015; Wong et al. 2015). Li et al. (2018) also found apparent
sex differences in PFOS elimination in male and female residents in Sweden exposed to contaminated
drinking water. The estimated t1/2 for PFOS were 4.6 years in males and 3.1 years in females. Zhang et
al. (2013) estimated serum t1/2 for various age and sex strata in a population of 86 individuals. Serum t1/2
for PFOS was lower for PFOS in younger females (≤50 years, t1/2 = 6.2 years±0.3 SE, n=66) compared to
males and older females (t1/2 =27 years±3.1 SE, n=20). Zhang et al. (2013) attributed the difference in
serum t1/2 to clearance in menstrual fluids. However, the estimated serum t1/2 of 27 years is much higher
than values calculated from other studies; Zhang et al. (2013) noted that the serum t1/2 should be
considered as an upper limit estimate. Estimated t1/2 for PFOA was not different in younger females
(2.1±0.3 SE, n=20) compared to males and older females (2.6±0.4 SE, n=66). Declines in serum PFOA
concentrations were observed in populations following initiation of activated carbon filtration of public
water supplies that had been contaminated with PFOA (Bartell et al. 2010). The estimated mean serum
t1/2 for a group of 200 adults followed for 1 year after filtration was initiated was 2.3 years (95% CI 2.1–
2.4). Elimination rates were not different in males and females. Serum PFOA concentration ranged from
16 to 1,200 ng/mL. A larger follow-up study measured serum PFOA concentrations in two populations
of former residents (n=1,672) of the same water districts (Seals et al. 2011). In one population (n=643),
the serum t1/2 increased with increasing elapsed time since leaving the water district. The t1/2 values were
2.9 years (95% CI 2.3–3.8) for elapsed time of <4 years and 10.1 years for elapsed time of >4 years. In a
second population with an elapsed time since residence of <9 years, the t1/2 was 8.5 years (95% CI 7.1–
10.1). Elimination rates (based on the annual percent decrease in serum concentrations) were faster in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

488

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

males (27%) compared to females (18%) for the first 4 years post-exposure; however, no difference was
evident between sexes when elapsed time from exposure was >4 years.

Bartell (2012) and Russell et al. (2015) point out that most studies examining PFOA elimination halflives fail to account for ongoing background exposure, which could result in an overestimation of
elimination half-lives. Bartell (2012) estimated that the bias from background exposure could resulted in
1–26% overestimation of calculated PFOA half-lives and that greater overestimations can occur for halflives based on longer follow-up times. Russell et al. (2015) estimated that the bias was greatest in
populations with serum PFOA levels closest to background levels. In a re-analysis of the Olsen et al.
(2007) occupational exposure data, Russell et al. (2015) estimated that overestimation was approximately
1.2% in workers with initial serum concentrations >500 ng/mL (100 times higher than NHANES general
population data) and 13% for workers with lower initial serum PFOA levels. Restricting the elimination
half-life calculation to workers with initial serum PFOA levels of >500 ng/mL would result in a half-life
of 3.0 years (Russell et al. 2015), compared to 3.8 years calculated for the whole cohort (Olsen et al.
2007).

Analysis of kinetics of serum PFOS concentrations in retired U.S. fluorochemical production workers
(24 males, 2 females) yielded a mean serum elimination t1/2 estimate of 5.4 years (95% CI 3.9–6.9;
geometric mean: 4.8 years, 95% CI 4.0–5.8) in subjects whose serum PFOS concentrations ranged from
37 to 3,490 ng/mL (Olsen et al. 2007a). Estimates for the two females in the same study were 4.9 and
6.8 years. Estimates based on renal clearance of PFOS from serum in subjects from the general
population of Japan ranged from 2.9 to 7.4 years; these subjects had serum PFOS concentrations that
ranged from 4 to 49 ng/mL (Harada et al. 2005a). Estimates in males (7.4, 2.9 years) were similar to
females (4.5, 4.6 years). This same study measured serum PFHxS concentrations in retired U.S.
fluorochemical production workers (24 males, 2 females) and yielded a mean estimate of 8.5 years (95%
CI 6.4–10.6; geometric mean: 7.3 years, 95% CI 5.8–9.2) for the serum elimination t1/2 in subjects whose
serum PFHxS concentrations ranged from 10 to 1,295 ng/mL (Olsen et al. 2007a). Estimates for the two
females in the same study were 12.2 and 13.3 years.

The elimination rate of PFBA was estimated in fluorochemical workers who may have been exposed to
various PFBA precursors (Chang et al. 2008a). In three male workers, the estimated mean t1/2 based on
serum PFBA kinetics was 81 hours (±41 SD). In a larger study of nine workers (seven males, two
females), the mean t1/2 was 72 hours (±38 SD). Estimates for the two female subjects were 56 and
118 hours. The combined mean value for the 12 estimates was 75 hours (±38 SD). Olsen et al. (2009)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

489

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

estimated serum t1/2 of PFBuS in six fluorochemical workers. The mean t1/2 was 27.4 days (±11.1 SD).
The group included a single female whose t1/2 was 45.7 days. Based on these observations, PFBA and
PFBuS are eliminated substantially faster in humans than perfluoroalkyls having longer carbon chain
lengths (e.g., PFHxS, PFOA, PFOS).

Temporal trends in perfluoroalkyl serum concentrations have also been used to estimate population
halving times (Glynn et al. 2012; Olsen et al. 2012; Spleithoff et al. 2008; Yeung et al. 2013). Population
halving times are influenced by temporal trends in intakes and may therefore not accurately reflect
clearance. Population halving times for PFOS ranged from 4 to 5 years (Olsen et al. 2012; Spleithoff et
al. 2008; Yeung et al. 2013). Glynn et al. (2012) monitored serum perfluoroalkyls in a population of
pregnant women (n=413) in Sweden over the period 1996–2010. Halving times were 22 years (95%
CI 16–38) for PFOA and 8.2 years (95% CI 6.3–12) for PFOS.

Elimination of Perfluoroalkyls in Nonhuman Primates. Elimination t1/2 values and systemic clearances
for PFOA, PFOS, PFHxA, PFHxS, PFBA, and PFBuS have been estimated in Cynomolgus monkeys
(Buttenoff et al. 2004c; Chang et al. 2012; Chengelis et al. 2009; Olsen et al. 2009; Seacat et al. 2002;
Sundström et al. 2012). Estimated terminal t1/2 values were 20–30 days for PFOA, 100–170 days for
PFOS, 90–140 days for PFHxS, 40 hours for PFBA and 8–95 hours for PFBuS. Elimination of
perfluoroalkyls in monkeys is multiphasic and, as a result, estimates of the terminal t1/2 can vary with the
duration of the observation period and assumptions made in modeling elimination kinetics (Chang et al.
2012; Chengelis et al. 2009; Olsen et al. 2009; Sundström et al. 2012). For example, the t1/2 values for
PFBuS were 8 and 15 hours in female and male monkeys, respectively, when monkeys were monitored
for 48 hours following a single intravenous dose (Chengelis et al. 2009), whereas the t1/2 values were
95 and 83 hours in male and female monkeys, respectively, when the monitoring period was extended to
14 days and a three-compartment model was used to estimate the terminal t1/2 (Olsen et al. 2009). Studies
in monkeys confirm general trends observed in humans that perfluoroalkyl sulfonates are more slowly
eliminated than perfluoroalkyl carboxylates and that elimination of longer-chain perfluoroalkyls occurs
more slowly than short-chain perfluoroalkyls. Systemic clearances were lower for PFOS, PFHxS, and
PFBuS compared to the corresponding carboxylates, PFOA, PFHxA, and PFBA (Table 3-6). Systemic
clearances were similar in male and female monkeys (Table 3-6).

Elimination of Perfluoroalkyls in Rats. Elimination t1/2 values and systemic clearances for PFOA,
PFOS, PFOSA, PFDeA, PFNA, PFHpA, PFHxA, PFHxS, PFBA, and PFBuS have been estimated in rats
(Benskin et al. 2009; Chang et al. 2008b, 2012; Chengelis et al. 2009; De Silva et al. 2009; Johnson and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

490

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Ober 1979; Kemper 2003; Kim et al. 2016b; Kudo et al. 2002; Ohmori et al. 2003; Olsen et al. 2009;
Seacat and Luebker 2000; Sundström et al. 2012; Vanden Heuvel et al. 1991b, 1991c; Ylinen et al. 1990).
Consistent with observations made in humans and Cynomolgus monkeys, perfluoroalkyl sulfonates are
more slowly eliminated than perfluoroalkyl carboxylates and short-chain perfluoroalkyls (e.g., PFBA,
PFBuS) are eliminated faster in rats than long-chain perfluoroalkyls (e.g., PFOA, PFOS, PFHxA,
PFHxS); Tables 3-5 and 3-6. Linear PFOA isomers tend to be eliminated more slowly than branched
isomers (Benskin et al. 2009; De Silva et al. 2009).

Elimination of perfluoroalkyls exhibits pronounced sex differences in rats, with faster elimination in
females than in males (Benskin et al. 2009; Chang et al. 2008b; Chengelis et al. 2009; Kemper 2003; Kim
et al. 2016b; Kudo et al. 2002; Ohmori et al. 2003; Sundström et al. 2012; Tatim-Gibbs et al. 2011;
Vanden Heuvel et al. 1991c; Ylinen et al. 1990). Estimates of systemic clearance for PFOA in male rats
ranged from 20 to 50 mL/day/kg, whereas estimates for female rats ranged from 600 to 2,200 mL/day/kg
(Kemper 2003; Kudo et al. 2002; Ohmori et al. 2003). Systemic clearances of PFOA, PFOS, PFNA,
PFHxA, PFHxS, PFBA, and PFBuS are also higher in female rats compared to male rats (Table 3-6).
Pronounced sex difference in elimination rates in rats (faster elimination in females) was observed in rats
following 30-minute nose-only exposures to aerosols (MMAD=1.9–2.1 µm) of 1–25 mg ammonium
PFOA/m3 (Hinderliter et al. 2006a). Plasma PFOA concentrations were not detectable 12 hours after
exposure of female rats, and were approximately 90% of peak plasma concentrations 24 hours after the
exposure in male rats. The slower elimination of PFOA in male rats resulted in steady-state plasma
concentrations within 3 weeks of repeated exposures (6 hours/day, 5 days/week) in male rats, whereas in
female rats, daily periodic oscillations of plasma concentrations from peak to below detection occurred on
each day of exposure. Steady-state plasma concentrations in male rats were approximately 10 times that
of daily peak concentrations in female rats.

Pronounced dose dependence appears in the t1/2 estimates for PFOA in female rats. With increasing dose,
plasma elimination kinetics in female rats converts from monophasic to biphasic. Following an oral dose
of PFOA of 0.1, 1, 5, or 25 mg/kg, the terminal t1/2 values in female rats were 3.2, 3.5, 4.6, or 16.2 hours,
respectively; no apparent dose dependence was observed in male rats over the same dose range (Kemper
2003). Dose-dependent elimination of PFOA has been attributed to a capacity-limited renal tubular
secretion of PFOA in female rats (see discussion below on Mechanisms of Excretion). The divergence in
elimination kinetics between male and female rats appears to be age-dependent, with faster elimination
becoming evident in female rats after 30 days of age, consistent with the timing of sexual maturation and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

491

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

involvement of sex hormones in the modulation of the renal excretion of PFOA in rats (Hinderliter et al.
2006b).

Elimination of Perfluoroalkyls in Mice. Elimination t1/2 values and systemic clearances for PFOS,
PFHxS, and PFBA have been estimated in mice (Chang et al. 2008a, 2012; Sundström et al. 2012).
Consistent with studies conducted in rats and monkeys, PFBA is eliminated more rapidly in mice than
PFOS and PFHxS. Systemic clearances ranged from 5 to 6 mL/day/kg for PFOS (Chang et al. 2012),
from 3 to 5 mL/day/kg for PFHxS (Sundström et al. 2012), and from 300 to 1,300 mL/day/kg for PFBA
(Chang et al. 2008a). Sex differences in elimination in mice were observed for PFBA, but not PFOS or
PFHxS. Systemic clearances of PFBA in female mice were approximately 2 times that of males (Chang
et al. 2008a). Systemic clearance of PFBA in male and female mice appeared to be dependent on dose.
Systemic clearance following a single oral dose of 100 mg PFBA/kg was approximately 2 times higher
than the systemic clearance following a dose of 10 or 30 mg PFBA/kg. Possible explanations for the
apparent dependence of clearance on dose are dose-dependent bioavailability or that the one-compartment
model used to estimate elimination rates and serum AUC did not adequately fit the serum kinetics
observed at the higher dose (Chang et al. 2008a). The latter could occur if renal tubular reabsorption of
PFBA or plasma protein binding of PFBA is saturable in mice. Systemic clearance rates for PFOA were
similar in male mice (13.1 mL/kg/day) and in female mice (9.0 mL/kg/day) (Fuji et al. 2015a, 2015b).

Elimination of Perfluoroalkyls in Other Species. Sex differences in elimination of PFOA have also been
observed in hamsters; unlike the rat, male hamsters excreted absorbed PFOA more rapidly than female
hamsters. Following a single gavage dose of 10 mg/kg as ammonium [14C]PFOA, cumulative excretion
of 14C in urine at 24 hours post-dosing was 96.4% of the dose in female rats and 8.7% in male rats; 24.6%
and 84.5% in female and male hamsters, respectively; 4.1% in male and female mice; and 90.5 and 80.2%
in female and male rabbits, respectively (Hundley et al. 2006).

Mechanisms of Excretion. Urinary excretion of perfluoroalkyls involves glomerular filtration and renal
tubular secretion and reabsorption (for PFOA, see Harada et al. 2005a; Kudo et al. 2002; Ohmori et al.
2003). Glomerular filtration of PFOA is limited by extensive binding of PFOA to albumin and other high
molecular weight proteins in plasma (Han et al. 2003, 2005; Ohmori et al. 2003; Kerstner-Wood et al.
2003; Vanden Heuvel et al. 1992a, 1992b; Ylinen and Auriola 1990). Elimination of PFOA and other
perfluoroalkyls shows pronounced sex differences in rats, with slower elimination in males for PFOA,
PFOS, PFNA, PFHxA, PFHxS, PFBA, and PFBuS (Chang et al. 2008a, 2012; Chengelis et al. 2009;
Kemper 2003; Kudo et al. 2002; Ohmori et al. 2003; Sundström et al. 2012). The sex difference in PFOA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

492

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

elimination in rats is dependent on testosterone (Hinderliter et al. 2006b; Kudo et al. 2002; Vanden
Heuvel et al. 1992a). The significantly slower elimination of PFOA in adult male rats compared to
female rats has been attributed to sex hormone modulation of organic anion transporters in kidney. At
similar doses administered to male and female rats, PFOA undergoes net tubular reabsorption in male rats
(i.e., urinary excretion rate < rate of glomerular filtration of PFOA) and net tubular secretion in female
rats (i.e., urinary excretion rate > rate of glomerular filtration of PFOA) (Harada et al. 2005a; Kudo et al.
2002; Ohmori et al. 2003). In rats, several transporters have been shown to have affinity for C7–C9
perfluoroalkyl carboxylates. The transporters, OAT1 and OAT3, located on the basolateral membrane of
the renal proximal tubule, appear to participate in secretion of C7–C9 perfluoroalkyl carboxylates into the
tubular fluid (Nakagawa et al. 2008; Weaver et al. 2010). The transporters, OATP1a1 (rat), OAT4
(human), and URAT1 (human), located on the apical membrane, appear to mediate reabsorption of C8–
C10 perfluoroalkyl carboxylates from the tubular fluid (Katakura et al. 2007; Nakagawa et al. 2009;
Weaver et al. 2010; Yang et al. 2009, 2010). In rats and mice, expression of OAT1, OAT3, and
OATP1a1 is controlled by male sex hormones and shows higher activities in males (Buist and Klaassen
2004; Gotoh et al. 2002; Kobayashi et al. 2002; Li et al. 2002; Lu et al. 1996; Lubojevic et al. 2004). The
slower elimination of PFOA (and other long-chain perfluoroalkyl carboxylates) in male rats has been
attributed to OATP1a1 (Weaver et al. 2010; Yang et al. 2009). Higher activity of OATP1a1 in male rats
results in higher reabsorptive transport and lower rates of urinary excretion. Affinities of OATP1a1 (rat),
OAT4 (human), and URAT1 (human) are highest for C7–C10 perfluoroalkyl carboxylates (Weaver et al.
2010; Yang et al. 2009, 2010). Affinity of rat OATP1a1 is strongly correlated with total clearance in rats
(r2=0.98; Yang et al. 2009).

Although sex differences for elimination of perfluoroalkyls have been detected in laboratory animals,
human monitoring studies have not consistently detected sex differences in elimination t1/2 of
perfluoroalkyls; this may reflect limitations in the studies, including numbers and age of subjects (Bartell
et al. 2010; Seals et al. 2011; Wong et al. 2014, 2015; Zhang et al. 2013). Menstruation may contribute to
faster elimination of PFOS in women (Wong et al. 2014, 2015; Zhang et al. 2013). The effect of
menstruation or other variables related to menstruation appear to contribute to faster elimination in
younger (≤50 years) women compared to men and older women (Zhang et al. 2013). Two studies have
found evidence for elimination of PFOS being affected by menstruation (Wong et al. 2014, 2015; Zhang
et al. 2013). The estimated t1/2 for PFOA was not different in younger females compared to males and
older females. Mechanisms by which menstruation could affect PFOS clearance are not understood.
Bulk elimination of blood would be expected to affect serum clearance of both PFOS and PFOA;
therefore, other mechanisms must contribute that discriminate between perfluoroalkyl species. A better

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

493

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

metric than serum t1/2 for evaluating sex differences in elimination for this would be systemic or renal
clearance of the perfluoroalkyl compound. Harada et al. (2005a) measured renal clearance in a small
sample of young adults (five males and five females, age 22–23 years) and found that renal clearance was
not different in males and females. Zhang et al. (2013) estimated renal clearance of PFOA and PFOS in a
population of younger females (≤50 years, n=20), older females (>50 years), younger males (≤50 years),
and older males (>50 years) and did not find significant sex or age differences. Studies that measured
systemic clearance in monkeys also have not found significant sex differences in systemic clearance of
PFOA (Buttenoff et al. 2004c) or PFOS (Chang et al. 2012).

Studies conducted in rats have shown that PFDeA, PFNA, PFOA, PFOS, and PFHxA are secreted in bile
and undergo extensive reabsorption from the gastrointestinal tract (Johnson et al. 1984; Kudo et al. 2001;
Vanden Heuvel et al. 1991b, 1991c). Biliary secretion rates of PFOA are similar in male and female rats
when renal excretion is blocked by ligation of the kidneys (Vanden Heuvel et al. 1991a, 1991b). This
lack of sex influence on biliary secretion (compared to the sex influence on renal clearance) may reflect a
relative sex insensitivity of OAT2 (or other organic anion transporter) expression in liver, compared to
kidney; the latter is approximately 7–8 times higher in adult female rats compared to male rats (Kudo et
al. 2002).
3.1.5

Physiologically Based Pharmacokinetic (PBPK)/Pharmacodynamic (PD) Models

PBPK models use mathematical descriptions of the uptake and disposition of chemical substances to
quantitatively describe the relationships among critical biological processes (Krishnan et al. 1994). PBPK
models are also called biologically based tissue dosimetry models. PBPK models are increasingly used in
risk assessments, primarily to predict the concentration of potentially toxic moieties of a chemical that
will be delivered to any given target tissue following various combinations of route, dose level, and test
species (Clewell and Andersen 1985). Physiologically based pharmacodynamic (PBPD) models use
mathematical descriptions of the dose-response function to quantitatively describe the relationship
between target tissue dose and toxic endpoints.
Several PBPK models of PFOA and PFOS have been reported. These include a human model for PFOA
and PFOS (Fàbrega et al. 2014, 2016; Loccisano et al. 2011; Worley et al. 2017b), models for PFOA and
PFOS in monkeys (Loccisano et al. 2011), models for PFOA and PFOS in rats (Harris and Barton 2008;
Loccisano et al. 2012a, 2012b; Tan et al. 2008; Worley and Fisher 2015a. 2015b), and a model for PFOA
in mice (Rodriguez et al. 2009). Models of PFOA and PFOS kinetics during gestation and lactation in
rats and mice also have been reported (Loccisano et al. 2012a, 2012b; Rodriguez et al. 2009). Various

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

494

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

empirical and compartmental models have also been reported (Hoffman et al. 2011; Lorber and Egeghy
2011; Lou et al. 2009; Thompson et al. 2010; Verner et al. 2016; Wambaugh et al. 2013; Wu et al. 2009).
Tardiff et al. (2009) utilized a human pharmacokinetic model to estimate an average daily oral dose
corresponding to a Reference Dose for PFOA plasma concentration in humans. Cheng and Ng (2017)
developed a permeability-limited PBPK model for PFOA in male rats that could be used for in vitro to
in vivo extrapolation. PBPK models were not identified for other perfluoroalkyls examined in this profile.
Given the toxicokinetic differences between compounds, the PFOA and PFOS PBPK models may not be
appropriate for other compounds.
3.1.5.1 Loccisano et al. (2012a, 2012b) Rat Models
Loccisano et al. (2012a) developed a model for simulating the kinetics of PFOA and PFOS in male and
female rats. The model was based, in part, on a multi-compartmental model developed by Tan et al.
(2008; Andersen et al. 2006). The female rat model (Loccisano et al. 2012a) was subsequently extended
to include gestation and lactation (Loccisano et al. 2012b). The general structures of the models are
depicted in Figures 3-3, 3-4, and 3-5. Complete lists of parameters and parameter values and the bases
for parameter values and evaluations of model predictions in comparison to observations are described in
Loccisano et al. (2012a, 2012b).

The basic (i.e., adult nonpregnant rat) model includes compartments representing plasma (including a
bound and free fraction), kidney and renal glomerular filtrate, liver, and a lumped compartment
representing all other tissues. Two storage compartments are included in the model: one receives
perfluoroalkyl from the gastrointestinal tract (unabsorbed) and liver (bile) and the other receives
perfluoroalkyl from the glomerular filtrate. The storage compartments were included in the model to
simulate time delays between elimination from plasma and appearance of perfluoroalkyl in feces or urine.
Absorption from the gastrointestinal tract is simulated as the balance between first-order absorption and
fecal excretion of unabsorbed chemical. Absorbed PFOA and PFOS are assumed to be delivered to the
liver where saturable binding of PFOS (but not PFOA) to liver proteins occurs. Saturable binding of
PFOS in liver was included to simulate the relatively long retention times of PFOS in liver that have been
observed in rats. Exchanges between PFOA or PFOS in liver (free fraction), kidney, and other tissues
with the free pool in plasma are assumed to be flow-limited (governed by blood flow) with equilibrium
determined by the tissue:blood partition coefficient. PFOA and PFOS in plasma are simulated as
instantaneous distributions into free and bound fractions. Extensive binding of PFOA and PFOS to
plasma proteins has been demonstrated in various animal species including rats (see Section 3.1.2). For

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

495

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-3. Structure of PBPK Model of PFOA and PFOS in the Rat

Intravenous

Oral,
diet

Gut
kabs

QLiv

Plasma

Liver
Bmax, Kb, koff

kunabs
kbile

storage

feces

Rest of body
QTis

Free
fraction

Kidney
QKid

QFil

Tm,Kt

Filtrate

storage

urine
Bmax = liver binding capacity; kabs = first-order absorption rate constant; Kb = liver binding affinity
constant; kbile = biliary excretion rate constant; Koff = liver binding dissocation constant; Kt = affinity
constant; kunabs = rate of unabsorbed dose to appear in feces; PBPK = physiologically based
pharmacokinetic; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; QFil = clearance
from plasma to glomerular filtrate; QKid = blood flow in and out of kidney; QLiv = blood flow in and out of
liver; QTis = blood flow in and out of tissues; Tm = transporter maximum
Source: Loccisano et al. 2012a (reproduced with permission of Elsevier Inc. in the format reuse in a
government report via Copyright Clearance Center; Reproductive Toxicology by Reproductive Toxicology
Center; Washington, DC)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

496

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-4. PBPK Model Structure for Simulating PFOA and PFOS Exposure
During Gestation in the Rat (Dam, Left; Fetus, Right)
Oral dose

Gut

QFBrn
kunabs

kabs
QLiv

QKid

kbile

Liver

storage

Fetal brain
feces

Fetal
plasma

Kidney

QFLiv

Fetal liver
QFTis

Tm,Kt
QFil

Filtrate

Plasma
free
fraction

storage

urine

Rest of fetal
tissues/body

Fetal structure for PFOS

ktrans1
QPla

QMam

QFat

Placenta
ktrans2

Amniotic
fluid

Mammary
Fat

Fetal
plasma

ktrans4

QFTis
QTis

ktrans3

Rest of fetal
tissues/body

Rest of body

kabs = first-order absorption rate constant; kbile = biliary excretion rate constant; Kt = affinity constant;
ktrans1/ktrans 2 = transfer between placenta and fetal plasma; ktrans3/ktrans4 = transfer between
amniotic fluid and rest of the body; kunabs = rate of unabsorbed dose to appear in feces;
PBPK = physiologically based pharmacokinetic; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane
sulfonic acid; QFat = blood flow in and out of fat; QFBrn = blood flow in and out of fetal brain;
QFil = clearance from plasma to glomerular filtrate; QFLiv = blood flow in and out of fetal liver;
QFTis = blood flow in and out of fetal tissue; QKid = blood flow in and out of kidney; QLiv = blood flow in
and out of liver; QMam = blood flow in and out of mammary tissue; QPla = blood flow in and out of
placenta; QTis = blood flow in and out of tissues; Tm = transporter maximum
Source: Loccisano et al. 2012b (reproduced with permission of Elsevier Inc. in the format reuse in a
government report via Copyright Clearance Center; Reproductive Toxicology by Reproductive Toxicology
Center; Washington, DC)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

497

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-5. PBPK Model Structure for Simulating PFOA/PFOS Exposure During
Lactation in the Rat (Dam, Left; Pup, Right)

QTis

Rest of body
PAMilk

QMam

QLiv

Pup gut

Pup feces

kabsP

Mammary
tissue
Liver

klac

Milk
Oral dose

Pup liver
kabs

QPLiv

Gut
Pup kidney

Plasma
kbile

storage

Pup urine

Free
fraction

QPKid

TmP,KtP
Pup filtrate

QFat

QKid

Fat

Pup
plasma

QPFil

feces

Kidney

Rest of body

QPTis

Tm,Kt
QFil

Filtrate
Transfer of PFOS to milk

Plasma
storage

QMam

Free
fraction

Milk

klac

Pup gut

urine

kabs = first-order absorption rate constant; kabsP = pup first-order absorption rate constant; kbile = biliary
excretion rate constant; klac = transfer to pup through milk; Kt = affinity constant; KtP = pup affinity
constant; PAMilk = transfer from mammary tissue to liver; PBPK = physiologically based pharmacokinetic;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; QFat = blood flow in and out of fat;
QFil = clearance from plasma to glomerular filtrate; QKid = blood flow in and out of kidney; QLiv = blood
flow in and out of liver; QMam = blood flow in and out of mammary tissue; QPFil = clearance from pup
plasma to glomerular filtrate; QPKid = blood flow in and out of pup kidney; QPLiv = blood flow in and out
of pup liver; QPTis = blood flow in and out of pup tissue; QTis = blood flow in and out of tissues;
Tm = transporter maximum; TmP = pup transporter maximum
Source: Loccisano et al. 2012b (reproduced with permission of Elsevier Inc. in the format reuse in a
government report via Copyright Clearance Center; Reproductive Toxicology by Reproductive Toxicology
Center; Washington, DC)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

498

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

PFOA, the free fraction is assigned a constant of 4.5% in females and 0.6% in males. These values were
optimized to fit observed kinetics of PFOA in plasma and urine of rats following intravenous and oral
exposures (Loccisano et al. 2012a). Adequate fit to observed PFOS plasma kinetics following single
doses of PFOS required introducing a time-dependence in binding of PFOS to protein (Loccisano et al.
2012a; Tan et al. 2008). The free fraction for PFOS in plasma decreases from an initial value (after
dosing) of 2.2% to a minimum of 0.1% with a t1/2 for the change of approximately 14 hours in a 0.25-kg
rat (k=0.035 hours-1/kg-0.25). The relatively short t1/2 for the change limits the effects of the timedependent plasma kinetics over the first 1–2 days of dosing (including peak concentrations) and has no
effect on longer-term kinetics or steady state. Although the time-dependence of the free fraction in
plasma was needed to simulate short-term plasma PFOS kinetics in rats, the physiological mechanism for
a dependence of plasma binding on the time following dosing (i.e., not on concentration of PFOS in
plasma or some other dose surrogate) has not been established. Elimination of absorbed chemical occurs
by biliary excretion and urinary excretion. Transfer from liver to feces (representing excretion following
biliary transfer) is represented as a first-order process acting on the free fraction in liver. Excretion in
urine is simulated as the balance between transfer from the free fraction to the glomerular filtrate and
renal tubular reabsorption, which removes PFOA and PFOS from the glomerular filtrate and returns it to
kidney tissue. Renal tubular reabsorption is simulated as a capacity-limited process with parameters Tm
(µg/hour/kg body weight), representing the maximum rate of transport, and KT (µg/L), representing
affinity for the transporter (the concentration in the glomerular filtrate at which reabsorptive transport rate
is half of maximum). This representation of renal tubular reabsorption is used to simulate observed sex
differences in elimination of PFOA from plasma, which have been attributed to higher reabsorptive
capacity in male rats (see Section 3.1.4). Values for the maximum and affinity parameters for PFOA
result in higher reabsorptive clearances from the glomerular filtrate (Tm/KT=4.1) in male rats compared to
female rats (Tm/KT=0.045), and correspondingly lower urinary clearance of PFOA from plasma in male
rats. Reabsorption parameters for PFOS are the same in both sexes and result in reabsorptive clearances
that are approximately twice that of PFOA in female rats (Tm/KT=7.2).

The basic rat model was extended to simulate gestation with inclusion of additional compartments
representing adipose and mammary tissue in the dam, placenta, and fetus (Figure 3-4); Loccisano et al.
2012b). Transfer of PFOA and PFOS to the fetus is simulated as a flow-limited transfer to the placenta,
with first-order exchange between the placenta and the free fraction in fetal plasma. The free fraction in
fetal plasma is simulated with as a constant fraction for PFOA and PFOS (i.e., no dependence on time as
in the adult). Within the fetus, PFOA in the free fraction of plasma exchanges with a single lumped
compartment representing the fetal body, which exchanges with PFOA in amniotic fluid. The fetal PFOS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

499

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

model subdivides fetal tissue into brain, liver, and a lumped compartment for other tissues, all of which
undergo flow-limited exchanges with the free fraction of PFOS in fetal plasma. Binding of PFOA and
PFOS in fetal liver is assumed to be negligible. Differences in the structure of the fetal models for PFOA
and PFOS reflect the differences in the availability of data of for estimating parameter values for the
various compartments (e.g., perfluoroalkyl concentrations in amniotic fluid, liver).

The lactation model extends the dam portion of the gestational model to include milk and pup
(Figure 3-5; Loccisano et al. 2012b). Transfer of PFOA to milk occurs through the mammary gland with
flow-limited exchange between plasma and mammary tissue and diffusion into milk from mammary
tissue. The model also includes transfer from the pup to the dam, which occurs during maternal
stimulation of the neonatal pup to induce elimination and during pup grooming. Data on PFOS in
mammary tissue of rodents were not available to establish parameters for a mammary tissue
compartment; therefore, the mammary tissue compartment was left out of the PFOS model, and transfer
of PFOS to milk is simulated as diffusion directly from plasma. The pup model includes compartments
representing the free fraction in plasma, liver, kidney, glomerular filtrate, and a lumped compartment
representing all other pup tissues. This structure is essentially identical to the nonpregnant rat model
(Loccisano et al. 2012a) with a few differences. Absorption from the gastrointestinal tract is assumed to
be complete in pups, and binding in pup liver is assumed to be negligible in pups. There are no storage
compartments for biliary or glomerular filtrate perfluoroalkyl in the pup model. Sex differences in renal
tubular reabsorption of PFOA are assumed to develop in response to sexual maturation and, therefore, are
not present during lactation (i.e., parameter values are allometrically scaled to pup body weight from the
male rat values). Reabsorptive transport parameters for PFOS are allometrically scaled from the lactating
dam. The liver/plasma partition coefficient for PFOS in the pups was set lower than that in the dam,
based on observations in rats. All other parameters for PFOA and PFOS in the pup were the same or
allometrically scaled from values for the dam.

Optimization of parameter values and evaluations of the rat models are described in Loccisano et al.
(2012a, 2012b). Data sets utilized in developing and evaluating the nonpregnant rat models included
single-dose intravenous and gavage studies and short-term feeding studies (Johnson and Ober 1979;
Kemper 2003; Kudo et al. 2007; Perkins et al. 2004). Data used in development and evaluation of the
gestation and lactation models included data from gestational and/or lactational exposure studies in rats
(Chang et al. 2009; Hinderliter et al. 2005; Kuklenyik et al. 2004; Luebker et al. 2002, 2005a, 2005b;
Thibodeaux et al. 2003).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

500

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Applications for Dosimetry Extrapolation and Risk Assessment. The wealth of data on
pharmacokinetics of PFOA and PFOS in rats allowed an extensive evaluation of the rat models for
predicting plasma urinary and liver PFOA and PFOS following single intravenous or single and repeated
oral dosing. Inclusion of renal tubular reabsorption parameters in the model provided accurate
simulations of sex differences in elimination rates of PFOA from plasma and excretion in urine, and
differences in rates of elimination of PFOA and PFOS. The gestation model successfully predicted fetal
plasma and liver PFOA and PFOS at the end (or near the end) of pregnancy. Consistent with
observations, the model predicts higher fetal plasma concentrations and lower fetal liver concentrations of
PFOS compared to maternal, and lower internal exposure (plasma concentrations) to PFOA in the fetus
compared to maternal (fetal liver data were not available for PFOA). The lactation model successfully
predicted PFOA and PFOS in pup plasma following dosing of the dam. Predicted plasma concentrations
of PFOA in nursing pups were approximately 10–50% lower than maternal concentrations, whereas
maternal and pup concentrations of PFOS were similar. The model could be used to estimate liver doses
and corresponding plasma profiles resulting from single or repeated dosing of adult male or female rats,
and maternal-fetal and maternal-pup transfer of PFOA and PFOS. The rat model was evaluated with data
from a 14-week oral dosing study and has not been tested for longer exposures. Harris and Barton (2008)
developed a PBPK model for PFOS in the rat and found that time adjustments that increased renal
clearance and decreased the liver-plasma partition coefficient as a function of time and dose improved
predictions of plasma and liver PFOS in adult rats exposed for a period of 105 weeks. Although the
Harris and Barton (2008) model is very different from the Loccisano et al. (2012a) model, these results
suggest the possibility that clearance of PFOS may be age- and/or dose-dependent in rats. This may
reflect age- or dose-related changes in kidney function, including tubular reabsorption or secretion of
PFOS.

3.1.5.2 Loccisano et al. (2011, 2013) Monkey and Human Models
Loccisano et al. (2011) developed a model for simulating the kinetics of PFOA and PFOS in monkeys
and humans. The human model described in Loccisano et al. (2011) was subsequently extended to
include simulations of pregnancy and lactation (Loccisano et al. 2013). The monkey model was based, in
part, on a multi-compartmental model developed by Tan et al. (2008; Andersen et al. 2006) for simulating
the kinetics of plasma and urinary PFOA in monkeys. The structures of the monkey and human models
are identical (Figure 3-6) and are very similar to the structure of the rat model (Loccisano et al. 2012a),
with inclusion of compartments representing fat and skin, and absence of a storage compartment for
biliary transfer. Complete lists of parameters and parameter values and the bases for parameter values

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

501

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-6. Structure of PBPK Model for PFOA and PFOS in Monkeys and
Humans

QGut

QLiv
QFat
Intravenous

Oral dose,
drinking water

Gut

Liver

Fat

Plasma
Free
fraction

QSkin
QR
QKid

Skin

Rest of body
Kidney
Tm,Kt

QFil

Filtrate
storage
kurine
Urine

Kt = half-saturation constant; kurine = urinary elmination rate; PBPK = physiologically based pharmacokinetic;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; QFat = blood flow in and out of fat;
Qfil = clearance from plasma to glomerular filtrate; QGut = blood flow in and out of gut; QKid = blood flow in and out
of kidney; QLiv = blood flow in and out of liver; QR = blood flow in and out of rest of body; QSkin = blood flow in and
out of skin; Tm = transport maximum
Source: Loccisano et al. 2011 (reproduced with permission of Academic Press in the format reuse in a government
report via Copyright Clearance Center; Regulatory Toxicology and Pharmacology: RTP by International Society of
Regulatory Toxicology and Pharmacology)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

502

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

and evaluations of model predictions in comparison to observations are reported in Loccisano et al.
(2011).

Parameters in the monkey and human models differ in several ways from the rat model. The free fraction
in plasma is represented as a constant for both PFOA and PFOS; time-dependency for PFOS in the rat
model is absent in the monkey and human models. The parameters for renal tubular reabsorption of
PFOA and PFOS are the same for males and females. This is consistent with the absence of evidence for
a sex difference in elimination kinetics in monkeys (Butenhoff et al. 2002, 2004a; Seacat et al. 2002).

Values for the affinity constant (KT) and maximum (Tm) for tubular reabsorption were optimized to
plasma concentration kinetics in monkeys. The value for KT in monkeys was used in the human model.
The value for Tm for PFOA in humans was set to yield a plasma elimination t1/2 of 2.3 or 3.8 years. The
latter two values derive from estimates of the serum t1/2 in populations exposed to PFOA in drinking
water (2.3 years; Bartell et al. 2010) or in retired fluorochemical workers (3.8 years; Olsen et al. 2007a).
The value for Tm for PFOS in humans was set to yield a plasma elimination t1/2 of 5.4 years, based on
observations in retired fluorochemical workers (Olsen et al. 2007a). Binding of PFOA and PFOS in the
liver was assumed to be negligible in monkeys and humans. Tissue-plasma partition coefficients used in
both models were derived from observations in rodents and were the same in the monkey and human
models.

Optimization of parameter values and evaluation of the monkey and human models are described in
Loccisano et al. (2011). Data sets utilized in developing and evaluating the monkey model included
single-dose intravenous and oral studies and repeated-dose oral studies conducted in Cynomolgus
monkeys (Butenhoff et al. 2004c; Noker and Gorman 2003; Seacat et al. 2002). Data used in evaluating
the human model consisted of serum measurements in people who experienced environmental exposures
(Emmett et al. 2006a; Hölzer et al. 2008; Steenland et al. 2009b), adult Red Cross donors (Olsen et al.
2003b, 2008), and retired fluorochemical workers (Olsen et al. 2007a). In general, PFOA and PFOS
intakes and exposure durations were not known with certainty in these populations and, as a result, these
data do not yield confident evaluations of the ability of the human model to predict intake-plasma level
relationships. Follow-up monitoring after a cessation or decrease in exposure can provide data that allow
evaluation of the ability of the model to accurately simulate elimination kinetics. Predicted declines in
serum PFOA concentrations encompassed observed group mean declines when the Tm for renal tubular
reabsorption was set to yield an elimination t1/2 of 2.3 or 3.8 years. Group mean declines in serum PFOS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

503

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

were predicted reasonably well for some populations, but not all populations, when the Tm for renal
tubular reabsorption was optimized to yield an elimination t1/2 of 5.4 years.

The human pregnancy model includes additional compartments representing the free fractions in plasma,
amniotic fluid, and a lumped compartment for fetal tissue (Loccisano et al. 2013). The same conceptual
approach was used in rat pregnancy model (Loccisano et al. 2012b, Figure 3-4). Rate constants for
placental transfer were initially those from the rat model, adjusted to yield predicated maternal/fetal
plasma ratios that agreed with observed maternal/fetal ratios in cord blood (Apelberg et al. 2007b; Fei et
al. 2007; Midasch et al. 2007; Washino et al. 2009). Transfers from amniotic fluid to fetus were the same
as those used in the rat model, as there were no data on which to base estimates for humans. The lactation
model included additional compartments for mammary milk and a lumped compartment representing the
infant. Transfer of PFOA to milk is simulated as flow-limited exchange between plasma and milk,
governed by mammary tissue blood flow and a milk/plasma partition coefficient. This structure obviated
the need to simulate mammary tissue kinetics, for which there were no data in humans. The milk/plasma
partition coefficient was calibrated to yield predictions of observed milk/plasma ratios (Fromme et al.
2010; Kärrman et al. 2007). Transfer from maternal milk to infants is the product of the milk
concentration and milk production rate (assumed to be equal to sucking rate). The pregnancy model was
evaluated by comparing predicted maternal/fetal plasma ratios for PFOA and PFOS with observations
from various human monitoring studies (Fei et al. 2007; Fromme et al. 2010; Hanssen et al. 2010; Inoue
et al. 2004; Kim et al. 2011; Midasch et al. 2007; Monroy et al. 2008; Tittlemier et al. 2004). The
lactation model was evaluated by comparing predicted maternal plasma/milk ratios for PFOA and PFOS
with observations from various human monitoring studies (Fromme et al. 2009; Kärrman et al. 2007; Liu
et al. 2011). In general, most model predictions were within plus or minus 2-fold of observations.

Applications for Dosimetry Extrapolation and Risk Assessment. The model predicts plasma
concentrations and tissue levels of PFOA and PFOS following intravenous or oral dosing. A skin
compartment is included in the model, which may serve for simulating absorption and distribution
following deposition onto the skin surface; however, the dermal absorption model was not evaluated in
Loccisano et al. (2011). The human model was calibrated to predict t1/2 values estimated for human
populations (e.g., 2.3 or 3.8 years for PFOA, 5.4 years for PFOS). As a result, comparisons made
between observed and predicted serum concentrations evaluate whether or not the populations actually
exhibit the t1/2 to which the model was calibrated, and not the validity of the model to predict the internal
distribution of PFOA or PFOS. It is not currently possible to assess with confidence whether the human
model can accurately predict doses to liver or any other tissues. Fábrega et al. (2014) applied the human

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

504

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

adult model to estimate plasma concentrations and tissue levels of PFOA and PFOS in human autopsy
samples. Exposure inputs to the model were intakes of PFOA and PFOS estimated from public water
supply concentrations in the local area where the subjects had resided (Catalonia, Spain) and
concentrations in local market basket foods (Domingo et al. 2012a, 2012b). The human model predicted
levels of PFOA in plasma and liver that were approximately 10- and 5-fold higher, respectively, than
observed. Predicted plasma levels of PFOS were approximately 2-fold higher than observed and
predicted levels of PFOS in kidney were approximately 25% of observed. Fábrega et al. (2014) explored
alternative values for tissue/plasma partition coefficients, determined from human autopsy issues (Maestri
et al. 2006). The adjusted partition coefficients improved predictions of observed tissue PFOA and PFOS
levels. Although the model could be applied to predicting plasma concentrations of PFOA and PFOS or
intakes associated with specific plasma concentrations (e.g., oral MRLs), it is not clear what advantages
the model offers over simpler empirical or compartmental models similarly calibrated to predict the serum
t1/2. The monkey model has been more thoroughly evaluated for predicting plasma and urinary kinetics of
PFOA and PFOS. This was possible because of the availability of more extensive experimental data on
plasma and urine PFOA and PFOS following intravenous and oral (single and repeated) dosing in male
and female monkeys. Nevertheless, data on internal distribution were not available to allow evaluation of
how well the monkey model predicts doses to the liver or other tissues. Predictions of plasma PFOA and
PFOS concentrations from the monkey (and human) model were highly sensitive to values assigned to the
maximum rate for tubular reabsorption (Tm) and other parameters that govern urinary elimination of
PFOA and PFOS (e.g., free fraction in plasma and glomerular filtration rate; Loccisano et al. 2011).
Optimization of the monkey models relied heavily on adjusting these same parameters and, for the human
model, the target plasma elimination t1/2 was achieved solely by adjusting Tm. Thus, despite the
complexity of the models, their potential to accurately predict plasma elimination kinetics and, therefore,
steady-state plasma concentrations and associated oral intakes, depends largely on how well they predict
plasma clearance. If plasma clearance and the free-fraction in plasma can be reliably predicted
empirically for the animal species of interest, then far simpler compartmental models can be used for
dosimetry extrapolation of steady-state free plasma concentrations.
3.1.5.3 Rodriguez et al. (2009) Mouse Model
Rodriguez et al. (2009) developed a model for simulating the maternal-fetal and maternal-pup kinetics of
PFOA in mice. The general structure of the model is depicted in Figure 3-7. Complete lists of
parameters and parameter values and the bases for parameter values and evaluations of model predictions
in comparison to observations are reported in Rodriguez et al. (2009). The maternal, fetal, and pup

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

505

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-7. Renal Resorption Pharmacokinetic Model of Gestation and Lactation
used in the Analysis of CD-1 Mice

Gestation days 1–18

Concepti
Vcon
Qcon

Qcon
Kidney

Dam

Oral dose
(mg/kg)
Lactation days 19–39

Ccon

Qur

kad

Cdam

Vdam

free
QGF

Filtrate

Qr-QGF

Renal
plasma

klac
Pup excreta
recirculation
(from birth to
postnatal day 4)

Urine

Tm; Kt

Milk
Qr-Qur

Cmilk

Vm
klac

Pups
Cpup

Vpup
kep

Urine
Ccon = concentration in concepti; Cdam = concentration in dam; Cmilk = concentration in milk; Cpup = concentration
in pup; kad = first-order absorption rate; kep = urinary excretion rate; klac = transfer rate via milk; Kt = half-saturation
constant; Qcon = blood flow to and from placenta; QGF = glomerular filtrate; Qr = renal plasma flow; Qur = urine flow;
Tm = transport maximum; Vcon = volume in concepti; Vdam = volume in dam; Vmilk = volume in milk; Vpup = volume
in pup
Source: Rodriguez et al. 2009 (reproduced with permission of Elsevier Inc. in the format reuse in a government
report via Copyright Clearance Center; Reproductive Toxicology by Reproductive Toxicology Center; Washington,
DC)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

506

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

systems are simulated as single well-mixed compartments. Absorption from the gastrointestinal tract is
simulated as first-order with complete absorption of the ingested dose. Elimination of absorbed PFOA
from the maternal system is simulated as the balance between glomerular filtration and renal tubular
reabsorption. The latter is represented as a saturable process with parameters Tm and KT. Transfer to the
fetus is flow-limited and governed by a fetus/maternal partition coefficient and placental blood flow.
Transfer from the maternal system to the pup by lactation is simulated as first-order governed by a
lactation transfer rate constant. Elimination of PFOA from the pup is first-order to urine. Data sets
utilized in developing and evaluating the mouse model included oral gestational dosing studies.

Applications for Dosimetry Extrapolation and Risk Assessment. The model predicted observed
concentrations of PFOA in maternal, fetal, and pup serum following oral gestational exposures to mice
(Abbott et al. 2007; Lau et al. 2006; White et al. 2007). Residuals for predictions are presented, which
provide a quantitative measure of how well the model predicted observations (Rodriguez et al. 2009).
Similar to the rat, the mouse model predicts higher internal exposure (serum PFOA concentrations) in the
maternal system compared to the fetus. It also predicts accelerated loss of PFOA from the maternal
system during lactation. The model simulates the maternal, fetal, and pup systems as single
compartments. Although this serves for simulating plasma concentrations (the main objective of the
modeling effort), it does not allow for simulation of tissue levels of PFOA in the maternal system, fetus,
or pup.
3.1.5.4. Wambaugh et al. 2013 (Andersen et al. 2006) Model
The Wambaugh et al. (2013) model is a three-compartment model based on the three-compartmental
monkey model of Andersen et al. (2006). The structure of the two models are identical (Figure 3-8).
Parameter values for the Wambaugh et al. (2013) model are presented in Table 3-7. The model includes a
central compartment, a secondary distribution compartment, and a renal glomerular filtrate compartment.
The central compartment (C1), which includes plasma, receives PFOA or PFOS from oral dosing (firstorder ka, hour-1) and exchanges perfluoroalkyl with the secondary compartment (C2, which lumps all
other tissues and distribution volumes into a single compartment) and with the glomerular filtrate (C3). A
fraction of the perfluoroalkyl in C1 is free (Free) and available for exchange with C2 and C3. Exchanges
between C1 and C2 are first order (k12, k21, hour-1) with k21 assigned a value equal to the RV2/V1, where
RV2/V1 is the ratio of the volumes of the two compartments (V2/V1). Transfer of perfluoroalkyl into the
glomerular filtrate is first order and governed by the glomerular filtration rate (Qfilc, L/hour). Transfer for

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

507

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-8. Andersen et al. (2006) Pharmacokinetic Model with Oral Absorption

Agut is the amount of chemical in the gut; ka is the first-order rate constant for absorption from the gut; Qfil is the flow
through the filtrate compartment; C1, C2, and C3 are the chemical concentrations in the central, second, and filtrate
compartments, respectively; Vc, Vt, and Vfil are the volumes of distribution of the central, second, and filtrate
compartments; free is the free fraction of compound in the central compartment; Qd is the flow between the central
and second compartments; the saturable resorption process from the filtrate back into the central compartment is
modeled with Michaelis-Menten kinetics, with a maximum rate Tmaximum and a half-maximum concentration KT.
Reprinted from Wambaugh et al. (2013) by permission of Oxford University Press

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

508
3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Table 3-7. Estimated and Assumed Pharmacokinetic Parameters for the Modified Andersen et al. (2006) Model
for PFOA and PFOS

Reference

Species

Cardiac outputa
BW (kg) (L/hour/kg0.74) Ka (hour-1)

Vcc (L/kg)

Parameter (units)
RV2:V1
Tmaxc
k12 (hour-1)
(unitless)
(µmole/hour) KT (µM)

Free
(unitless)

Qfilc (L/hour) Vfilc (L/kg)

Lou et al.
(2009)

Mouse: CD1 0.02
(F)

8.68

290 (0.6–
73,000)

0.18 (0.16–
2.0)

Dewitt et al.
(unpublished)

Mouse:
0.02
C57Bl/6 (F)

8.68

340 (0.53–
69,000)

0.17 (0.13–
2.3)

PFOAb
0.021
(3.1x10-10 to
3.8x104)
0.35 (0.058–
52)

12.39

1.7 (1.1–3.1)

0.14 (0.11–
0.17)

0.098 (0.039– 9.2 (3.4–28)
0.27)

0.077 (0.015– 9.7x10-4
0.58)
(3.34x10-9–
7.21)
0.017 (0.010– 7.6x10-5
0.081)
(2.7x10-10–
6.4)
1.1 (0.25–9.6) 1.1 (0.27–4.5) 0.086 (0.031– 0.039 (0.014– 2.6x10-5
0.23)
0.13)
(2.9x10-10–28)

12.39

1.1 (0.83–1.3) 0.15 (0.13–
0.16)

0.028
8.4 (3.1–23)
(0.0096–0.08)

190 (5.5–
50,000)

0.092
0.08 (0.03–
(3.4x10-4–1.6) 0.22)

19.8

230 (0.27–
73,000)

0.0011
(2.4x10-10 to
3.5x104)

3.9 (0.65–
9,700)

0.043
(4.3x10-5–
0.29

Kemper (2003) Rat:
SpragueDawley (F)
Kemper (2003) Rat:
SpragueDawley (M)
Butenhoff et al. Monkey:
(2004b)
Cynomolgus
(M/F)

Reference

Species

0.20
(0.16–
0.23)c
0.24
(0.21–
0.28)c
7 (m),
4.5 (f)

Cardiac outpute
BW d (kg) (L/hour/kg0.74) Ka (hour-1)

Chang et al.
(2012)

Mouse: CD1 0.02
(F)

8.68

Chang et al.
(2012)

Mouse: CD1 0.02
(M)

8.68

Chang et al.
(2012)

Rat:
0.203
SpragueDawley (F)
Rat:
0.222
SpragueDawley (M)
Monkey:
3.42
Cynomolgus
(M/F)

12.39

Chang et al.
(2012)
Seacat et al.
(2002) and
Chang et al.
(2012)

12.39

19.8

1.16 (0.617–
42,400)

0.4 (0.29–
0.55)

Vcc (L/kg)

1.07 (0.26–
5.84)
53 (11–97)

0.98 (0.25–
3.8)

4.91 (1.75–
2.96)

0.037
0.011
(0.0057–0.17) (0.0026–
0.051
2.7 (0.95–22) 0.12 (0.033– 0.034 (0.014–
0.24)
0.17)

Parameter (units)
Tmaxc
RV2:V1
k12 (hour-1)
(unitless)
(µmol/hour)

PFOS
0.264 (0.24– 0.0093
0.286)
(2.63e-10–
38,900)
0.292 (0.268– 2,976 (2.8e0.317
10–4.2e4)

1.01 (0.251–
4.06)

0.0109
(1.44x10-5–
1.45)
433.4 (0.51–
1.29 (0.24–
1.1e4 (2.1–
381
803.8
4.09)
7.9e4)
(2.6x10-5–
2,900)
4.65 (3.02–
0.535 (0.49– 0.0124 (3.1e- 0.957 (0.238– 1,930 (4.11– 9.49
1,980)
0.581)
10–46 800)
3.62)
83,400)
(0.00626–
11,100)
0.836 (0.522– 0.637 (0.593– 0.00524
1.04 (0.256– 1.34e-06
2.45
1.51)
0.68)
(2.86e-10–
4.01)
(1.65e-10–44) (4.88x10-10–
43,200)
60 300)
132 (0.225–
0.303 (0.289– 0.00292
1.03 (0.256– 15.5 (0.764– 0.00594 (2.34
72,100)
0.314)
(2.59e-10–
4.05)
4,680)
x10-5–0.0941)
34,500)

a

57.9 (0.671–
32,000)

KT (µM)

0.01 (0.0026– 0.15 (0.02–
0.038)
24)

c

Source: Wambaugh et al. (2013)

1

***DRAFT FOR PUBLIC COMMENT***

0.0082
(1.3x10-8–7.6)
0.0021
(3.3x10-9–6.9)

Free
(unitless)

Qfilc (L/hour) Vfilc (L/kg)

0.00963
(0.00238–
0.0372)
0.012
(0.0024–
0.038)
0.00807
(0.00203–
0.0291)
0.00193
(0.000954–
0.00249)
0.0101
(0.00265–
0.04)

0.439
0.00142
(0.0125–307) (4.4x10-10–
6.2)
27.59 (0.012– 0.51
283)
(3.5x10-10–
6.09)
0.0666
0.0185
(0.0107–8.95) (8.2x10-7–
7.34)
0.0122
0.000194
(0.0101–
(1.48x10-9–
0.025)
5.51)
0.198 (0.012– 0.0534
50.5)
(1.1x10-7–
8.52)

Cardiac outputs obtained from Davies and Morris (1993).
Means and posterior distributions from the Bayesian Markov Chain Monte Carlo (MCMC) analysis (95% credible interval in parentheses) are reported.
Estimated average body weight (BW) for species used except with Kemper (2003) study where individual rat weights were available and assumed to be constant.
d
Average BW for species: individual-specific BWs.
e
Cardiac outputs obtained from Davies and Morris (1993).
b

0.22 (0.011–
58)

 PERFLUOROALKYLS

509

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

perfluoroalkyl from the glomerular filtrate to C1 (representing renal tubular reabsorption) is capacity
limited (Tmaxc, µmol/hr; KT, µM). Perfluoroalkyl that is not reabsorbed is excreted.

Parameter values for the various species and strains were estimated from experimental pharmacokinetic
data for each species and strain using Bayesian Markov Chain Monte Carlo (MCMC) analysis. Studies
that provided data used to estimate parameter values are listed in Wambaugh et al. (2013). The parameter
values shown in Table 3-7 are the mean values and posterior distributions (95% credible interval) from
the MCMC analyses.

Applications for Dosimetry Extrapolation and Risk Assessment. Wambaugh et al. (2013) applied the
model to predicting internal doses (mean and maximum serum concentrations and plasma AUC) for
Benchmark Dose Software (BMDS) modeling and for comparing internal dosimetry from in vivo toxicity
studies to estimates of potency (AC50, maximum Efficacy) from in vitro studies. EPA applied the
Wambaugh et al. (2013) model to deriving chronic oral reference doses (RfDs) for PFOA and PFOS
(EPA 2016e, 2016f). The model was used to predict internal doses (time-integrated plasma PFOA or
PFOS concentrations) achieved in toxicity studies conducted in various laboratory animal models (CD-1
mouse, C57Bl/6 mouse, Sprague-Dawley rat, Cynomolgus monkey). Plasma concentrations were then
extrapolated to equivalent steady-state concentrations in humans using a model of first-order elimination
of PFOA and PFOS from plasma. The same approach was used to derive MRLs for PFOA and PFOS
(see Appendix A).
3.1.5.5 Harris and Barton (2008) Rat Model
Harris and Barton (2008) developed a model for simulating PFOS kinetics in adult rats. The general
structure of the model is depicted in Figure 3-9. Complete lists of parameters and parameter values and
the bases for parameter values and evaluations of model predictions in comparison to observations are
reported in Harris and Barton (2008). The model includes systemic compartments representing blood
(including a bound and free fraction of plasma and red blood cells), liver, and a lumped compartment
representing all other tissues. The gastrointestinal tract is simulated as separate compartments representing
the upper and lower tracts. Absorption occurs from both the upper and lower tracts, with distinct firstorder rate constants assigned to each. Biliary PFOS is transferred from liver to the lower tract. Absorbed
PFOS is delivered to the liver where it enters plasma to be distributed to other tissues. Exchanges between
PFOS in plasma and all tissues are assumed to be diffusion-limited, with the free pool in plasma
participating in the exchange with red blood cells, and the total plasma pool exchanging with liver and all

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

510

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Figure 3-9. Conceptual Representation of a Physiologically Based
Pharmacokinetic Model for PFOS Exposure in Rats
intravenous

P + A ↔ P:A Plasma

QPlas

urine

Red blood cells
Rest of body plasma

QR

Rest of body tissue
QL

Liver plasma

Liver tissue
kau

oral

kal

Upper
gastrointestinal ktl
tract

Lower
gastrointestinal
tract

kb

feces

kal = rate of absorption from the lower gastrointestinal tract; kau = rate of absorption from the upper gastrointestinal
tract; kb = maximum rate of biliary elimination; ktl = rate of transfer from upper-lower gastrointestinal tract;
P:A = PFOS-bound albumin in plasma; PFOS = perfluorooctane sulfonic acid QL = plasma flow rate to the liver;
QPlas = plasma flow rate by the heart; QR = plasma flow rate to the rest of body
Source: Harris and Barton 2008 (reproduced with permission of Elsevier Ireland Ltd. in the format reuse in a
government report via Copyright Clearance Center; Toxicology Letters by European Societies of Toxicology)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

511

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

other tissues. Binding of PFOA to plasma albumin is assumed to be saturable, with a dissociation constant
10-7 M and a maximum capacity 4.1x 10-4 M. This is implemented by assigning bound PFOA to a
subcompartment of plasma in which PFOA enters (binds) or exits (unbinds) at rates governed by binding
on and off rates, respectively, that yield a dissociation constant of 10-7 M. Elimination of absorbed
chemical occurs by biliary excretion and urinary excretion. Transfer from liver to the lower
gastrointestinal tract (representing excretion following biliary transfer) is represented as a first-order
process acting on the total amount of PFOS in liver. PFOA is transferred to urine from the free fraction of
plasma at a rate governed by a urinary clearance parameter, which is assigned value of 28% of renal
plasma flow.

In evaluating performance of the model for simulating PFOS concentrations in a chronic rat feeding
study, Harris and Barton (2008) found that the model predicted plasma and liver concentrations measured
at 4 and 16 weeks, but over-predicted both at 104 weeks. Performance of the model was improved by
having renal clearance increase and the liver/plasma partition coefficient decrease as a function of time
(i.e., study duration). These results suggest the possibility that clearance of PFOS may be dependent on
age and/or a metric of dose (e.g., cumulative internal dose). This may reflect age- or dose-related changes
in kidney function, including tubular reabsorption or secretion of PFOS.

Applications for Dosimetry Extrapolation and Risk Assessment. The model simulates kinetics of PFOS
following oral or intravenous dosing in adult rats and includes several features that are different from
other PBPK models of perfluoroalkyls. The Harris and Barton (2008) model includes a red cell
compartment that allows predictions of whole-blood concentrations. The utility of this feature remains to
be determined, since PFOS does not appreciably concentrate in red blood cells and PFOS (and other
perfluoroalkyls) is typically monitored in the central compartment with measurements of plasma or serum
concentrations. The model assumes that the total concentration of PFOS (not just the free concentration)
in plasma is available for distribution to liver and other tissues, whereas other models assume that only
the free pool in plasma exchanges with tissues. The practical consequence of this difference may not be
significant in terms of the toxicokinetics of PFOS if the tissue/plasma partition coefficients in the various
models were estimated based on the relevant perfluoroalkyl pool in plasma. However, without basing
distribution kinetics on the free concentration, it is not possible for concentration-dependent free fraction
to be modeled. The model assumes time-dependence in the liver uptake and urinary excretion of PFOS,
which were needed to improve predictions of plasma and liver concentrations of PFOS during chronic
exposures. Other rat models (Loccisano et al. 2012a) have not been similarly evaluated. A mechanistic

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

512

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

understanding of the time-dependent changes in PFOS kinetics will be important for applications of these
models for dosimetry extrapolation across exposure durations.
3.1.5.6 Worley and Fisher (2015a, 2015b) Rat Model
Worley and Fisher (2015a, 2015b) expanded the Loccisano et al. (2012a) adult rat model to include
simulation of renal proximal tubule apical (tubule-lumen) and basolateral (tubule-plasma) PFOA
transport. This configuration allowed the use of data from in vitro studies of kinetics of specific
transporters thought to be involved in proximal tubular transport of PFOA in the parametrization of the
model. The kidney compartment was expanded to include compartments representing the proximal
tubule lumen (glomerular filtrate) and proximal tubule cells. In the model, transfer of PFOA to the tubule
lumen is governed by the glomerular filtration rate, represented by a clearance parameter (L/hour/kg
kidney). PFOA in the tubule lumen can undergo first-order transfer to urine or saturable transport into the
tubule cell (Km, Vmax). PFOA in the tubule exchanges with PFOA in plasma by three mechanisms:
saturable transport from plasma into the cell (Km, Vmax), first-order transport from the cell to plasma
(kefflux), or bidirectional diffusion between the cell and plasma (kdif). Parameter values (Km, Vmax) for
apical and basolateral transport of PFOA were derived from in vitro estimates for OATP1a1 (apical) and
OAT1 and OAT3 (basolateral) (Nakagawa et al. 2008; Weaver et al. 2010; Yamada et al. 2007). These
estimates were scaled to kidney proximal tubule cell mass (Hsu et al. 2014) and the mass-scaled estimates
of Vmax were adjusted with relative activity factors, which were calibrated to in vivo observations of
plasma PFOA elimination kinetics in rats (Kemper 2003). Values for kefflux (proximal tubule cell to
kidney plasma) and kdif (diffusion between kidney plasma and the tubule cell) were also calibrated with
in vivo data (Kemper 2003; Kudo et al. 2007).

Calibration of the relative activity factor for apical and basolateral membrane transport of PFOA to serum
observations made in male and female rats resulted in lower values for activity of both transporters in
females compared to males. This resulted in the model predicting lower rates of reabsorptive transfer of
filtered PFOA to plasma, and higher renal and systemic (plasma) clearance in females compared to males.
Because proximal tubule transporters were assumed to be saturable, the model predicts an increase in
clearance with increasing PFOA dose, with larger increases in clearance at lower doses in females
compared to males. The model simulated the observed dose-dependent increase in serum clearance
(decreasing serum t1/2) and higher serum clearance of PFOA (lower t1/2) in female rats compared to males
(Kemper 2003).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

513

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

3.1.5.7 Worley et al. (2017b) Human Model
Worley et al. (2017b) scaled and calibrated the Worley and Fisher (2015a, 2015b) rat model to simulate
PFOA kinetics in humans exposed to PFOA in drinking water. Physiological parameters were
allometrically scaled to the human. Tissue-plasma partition coefficients were derived from human
autopsy data (kidney, liver) or studies of distribution of PFOA in rats (Fabrega et al. 2014; Kudo et al.
2007; Perez et al. 2013). Parameter values (Km, Vmax) for apical and basolateral transport of PFOA
were derived from in vitro estimates for OAT4 (apical) and OAT1 and OAT3 (basolateral) (Nakagawa et
al. 2008; Weaver et al. 2010; Yang et al. 2010; Yamada et al. 2007). These estimates were scaled to
kidney proximal tubule cell mass (Hsu et al. 2014) and the mass-scaled estimates of Vmax were adjusted
with relative activity factors. Parameters that control apical and basolateral transfers of PFOA in the
proximal tubule and absorption in the gastrointestinal tract were calibrated against data on serum PFOA
concentrations measured in people who drank water from a municipal water supply (Worley et al. 2017b).
Model parameter values were adjusted to achieve agreement with geometric mean serum PFOA
concentrations measured at two times separated by 6 years. The model was evaluated by comparing
predicted and observed serum PFOA concentrations in populations exposed to PFOA in drinking water
(Bartell et al. 2010; Emmett et al. 2006b; Steenland et al. 2009a, 2009b). A sensitivity analysis of the
model identified that following biokinetic parameters that had standardized sensitivity coefficients >0.1:
parameters controlling proximal tubule transport and urinary excretion, plasma-liver partition coefficient,
biliary excretion and protein binding. These parameters, along with drinking water consumption, were
assigned probability distributions to conduct a Monte Carlo analysis of predicted serum PFOA predictions
associated with exposures to PFOA in drinking water. The probabilistic model simulated interindividual
variability in serum PFOA concentrations observed in exposed populations (Bartell et al. 2010; Emmett et
al. 2006b; Steenland et al. 2009a, 2009b). These results suggest that that biokinetic variability, as well as
exposure variability, may contribute to variability in serum PFOA concentrations observed in
populations.
3.1.5.8 Fàbrega et al. (2014, 2016) Human Model
Fàbrega et al. (2014, 2016) modified the Loccisano et al. (2011, 2013) human models for PFOA and
PFOS with inclusion of brain and lung compartments and removal of the skin compartment. Tissueplasma partition coefficients were re-estimated using data from human cadavers (Maestri et al. 2006) in
place of estimates based on rat data (Loccisano et al. 2011). The major differences in the partition
coefficients for PFOA were lower values for liver in humans (1.03) compared to rats (2.20), higher values
for fat in humans (0.47) compared to rats (0.04), and inclusion of partition coefficients for brain (0.17)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

514

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

and lung (1.27). For PFOS, the major differences in the partition coefficients were lower values for liver
in humans (2.67) compared to rats (3.72) and higher values for fat in humans (0.33) compared to rats
(0.14). Values for parameters that control urinary excretion (Tm and Km for reabsorptive transport from
glomerular filtrate to kidney tissue) were recalibrated based on plasma concentration data (Ericson et al.
2007). Fàbrega et al. (2014) compared predictions to observed concentrations of PFOA and PFOS in
cadaver samples (from Tarragona County, Spain) for constant intakes of 0.11 µg/day for PFOA or
0.13µg/day for PFOS. Better agreement with observations was achieved with partition coefficients based
on cadaver data. Fàbrega et al. (2016) performed a quantitative uncertainty analysis of predictions of
tissue PFOA and PFOS concentrations by assigning lognormal probability distributions to renal transport
parameters, the unbound fraction in plasma, and intake. Probability distributions for PFOA and PFOS
intakes were based on data from Domingo et al. (2012a, 2012b). Distributions for biokinetic parameters
were established to achieve a coefficient of variation of 0.3 (Allen et al. 1996; Brochot et al. 2007;
Sweeney et al. 2001). Observations of tissue PFOA and PFOS were within uncertainty bounds on
predictions.

3.1.6

Animal-to-Human Extrapolations

Interspecies differences in the toxicokinetics of perfluoroalkyls and possible differences in the
mechanisms of toxicity have been found. The elimination rate for PFOA in female rats is approximately
45 times faster than in male rat, 150 times faster than in Cynomolgus monkeys, and approximately 5,000–
9,000 times faster than in humans (Bartell et al. 2010; Butenhoff et al. 2004c; Kemper 2003; Olsen et al.
2007a). Elimination of PFOS in male rats is approximately 3 times faster than in Cynomolgus monkeys
and approximately 40 times faster than in humans (Chang et al. 2012; De Silva et al. 2009; Olsen et al.
2007a; Seacat et al. 2002). These large differences in elimination rates imply that similar external PFOA
or PFOS dosages (i.e., mg/kg/day) in rats, monkeys, or humans would be expected to result in
substantially different steady-state internal doses (i.e., body burdens, serum concentrations) of these
compounds in each species. In addition, exposure durations required to achieve steady state would be
expected to be much longer in humans than in monkeys or rats. Assuming a terminal elimination t1/2 of
1,400 days for PFOA in humans (Olsen et al. 2007a), a constant rate of intake for 17 years would be
required to achieve 95% of steady state. Steady state (i.e., 95%) would be achieved in approximately
110 days in monkeys (t1/2=25 days, Butenhoff et al. 2004c), 30 days in male rats (t1/2=7 days; Kemper
2003), and 1 day in female rats (t1/2=0.2 days; Kemper 2003). Using an internal dose metric such as
serum perfluoroalkyl concentration and PBPK models that can account for these differences in
elimination rates can decrease the uncertainty in extrapolating from animals to humans.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

515

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Many perfluoroalkyl-induced effects in rats and mice are mediated through the PPARα and it is generally
agreed that humans and nonhuman primates are refractory, or at least less responsive than rodents, to
PPARα-mediated effects (Corton et al. 2014; Klaunig et al. 2003; Maloney and Waxman 1999). While
studies in mice have identified specific effects that require PPARα activation, for example, postnatal
viability (Abbott et al. 2007) and some immunological effects (Yang et al. 2002b), other effects such as
hepatomegaly and antigen-specific antibody response (DeWitt et al. 2016) were reported to be PPARαindependent (Yang et al. 2002b). Therefore, further studies are needed to expand the knowledge
regarding PPARα-dependent and -independent effects that would allow selection of an appropriate animal
model for perfluoroalkyls toxicity.
3.2

CHILDREN AND OTHER POPULATIONS THAT ARE UNUSUALLY SUSCEPTIBLE

This section discusses potential health effects from exposures during the period from conception to
maturity at 18 years of age in humans. Potential effects on offspring resulting from exposures of parental
germ cells are considered, as well as any indirect effects on the fetus and neonate resulting from maternal
exposure during gestation and lactation. Children may be more or less susceptible than adults to health
effects from exposure to hazardous substances and the relationship may change with developmental age.

This section also discusses unusually susceptible populations. A susceptible population may exhibit
different or enhanced responses to certain chemicals than most persons exposed to the same level of these
chemicals in the environment. Factors involved with increased susceptibility may include genetic
makeup, age, health and nutritional status, and exposure to other toxic substances (e.g., cigarette smoke).
These parameters can reduce detoxification or excretion or compromise organ function.

Populations at greater exposure risk to unusually high exposure levels to perfluoroalkyls are discussed in
Section 5.7, Populations with Potentially High Exposures.

The possible association between serum perfluoroalkyl levels in children and health effects has been
examined in participants of the C8 Health Project and in the general population. The studies examined a
number of health effects including alterations in serum lipid levels, adverse renal outcomes,
neurodevelopmental alterations, and reproductive development. Immunotoxicity has been examined in
children in several general population studies. Additionally, a large number of studies have examined the
possible association of elevated serum perfluoroalkyl levels and adverse birth outcomes.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

516

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

Similar to adults, associations between serum PFOA and PFOS and serum cholesterol levels were
observed in a study of over 12,000 children (Frisbee et al. 2010); an increased risk of high cholesterol was
also observed in children with higher serum PFOA and PFOS levels. A smaller study of children (n=43)
living in the Mid-Ohio Valley did not find associations between serum PFOA levels and hematology
parameters, total cholesterol and liver enzymes, indices of kidney function, or serum TSH levels (Emmett
et al. 2006b). Another study of highly-exposed residents did not find any associations between serum
PFOA levels in children aged 6–12 years and IQ, reading and math skills, language, memory, learning, or
attention (Stein et al. 2013). Similarly, no association between serum PFOA, PFOS, or PFNA levels in
children 5–18 years old and the likelihood of ADHD diagnosis was observed in a study of highly-exposed
residents, although the study did find an increased risk associated with higher PFHxS levels (Stein and
Savitz 2011). A general population study that utilized the NHANES data found an s association between
serum PFOA, PFOS, and PFHxS levels and the risk of ADHD diagnosis (as reported by the parent)
(Hoffman et al. 2010). Another smaller-scale study found associations between serum PFOS, PFNA,
PFDeA, PFHxS, and PFOSA and impulsivity; no association with PFOA was found (Gump et al. 2011).
A study of children 8–18 years of age participating in the C8 studies found reduced odds of reaching
puberty at higher serum PFOA levels (Lopez-Espinosa et al. 2011); however, the biological significance
of the short delay (4–5 months) is not known.

Several studies have evaluated immunotoxicity in children and adolescents. These studies have found
impaired antibody responses associated with serum PFOA, PFOS, PFHxS, and PFDeA (Grandjean et al.
2012, 2017; Granum et al. 2013; Mogensen et al. 2015a; Stein et al. 2016a). An increased asthma
diagnosis was also associated with serum PFOA levels (Dong et al. 2013; Humblet et al. 2014; Zhu et al.
2016). Marginal evidence was also found for PFOS, PFHxS, PFNA, PFDeA, PFBuS, and PFDoA (Dong
et al. 2013; Zhu et al. 2016), although some studies found no associations for these compounds (Humblet
et al. 2014; Smit et al. 2015; Stein et al. 2016a).

Hines et al. (2009) showed that in utero exposure (GDs 1–17) to low levels of PFOA (0.01–
0.3 mg/kg/day) resulted in increases in body weight gain in 10–40-week-old mice; by 18 months of age,
the body weights in these mice were similar to controls. Increases in serum insulin and leptin levels were
also observed in the mice exposed to 0.01 and 0.1 mg/kg/day. The study also compared body weight and
body composition of in utero exposed mice (exposed on GDs 1–17) and adult exposed mice (exposed for
17 days starting at 8 weeks of age) and found that in utero exposure to 1 mg/kg/day resulted in
significantly higher body weight, brown fat weight, and white fat weight; this was not observed in mice

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

517

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

exposed to 5 mg/kg/day. The results of the study suggest that gestational exposure to low doses of PFOA
may result in increased susceptibility to PFOA toxicity.

A number of studies of highly exposed residents and the general population have examined the potential
associations between serum perfluoroalkyl levels and alterations in birth weight. Decreases in birth
weight have been found to be associated with higher PFOA (Fei et al. 2007; Lee et al. 2013; Maisonet et
al. 2012; Savitz et al. 2012b) or PFOS levels (Maisonet et al. 2012), but not with lower levels of
perfluoroalkyls (Fei et al. 2007; Hamm et al. 2010; Inoue et al. 2004; Kim et al. 2011; Monroy et al.
2008; Washino et al. 2009; Whitworth et al. 2012b). The decreases in birth weight were small and not
likely biological relevant. Additionally, no increases in the risk of low birth weight infants were found in
highly exposed populations (Darrow et al. 2013; Nolan et al. 2009; Savitz et al. 2012b; Stein et al. 2009).
No apparent alterations in the risk of birth defects were found in C8 Health Studies (Darrow et al. 2013;
Savitz et al. 2012b; Stein et al. 2009) or in another study of these communities (Nolan et al. 2009).

The developmental toxicity of PFOA and PFOS has been investigated in a number of rat and mouse
studies. The observed effects include PFOA- and PFOS-induced increases in prenatal losses and
decreases in pup survival, decreases in pup body weight, and neurodevelopmental toxicity (Abbott et al.
2007; Albrecht et al. 2013; Case et al. 2001; Chen et al. 2012b; Era et al. 2009; Fuentes et al. 2006,
2007a, 2007b; Grasty et al. 2003; Hu et al. 2010; Johansson et al. 2008; Lau et al. 2003, 2006; Luebker et
al. 2005a, 2005b; Onishchenko et al. 2011; Thibodeaux et al. 2003; White et al. 2007, 2009, 2011; Wolf
et al. 2007; Xia et al. 2011; Yahia et al. 2008, 2010). Additionally, delays in mammary gland
development were observed in mice exposed to PFOA (Macon et al. 2011; White et al. 2007, 2009,
2011). A limited number of developmental endpoints have been examined in rats and mice exposed to
PFDeA, PFHxS, or PFBA (Butenhoff et al. 2009a; Das et al. 2008; Harris and Birnbaum 1989; Hoberman
and York 2003; Johansson et al. 2008; Viberg et al. 2013). A more in-depth discussion of the
developmental toxicity of perfluoroalkyls in animals is included in Section 2.17.

PFOA and PFOS, as well as other perfluoroalkyl compounds, are valid biomarkers of exposure to these
compounds in children, as they are in adults. No relevant studies were located regarding interactions of
perfluoroalkyl compounds with other chemicals in children or adults. No information was located
regarding pediatric-specific methods for reducing peak absorption following exposure to perfluoroalkyl
compounds, reducing body burden, or interfering with the mechanism of action for toxic effects.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

518

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

No studies examining increased susceptibility to the toxicity of perfluoroalkyls were identified. The
available epidemiology data identify several potential targets of toxicity of perfluoroalkyls, and
individuals with pre-existing conditions may be unusually susceptible. For example, it appears that
exposure to PFOA or PFOS can result in increases in serum lipid levels, particularly cholesterol levels.
Thus, an increase in serum cholesterol may result in a greater health impact in individuals with high levels
of cholesterol or with other existing cardiovascular risk factors. Similarly, increases in uric levels have
been observed in individuals with higher perfluoroalkyl levels; increased uric acid may be associated with
an increased risk of high blood pressure and individuals with hypertension may be at greater risk.
Additionally, associations have been found between PFOA and PFOS levels and an increased risk of
hypertension/pre-eclampsia in pregnant women. The liver has been shown to be a sensitive target in a
number of animal species and there is some indication that it is also a target in humans. Therefore,
individuals with compromised liver function may represent a susceptible population. Likewise,
individuals with a compromised immune system may have an increased risk of perfluoroalkyl-induced
immunotoxicity.

3.3

BIOMARKERS OF EXPOSURE AND EFFECT

Biomarkers are broadly defined as indicators signaling events in biologic systems or samples. They have
been classified as biomarkers of exposure, biomarkers of effect, and biomarkers of susceptibility
(NAS/NRC 1989).

A biomarker of exposure is a xenobiotic substance or its metabolite(s) or the product of an interaction
between a xenobiotic agent and some target molecule(s) or cell(s) that is measured within a compartment
of an organism (NAS/NRC 1989). The preferred biomarkers of exposure are generally the substance
itself, substance-specific metabolites in readily obtainable body fluid(s), or excreta. Biomarkers of
exposure to perfluoroalkyls are discussed in Section 3.3.1. The National Report on Human Exposure to
Environmental Chemicals provides an ongoing assessment of the exposure of a generalizable sample of
the U.S. population to environmental chemicals using biomonitoring (see http://www.cdc.gov/
exposurereport/). If available, biomonitoring data for perfluoroalkyls from this report are discussed in
Section 5.6, General Population Exposure.

Biomarkers of effect are defined as any measurable biochemical, physiologic, or other alteration within an
organism that (depending on magnitude) can be recognized as an established or potential health
impairment or disease (NAS/NRC 1989). This definition encompasses biochemical or cellular signals of

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

519

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

tissue dysfunction (e.g., increased liver enzyme activity or pathologic changes in female genital epithelial
cells), as well as physiologic signs of dysfunction such as increased blood pressure or decreased lung
capacity. Note that these markers are not often substance specific. They also may not be directly
adverse, but can indicate potential health impairment (e.g., DNA adducts). Biomarkers of effect caused
by perfluoroalkyls are discussed in Section 3.3.2.

A biomarker of susceptibility is an indicator of an inherent or acquired limitation of an organism's ability
to respond to the challenge of exposure to a specific xenobiotic substance. It can be an intrinsic genetic or
other characteristic or a preexisting disease that results in an increase in absorbed dose, a decrease in the
biologically effective dose, or a target tissue response. If biomarkers of susceptibility exist, they are
discussed in Section 3.2, Children and Other Populations that are Unusually Susceptible.
3.3.1

Biomarkers of Exposure

Measurement of serum or whole-blood perfluoroalkyl concentrations is the standard accepted biomarker
of perfluoroalkyl exposure in humans. Perfluoroalkyl compounds have been detected in the serum of
workers, residents living near perfluoroalkyl facilities, and the general population. As part of NHANES,
CDC has been measuring serum levels of perfluoroalkyls in the U.S. general population since 1999. Of
the 14 perfluoroalkyls examined in this toxicological profile, blood concentrations of 7 compounds
(PFOA, PFOS, PFDeA, PFHxS, PFNA, Me-PFOSA-AcOH, and PFUA) were detected in enough subjects
to allow for estimation of the geometric mean. As compared to the general population, serum PFOA and
PFOS levels are much higher in individuals with occupational exposure to these compounds (Olsen et al.
2003a; Sakr et al. 2007a) and PFOA levels are much higher in individuals living near a PFOA
manufacturing facility (Emmett et al. 2006a; Hölzer et al. 2008; Steenland et al. 2009a), suggesting that
serum levels are a good biomarker of exposure. Due to the long half-life of some perfluoroalkyl
compounds, particularly PFOA and PFOS, elevated serum levels may not be indicative of recent
exposure. Although elevated serum levels are likely to be indicative of exposure to the parent compound,
their presence in blood can also indicate exposure to other perfluoroalkyl compounds. For example,
PFOS can be derived from metabolism of Et-PFOSA-AcOH, Me-PFOSA-AcOH, or PFOSA (Olsen et al.
2005; Seacat and Luebker 2000). PFOA can be derived from metabolism of 8-2 fluorotelomer alcohol
(Fasano et al. 2006; Henderson and Smith 2007; Kudo et al. 2005; Nabb et al. 2007). Exposure of mice
to 8–2 telomer alcohol also generated PFNA as a metabolite (Kudo et al. 2005). Because Et-PFOSAAcOH and Me-PFOSA-AcOH are metabolized fairly rapidly and have a relatively short serum t1/2, their

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

520

3. TOXICOKINETICS, SUSCEPTIBLE POPULATIONS, BIOMARKERS, CHEMICAL INTERACTIONS

presence in serum should indicate only recent non-occupational exposure in the general population (Olsen
et al. 2005).

Two studies have also evaluated the use of perfluoroalkyl levels in hair as a biomarker of exposure. In
rats administered PFOA, PFOS, or PFNA in the drinking water for 90 days, significant correlations
between hair perfluoroalkyl levels and serum and tissue (liver, heart, lung, kidney) levels were found,
suggesting that hair perfluoroalkyl levels may be a reliable biomarker of exposure (Gao et al. 2015). A
study in humans (Alves et al. 2015) has also found detectable levels of PFBA, PFHxA, PFOA, PFBuS,
and PFHxS in hair samples, but PFHpA, PFNA, and PFOS were not detected in hair samples. The study
did not evaluate the potential relationship between serum perfluoroalkyl levels and hair levels, which does
not allow for an assessment of whether it is a viable biomarker of exposure.
3.3.2

Biomarkers of Effect

There are no specific biomarkers of effect caused by perfluoroalkyl compounds.
3.4

INTERACTIONS WITH OTHER CHEMICALS

There are limited data on the interactions of perfluoroalkyls with other chemicals. Particularly absent are
studies examining toxicological and toxicokinetic interactions of a perfluoroalkyl compound with other
perfluoroalkyl compounds. Olestra decreased the absorption of PFOA from the gastrointestinal tract of
mice (Jandacek et al. 2010). No additional information was located regarding interactions among
chemicals of this class or between perfluoroalkyl compounds and other chemicals. Both PFOA and
PFOS (and many other diverse chemicals) can activate the PPARα, as well as other PPARs to a lesser
extent (Takacs and Abbott 2007; Vanden Heuvel et al. 2006). Therefore, it is not unreasonable to
speculate that interactions at the receptor level might occur; however, there are no experimental data to
support or rule out this presumption. Given that the PPARα receptor is much less responsive in humans
than it is in rodents, it is unclear if this type of possible interaction would be relevant to humans.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

521

CHAPTER 4. CHEMICAL AND PHYSICAL INFORMATION
4.1

CHEMICAL IDENTITY

Information regarding the chemical identity of perfluoroalkyls is located in Table 4-1. This information
includes synonyms, chemical formulas and structures, and identification numbers. The perfluoroalkyls
discussed in this profile exist as linear and branched isomers depending upon the method of production
(see Chapter 5) and the reported values for the physical-chemical properties are typically reflective of the
mixtures rather than a single specific isomer.
4.2

PHYSICAL AND CHEMICAL PROPERTIES

Information regarding the physical and chemical properties of perfluoroalkyls is located in Table 4-2.

Perfluoroalkyl compounds are very stable, owing to the strength of the carbon-fluorine bonds, the
presence of the three electron pairs surrounding each fluorine atom, and the shielding of the carbon atoms
by the fluorine atoms (3M 1999; Kissa 2001; Schultz et al. 2003). Perfluoroalkyl carboxylates and
sulfonates are resistant to direct photolysis and reaction with acids, bases, oxidants, and reductants (3M
2000; EPA 2008a; OECD 2002, 2006a, 2007; Schultz et al. 2003).

APFO was shown to decompose starting at 196°C (Krusic and Roe 2004) and PFOA was shown to
decompose rapidly in the presence of crushed borosilicate glass at 307°C (Krusic et al. 2005).
1-H perfluoroheptane and perfluoroheptene are noted degradation products.

Perfluoroalkyl carboxylates and sulfonates consist of a perfluorocarbon tail that is both hydrophobic and
oleophobic and a charged end that is hydrophilic (3M 1999; de Vos et al. 2008; Kissa 2001; Schultz et al.
2003). This combination of hydrophobic and oleophobic characteristics makes these substances very
useful as surfactants. The ability of these substances to repel oil, fat, and water has resulted in their use in
surface protectants (Kissa 2001). Their ability to reduce the surface tension of aqueous systems to
<20 mN/m has resulted in their use as wetting agents (Kissa 2001). Neutral or uncharged perfluoroalkyls
or very long chain constituents are expected to form separate layers when mixed with hydrocarbons and
water. Conversely, charged species, salts, and ionized species at relevant pH (i.e., PFOS, PFOA, PFHpA,
PFNA) and short-chain species (i.e., PFBA, PFBuS) have relatively good solubility in water and alcohol.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

522
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-1. Chemical Identity of Perfluoroalkyls
Characteristic

Information

Chemical name

Perfluorooctanoic acid

Ammonium perfluorooctanoate

Perfluorobutyric acid

Synonym(s)

PFOA; pentadecafluoro-1-octanoic
acid; pentadecafluoro-n-octanoic acid;
pentadecafluoroctanoic acid;
perfluorocaprylic acid; perfluoroctanoic
acid; perfluoroheptanecarboxylic acid;
octanoic acid,
2,2,3,3,4,4,5,5,6,6,7,7,8,8,
8-pentadecafluoro-

APFO; ammonium
pentadecafluorooctanoate; octanoic
acid, 2,2,3,3,4,4,5,5,6,6,7,7,8,8,8pentadecafluoro-ammonium salt (1:1)

PFBA; heptafluoro-1-butanoic acid;
heptafluorobutanoic acid;
heptafluorobutyric acid;
perfluorobutanoic acid;
perfluorobutyric acid;
perfluoropropanecarboxylic acid;
2,2,3,3,4,4,4-heptafluorobutanoic
acid

Registered trade name(s)

No data

No data

No data

Chemical formula

C8HF15O2

C8H4F15NO2

C4HF7O2

Chemical structure

F
F
CAS Registry Number

335-67-1

F
F

F
F

F
F

F
F

F
F

F

O

O

F

F

O

F

OH

F

F

F
F

F
F

F

F

F

F

F

F

F

F
F

F

3825-26-1

***DRAFT FOR PUBLIC COMMENT***

375-22-4

F
F

F

O
OH

F

F

 PERFLUOROALKYLS

523
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-1. Chemical Identity of Perfluoroalkyls
Characteristic

Information

Chemical name

Perfluorohexanoic acid

Perfluoroheptanoic acid

Synonym(s)

PFHxA; undecafluoro-1-hexanoic acid; PFHpA; perfluoro-n-heptanoic acid;
hexanoic acid,
tridecafluoro-1-heptanoic acid;
2,2,3,3,4,4,5,5,6,6,6-undecafluoroheptanoic acid, 2,2,3,3,4,4,5,5,6,6,7,7,
7-tridecafluoro-

PFNA; perfluoro-n-nonanoic acid;
perfluorononan-1-oic acid; heptadecafluoro-nonanoic acid;
nonanoic acid,
2,2,3,3,4,4,5,5,6,6,7,7,8,8,
9,9,9-heptadecafluoro-

Registered trade name(s)

No data

No data

No data

Chemical formula

C6HF11O2

C7HF13O2

C9HF17O2

O

Chemical structure
F

OH

F
F

F
F

F
F

F
F
F

CAS Registry Number

307-24-4

F

F
F

F
F

F
F

F
F

375-85-9

***DRAFT FOR PUBLIC COMMENT***

F
F

F

Perfluorononanoic acid

O
OH

F

F
F
F

F
F

375-95-1

F
F

F
F

F
F

F
F

F
F

F

O
OH

F

F

 PERFLUOROALKYLS

524
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-1. Chemical Identity of Perfluoroalkyls
Characteristic

Information

Chemical name

Perfluorodecanoic acid

Perfluoroundecanoic acid

Perfluorododecanoic acid

Synonym(s)

PFDA; PFDeA; Ndfda;
nonadecafluoro-n-decanoic acid;
nonadecafluorodecanoic acid;
perfluoro-n-decanoic acid; decanoic
acid, 2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,
10,10,10-nonadecafluoro-

PFUA; perfluoro-n-undecanoic acid;
henicosafluoroundecanoic acid;
2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,
11,11-heneicosafluoroundecanoic acid

PFDoA; tricosafluorododecanoic
acid; dodecanoic acid,
2,2,3,3,4,4,5,5,6,6,7,7,8,
8,9,9,10,10,11,11,12,12,
12-tricosafluoro-

Registered trade name(s)

No data

No data

No data

Chemical formula

C10HF19O2

C11HF21O2

C12HF23O2

Chemical structure

F
F

CAS Registry Number

335-76-2

F
F

F
F

F
F

F
F

F
F

F
F

F
F

F

O
OH

F

F

F
F

F
F

F
F

F
F

F
F

F
F

2058-94-8

***DRAFT FOR PUBLIC COMMENT***

F
F

F
F

F
F

F

O
OH

F

F

F
F

F
F

F
F

307-55-1

F
F

F
F

F
F

F
F

F
F

F
F

F
F

F

O
OH

F

F

 PERFLUOROALKYLS

525
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-1. Chemical Identity of Perfluoroalkyls
Characteristic

Information

Chemical name

Perfluorooctane sulfonic acid

Perfluorohexane sulfonic acid

Perfluorobutane sulfonic acid

Synonym(s)

PFOS; 1-perfluorooctanesulfonic acid;
heptadecafluoro-1-octanesulfonic acid;
heptadecafluorooctan-1-sulphonic
acid; perfluorooctane sulfonate;
perfluorooctylsulfonic acid; 1-octanesulfonic acid,
1,1,2,2,3,3,4,4,5,5,6,6,7,7,
8,8,8-heptadecafluoro-

PFHxS; perfluorohexane-1-sulphonic
acid; 1-hexanesulfonic acid,
1,1,2,2,3,3,4,4,5,5,6,6,6-tridecafluoro-;
1,1,2,2,3,3,4,4,5,5,6,6,6-tridecafluorohexane-1-sulfonic

PFBuS; 1-perfluorobutanesulfonic
acid; nonafluoro-1-butanesulfonic
acid; nonafluorobutanesulfonic
acid; pentyl perfluorobutanoate;
1,1,2,2,3,3,4,4,4-nonafluoro1-butanesulfonic acid;
1,1,2,2,3,3,4,4,4-nonafluorobutane1-sulphonic acid; 1-butanesulfonic
acid, nonafluoro- (6Cl,7Cl,8Cl)

Registered trade name(s)

No data

No data

No data

Chemical formula

C8HF17O3S

C6HF13O3S

C4HF9O3S

Chemical structure

F
F

CAS Registry Number

F
F

1763-23-1

F
F

F
F

F
F

F
F

F
F

F
F

O
OH
S
O
F

F
F

F
F

F
F

F
F

355-46-4

***DRAFT FOR PUBLIC COMMENT***

F
F

F
F

O
OH
S
O
F

F
F

375-73-5

F
F

F
F

F
F

O
OH
S
O
F

 PERFLUOROALKYLS

526
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-1. Chemical Identity of Perfluoroalkyls
Characteristic

Information

Chemical name

Perflurorooctanesulfonamide

Synonym(s)

PFOSA; perfluoroctylsulfonamide;
Me-PFOSA-AcOH
perfluorooctanesulfonic acid amide;
heptadecafluorooctanesulphonamide;
1-octanesulfonamide, 1,1,2,2,3,3,4,4,
5,5,6,6,7,7,8,8,8-heptadecafluoro-

Et-PFOSA-AcOH

Registered trade name(s)

No data

No data

No data

Chemical formula

C8H2F17NO2S

C11H6F17NO4S

C12H8F17NO4S

Chemical structure

F

F

F

CAS Registry Number

754-91-6

F

F
F

F
F

F
F

F
F

F
F

F
F

O
NH2
S
O
F

2-(N-methyl-perfluorooctane
sulfonamide) acetic acid

CH3 O
O
N
F
OH
S
F
F
O
F
F
F
F
F
F
F
F
F
F
F
F
F
F

2355-31-9

CAS = Chemical Abstracts Services
Sources: Calafat et al. 2007a, 2007b; CAS 2008; ChemIDplus 2008, 2017; RTECS 2008

1

***DRAFT FOR PUBLIC COMMENT***

2-(N-ethyl-perfluorooctane
sulfonamide) acetic acid

H3C
O
N
F
S
F
F
O
F
F
F
F
F
F
F
F
F
F
F
F
F
F

2991-50-6

O
OH

 PERFLUOROALKYLS

527
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-2. Physical and Chemical Properties of Perfluoroalkyls
Property

PFOA

Molecular weight

414.069

Color

No data

Physical state

Solid

APFO
a

431.1

b

No data

d

Solid
a

b

PFBA
214.039

PFHxA
a

314.06c

No data
Liquid

No data

a

No data

Melting point

54.3°C

Decomposition starts -17.5°Ca
above 105°Cb

Boiling point

188°Ca

No data

121°Ca

168°C at 742 mm
Hge

Density at 20°C

1.8 g/cm3f

No data

1.651 g/cm3a

1.789e

Odor

No data

No data

No data

No data

Water

No data

No data

No data

No data

Air

No data

No data

No data

No data

Water

9.5x103 mg/L at
25°Cg

>500 g/Lb

2.14x105 mg/L at
25°Ch

15,700 mg/Li

Organic solvents

No data

No data

Soluble in ethanol
and toluene;
insoluble in
petroleum ethera

No data

Log Kow

Not applicablej

No data

Not applicablej

Not applicablej

Koc

17–230k

No data

No data

No data

pKa

3.8l

No data

0.08 (estimated)m

-0.16i

Vapor pressure

0.017 mm Hg at
0.0081 Pa at 20°Cb
20°C (extrapolated);
0.962 mm Hg at
59.25°C (measured)n

44 mm Hg at 56°Cg

No data

Henry's law constant

0.362 Pa-m3/molh

1.24Pa-m3/molh

No data

No data

Odor threshold:

Solubility:

Partition coefficients:

Autoignition
temperature

Not applicable

o

Flashpoint

No data
No data

Not applicable

o

Not applicableo

Not applicableo

No data

Not applicableo

Not applicableo

Flammability limits

Not applicableo

No data

Not applicableo

Not applicableo

Conversion factors

1 ppm=17.21 mg/m3; 1 ppm=17.63 mg/m3; 1 ppm=8.90 mg/m3; 1 ppm=12.84 mg/m3;
1 mg/m3=0.058 ppmp 1 mg/m3=0.06 ppmp 1 mg/m3=0.11 ppmp 1 mg/m3=0.08 ppmp

Explosive limits

Not applicableo

Not applicableo

Not applicableo

***DRAFT FOR PUBLIC COMMENT***

Not applicableo

 PERFLUOROALKYLS

528
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-2. Physical and Chemical Properties of Perfluoroalkyls
Property

PFHpA

PFNA

c

Molecular weight

364.06

Color

No data

Physical state

No data

PFDeA
514.084

No data

No data

No data

464.08

a

PFUA

c

564.085q

No data

No data

No data

Melting point

r

24–30°C

No data

No data

97.9–100.3°Cr

Boiling point

175°C at 742 mm
Hgg

No data

219°C

No data

Density at 20°C

No data

No data

No data

No data

Odor

No data

No data

No data

No data

Water

No data

No data

No data

No data

Air

No data

No data

No data

No data

Water

4.37x105 mg/L at
25°Ch

No data

No data

No data

Organic solvents

No data

No data

No data

No data

Log Kow

Not applicablej

Not applicablej

Not applicablej

Not applicablej

Koc

No data

No data

No data

No data

-0.15 (estimated)m

-0.17 (estimated)m

-0.17 (estimated)m

Odor threshold:

Solubility:

Partition coefficients:

pKa

-3

4.6 mm Hg at 25 °C 4.83x10 mm Hg at
20°C (extrapolated);
8.4 mm Hg at
99.63°C (measured)n

7.62x10 mm Hg at
20°C (extrapolated);
23.5 mm Hg at
129.56°C
(measured)s

3.44x10-4 mm Hg at
20°C (extrapolated);
4.62 mm Hg at
112.04°C
(measured)s

Henry's law constant
at 25°C

0.573 Pa-m3/molh

No data

No data

No data

Autoignition
temperature

Not applicableo

Not applicableo

Not applicableo

Not applicableo

Flashpoint

Not applicableo

Not applicableo

Not applicableo

Not applicableo

Flammability limits

Not applicableo

Not applicableo

Not applicableo

Not applicableo

Conversion factors

1 ppm=15.14 mg/m3; 1 ppm=19.29 mg/m3; 1 ppm=21.37 mg/m3; 1 ppm=23.45 mg/m3;
1 mg/m3=0.07 ppmp 1 mg/m3=0.05 ppmp 1 mg/m3=0.05 ppmp 1 mg/m3=0.04 ppmp

Explosive limits

Not applicableo

Not applicableo

-4

-0.17 (estimated)m

Vapor pressure

h

Not applicableo

***DRAFT FOR PUBLIC COMMENT***

Not applicableo

 PERFLUOROALKYLS

529
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-2. Physical and Chemical Properties of Perfluoroalkyls
Property

PFDoA

PFOS

c

500.03

PFHxS

c

400.12

PFBuS

c

300.1c

Molecular weight

614.1

Color

No data

No data

No data

No data

Physical state

No data

No data

No data

No data

Melting point

No data

≥400°C (potassium
salt)s

No data

No data

Boiling point

No data

No data

No data

No data

Density at 20°C

No data

No data

No data

No data

Odor

No data

No data

No data

No data

Water

No data

No data

No data

No data

Air

No data

No data

No data

No data

Water

No data

570 mg/L (potassium No data
salt in pure water)s

No data

Organic solvents

No data

No data

No data

No data

Log Kow

Not applicablej

Not applicablej

Not applicablej

Not applicablej

Koc

No data

No data

No data

No data

-0.17 (estimated)m

0.14 (estimated)m

0.14 (estimated)m

0.14 (estimated)m

-60

-6

Odor threshold:

Solubility:

Partition coefficients:

pKa
Vapor pressure

5.11x10 mm Hg at 2.48x10 mm Hg at
20°C (extrapolated)s 20°C (potassium
salt)d

No data

No data

Henry's law constant
at 25°C

No data

No data

No data

No data

Autoignition
temperature

Not applicableo

Not applicableo

Not applicableo

Not applicableo

Flashpoint

Not applicableo

Not applicableo

Not applicableo

Not applicableo

o

o

o

Not applicableo

Flammability limits

Not applicable

Not applicable

Conversion factors

1 ppm=25.53 mg/m3; 1 ppm=20.79 mg/m3; 1 ppm=16.63 mg/m3; 1 ppm=12.48 mg/m3;
1 mg/m3=0.04 ppmp 1 mg/m3=0.05 ppmp 1 mg/m3=0.06 ppmp 1 mg/m3=0.08 ppmp

Explosive limits

Not applicableo

Not applicableo

Not applicable

Not applicableo

***DRAFT FOR PUBLIC COMMENT***

Not applicableo

 PERFLUOROALKYLS

530
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-2. Physical and Chemical Properties of Perfluoroalkyls
Property

PFOSA
c

Me-PFOSA-AcOH

Et-PFOSA-AcOH

571.21 (from structure)

585.24c

Molecular weight

499.15

Color

No data

No data

No data

Physical state

No data

No data

No data

Melting point

No data

No data

No data

Boiling point

No data

No data

No data

Density at 20°C

No data

No data

No data

Odor

No data

No data

No data

Water

No data

No data

No data

Air

No data

No data

No data

Water

No data

No data

No data

Organic solvents

No data

No data

No data

Log Kow

Not applicablej

Not applicablej

Not applicablej

Koc

No data

Odor threshold:

Solubility:

Partition coefficients:
No data
m

3.92 (estimated)

No data
m

3.92 (estimated)m

pKa

6.24 (estimated)

Vapor pressure

No data

No data

No data

Henry's law constant

No data

No data

No data

Autoignition
temperature

Not applicableo

Not applicableo

Not applicableo

Flashpoint

Not applicableo

Not applicableo

Not applicableo

Flammability limits

Not applicableo

Not applicableo

Not applicableo

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

531
4. CHEMICAL AND PHYSICAL INFORMATION

Table 4-2. Physical and Chemical Properties of Perfluoroalkyls
Property

PFOSA

Me-PFOSA-AcOH
3

3

Et-PFOSA-AcOH

Conversion factors

1 ppm=20.75 mg/m ;
1 mg/m3=0.05 ppmp

1 ppm=23.75 mg/m ;
1 mg/m3=0.04 ppmp

1 ppm=24.33 mg/m3;
1 mg/m3=0.04 ppmp

Explosive limits

Not applicableo

Not applicableo

Not applicableo

aLide

2005.
2014.
cEPA 2008c.
d3M 2008c.
eSavu 1994a.
fKroschwitz and Howe-Grant 1994.
gKauck and Diesslin 1951.
hKwan 2001.
iZhao et al. 2014.
jThe log K
ow is not measureable since these substances are expected to form multiple layers in an octanol-water
mixture (3M 1999, 2008c; EPA 2005a).
kPrevedouros et al. 2006.
lBurns et al. 2008.
mSPARC 2008.
nKaiser et al. 2005.
oPerfluorocarboxylates and perfluorosulfonates are nonflammable (3M 1999, Kissa 2001, OECD 2007). However,
they readily degrade via incineration (Krusic and Roe 2004; Krusic et al. 2005; Yamada et al. 2005).
pCalculated using molecular weight.
qChemID Plus 2008.
rKunleda and Shinoda 1976.
s
3M 2000.
bEPA

Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; Me-PFOSA-AcOH = 2-(N-methylperfluorooctane sulfonamide) acetic acid; PFBA = perfluorobutyric acid; PFBuS = perfluorobutane sulfonic acid;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxA = perfluorohexanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide;
PFUA = perfluoroundecanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

532
4. CHEMICAL AND PHYSICAL INFORMATION

Both the potential to form separate layers when mixed with hydrocarbons and water and the propensity
for charged or ionized perfluoroalkyls to concentrate at interfaces make the measurement of the n-octanol
water partition coefficient impractical (3M 1999; EPA 2005a).

With the exception of PFOSA, estimated pKa values for the perfluoroalkyls listed in Table 4-2 range
from -0.17 to 3.92 (SPARC 2008). This pKa range indicates that these substances will exist in anion
form when in contact with water at environmental and physiologically relevant pHs. An estimated pKa of
6.24 indicates that PFOSA will exist as both the anion and the neutral species (SPARC 2008).
Perfluoroalkyl salts, such as APFO, will form the corresponding anions when dissolved in water.
Prevedouros et al. (2006) reported a Krafft point of 22°C and critical micelle concentration of
3.7x103 mg/L for the perfluorooctanoate anion (PFO). At temperatures above the Krafft point, the
solubility of PFO is expected to increase abruptly due to the formation of micelles.

Vapor pressures at 25°C were extrapolated for PFOA, PFNA, PFDeA, PFUA, and PFDoA using Antoine
coefficients. Experimental vapor pressures were as follows: 0.962–724 mm Hg (59.25–190.80°C) for
PFOA; 8.40–750 mm Hg (99.63–203.12°C) for PFNA; 23.5–750 mm Hg (129.56–218.88°C) for PFDeA;
4.62–750 mm Hg (112.04–237.65°C) for PFUA; and 6.42–750 mm Hg (127.58–247.36°C) for PFDoA
(Kaiser et al. 2005).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

533

CHAPTER 5. POTENTIAL FOR HUMAN EXPOSURE
5.1

OVERVIEW

Perfluoroalkyls has been identified in at least 4 of the 1,854 hazardous waste sites that have been
proposed for inclusion on the EPA National Priorities List (NPL) (ATSDR 2017). However, the number
of sites in which substance perfluoroalkyls has been evaluated is not known. The number of sites in each
state is shown in Figure 5-1. Of these sites, 4 are located within the United States.

Figure 5-1. Number of NPL Sites with Perfluoroalkyls Contamination

•
•
•

The general population is exposed to the perfluoroalkyls through food and water ingestion, dust
ingestion, inhalation exposure, and hand-to-mouth transfer of materials containing these
substances.
PFOA, PFOS, and their precursor substances are no longer produced or used in the United States
or most other industrialized nations; however, these substances are persistent in the environment
and exposure near highly contaminated sites may continue to occur.
Serum levels of PFOA and PFOS in the general population of the United States have declined
dramatically since 2000.

Perfluoroalkyls have been released to air, water, and soil in and around fluorochemical facilities located
within the United States (3M 2007b, 2008a, 2008b; Barton et al. 2007; Davis et al. 2007; DuPont 2008;
EPA 2008a). Since the early 2000s, companies in the fluorochemical industry have been working in
concert with the EPA to phase out the production and use of long-chain perfluoroalkyl compounds and
their precursors (3M 2008a; DuPont 2008; EPA 2007a, 2008a, 2016a). Perfluorocarboxylic acids

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

534
5. POTENTIAL FOR HUMAN EXPOSURE

containing seven or more perfluorinated carbon groups and perfluoroalkyl sulfonic acids containing six or
more perfluorinated carbon units are considered long-chain substances. Perfluorinated carboxylic acids
and sulfonic acids containing less than seven and six perfluorinated carbons, respectively, are considered
short-chain substances. The short-chain substances are not as bioaccumulative as the longer-chain
substances such as PFOA and PFOS. PFOA, PFOA precursors, and higher homologues have been
phased out by the eight corporations in the perfluorotelomer/fluorotelomer industry (Arkema, Asahi,
BASF [successor to Ciba], Clariant, Daikin, 3M/Dyneon, DuPont, and Solvay Solexis) as part of the
EPA’s PFOA Stewardship Program (DuPont 2008; EPA 2008a, 2016a). Industrial releases of these
compounds in the United States have declined or been totally eliminated based on company reports
submitted to EPA (EPA 2008a, 2016a). Although the United States and most industrialized nations have
stopped producing PFOA and PFOS, China remains a major producer and user of both substances, and its
production has increased as production in the rest of the world has declined (HAES 2017; Li et al. 2015;
Lim et al. 2011).

Perfluoroalkyl carboxylic acids and sulfonic acids are expected to dissociate in the environment based on
their low pKa values (Kissa 2001; SPARC 2008), and anions will not volatilize from water or soil
surfaces (Prevedouros et al. 2006). The unique surfactant properties of these substances may prevent total
dissociation of perfluoroalkyls in water (EPA 2005a; Kissa 2001; Prevedouros et al. 2006); therefore,
some volatilization of perfluoroalkyls may occur since the neutral forms of these substances are
considered to be volatile (Barton et al. 2007; EPA 2005a; Kim and Kannan 2007). Perfluoroalkyls have
been detected in air both in the vapor phase and as adsorbed to particulates (Kim and Kannan 2007).
Perfluoroalkyls are very stable compounds and are resistant to biodegradation, direct photolysis,
atmospheric photooxidation, and hydrolysis (3M 2000; EPA 2008a; OECD 2002, 2007; Schultz et al.
2003). Perfluoroalkyls released to the atmosphere are expected to adsorb to particles and settle to the
ground through wet or dry deposition (Barton et al. 2007; Hurley et al. 2004; Prevedouros et al. 2006).
The chemical stability of perfluoroalkyls and the low volatility of these substances in ionic form indicate
that perfluoroalkyls will be persistent in water and soil (3M 2000; Prevedouros et al. 2006). Soil
adsorption coefficient data as well as monitoring studies suggest that perfluoroalkyls such as PFOA are
mobile in soil and can leach into groundwater (Davis et al. 2007; Prevedouros et al. 2006).

Perfluoroalkyls have been detected in environmental media and biota of the Arctic region and in other
remote locations such as open ocean waters (Barber et al. 2007; Prevedouros et al. 2006; Wei et al. 2007a;
Yamashita et al. 2005, 2008). Proposed source pathways include long-range atmospheric transport of
precursor compounds followed by photooxidation to form perfluoroalkyls, direct long-range transport of

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

535
5. POTENTIAL FOR HUMAN EXPOSURE

perfluoroalkyls via oceanic currents, and transport of perfluoroalkyls in the form of marine aerosols
(Armitage et al. 2006; Barber et al. 2007; Prevedouros et al. 2006; Wania 2007). Direct transport of
perfluoroalkyls in the atmosphere has also been proposed as a source pathway since these substances have
been detected in the vapor phase in outdoor air samples (CEMN 2008; Prevedouros et al. 2006). The
actual source of perfluoroalkyls in remote locations is likely to be a combination of these pathways.

The highest concentrations of PFOA and PFOS were in apex predators, such as polar bears, which
indicates that these substances biomagnify in food webs (de Vos et al. 2008; Houde et al. 2006b; Kannan
et al. 2005; Kelly et al. 2007; Smithwick et al. 2005a, 2005b, 2006). The bioaccumulation potential of
perfluoroalkyls is reported to increase with increasing chain length (de Vos et al. 2008; Furdui et al. 2007;
Martin et al. 2004b). In living organisms, perfluoroalkyls bind to protein albumin in blood, liver, and
eggs and do not accumulate in fat tissue (de Vos et al. 2008; Kissa 2001).

The levels of PFOA and PFOS in serum samples of U.S. residents have decreased appreciably since the
phase out of these substances in the United States. The geometric mean serum levels of PFOS have
declined over 80% from NHANES survey years 1999–2000 (30.4 ng/mL) to 2013–2014 (4.99 ng/mL)
and the geometric mean serum levels of PFOA have declined almost 70% over the same temporal period,
decreasing from 5.2 ng/mL in years 1999–2000 to 1.94 ng/mL for 2013–2014 (CDC 2018).

Mean concentrations of PFHpA, PFNA, PFDeA, PFUA, PFDoA, PFBuS, PFBA, PFOSA, Me-PFOSAAcOH, and Et-PFOSA-AcOH are generally <1 ng/mL (Calafat et al. 2006b, 2007a, 2007b; CDC 2015;
De Silva and Mabury 2006; Kuklenyik et al. 2004; Olsen et al. 2003a, 2003b, 2004c, 2005, 2007a).
Major PFOS exposure pathways proposed for the general population include food and water ingestion,
dust ingestion, and hand-to-mouth transfer from mill- or home-treated carpets (Trudel et al. 2008). For
PFOA, the major exposure pathways are proposed to be oral exposure resulting from general food and
water ingestion, inhalation from impregnated clothes, and dust ingestion. While migration of residual
PFOA in paper packaging and wrapping into food is also a potential route of exposure (Trudel et al.
2008), polyfluoroalkyl phosphoric acids in food packaging can also be metabolized in the body to PFOA
(D’eon and Mabury 2007; D’eon et al. 2009). Polyfluoroalkyl phosphoric acids are fluorinated surfactant
substances used to greaseproof food-containing paper products. Biotransformation of the
8:2 polyfluoroalkyl phosphoric acid and the 8:2 fluorotelomer alcohol into PFOA has been demonstrated
(D’eon et al. 2009). Based on these proposed exposure pathways, Trudel et al. (2008) estimated that adult
uptake doses for high-exposure scenarios were approximately 30 and 47 ng/kg body weight/day for PFOS
and PFOA, respectively. The estimated dosage for children under the age of 12 under a high-exposure

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

536
5. POTENTIAL FOR HUMAN EXPOSURE

scenario were estimated to be 101–219 and 65.2–128 ng/kg body weight/day for PFOS and PFOA,
respectively (Trudel et al. 2008). Estimated daily doses for the general population were also estimated by
Vestergren et al. (2008) to range from 3.9 to 520 ng/kg body weight/day for PFOS and from 0.3 to
150 ng/kg body weight/day for PFOA. Infants and toddlers had the highest estimated dosages due to
greater hand-to-mouth contact with treated carpeting, mouthing activities of clothes, and greater dust
ingestion. While conversion of precursor compounds to PFOA and PFOS was generally considered as a
minor contribution to the total exposure, under certain scenarios, it was estimated that up to 80% of the
intake could be attributable to exposure to precursor substances followed by subsequent metabolism to
PFOS or PFOA.

Perfluoroalkyls have been detected in human breast milk and umbilical cord blood. The reported
maximum concentrations of PFOS and PFOA measured in human breast milk samples were 0.360–
0.685 and 0.210–0.609 ng/mL, respectively (Kärrman et al. 2007; Llorca et al. 2010; So et al. 2006b;
Völkel et al. 2008). Maximum concentrations of other perfluoroalkyl compounds were <0.18 ng/mL
(Kärrman et al. 2007). PFOS and PFOA have been detected in most umbilical cord blood samples with
reported concentrations of 4.9–11.0 and 1.6–3.7 ng/mL, respectively (Apelberg et al. 2007a, 2007b; Fei et
al. 2007; Inoue et al. 2004; Midasch et al. 2007). Other perfluoroalkyls have been detected less
frequently, with maximum concentrations of <2.6 ng/mL.

Individuals who perform jobs that require frequent contact with perfluoroalkyl-containing products, such
as individuals who install and treat carpets or firefighters, are expected to have occupational exposure to
these substances. Individuals who work at fluorochemical facilities generally have had higher
perfluoroalkyl serum levels than the general population based on exposures in the work environment (3M
2007b, 2008b, 2008c; Barton et al. 2006; Davis et al. 2007). Studies of individuals living near
fluorochemical facilities indicate that drinking water is the major exposure pathway (Emmett et al. 2006a;
Hölzer et al. 2008; Wilhelm et al. 2009). 3M conducted an exposure assessment to estimate the
cumulative exposure to PFOA due to activities at the Decatur, Alabama facility. On-site exposure to
groundskeepers, maintenance workers, construction workers, and on-site trespassers were considered.
Off-site exposures to anglers, boaters, and residential individuals were also estimated. Various plausible
exposure scenarios were considered, and the highest PFOA exposure doses by receptor and pathway
occurred for local residents from groundwater followed by residents consuming drinking water from the
West Morgan/East Lawrence (WM/EL) public drinking water supply (3M 2008c).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

537
5. POTENTIAL FOR HUMAN EXPOSURE

5.2
5.2.1

PRODUCTION, IMPORT/EXPORT, USE, AND DISPOSAL
Production

Perfluoroalkyls been manufactured industrially by electrochemical fluorination (ECF), fluorotelomer
iodide oxidation, fluorotelomer olefin oxidation, and fluorotelomer iodide carboxylation (Prevedouros et
al. 2006; Schultz et al. 2003). During the ECF process, an organic acyl or sulfonyl fluoride backbone
structure is dissolved in a solution of aqueous hydrogen fluoride (Savu 1994b; Siegemund et al. 2005). A
direct electrical current is then passed through the solution, which replaces all of the hydrogens on the
molecule with fluorines. Perfluoroacyl fluorides produced by ECF are hydrolyzed to form the
perfluorocarboxylic acid, which is then separated via distillation. This method was used extensively by
3M in the production of perfluoroalkylsulfonates such as PFOS (3M 1999; Hekster et al. 2003; Schultz et
al. 2003).
RhCOF + HF 
→ RfCOF + H2 + byproducts
RfCOF + H2O 
→ RfCOOH + HF

Perfluoroalkanesulfonyl fluorides produced by ECF are hydrolyzed under alkaline conditions to form the
corresponding salt (Savu 1994b; Siegemund et al. 2005). Acidification followed by distillation yields the
anhydrous perfluoroalkanesulfonic acid.
RhSO2F + HF 
→ RfSO2F + H2
RfSO2F + KOH 
→ RfSO3K + HF
RfSO3K + H2SO4 
→ RfSO3H + KHSO4

Perfluorosulfonamido compounds, such as PFOSA, can be formed by reacting the perfluoroalkanesulfonyl fluoride with a primary or secondary amine (3M 1999; Hekster et al. 2003; Siegemund et al.
2005).

The fluorotelomer iodide oxidation process was developed by DuPont and has served as the basis for their
fluoropolymer production chemistry (Buck et al. 2011; Hekster et al. 2003; Savu 1994a; Siegemund et al.
2005). It begins with the preparation of pentafluoroiodoethane from tetrafluoroethene. Tetrafluoroethene
is then added to this product at a molar ratio that gives a product of desired chain length. Finally, the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

538
5. POTENTIAL FOR HUMAN EXPOSURE

product is oxidized to form the carboxylic acid. The process produces linear perfluorocarboxylic acids of
even carbon numbers as illustrated below.
5C2F4 + IF5 + 2I2 catalyst
→ 5C2F5I
C2F5I + nC2F4 
→ C2F5(C2F4)nI
C2F5(C2F4)n-1CF2COOH
C2F5(C2F4)nI SO
→
3

Fluorotelomer alcohols are created similarly, but the perfluoroalkyl iodide (telomer A) is reacted with
ethylene to create F(CF2)nCH2CH2I (telomer B), which is converted to the alcohol. The ECF process
resulted in a mixture of linear and branched isomers, whereas the telomerization processes yielded
predominantly linear products. It has been reported that the 3M ECF process resulted in approximately
70% linear and 30% branched isomers for PFOS and 78% linear and 22% branched isomers for PFOA
(Benskin et al. 2009).

No information is available in the TRI database on facilities that manufacture, process, or otherwise use
perfluoroalkyls because this class of substances is not required to be reported under Section 313 of the
Emergency Planning and Community Right-to-Know Act (Title III of the Superfund Amendments and
Reauthorization Act of 1986) (EPA 2005a, 2016g). The Chemical Data Reporting (CDR) rule, under the
Toxic Substances Control Act (TSCA), requires manufacturers (including importers) to provide EPA with
information on the production and use of chemicals in commerce in large quantities. Information on
perfluoroalkyls can be found at (https://www.epa.gov/chemical-data-reporting/basic-informationchemical-data-reporting#what).

Perfluoroalkyls have been manufactured for their direct use in commercial products as well as for their
use in industrial process streams. Two important chemicals that have resulted from manufacturing
involving perfluoroalkyls, namely PFOS and PFOA, are of worldwide interest given their detection in
multiple media in the environment. However, these substances and related long-chain perfluoroalkyl
compounds have been essentially phased out as a joint effort by EPA and industry (Lindstrom et al.
2011). The timeline for history of perfluorinated compound production, use, and phase out is presented in
Figure 5-2. Given their unique properties, certain narrow exceptions exist for specific applications.
Additionally, many of the substances that were used in the production of stain resistant or anti-sticking
products that could break down into PFOA and PFOS have also been replaced.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

539
5. POTENTIAL FOR HUMAN EXPOSURE

Figure 5-2. Timeline of Important Events in the History of Polyfluorinated Compounds

EPA = U.S. Environmental Protection Agency; PFC = perfluorinated compound; PFOA = perfluorooctanoic acid; PTFE = polytetrafluoroethylene;
SNUR = significant new use rule; WV = West Virginia
Source: Reprinted (adapted) with permission from Lindstrom et al. 2011 (Environ Sci Technol 45:7954-7961). Copyright 2011 American Chemical
Society.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

540
5. POTENTIAL FOR HUMAN EXPOSURE

The 3M Company was the principal worldwide manufacturer of PFOS and related chemicals. As a result
of its phase-out decision in May 2000, 3M no longer manufactures perfluorooctanyl compounds (PFOA
and PFOS). The company ceased manufacturing and using the vast majority of these compounds within
approximately 2 years of the phase-out announcement, and ceased all manufacturing and the last
significant use of this chemistry by the end of 2008 (3M 2008a; EPA 2007a). In 2000, EPA finalized the
significant new use rule (SNUR) for 88 perfluoroalkyl sulfonate compounds, which requires
manufacturers to notify EPA 90 days prior to commencing manufacture or import of these substances for
a significant new use to allow time for evaluation (EPA 2002, 2007a, 2008a). The purpose of this rule
was to limit future manufacturing and importation of these substances. According to EPA, the rule
allowed for the continuation of a few limited, highly technical uses for which no alternatives are
available, and which are characterized by very low volume, low exposure, and low releases. The SNUR
was amended in 2007 to include 183 additional perfluoroalkyl sulfonate compounds (EPA 2007a, 2008a).

Included on the current list are PFOS, PFHxS, PFOSA, and Et-PFOSA-AcOH. EPA believed that the
perfluoroalkyl sulfonate compounds listed under the SNUR were no longer manufactured in the United
States; however, during the comment period of the 2007 amendment, EPA learned of the ongoing use of
tetraethylammonium perfluorooctanesulfonate as a fume/mist suppressant in metal finishing and plating
baths (EPA 2007a). EPA has since excluded this from the list of significant uses. This rule has been
amended again by the EPA to designate the processing, use, or importation of long-chain perfluoroalkyls
as a significant new use if there are no current ongoing uses, or for uses that were scheduled to end
December 31, 2015 (EPA 2015). As part of this amendment, EPA proposed to amend a SNUR for
perfluoroalkyl sulfonate chemical substances that would make the exemption inapplicable for persons
who import perfluoroalkyl sulfonate chemical substances as part of carpets or any articles that contain
long-chain perfluoroalkyls.

In 2006, the eight major companies of the perfluoropolymer/fluorotelomer industry agreed to participate
in EPA's PFOA Stewardship Program (EPA 2008a). All public documents and reports from the PFOA
Stewardship Program may be reviewed at the EPA docket (EPA-HQ-OPPT-2006-0621). This program
included voluntary commitments from these companies to reduce facility emissions and product content
of PFOA and related chemicals on a global basis by 95% no later than 2010, and to work toward
elimination of these substances in products by 2015. Progress reports have been submitted annually
beginning in 2007 and are available from the EPA PFOA Stewardship Program website. Data from these
reports that list the content and percent reduction of PFOA, PFOA precursors, and higher PFOA
homologues in products are listed in Table 5-1. Nonconfidential emission reports from 2014 and 2015

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

541
5. POTENTIAL FOR HUMAN EXPOSURE

obtained from the EPA docket indicate that the companies have met the goals of the program. According
to its 2015 emission report to EPA, Solvay Specialty Polymers (formerly Solvay Solexis) ceased using
PFOA and related higher homologs in 2010 and ceased importation of products containing PFOA
residuals in December 2015 (Jones 2016). DuPont does not use PFOA any longer in the emulsion
polymerization of polytetrafluoroethylene (PTFE), replacing it with a processing aid consisting of the
ammonium salt of 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)-propanoate, also known as GenX (Wang et
al. 2013b). Similarly, 3M and other manufacturers are using various perfluoropolyethers in
fluoropolymer manufacturing and have reformulated surface treatment products to employ short-chain
substances that are not as bioaccumulative as the long-chain perfluoroalkyls. Wang et al. (2013b) provide
a comprehensive review of the newer substances that manufacturers are using as replacements for legacy
perfluoroalkyls in fluoropolymer manufacturing, metal plating, firefighting foams, and other
miscellaneous uses such as food packaging materials.

Table 5-1. Content (ppm) and Percent Reduction of PFOA, PFOA Homologues, or
PFOA Precursors in Products from 2006 and 2013 U.S. Operations of
Fluoropolymer/Fluorotelomer Companies
Dispersions
Company

Chemicals

Content

Other fluoropolymers

Percent
reductiona Content

Telomers

Percent
reductiona Content

Percent
reductiona

2006 (EPA 2008a)
Arkema, Inc.

Asahi Glass
Company

PFOA and
higher
homologues

>500–
1,000

0%

Precursors

Not
Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable applicable

PFOA, PFOA
500–1,570 12%
salts, and higher
homologues
Precursors

Ciba Specialty PFOA
Chemicals
Higher
Corporation
homologues
Precursors
Clariant Inter- Not applicable
national Ltd.

>70–150

0.12

30

Not
Not
applicable applicable

Not
Not
Not
applicable applicable applicable

Not
Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable applicable
0.05 kg

>99%

0.05 kg

>99%

0.05 kg

>99%

0.05 kg

>99%

0.05 kg

>99%

0.05 kg

>99%

0

>99%

0

>99%

0

>99%

Not
Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable applicable

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

542
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-1. Content (ppm) and Percent Reduction of PFOA, PFOA Homologues, or
PFOA Precursors in Products from 2006 and 2013 U.S. Operations of
Fluoropolymer/Fluorotelomer Companies
Dispersions
Company

Chemicals

Content

Other fluoropolymers

Percent
reductiona Content

Telomers

Percent
reductiona Content

Percent
reductiona

Daikin
PFOA
280
34%
2; 300
0%
0.28
America, Inc. Precursors and Not
Not
Not
Not
2,500
higher
applicable applicable applicable applicable
homologues

72%

E.I. DuPont
PFOA, PFOA
de Nemours salts
and Company Direct
precursors

547

50%

3M/Dyneon

PFOA

0

100%

Not
reported

Solvay
Solexis

PFOA and
PFOA salts

600–700

59%

Not
Not
Not
Not
applicable applicable applicable applicable

Higher
homologues

Not
Not
170–200
applicable applicable

Precursors

0

44%

69

80%

246 kg

Not
Not
Not
Not
57 kg
applicable applicable applicable applicable
Not
reported

0%

78%

14%

Not
Not
applicable applicable

Not
Not
applicable applicable

Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable
2013 (EPA 2016a)

Arkema, Inc.

Asahi Glass
Company

BASF (Ciba
Specialty
Chemicals
Corporation)

PFOA and
higher
homologues

Not
100%
applicable

Precursors

Not
Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable applicable

PFOA, PFOA
0
salts, and higher
homologues

100%

>25–75

18

96%

100%

Not
Not
applicable applicable

Negligible

Not
applicable

Precursors

Not
Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable applicable

PFOA and
higher
homologues

Not
100%
applicable

Not
100%
applicable

Not
100%
applicable

Precursors

Not
100%
applicable

Not
100%
applicable

Not
100%
applicable

Clariant Inter- PFOA, PFOA
None
national Ltd. salts, and higher reported
homologues

Not
None
applicable reported

Not
<1 kg
applicable

Not
applicable

Precursors

Not
None
applicable reported

Not
<7 kg
applicable

Not
applicable

None
reported

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

543
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-1. Content (ppm) and Percent Reduction of PFOA, PFOA Homologues, or
PFOA Precursors in Products from 2006 and 2013 U.S. Operations of
Fluoropolymer/Fluorotelomer Companies
Dispersions
Company

Percent
reductiona Content

Percent
reductiona

100%

100%

None
reported

100%

Not
None
applicable reported

Not
None
applicable reported

>99%

1

99%

0

99%

9 kg

>99%

Higher
homologues

1

99%

0

99%

None
reported

99%

Precursors

None
reported

None
reported

None
reported

None
reported

None
reported

98%

100%

0

Not
0
applicable

Content
None
reported

Precursors and None
higher
reported
homologues
E.I. DuPont
PFOA and
de Nemours PFOA salts
and Company

PFOA, PFOA
0
salts, and higher
homologues
Precursors

Solvay
Solexis

Telomers

Percent
reductiona Content

Chemicals

Daikin
PFOA
America, Inc.

3M/Dyneon

Other fluoropolymers

None
reported

No
telomere
production

No
No
No
No
No
No
precursor precursor precursor precursor precursor precursor
production production production production production production

PFOA, PFOA
Not
>99.999% Not
>99.999% Not
>99.999%
salts, and higher applicable
applicable
applicable
homologues
Precursors

Not
Not
Not
Not
Not
Not
applicable applicable applicable applicable applicable applicable

aPercent

reduction in product content of these compounds from baseline year levels. The baseline year is the year
nearest to the year 2000 for which company data are available.
PFOA precursors include: Octane, 1,l,1,2,2,3,3,4,4,5,5,6,6,7,7, 8,8-heptadecafluoro-8-iodo- (CAS 507-63-1);
1-Decanol, 3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-heptadecafluoro-(CAS 678-39-7); 1-Decene,
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-heptadecafluoro- (CAS 21652-58-4); 2-Propenoic acid,
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-heptadecafluorodecyl ester (CAS 27905-45-9); 2-Propenoic acid, 2-methy 1-,
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-heptadecafluorodecyl ester (CAS 1996-88-9); 2-Decenoic acid,
3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-hexadecafluoro- (CAS 70887-84-2); and Decanoic acid, 3,3,4,4,5,5,6,6,7,
7,8,8,9,9,10,10,10-heptadecafluoro- (CAS 27854-31-5).

Higher homologues include: Dodecane, 1,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12-pentacosafluoro12-iodo (CAS 307-60-8); Decane, 1,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10-heneicosafluoro-10-iodo(CAS 423-62-1); Nonanoic acid, heptadecafluoro- (CAS 375-95-1); Decanoic acid, nonadecafluoro- (CAS 335-76-2);
1-Decanol, 3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-heptadecafluoro- (CAS 678-39-7); Decane,
1,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8-heptadecafluoro-10-iodo- (CAS 2043-53-0); Dodecane,
1,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10-heneicosafluoro-12-iodo- (CAS 2043-54-1); 2-Propenoic acid,
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-heptadecafluorodecyl ester (CAS 4980-53-4); and 2-Propenoic acid, 2-methyl-,
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,12-heneicosafluorododecyl ester (CAS 17741-60-5).
CAS = Chemical Abstracts Service; PFOA = perfluorooctanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

544
5. POTENTIAL FOR HUMAN EXPOSURE

China is one of the few remaining producers and consumers of PFOA and its salts, with a total of
480 metric tons produced from 2004 to 2012 (Li et al. 2015). China also continues to be a producer of
PFOS. Growth in production volumes in China have coincided with decreases in production in the west.
For example, China produced approximately 30 metric tons of PFOS in 2001; however, as 3M ceased
production, China’s production of PFOS increased to 91, 165, and 247 metric tons in 2004, 2005, and
2006, respectively (Lim et al. 2011). PFOA production in China was approximately 30 metric tons in
2004, but increased to approximately 90 metric tons in 2012 (Li et al. 2015). According to a report from
the Hubei Academy of Environmental Sciences (HAES), China is planning to gradually phase out the
production of some PFOS uses before 2019 and conduct a best available technology (BAT)/best
environmental practice (BEP) analysis with the ultimate goal of completely phasing out the production
and use of PFOS and potential precursors (HAES 2017).

Historical U.S. production volume data for PFOA, PFBA, and PFOS reported by manufacturers under the
EPA Inventory Update Rule (IUR) are provided in Table 5-2. Production volume ranges for the
ammonium salt of PFOA, APFO, are also listed. During the reporting year 2002, manufacturers reported
that the production volumes were within the range of 10,000–500,000 pounds (6–227 metric tons) for
PFOS and PFOA and within the range of 500,000–1,000,000 pounds (227–454 metric tons) for APFO
(EPA 2008b). PFBA was reported as having a production volume within the range of 10,000–
500,000 pounds (6–227 metric tons) during 1986; however, PFBA production volumes were not reported
for subsequent years (EPA 2008b). None of the other perfluoroalkyl compounds were listed in EPA’s
IUR database. Current U.S. production volume data for perfluoroalkyl compounds are limited. The IUR
database has been superseded by the Chemical Data Reporting (CDR) database. Data for 2012 indicated
that PFOA was not imported into the United States, but any use or production volume data were reported
as confidential business information. No data were located in the CDR for the other substances listed in
Table 5-2. Nonconfidential emission reports from 2015 obtained from the EPA docket indicate that there
is no current production of PFOA or PFOS in the United States.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

545
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-2. U.S. Production Volume Ranges for Perfluoroalkyls (1986–2002)
Reported under the EPA Inventory Update Rule
Reporting year production volume range (pounds)

Perfluoroalkyl

1986

PFOA

10,000–500,000 Not reported

APFO

10,000–500,000 10,000–500,000 10,000–500,000 10,000–500,000 500,000–1,000,000

PFBA

10,000–500,000 Not reported

Not reported

PFOS

Not reported

10,000–500,000 Not reported

1990

1994

Not reported

1998

2002

10,000–500,000 10,000–500,000 10,000–500,000
Not reported

Not reported
10,000–500,000

APFO = ammonium perfluorooctanoate; EPA = Environmental Protection Agency; PFBA = perfluorobutyric acid;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid
Source: EPA 2008b

5.2.2

Import/Export

The SNURs cited in Section 5.2.1 severely limits the production, import, or export of the long-chain
perfluoroalkyls. There were no reported imports of chemicals listed in Table 4-1 in 2012 from the CDR
database. Production volumes and import volumes of the 8:2 fluorotelomer alcohol were listed as
confidential business information by DuPont. Shaw Industries Group reported that they imported 37,478
pounds of 6:2 fluorotelomer alcohol into the United States in 2012, but DuPont declared both production
volume and import volumes as confidential business information (EPA 2016d).
5.2.3

Use

Applications of perfluoroalkyl compounds have made use of their unique surfactant properties (Schultz et
al. 2003). The alkyl tails of perfluoroalkyls make these substances both hydrophobic (water-repelling)
and oleophobic (oil-repelling) (3M 1999; Kissa 2001; Schultz et al. 2003). Because of these properties,
perfluoroalkyls have been used extensively in surface coating and protectant formulations (Kissa 2001).
Major applications have included protectants for paper and cardboard packaging products, carpets, leather
products, and textiles that enhance water, grease, and soil repellency (Hekster et al. 2003; Schultz et al.
2003). These compounds have been widely used in industrial surfactants, emulsifiers, wetting agents,
additives, and coatings as well (3M 1999; Schultz et al. 2003). Perfluoroalkyls have been used in firefighting foams since they are effective in extinguishing hydrocarbon fueled fires (Schultz et al. 2003).
Perfluoroalkyls have also been used as processing aids in the manufacture of fluoropolymers such as
nonstick coatings on cookware, membranes for clothing that are both waterproof and breathable, personal

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

546
5. POTENTIAL FOR HUMAN EXPOSURE

care products (such as dental floss, cosmetics, sunscreens), electrical wire casing, fire and chemical
resistant tubing, and plumbing thread seal tape (DuPont 2008; EPA 2008a).
5.2.4

Disposal

Information concerning disposal of individual perfluoroalkyl products may be found on Material Safety
Data Sheets (MSDS) or Safety Data Sheets (SDS) from the manufacturers of the chemicals. Two
methods are generally recommended for the disposal of fluoropolymer dispersions. The first method
involves precipitation, decanting, or filtering to separate solids from liquid waste. The dry solids are then
disposed of in an approved industrial solid waste landfill or incinerated, while the liquid waste is
discharged to a waste water treatment facility (Plastics Europe 2012). The second method involves
incineration at temperatures >800°C using a scrubber to remove hydrogen fluoride (Plastics Europe
2012). According to perfluorochemical facility assessment reports, historical disposal of perfluoroalkyl
containing waste has been through on- and off-site landfills, through sludge incorporation (subsurface
injection), and through incineration (3M 2007b, 2008a; ATSDR 2005). Pilot scale studies in which
carpet samples were incinerated using a rotary kiln incinerator indicated that most perfluoroalkyls were
effectively destroyed in combustors (Lemieux et al. 2007). Similar conclusions were reached by Yamada
et al. (2005) when studying the incineration of textiles and paper treated with fluorotelomer-based acrylic
polymers. Incineration at conventional temperatures is a proven technology for treating wastes containing
perfluoroalkyls.

5.3

RELEASES TO THE ENVIRONMENT

There is no information listed in EPA’s Toxic Release Inventory (TRI) on releases of perfluoroalkyls to
the environment from facilities manufacturing, processing, or otherwise using perfluoroalkyls because
these releases are not required to be reported within this program (EPA 2005b, 2016g).

Perfluoroalkyls are man-made compounds that are not naturally occurring in the environment.
Perfluoroalkyls such as PFOS and PFOA have been widely used in the manufacturing of many consumer
products (Hekster et al. 2003; Schultz et al. 2003). These substances are still detected in both
environmental and biological media around the world as well as in serum samples collected from the
general population (Calafat et al. 2006b, 2007a, 2007b; CDC 2018; De Silva and Mabury 2006;
Kuklenyik et al. 2004; Olsen et al. 2003b, 2003c, 2004b, 2004c, 2005, 2007a; Prevedouros et al. 2006).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

547
5. POTENTIAL FOR HUMAN EXPOSURE

In 2006, the eight major companies of the perfluoropolymer/perfluorotelomer industry agreed to
participate in EPA's PFOA Stewardship Program (EPA 2008a). This included voluntary commitments
from these companies to reduce facility emissions and product content of PFOA and related chemicals on
a global basis by 95% no later than 2010, and to work toward elimination of these substances by 2015
(EPA 2008a). Data from 2007 and 2013 progress reports regarding releases of PFOA, PFOA precursors,
and higher PFOA homologues to all media as well as percent reduction in releases are listed in Table 5-3.

Table 5-3. Reported Emissions of PFOA, PFOA Homologues, or PFOA
Precursors in Products from the 2006 and 2013 U.S. Operations of
Fluoropolymer/Fluorotelomer Companies
Releases to all media from fluorotelomer and telomer manufacturing
Company

Chemicals

kg

kg of release/100 kg of Percent reduction
product produced
in emissionsa

2006 Data (EPA 2008a)
Arkema, Inc.

Asahi Glass
Company

BASF (Ciba
Specialty
Chemicals
Corporation)

PFOA and higher
homologues

>1,000–10,000

For fluorotelomer
production: >0.1–1

22%

Precursors

Not applicable

Not applicable

Not applicable

PFOA, PFOA salts, 4,922
and higher
homologues

For fluorotelomer
production: <1

6%

Precursors

Not applicable

Not applicable

Not applicable

PFOA

0.05b

>99%

Higher homologues 0.05b

>99%

0b

Precursors

>99%

Clariant
Not applicable
International Ltd.

Not applicable

Not applicable

Daikin America,
Inc.

Confidential
business
information

For fluorotelomer
94% for FP
production: 8.0x10-3; for production; 92% for
telomer production:
telomer production
6.4x10-7

E.I. DuPont de
Nemours and
Company

PFOA

Not applicable

Precursors and
Confidential
higher homologues business
information

For production: 6.4x10-7 22% for telomer
production

PFOA, PFOA salts

1,100

Not reported

98%

Direct precursors

Confidential
business
information

Not reported

Confidential
business
information

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

548
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-3. Reported Emissions of PFOA, PFOA Homologues, or PFOA
Precursors in Products from the 2006 and 2013 U.S. Operations of
Fluoropolymer/Fluorotelomer Companies
Releases to all media from fluorotelomer and telomer manufacturing
Company

Chemicals

kg

kg of release/100 kg of Percent reduction
product produced
in emissionsa

3M/Dyneon

PFOA

0

0

100%

Solvay Solexis

PFOA and PFOA
salts

Not applicable

Not applicable

Not applicable

Higher homologues >1,000–10,000

For fluorotelomer
production: 0.161

28%

Precursors

Not applicable

Not applicable

Not applicable

2013 data (EPA 2016a)
Arkema, Inc.

PFOA and higher
homologues

>500–2,000

>0.001–0.005

91%

Precursors

Not applicable

Not applicable

Not applicable

PFOA, PFOA salts, 0
and higher
homologues

Not applicable

100%

Precursors

Not applicable

Not applicable

Not applicable

PFOA

Not applicable

Not applicable

100%

Precursors

Not applicable

Not applicable

100%

Not applicable

Not applicable

Not applicable

Precursors

Not applicable

Not applicable

Not applicable

Daikin America,
Inc.

PFOA

0

0

100%

E.I. DuPont de
Nemours and
Company

Precursors and
0
higher homologues

0

100%

PFOA, PFOA salts

None reported

99.8%

Higher homologues None reported

None reported

None reported

Precursors

None reported

Confidential
business
information

PFOA, PFOA salts, 0
and higher
homologues

0

100%

Precursors

0

Not applicable

Asahi Glass
Company

BASF (Ciba
Specialty
Chemicals
Corporation)

Clariant
PFOA, PFOA salts
International Ltd.

3M/Dyneon

90
Confidential
business
information

0

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

549
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-3. Reported Emissions of PFOA, PFOA Homologues, or PFOA
Precursors in Products from the 2006 and 2013 U.S. Operations of
Fluoropolymer/Fluorotelomer Companies
Releases to all media from fluorotelomer and telomer manufacturing
kg

kg of release/100 kg of Percent reduction
product produced
in emissionsa

Company

Chemicals

Solvay Solexis

PFOA, PFOA salts, 0
and higher
homologues

0

>99.999%

Precursors

0

Not applicable

0

aPercent

reduction in product content of these compounds from baseline year levels. The baseline year is the year
nearest to the year 2000 for which company data are available.
bTotal for emissions and product content
PFOA = perfluorooctanoic acid

While the United States and most industrialized countries around the world have ceased production of
PFOS and PFOA, China is still a major producer of both substances (Li et al. 2015; Lim et al. 2011).
Over the period from 2004 to 2012, it was estimated that 250 metric tons of PFOA were released to the
environment from production in China (Li et al. 2015). Fluoropolymer manufacturing and processing
was considered the dominant source of environmental releases, accounting for >80% of the total, while
PFOA releases related to end use consumer products accounted for 6% of the total.

Prevedouros et al. (2006) estimated the total global historical emissions of perfluoroalkyl carboxylates
into the environment from both direct and indirect sources from the time period of 1951–2004. These
data are provided in Table 5-4. Based on these estimations, direct emissions (3,200–6,900 metric tons)
have far exceeded indirect emissions (30–350 metric tons). The largest direct emissions identified are
from industrial processes such as the manufacture of perfluoroalkyl carboxylates (470–900 metric tons),

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

550
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-4. Global Historical PFCA Production and Emissions Estimates from
1951 to 2004a

Historical time
period (years)

Estimated total global
historical PFCA emissions
(tonnes)

PFO/APFO

1951–2004

400–700

PFN/APFN

1975–2004

70–200

Environmental input source
Direct PFCA sources
PFCA manufacture

Total manufactured

470–900

Industrial and consumer uses
Fluoropolymer manufacture (APFO)

1951–2004

2,000–4,000

Fluoropolymer dispersion processing (APFO)

1951–2004

200–300

Fluoropolymer manufacture (APFN)

1975–2004

400–1,400

Fluoropolymer processing (APFN)

1975–2004

10–20

Aqueous firefighting foams (AFFF)

1965–1974

50–100

Consumer and industrial products

1960–2000

40–200

Total direct

3,200–6,900

Indirect PFCA sources
POSF-based products
PFCA residual impuritiesb

1960–2002

20–130

POSF-based precursor degradation

1960–2002

1–30

POSF-based AFFF

1970–2002

3–30

PFCA residual impuritiesb

1974–2004

0.3–30

Fluorotelomer-based precursor degradation

1974–2004

6–130

Fluorotelomer-based AFFF

1975–2004

<1

Fluorotelomer-based products

Total indirect

30–350

Total source emissions (direct and indirect)

3,200–7,300

aLow

and high estimated values as well as the period of use/production for each source are based upon publicly
available information cited in the text.
bSome authors classify residual impurities as a direct emission source rather than indirect emission source (Buck et
al. 2011).
AFFF = aqueous firefighting foams; APFN = ammonium perfluorononanoate; APFO = ammonium perfluorooctanoate; PFCA = perfluorinated carboxylic acid; PFN = perfluorononanoate; PFO = perfluorooctanoate;
POSF = perfluorooctanesulfonyl fluoride
Source: Prevedouros et al. 2006

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

551
5. POTENTIAL FOR HUMAN EXPOSURE

fluoropolymer manufacture (2,400–5,400 metric tons), and fluoropolymer processing (210–320 metric
tons). Direct release of perfluoroalkyl carboxylates from use of aqueous firefighting foams and consumer
and industrial products were estimated to be 50–100 and 40–200 metric tons, respectively. The largest
indirect emissions identified were from perfluoroalkyl carboxylate residual impurities in
perfluorooctylsulfonyl fluoride products (20–130 metric tons) and fluorotelomer-based precursor
degradation (6–130 metric tons). Wang et al. (2014) expanded upon the work of Prevedouros et al.
(2006) by considering additional emission sources of these substances and estimating emissions from
2003 to 2015 and projecting future emissions. These authors estimated emissions of 820–7,180 metric
tons for 2003–2015 and projected between 20 and 6,420 metric tons for years 2016–2030. The estimates
by Prevedouros et al. (2006) and Wang et al. (2014) contain a great degree of uncertainty as demonstrated
by the wide range of values presented in the data. Wang et al. (2014) stated that uncertainty analysis
using Monte Carlo methods is not possible because there is insufficient information available with respect
to the range or distribution of the emissions. Instead, they introduced a scoring system to provide a
qualitative description of the accuracy of the estimates that ranged from 0–1 (low uncertainty) to 2–3
(high uncertainty where estimates were based on crude assumptions or extrapolations).
5.3.1

Air

There is no information listed in the TRI on releases of perfluoroalkyls to the atmosphere from facilities
manufacturing, processing, or otherwise using perfluoroalkyls because these releases are not required to
be reported (EPA 2005b, 2016g).

According to 3M, low levels of PFOA were released to air during manufacturing processes at the Decatur,
Alabama facility until use of this substance ceased in 2004 (3M 2008b). This company states that there
are currently no process-related air emissions of PFOA at this facility (3M 2008b). PFOA concentrations
as high as 75,000–900,000 pg/m3 were measured at the fence line of the DuPont Washington Works
facility near Parkersburg, West Virginia in 2004 (Barton et al. 2006; Davis et al. 2007; Prevedouros et al.
2006). High volume air samples collected at several monitoring stations near the Washington Works
facility during nine events between August and October of 2005 contained PFOA at reported
concentrations ranging from 10 to 75,900 pg/m3 (EPA 2007b). The mean and median of these reported
concentrations are 5,500 and 240 pg/m3.

The presence of perfluoroalkyl compounds in indoor air and dust indicates that perfluoroalkyl-containing
consumer products such as treated carpets and textiles may be sources of release to air (Barber et al. 2007;
Jahnke et al. 2007b; Kubwabo et al. 2005; Moriwaki et al. 2003; Prevedouros et al. 2006; Shoeib et al.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

552
5. POTENTIAL FOR HUMAN EXPOSURE

2004; Strynar and Lindstrom 2008). Perfluoroalkyl compounds have also been identified on both indoor
and outdoor window films (Gewurtz et al. 2009). Disposal of perfluoroalkyl-containing consumer
products is also expected to be a source of release to air (Prevedouros et al. 2006). Harada et al. (2005a,
2006) proposed that automobiles may be a source of PFOA in urban air based on elevated levels
measured near heavy traffic areas and the widespread use of this substance in automobile materials.

Perfluoroalkyl carboxylic acids and perfluoroalkyl sulfonic acids are formed by the atmospheric
photooxidation of precursor compounds such as fluorotelomer alcohols and perfluoroalkyl sulfonamides
(D'eon et al. 2006; Ellis et al. 2004; Martin et al. 2006; Wallington et al. 2006; Wania 2007).
Perfluoroalkyl carboxylic acids including PFOA, PFNA, PFHpA, and PFBA were observed as products
during a laboratory study involving the photooxidation of 4:2, 6:2, and 8:2 fluorotelomer alcohols (Ellis et
al. 2004). D'eon et al. (2006) observed both perfluoroalkyl carboxylic acids and perfluorobutane
sulfonate among products of the photooxidation of N-methyl perfluorobutane sulfonamidoethanol.

5.3.2

Water

There is no information listed in the TRI on releases of perfluoroalkyls to water from facilities
manufacturing, processing, or otherwise using perfluoroalkyls because these releases are not required to
be reported (EPA 2005b, 2016g).

There are a number of sources of perfluoroalkyl release to surface water and groundwater, including
release from manufacturing sites, industrial use, use and disposal of perfluoroalkyl-containing consumer
products, fire/crash training areas, waste water treatment facilities, and from the use of contaminated
biosolids (3M 2008b; Clara et al. 2009; Davis et al. 2007; Eggen et al. 2010; EPA 2009a; Kelly and
Solem 2009; Moody and Field 1999; Moody et al. 2003; Sinclair and Kannan 2006; Prevedouros et al.
2006).

Waste water discharge was identified as a release pathway for APFO from the DuPont Washington
Works facility in West Virginia (Davis et al. 2007). The average monthly concentrations of APFO
measured in surface water from three outlets at the facility during 2007 and early 2008 ranged from
3.65 to 377 µg/L (EPA 2008d). Reported concentrations of APFO and PFOA measured in surface water
from four separate outlets at this facility during the same period were 3–64 and 2.3–61 µg/L, respectively.
During perfluorochemical operations at the 3M Cottage Grove facility in Minnesota, waste water
treatment plant effluent containing perfluoroalkyl compounds was discharged to the Mississippi River.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

553
5. POTENTIAL FOR HUMAN EXPOSURE

Discharge into Bakers Creek from the waste water treatment plant at the 3M Decatur facility was a
principal source of PFOA release from this facility (3M 2008b).

Elevated levels of perfluoroalkyls, such as PFOA, PFOS, and PFHxS, measured in groundwater near firetraining areas are attributed to the use of these substances in aqueous firefighting foams (Moody and Field
1999; Moody et al. 2003). The concentrations of these three perfluoroalkyls in groundwater near a
military fire-training site in Michigan were 8–105, 4.0–110, and 9–120 µg/L, respectively (Moody et al.
2003). A study of landfill leachates showed that perfluoroalkyls were primarily distributed to the water
phase of leachates, which could eventually contaminate ground water (Eggen et al. 2010).

Waste water treatment plants have been shown to be significant contributors to perfluoroalkyls
contamination of surface and ground water (Clara et al. 2009; EPA 2009a; Kelly and Solem 2009;
Loganathan et al. 2007; Sinclair and Kannan 2006; Yu et al. 2009c). Influent, effluent, and sludge
samples from 28 public and private waste water treatment plants in Minnesota were analyzed for
13 perfluoroalkyl compounds; detectable concentrations of perfluoroalkyl compounds were found in
several facilities, primarily urban treatment plants (Kelly and Solem 2009). Elevated levels of PFOS at
one facility (1.51 µg/L in effluent) were attributed to a chrome plating facility using a surfactant
containing fluorosulfonate to control hexavalent chromium emissions. Another study of chromium
electroplating facilities in Chicago, Illinois and Cleveland, Ohio also found them to be significant sources
of PFOS and other perfluoroalkyls in the environment (EPA 2009a). It was determined that
perfluoroalkyls were being discharged from all 11 facilities at quantifiable levels and that PFOS was
detected in waste water from 10 out of 11 facilities at levels of 0.0314–39 µg/L. PFOA and PFOS were
detected in effluents of six waste water treatment plants located in New York at levels of 0.058–1.05 and
0.003–0.068 µg/L, respectively (Sinclair and Kannan 2006). PFOS and PFOA were detected in effluents
of two waste water treatment plants located in Singapore at levels of 0.0053–0.5609 and 0.0112–
1.057 µg/L, respectively (Yu et al. 2009c). PFOA, PFOS, and several other perfluoroalkyls were detected
in effluent samples of 21 waste water treatment plants and 9 industrial point sources; PFOA and PFOS
were reportedly identified in the effluents of all of the facilities monitored at an average level of 0.060
µg/L for both substances (Clara et al. 2009).

Studies comparing perfluoroalkyl levels in influent and effluent from municipal waste water treatment
facilities have found higher levels of some perfluoroalkyls, such as PFNA, PFOA, PFOS, and PFOSA, in
the effluent, as compared to the influent (Loganathan et al. 2007; Schultz et al. 2006b). For others, such
as PFHxS and PFDeA, waste water treatment resulted in lower concentrations or no change in the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

554
5. POTENTIAL FOR HUMAN EXPOSURE

concentrations. Increases in perfluoroalkyl concentrations are likely due to the breakdown of
perfluoroalkyl precursors such as polyfluoroalkyl phosphoric acids or fluorotelomer alcohols (D’eon et al.
2009; Gauthier and Mabury 2005; Wang et al. 2005a, 2005b).

Land application of biosolids (treated sewage sludge) can also result in the release of perfluoroalkyls to
surface and groundwater (Clark and Smith 2011; Lindstrom et al. 2011; Sepulvado et al. 2011). There
appears to be some differences in the distribution of PFOA and PFOS in waste water effluent and
biosolids, with higher levels of PFOA in waste water and higher PFOS levels in biosolids (Guo et al.
2010).
5.3.3

Soil

There is no information listed in the TRI on releases of perfluoroalkyls to soil from facilities
manufacturing, processing, or otherwise using perfluoroalkyls because these releases are not required to
be reported (EPA 2005b, 2016g).

Perfluoroalkyls can be inadvertently released to soils through the use of biosolids applied as fertilizer to
help maintain productive agricultural soils and stimulate plant growth. PFOA and PFOS were detected in
both biosolids and biosolid-amended soils (Sepulvado et al. 2011). Six samples of biosolids obtained
from the Metropolitan Water Reclamation District of Greater Chicago had levels of PFOS, Me-PFOSA,
Et-PFOSA, and PFOA of 80–219, 63–143, 42–72, and 8–68 ng/g, respectively (Sepulvado et al. 2011).
The mean sum (±SD) of all perfluoroalkyls in the biosolids was 433±121 ng/g, with PFOS being most
prominent. Perfluoroalkyls can also be released into soil due to atmospheric transport and wet/dry
deposition (Rankin et al. 2016; Strynar et al. 2012).

Liu et al. (2007) measured PFOA as a product of the biodegradation of 8:2 fluorotelomer alcohol in soil.
This result, along with similar findings in activated sludge tests, indicates that biodegradation of
fluorotelomer alcohols may result in the formation of perfluoroalkyl carboxylic acids in soil (Liu et al.
2007; Rankin et al. 2014; Wang et al. 2005a, 2005b).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

555
5. POTENTIAL FOR HUMAN EXPOSURE

5.4
5.4.1
Air.

ENVIRONMENTAL FATE
Transport and Partitioning
Barton et al. (2007) investigated the atmospheric partitioning of PFOA during rain events near an

industrial facility and concluded that this substance will be primarily adsorbed to particles in the air since
PFOA was not detected in the vapor phase (detection limit of 0.2 ng/m3). Concentrations of PFOA in
raindrops and as particulates were 11.3–1,660 ng/L and 0.09–12.40 ng/m3, respectively. The authors
proposed that PFOA or APFO released into air from industrial facilities will be scavenged by atmospheric
particles (including aqueous aerosols and raindrops) and dissociate to form the perfluorooctanoate anion.
Although Barton et al. (2007) did not detect PFOA in the vapor phase during rain events, low
concentrations (<0.12–3.16 pg/m3) of vapor-phase perfluoroalkyl compounds measured by Kim and
Kannan (2007) in urban air provide evidence of a partitioning equilibrium. Wet and dry deposition are
expected to be the principal removal mechanisms for perfluoroalkyl carboxylic acids and sulfonic acids in
particulate form from the atmosphere. Residence times with respect to these processes are expected to be
days to weeks (Barton et al. 2007; Hurley et al. 2004; Kim and Kannan 2007).

Long-range atmospheric transport of precursor compounds such as fluorotelomer alcohols and
perfluoroalkyl sulfonamides followed by the atmospheric photooxidation of these substances to form
perfluoroalkyl carboxylic acids and perfluoroalkyl sulfonic acids resulted in PFOA and PFOS
contamination in remote locations with no direct point sources for these compounds (Barber et al. 2007;
D'eon et al. 2006; Dinglasan-Panlilio and Mabury 2006; Ellis et al. 2004; Martin et al. 2006; Simcik
2005; Small 2009; Wallington et al. 2006; Wania 2007). Fluorotelomer alcohols and perfluoroalkyl
sulfonamides are volatile and possess long enough atmospheric residence times for long-range transport
to occur (Barber et al. 2007; Yarwood et al. 2007). The presence of fluorotelomer alcohols and
perfluoroalkyl sulfonamides in urban and Arctic air offers evidence of long-range atmospheric transport
(Loewen et al. 2005; Shoeib et al. 2006; Stock et al. 2004). Photooxidation studies have demonstrated the
conversion of these substances to perfluoroalkyl carboxylic acids and sulfonates. According to Young et
al. (2007), the presence of perfluorodecanoic acid and perfluoroundecanoic acid in an Arctic ice cap
indicates atmospheric oxidation as a source.
Water. With the exception of PFOSA, estimated pKa values for the perfluoroalkyls discussed in this
profile range from -0.17 to 3.92 (SPARC 2008). This pKa range indicates that these substances will exist
primarily as the dissociated conjugate base (anion) when in contact with water at environmental pH
(pH 5–9). Volatilization will not be an important environmental fate process when the substances exist as

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

556
5. POTENTIAL FOR HUMAN EXPOSURE

anions; however, under acidic conditions, undissociated perfluoroalkyls may volatilize into the
atmosphere (Martin et al. 2006). Perfluoroalkyls may be transported to remote areas by direct oceanic
advection of these substances (Armitage et al. 2006; Barber et al. 2007; Simcik 2005; Wania 2007;
Yamashita et al. 2005, 2008). Perfluoroalkyls may also be transported over long distances in the form of
marine aerosols (Barber et al. 2007; CEMN 2008; Prevedouros et al. 2006). This transport mechanism
may be especially relevant since surfactants have been shown to accumulate in upper sea layers and at
water surfaces (Prevedouros et al. 2006).

Perfluoroalkyl compounds have been measured in invertebrates, fish, amphibians, reptiles, birds, bird
eggs, and mammals located around the world (Dai et al. 2006; Giesy and Kannan 2001; Houde et al.
2005, 2006a, 2006b; Keller et al. 2005; Kannan et al. 2001a, 2001b, 2002a, 2002b, 2002c, 2002d, 2005,
2006; Sinclair et al. 2006; So et al. 2006a; Wang et al. 2008). The highest concentrations of PFOA and
PFOS in animals are measured in apex predators, such as polar bears, which indicates that these
substances biomagnify in food webs (de Vos et al. 2008; Houde et al. 2006b; Kannan et al. 2005; Kelly et
al. 2007). Table 5-5 shows levels of PFOA and PFOS measured in Arctic organisms. The
bioaccumulation potential of perfluoroalkyls increases with increasing chain length from 4 to 8 carbon
units and then declines with further increases in chain length (Conder et al. 2008; de Vos et al. 2008;
Furdui et al. 2007; Martin et al. 2004b). In living organisms, perfluoroalkyls bind to protein albumin in
blood, liver, and eggs and do not accumulate in fat tissue, which may explain why bioconcentration
factors (BCFs) are lower than expected in aquatic organisms (de Vos et al. 2008; Kissa 2001).

Table 5-5. Biological Monitoring of PFOA and PFOS in the Arctic
Concentration (ng/g)
Location and organism

PFOA

Northeastern Canada, 1996–2002; wet
Zooplankton (n=5)
Clams (n=5)

PFOS

weighta
2.6

ND

Tomy et al. 2004
1.8
0.28

Shrimp (n=7)

0.17

0.35

Arctic cod (n=6)

0.16

1.3

Redfish (n=7)

1.2

1.4

Walrus (n=5)

0.34

2.4

Narwhal (n=5)

0.9

10.9

Beluga (n=5)

1.6

12.6

Black-legged kittiwake (n=4)
Glaucous gulls (n=5)

ND

10.0
0.14

Reference

20.2

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

557
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-5. Biological Monitoring of PFOA and PFOS in the Arctic
Concentration (ng/g)
Location and organism

PFOA

PFOS

Northern Canada, 1992–2002a

Martin et al. 2004a

Polar bear (n=7)

8.6

3,100

Arctic fox (n=10)

<2

250

Ringed seal (n=9)

<2

16

Mink (n=10)

<2

Common loon (n=5)

<2

Northern fulmar (n=5)

<2

Black guillemot (n=5)

<2

White sucker (n=3)

<2

Brook trout (n=2)

<2

39

Lake whitefish (n=2)

<2

12

Lake trout (n=1)

<2

31

Northern pike (n=1)

<2

Arctic sculpin (n=1)

<2

8.7
20
1.3
ND
7.6

5.7
12

Northwestern Canada, 2004

Powley et al. 2008

Zooplankton (n=3)

ND

ND–0.2

Arctic cod (n=5)

ND

0.3–0.7

Ringed seal (n=5)

2.5–8.6

Bearded seal (n=1)
Northern Norway; ng/g wet

Reference

ND

1.3

weighta

Herring gull eggs

Verreault et al. 2005, 2007
<0.091–0.652

21.4–42.2

Glaucous gulls
Eggs (n=10)
Plasma (n=20)

<0.70

104

<0.70–0.74

134

Nanavut, Canada

Butt et al. 2007a, 2007b

Thick-billed murres

<MDLb–0.16

<0.40–0.76

Northern fulmars

<MDLb–0.09

<0.40–0.60

Ringed seals

<0.85–6.2

2–20

Northern Canada, 2002–2005
Ringed seal livers (n=110)

Butt et al. 2008
<0.7–13.9

0.89–189

Greenland

Bossi et al. 2005

Ringed seals

<1.2

12.5–95.6

North American and European Arctic, 1999–2002
Polar bears (n>72)

<2.3–57.1

Smithwick et al. 2005a
263–6,340

Greenland, 1999–2001
Polar bears

(n=29)a

Smithwick et al. 2005b
10

2,470

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

558
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-5. Biological Monitoring of PFOA and PFOS in the Arctic
Concentration (ng/g)
Location and organism

PFOA

PFOS

Greenland, 1972–2002
Polar bears

Reference
Smithwick et al. 2006

1.6–4.4

120–1,400

aReported

as mean values
detection limits for study analytes ranged from 0.03 to 2.3 ng/g. To calculate means, concentrations less
than the MDL were replaced with a random value that was less than half the MDL.

bMinimum

MDL = minimum detection limit; ND = not detected; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic
acid

Sediment and Soil. Koc values of 17–230 measured for PFOA in soils of various organic carbon
content indicate that PFOA will be mobile in soil and will not adsorb to suspended solids and sediment in
the water column (Davis et al. 2007; Prevedouros et al. 2006). This is supported by the presence of
PFOA in groundwater at the Decatur, Cottage Grove, and Washington Works fluorochemical industrial
facilities (3M 2007b, 2008b; Davis et al. 2007). In addition to migration to groundwater from plumes
near industrial facilities, other sources of perfluoroalkyls that may contaminate nearby groundwater
include air emissions followed by atmospheric deposition to soils and subsequent leaching (Davis et al.
2007). Low volatility, high water solubility (9,500 mg/L at 25°C), and low sorption to solids indicate that
the perfluorooctanoate anion will accumulate in surface waters, especially oceans (Armitage et al. 2006;
Kauck and Diesslin 1951; Prevedouros et al. 2006; Wania 2007).

Perfluoroalkyl compounds can be taken up by plants in contaminated soils. Laboratory studies have
suggested that short-chain perfluoroalkyls such as PFBA are more concentrated in edible portions of
plants when compared to longer carbon chain substances such as PFOA or PFOS (Blaine et al. 2013;
2014a, 2014b; MDH 2014). Yoo et al. (2011) studied the accumulation of perfluoroalkyl carboxylic
acids, perfluorosulfonic acids, and fluorotelomer alcohols in grass samples collected near Decatur,
Alabama and calculated the grass-soil accumulation factor (GSAF), which is the concentration of
perfluoroalkyl in grass divided by the concentration of perfluoroalkyl in soil. The shortest chain
compounds had the largest GSAFs, and accumulation factors decreased rapidly with chain length. The
mean (±SD) GSAF values were 3.4±2.6, 0.90±0.66, 0.25±0.23, 0.12±0.08, 0.10±0.08, 0.11±0.09, and
0.10±0.09 for PFHxA, PFHpA, PFOA, PFNA, PFDeA, PFUA, and PFDoA, respectively. The GSAF for
PFOS was 0.07±0.04. Increasing salinity and temperature was shown to increase uptake and transport
from the roots into the shoots by the wheat plants grown in hydroponic systems spiked with
perfluorocarboxylic acids (Zhao et al. 2016). Transport into the shoots also increased with decreasing

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

559
5. POTENTIAL FOR HUMAN EXPOSURE

carbon chain length. Concentrations in the shoots of the wheat plants increased in the following order:
PFBA > PFHpA > PFOA > PFDoA.
Other Media.

Data are not available regarding the transport and partitioning of perfluoroalkyl

compounds in other media.
5.4.2

Transformation and Degradation

Perfluoroalkyl compounds are considered to be environmentally persistent chemicals (EPA 2008a; OECD
2002, 2007; Schultz et al. 2003). The carbon atoms of the perfluoroalkyl chain are protected from attack
by the shielding effect of the fluorine atoms; furthermore, environmental degradation processes generally
do not possess the energy needed to break apart the strong fluorine-carbon bonds (3M 2000; Hekster et al.
2003; Schultz et al. 2003). Perfluoroalkyl compounds are resistant to biodegradation, direct photolysis,
atmospheric photooxidation, and hydrolysis (OECD 2002, 2007; Prevedouros et al. 2006).
Air.

Although transport and partitioning information indicates that air will not be a sink for

perfluoroalkyl compounds in the environment, low concentrations of perfluoroalkyl carboxylic acids,
sulfonic acids, and sulfonamides have been measured in air both in the vapor phase and as bound to
particulates (Barton et al. 2007; Kim and Kannan 2007). Available information indicates that
photodegradation will not compete with wet deposition as an atmospheric removal process for
perfluoroalkyls (Barton et al. 2007; Hurley et al. 2004; Prevedouros et al. 2006). However,
photooxidation may be an important degradation mechanism for perfluoroalkyl sulfonamides (D'eon et al.
2006; Martin et al. 2006).

PFOA does not absorb UV light at environmentally relevant wavelengths (>290 nm); Hori et al. (2004a)
reported a weak absorption band for PFOA that ranged from 220 to 270 nm. Based on the measured
absorption wavelength of PFOA, perfluoroalkyl carboxylic acids are not expected to undergo direct
photolysis. Following irradiation of the potassium salt of PFOS with light of wavelength 290–800 nm for
67–167 hours, it was concluded that there was no evidence of direct photolysis of PFOS under any of the
test conditions (OECD 2002). Based on these test results for PFOS, perfluoroalkyl sulfonic acids are not
expected to undergo direct photolysis in the atmosphere. Direct photolysis data were not located for
perfluoroalkyl sulfonic acids.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

560
5. POTENTIAL FOR HUMAN EXPOSURE

A measured photooxidation rate constant is not available for PFOA. Hurley et al. (2004) measured the
reaction of short-chain (C1–C4) perfluoroalkyl carboxylic acids with photochemically generated hydroxyl
radicals. The proposed mechanism begins with abstraction of the carboxyl hydrogen, which is followed
by the removal of the carboxyl group and generation of a perfluoroalkyl radical. Finally, the
perfluoroalkyl chain is broken down one carbon atom at a time through an unzipping sequence. The same
rate constant, 1.69x10-13 cm3/molecule-second, was measured for the photooxidation of the C2, C3, and
C4 molecules, indicating that the chain length may have little effect on the reactivity of perfluoroalkyls
with hydroxyl radical. According to the authors, this rate constant corresponds to a half-life of 130 days.
Based on the data for the short-chain structures, the authors concluded that atmospheric photooxidation of
perfluoroalkyl carboxylic acids is not expected to compete with wet and dry deposition, which is
predicted to occur on a time scale of the order of 10 days.

Atmospheric photooxidation data are not available for perfluoroalkyl sulfonic acids. Atmospheric
photooxidation studies involving n-methyl perfluorobutane sulfonamidoethanol (Me-FBSE) and n-ethyl
perfluorobutanesulfonamide (Et-FBSA) indicate possible mechanisms for the reaction of these substances
with atmospheric hydroxyl radicals (D'eon et al. 2006; Martin et al. 2006). Products observed from the
photooxidation of these compounds indicate the following pathways: removal of an alkyl from the amide
(cleavage of the N-C bond); removal of the amido group (cleavage of the S-N bond); and removal of the
sulfonamido group (cleavage of the S-C bond) (D'eon et al. 2006; Martin et al. 2006). Each of these
pathways would be applicable to the photooxidation of Me- and Et-PFOSA-AcOH. The last two
pathways indicate that PFOSA may be photooxidized through removal of the amido or sulfonamido
group. The third pathway, cleavage of the S-C bond, also indicates a photooxidation mechanism for
perfluoroalkyl sulfonic acids. Martin et al. (2006) proposes an unzipping sequence for the perfluoroalkyl
chain following removal of the sulfonyl group.

Measured rate constants for the reaction of Me-FBSE and Et-FBSA with atmospheric hydroxyl radicals
are 5.8x10-12 and 3.74x10-13 cm3/molecule-second, respectively (D'eon et al. 2006; Martin et al. 2006).
Atmospheric half-lives calculated using these rate constants were 2 days for Me-FBSE and 20–50 days
for Et-FBSA.
Water.

PFOS and PFOA are expected to be stable to hydrolysis in the environment based on half-lives

of 41 and 92 years, respectively, calculated from experimental hydrolysis data that were measured at pH
5, 7, and 9 (OECD 2002, 2006b). Based on the data for PFOS and PFOA, hydrolysis is not expected to

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

561
5. POTENTIAL FOR HUMAN EXPOSURE

be an important degradation process for perfluorinated carboxylates and sulfonates in the environment.
Hydrolysis data were not located for perfluoroalkyl sulfonamides.

Available information indicates that perfluoroalkyl compounds are resistant to aerobic biodegradation.
PFOA and PFNA were not biodegraded using an Organisation for Economic Co-operation and
Development (OECD) guideline (301F) manometric respirometry screening test for ready
biodegradability; 0% of the theoretical oxygen demand was reached after 28 days (Stasinakis et al. 2008).
Meesters and Schröder (2004) reported that PFOA and PFOS were not degraded from an initial
concentration of 5 mg/L in aerobic sewage sludge in a laboratory scale reactor.

Substances such as fluorotelomer alcohols and perfluoroalkyl sulfonamides are degraded to other
substances such as PFOA and PFOS in water and can be considered a source of these substances in the
environment (Liu et al. 2007).

Sediment and Soil.

Data are not available regarding the transformation and degradation of

perfluoroalkyl compounds in sediment and soil. Based on the chemical stability of these substances and
their resistance to biodegradation in screening tests, environmental degradation processes are not
expected to be important removal mechanisms for perfluoroalkyl compounds in sediment and soil (3M
2000; EPA 2008a; Hekster et al. 2003; OECD 2002, 2007; Prevedouros et al. 2006; Schultz et al. 2003).

Substances such as fluorotelomer alcohols and perfluoroalkyl sulfonamides are degraded to other
substances such as PFOA and PFOS in soil and sediment and can be considered a source of these
substances in the environment (Liu et al. 2007; Washington and Jenkins 2015; Washington et al. 2015).
Other Media.

Data are not available regarding the transformation and degradation of perfluoroalkyl

compounds in other media.
5.5

LEVELS IN THE ENVIRONMENT

Reliable evaluation of the potential for human exposure to perfluoroalkyls depends, in part, on the
reliability of supporting analytical data from environmental samples and biological specimens.
Concentrations of perfluoroalkyls in unpolluted atmospheres and in pristine surface waters are often so
low as to be near the limits of current analytical methods. In reviewing data on perfluoroalkyls levels

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

562
5. POTENTIAL FOR HUMAN EXPOSURE

monitored or estimated in the environment, it should also be noted that the amount of chemical identified
analytically is not necessarily equivalent to the amount that is bioavailable.

Table 5-6 shows the limits of detection typically achieved by analytical analysis in environmental media.
An overview summary of the range of concentrations detected in environmental media is presented in
Table 5-7.

Table 5-6. Lowest Limit of Detection Based on Standardsa
Media

Detection limit

Reference

pg/m3

Air
Drinking water
Surface water and
groundwater
Soil

0.1
0.5–6.5 ng/L
No data

Harada et al. 2006
EPA 2009c (Method 537)

0.11–0.75 µg/kg (median reporting limits)

Anderson et al. 2016

Sediment
Whole blood
Serum

0.21–1.2 µg/kg (median reporting limits)
0.1–2 ng/mL
0.082–0.2 ng/mL

Anderson et al. 2016
Kärrman et al. 2005
CDC 2015

aDetection

limits based on using appropriate preparation and analytics. These limits may not be possible in all

situations.

Table 5-7. Summary of Environmental Levels of Perfluoroalkylsa
Low

Highb

For more information

Outdoor air
Indoor air (pg/m3)
Dust (ng/g)

<1
<5
<2.6

900,000
12,100
43,765

Table 5-9
Table 5-10
Table 5-11

Surface water (ppb)
Groundwater (ppb)
Drinking water (ppb)
Ocean water (pg/L)
Food (ppb)
Soil (ppb)

<1
<0.1
<0.0023
<0.2
<1
<0.17

67,000
619
~100
192,000
118.29
104,000

Tables 5-12, 5-13, 5-14, 5-17
Tables 5-14, 5-17
EPA 2010
Table 5-15
Fromme et al. 2007b
Tables 5-16, 5-17

Media
(pg/m3)

aFor

PFOA or PFOS only.
levels are representative of monitoring data at localized contaminated sites and are not reflective of
background environmental levels.
bHigh

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

563
5. POTENTIAL FOR HUMAN EXPOSURE

Detections of perfluoroalkyls in air, water, and soil at NPL sites are summarized in Table 5-8.

Table 5-8. Perfluoroalkyls Levels in Water, Soil, and Air of National Priorities List
(NPL) Sites

Mediana

Geometric
meana

Geometric
standard
deviationa

Number of
quantitative
measurements NPL sites

Water (ppb)

0.35

0.25

6,064

5

4

Soil (ppb)

18,050

18,050

1,000

2

2

Medium
PFOA

Air (ppbv)

No data

PFOS
Water (ppb)

0.91

0.35

9,089

4

3

Soil (ppb)

108,000

108,000

1,000

2

2

Air (ppbv)

No data

PFBA
Water (ppb)

2.15

1.03

28,192

3

3

Soil (ppb)

1,600

1,600

1,000

2

2

Air (ppbv)

No data

PFBuS
Water (ppb)

0.05

0.02

6,770

2

2

Soil (ppb)

224

224

1,000

2

2

No data

Air (ppbv)

PFHpA
Water (ppb)

0.07

0.04

3,169

3

2

Soil (ppb)

1,275

1,275

1,000

2

2

Air (ppbv)

No data

PFHxA
Water (ppb)

0.25

0.10

8,444

2

2

Soil (ppb)

1,175

1,175

1,000

2

2

Air (ppbv)

No data

PFHxS
Water (ppb)

0.26

1.12

52,496

4

3

Soil (ppb)

5,585

5,585

1,000

2

2

2

2

Air (ppbv)

No data

PFNA
Water (ppb)
Soil (ppb)
Air (ppbv)

No data
27.2

27.2

1,000
No data

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

564
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-8. Perfluoroalkyls Levels in Water, Soil, and Air of National Priorities List
(NPL) Sites

Mediana

Geometric
meana

Geometric
standard
deviationa

Number of
quantitative
measurements NPL sites

Water (ppb)

0.18

0.11

3,465

3

3

Soil (ppb)

178

178

1,000

2

2

Medium
PFPeA

Air (ppbv)

No data

aConcentrations

found in ATSDR site documents from 1981 to 2017 for 1,854 NPL sites (ATSDR 2017). Maximum
concentrations were abstracted for types of environmental media for which exposure is likely. Pathways do not
necessarily involve exposure or levels of concern.

5.5.1

Air

Perfluoroalkyl levels have been measured in outdoor air at locations in the United States, Europe, Japan,
and over the Atlantic Ocean (Barber et al. 2007; Barton et al. 2006; Harada et al. 2005a, 2006; Kim and
Kannan 2007). Concentrations reported in these studies are provided in Table 5-9.
Mean PFOA levels ranged from 1.54 to 15.2 pg/m3 in air samples collected in the urban locations in
Albany, New York; Fukuchiyama, Japan; and Morioka, Japan and in the rural locations in Kjeller,
Norway and Mace Head, Ireland. Higher mean concentrations (101–552 pg/m3) were measured at the
urban locations in Oyamazaki, Japan and Manchester, United Kingdom, and semirural locations in
Hazelrigg, United Kingdom. Maximum reported concentrations at Oyamazaki and Hazelrigg were
919 and 828 pg/m3, respectively. The authors attributed the elevated concentrations at the Hazelrigg
location to emissions from a fluoropolymer production plant located 20 km upwind of this semirural
community.

Elevated levels of PFOA were observed in air samples collected along the fence line of the DuPont
Washington Works fluoropolymer manufacturing facility, which is located near Parkersburg, West
Virginia, in the Ohio River valley (Barton et al. 2006).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

565
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-9. Concentrations of Perfluoroalkyl in Outdoor Air

Location

Mean (range) concentration (pg/m3)
PFOA
PFHpA
PFNA

Reference

Urban
Albany, New York
Gas phase (n=8)

3.16 (1.89–6.53)

0.26 (0.13–0.42)

0.21 (0.16–0.31)

Kim and
Kannan 2007

Particulate phase (n=8)

2.03 (0.76–4.19)

0.37 (<0.12–0.81) 0.13 (<0.12–0.40) Kim and
Kannan 2007

Oyamazaki, Japan (n=12)

262.7 (72–919);
3,412.8 ng/g in
dust

—

—

Harada et al.
2005b

Fukuchiyama, Japan

15.2; 314 ng/g in
dust

—

—

Harada et al.
2006

Morioka, Japan (n=8)

2.0 (1.59–2.58)

—

—

Harada et al.
2005b

Manchester, United
Kingdom (n=2,1)a

341, 15.7

8.2, 0.2

<26.6, 0.8

Barber et al.
2007

Kjeller, Norway (n=2)

1.54

0.87

0.12

Barber et al.
2007

Mace Head, Ireland (n=4)

8.9

<0.001

<3.3

Barber et al.
2007

1.6, 14.4b

0.9

Barber et al.
2007

Rural

Hazelrigg, United Kingdom 101, 552b,c
(semi-rural) (n=10)
Marine air
Near Europe (northwest)
(n=3)

1.22 (0.5–2.0)

<0.6 (ND–<0.6)

0.3 (ND–0.5)

Jahnke et al.
2007a

Near Africa (east coast)
(n=5)

<0.5 (ND–0.7)

ND

<0.2 (ND–0.3)

Jahnke et al.
2007a

DuPont Washington Works 430,000 (75,000– —
Facility; Parkersburg, West 900,000)d
Virginia (n=28)

—

Barton et al.
2006

DuPont Washington Works 5,500 (10–75,900) —
Facility; Parkersburg, West
Virginia (n=90)

—

EPA 2007b

Source dominated

Location

Mean (range) concentration (pg/m3)
PFDeA
PFUA
PFDoA

Reference

Urban
Albany, New York
Gas phase (n=8)

0.63 (0.24–1.56)

<0.12 (ND–0.16)

0.27 (0.14–0.43)

Particulate phase (n=8)

0.27 (0.13–0.49)

ND

0.12 (<0.12–0.38) Kim and
Kannan 2007

***DRAFT FOR PUBLIC COMMENT***

Kim and
Kannan 2007

 PERFLUOROALKYLS

566
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-9. Concentrations of Perfluoroalkyl in Outdoor Air
Oyamazaki, Japan (n=12)

—

—

—

Harada et al.
2005b

Fukuchiyama, Japan

—

—

—

Harada et al.
2006

Morioka, Japan (n=8)

—

—

—

Harada et al.
2005b

Manchester, United
Kingdom (n=2,1)a

5.4, <0.8

<0.01, <0.4

<0.01, <0.01

Barber et al.
2007

Kjeller, Norway (n=2)

<0.15

<0.12

<0.12

Barber et al.
2007

Mace Head, Ireland (n=4)

<2.8

<0.002

<0.003

Barber et al.
2007

0.7

<0.01

Barber et al.
2007

Rural

Hazelrigg, United Kingdom 1.0, 8.3b
(semi-rural) (n=10)
Marine air
Near Europe (northwest)
(n=3)

<0.6 (ND–0.6)

ND

<0.14 (ND–0.17)

Jahnke et al.
2007a

Near Africa (east coast)
(n=5)

ND

0.03 (ND–0.2)

ND

Jahnke et al.
2007a

DuPont Washington Works —
Facility; Parkersburg, West
Virginia (n=28)

—

—

Barton et al.
2006

DuPont Washington Works —
Facility; Parkersburg, West
Virginia (n=90)

—

—

EPA 2007b

Source dominated

Location

Mean (range) concentration (pg/m3)
PFOS
PFBuS PFHxS
PFOSA

Reference

Urban
Albany, New York
Gas phase (n=8)

1.70 (0.94–3.0)

Particulate phase (n=8)

—

0.31 (0.13– 0.67 (0.22–
0.44)
2.26)

Kim and
Kannan 2007

0.64 (0.35–1.16) —

<0.12

0.29 (<0.12–
0.79)

Kim and
Kannan 2007

Oyamazaki, Japan (n=12)

5.2 (2.51–-9.80); —
72.2 ng/g in dust

—

—

Harada et al.
2005b

Fukuchiyama, Japan

2.2; 46.0 ng/g in —
dust

—

—

Harada et al.
2006

Morioka, Japan (n=8)

0.7 (0.46–1.19)

—

—

—

Harada et al.
2005b

Manchester, United
Kingdom (n=2,1)a

46, 7.1

2.2, <1.6

1.0, 0.1

<1.6, <0.2

Barber et al.
2007

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

567
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-9. Concentrations of Perfluoroalkyl in Outdoor Air
Rural
Kjeller, Norway (n=2)

1.0

<0.09

0.05

0.78

Barber et al.
2007

Mace Head, Ireland (n=4)

<1.8

<1.0

0.07

<0.56

Barber et al.
2007

2.6

0.04

0.2

Barber et al.
2007

Hazelrigg, United Kingdom 1.6
(semi-rural) (n=10)
Marine air
Near Europe (north west)
(n=3)

1.36 (0.4–2.5)

ND

0.12 (0.02– ND
0.3)

Jahnke et al.
2007a

Near Africa (east coast)
(n=5)

0.544 (0.05–1.9) ND

0.013 (ND– ND
0.05)

Jahnke et al.
2007a

Source dominated
DuPont Washington Works —
Facility; Parkersburg, West
Virginia (n=28)

—

—

—

Barton et al.
2006

DuPont Washington Works —
Facility; Parkersburg, West
Virginia (n=90)

—

—

—

EPA 2007b

aMean

values were reported for separate sampling sessions.
second concentration reported was measured during an earlier sampling session (n=2).
cA maximum PFOA concentration of 828 pg/m3 was measured in air at Hazelrigg, United Kingdom.
dAverage and range of concentrations in 6 out of 28 samples that contained PFOA.
bThe

“—“ indicates no available data; ND = not detected; PFBuS = perfluorobutane sulfonic acid; PFDeA =
perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid; PFHxS =
perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS =
perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid

The reported concentrations in these samples ranged from 75,000 to 900,000 pg/m3. The highest
concentrations were measured at locations downwind of the facility. High volume air samples collected
at several monitoring stations near the Washington Works facility contained PFOA at reported
concentrations ranging from 10 to 75,900 pg/m3 (EPA 2007b). The mean and median of these reported
concentrations are 5,500 and 240 pg/m3.

PFOS was detected above quantitation limits in most of the studies, but concentrations were generally
below 5 pg/m3. A concentration of 46 pg/m3 was reported in samples from Manchester, United Kingdom.
Reported concentrations of other perfluoroalkyls (PFHpA, PFNA, PFDeA, PFUA, PFDoA, PFBuS,
PFHxS, and PFOSA) were generally <1 pg/m3 in these studies. PFHpA was detected at slightly higher
concentrations (8.2 and 14.4 pg/m3) at Manchester and Hazelrigg, United Kingdom, respectively.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

568
5. POTENTIAL FOR HUMAN EXPOSURE

Jahnke et al. (2007a) collected eight marine air samples during a cruise between Germany and South
Africa (53°N to 33°S). Perfluoroalkyl concentrations steadily declined as the sampling moved further
from Europe and toward less industrialized regions. Only PFOS was detected in the two samples
collected over the Atlantic Ocean east of southern Africa.

Measurements of perfluoroalkyls in snow samples collected from Canadian Arctic ice caps indicate that
these substances may be generated in the atmosphere at these locations (Young et al. 2007). Reported
concentrations in these snow samples were 2.6–86 pg/L for PFOS, 12–147 pg/L for PFOA, 5.0–246 pg/L
for PFNA, <8–22 pg/L for PFDeA, and <6–27 pg/L for PFUA.

The concentration of PFOS measured in rainwater collected during a rain event in Winnipeg, Manitoba
was 0.59 ng/L (Loewen et al. 2005). PFOA, PFNA, PFDeA, PFUA, and PFDoA were not detected in the
rainwater. Reported method detection limits for these compounds were 7.2, 3.7, 1.7, 1.2, and 1.1 ng/L,
respectively.

Studies of perfluoroalkyl concentrations in indoor environments are available (Table 5-10). The reported
mean concentrations of perfluoroalkyls measured in four indoor air samples collected from Tromso,
Norway were 0.2 pg/m3 for PFOSA, <0.5 pg/m3 for PFBuS, <4.1 pg/m3 for PFHxS, <47.4 pg/m3 for
PFOS, 0.8 pg/m3 for PFHpA, 4.4 pg/m3 for PFOA, 2.7 pg/m3 for PFNA, 3.4 pg/m3 in PFDeA, <1.3 pg/m3
for PFUA, and 1.2 pg/m3 for PFDoA (Barber et al. 2007).

Table 5-10. Concentrations of Perfluoroalkyl in Indoor Air

Location

Concentration: mean (range); median (pg/m3)
PFOA
PFHpA
PFNA

Indoor air (pg/m3)
Tromso, Norway (n=4) 4.4
Indoor dust (ng/g)
Ottawa, Canada (n=67) 106.00 (<2.29–
1,234); 19.72a
Japan (n=16)
380 (70–3,700); 165
North Carolina and Ohio 296 (<10.2–1,960);
(n=112)
142b

Reference

0.8

2.7

Barber et al. 2007

—

—

Kubwabo et al. 2005

—
109 (<12.5–
1,150); 50.2b

—
22.1 (<11.3–
263); 7.99b

Moriwaki et al. 2003
Strynar and
Lindstrom 2008

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

569
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-10. Concentrations of Perfluoroalkyl in Indoor Air

Location

Concentration: mean (range) median (pg/m3)
PFDeA
PFUA
PFDoA

Reference

3.4

<1.3

1.2

Barber et al. 2007

—
—
15.5 (<9.40–267);
6.65b

—
—
30.4 (<10.7–
588); 7.57b

—
—
18.0 (<11.0–
520); 7.78b

Kubwabo et al. 2005
Moriwaki et al. 2003
Strynar and
Lindstrom 2008

(pg/m3)

Indoor air
Tromso, Norway (n=4)
Indoor dust (ng/g)
Ottawa, Canada (n=67)
Japan (n=16)
North Carolina and Ohio
(n=112)

Location

Concentration: mean (range); median (pg/m3)
PFOS
PFBuS
PFHxS
PFOSA Reference

Indoor air (pg/m3)
Tromso, Norway (n=4) <47.4
<0.5
Indoor dust (ng/g)
Ottawa, Canada (n=67) 443.68 (<4.56– NDa
5,065); 37.8a
Japan (n=16)

200 (11–
2,500); 24.5
North Carolina and Ohio 761 (<8.93–
(n=112)
12,100); 201b

—

<4.1

2.8

Barber et al. 2007

391.96
(<4.56–
4,305); 23.1a
—

<0.99a

Kubwabo et al. 2005

—

Moriwaki et al. 2003

41.7 (<12.5– 874 (<12.9– —
1,150); 9.11b 35,700); 45.5b

Strynar and
Lindstrom 2008

aMethod

detection limits (MDL) and percent below MDL are as follows: PFOA (2.29 ng/g, 37%), PFOS (4.56 ng/g,
33%), PFBuS (1.38 ng/g, 100%), PFHxS (4.56, 15%), and PFOSA (0.99 ng/g, 90%).
bLimit of quantitation (LOQ) and percent above LOQ are as follows: PFHpA (12.5 ng/g, 74.1%), PFOA (10.2 ng/g,
96.4%), PFNA (11.3 ng/g, 42.9%), PFDeA (9.40 ng/g, 30.4%), PFUA (10.7 ng/g, 36.6%), PFDoA (11.0 ng/g,
18.7%), PFOS (8.93 ng/g, 94.6%), PFHxS (12.9 ng/g, 77.7%), PFBuS (12.5 ng/g, 33.0%). Values below the LOQ
were assigned a value of LOQ/1.412 when calculating the median and mean.
“—“ indicates no available data; ND = not detected; PFBuS = perfluorobutane sulfonic acid; PFDeA =
perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid; PFHxS =
perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid; PFOS =
perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide

Kubwabo et al. (2005) measured the concentrations of selected perfluoroalkyls in dust samples from
67 Canadian homes. PFOA, PFOS, and PFHxS were each detected in 37, 33, and 15% of these samples,
respectively (detection limits of 2.29, 4.56, and 4.56 ng/g, respectively). Mean, median, and range of
concentrations in these samples were 106, 19.72, and 1.15–1,234 ng/g, respectively, for PFOA; 443.68,
37.8, and 2.28–5,065 ng/g, respectively, for PFOS; and 391.96, 23.1, and 2.28–4,305 ng/g, respectively,
for PFHxS. Concentrations were not reported for PFOSA, which was detected above 0.99 ng/g in 10% of
the samples. PFBuS was not detected in any of the samples. Moriwaki et al. (2003) measured PFOS and
PFOA concentrations in vacuum cleaner dust samples collected from 16 Japanese homes. PFOS and
PFOA were detected in every sample with reported concentrations of 11–140 and 69–380 ng/g,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

570
5. POTENTIAL FOR HUMAN EXPOSURE

respectively, in 15 of the 16 samples. One of the samples contained 2,500 ng/g PFOS and 3,700 ng/g
PFOA.

Strynar and Lindstrom (2008) measured perfluoroalkyl levels in 112 indoor dust samples collected from
homes and daycare centers in North Carolina and Ohio. These authors detected PFHpA, PFOA, PFNA,
PFDeA, PFUA, PFDoA, PFOS, PFHxS, and PFBuS. Mean values ranged from 15.5 to 874 ng/g. PFOS
and PFOA were detected in 94.6 and 96.4% of the samples, respectively. Maximum detections in the
samples were as high as 12,100 ng/g for PFOS and 35,700 ng/g for PFHxS. Household dust samples
collected from the United Kingdom, Australia, Germany, and the United States showed the presence of
perfluoroalkyls (Kato et al. 2009a). These data are summarized in Table 5-11.

Table 5-11. Concentration (ng/g) of Perfluoroalkyls in 39 Dust Samplesa

Analyte

25th percentile

PFBuS

86.3

PFHxS
PFOS
PFHpA
PFOA
PFNA
PFDeA
PFUA
PFDoA
PFOSA
Me-PFOSA-AcOH
Et-PFOSA-AcOH
aThe

50th percentile

Frequency of
75th percentile Maximum detection (%)

359.0

782.1

7,718

92.3

47.7
31.7
33.9
<LOQ
<LOQ
<LOQ

185.5
479.6
97.3
96.5
<LOQ
<LOQ

632.2
1,456.6
532.5
667.7
26.2
61.0

43,765
18,071
5,195
9,818
832
1,965

79.5
74.4
61.2
64.1
25.6
38.5

<LOQ
<LOQ
<LOQ
<LOQ
92.4

<LOQ
<LOQ
<LOQ
<LOQ
243.5

<LOQ
37.6
16.1
110.3
417.9

732
1,048
184
4,520
3,795

20.5
43.6
23.1
33.3
87.2

LOQs are 2.6 ng/g except for PFHpA, which is 4.0 ng/g.

Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; LOQ = limit of quantification; Me-PFOSAAcOH = 2-(N-methyl-perfluorooctane sulfonamide) acetic acid; PFBuS = perfluorobutane sulfonic acid;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid
Source: Kato et al. 2009a

Median levels of PFOS in dust samples collected in homes, apartments, daycare centers, offices, and cars
in Sweden were 39, 85, 31, 110, and 12 ng/g, respectively (Björklund et al. 2009). Median PFOA levels
in dust samples from the same study were 54, 93, 41, 70, and 33 ng/g in homes, apartments, daycare
centers, offices, and cars, respectively. The authors concluded that while dietary intake was the major

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

571
5. POTENTIAL FOR HUMAN EXPOSURE

PFOA/PFOS exposure pathway for adults and toddlers in the general population, dust ingestion could
become an important pathway under a worst-case scenario (high dust ingestion and maximum dust
levels).
5.5.2

Water

PFOS and PFOA have been widely detected in surface water samples collected from various rivers, lakes,
and streams in the United States (Boulanger et al. 2004; DRBC 2013; Kannan et al. 2005; Kim and
Kannan 2007; Nakayama et al. 2007; Simcik and Dorweiler 2005; Sinclair et al. 2004, 2006). Levels of
these substances in surface water appear to be declining since the phase out of these two substances.
Zhang et al. (2016) measured surface water levels of 14 perfluoroalkyls in 2014 at 37 sites across the
northeastern United States. Although PFOA and PFOS were detected in 100 and 95% of the samples
collected, respectively, the levels were lower when compared to measurements of U.S. surface waters
from other regions of the country from 2000 to 2009. The maximum concentration of PFOA was 56 ng/L
measured in the Passaic River, New Jersey and the maximum PFOS concentration was 27.5 ng/L
measured in the Woonasquatucket River, Rhode Island (Zhang et al. 2016).

Less data are available regarding the concentrations of other perfluoroalkyl compounds in surface water.
PFHpA and PFHxS were commonly detected in the few studies that analyzed surface water for these
compounds (DRBC 2013; Kim and Kannan 2007; Nakayama et al. 2007; Simcik and Dorweiler 2005).
Concentrations of PFOA and PFOS measured in surface water are presented in Table 5-12 and other
perfluoroalkyls are summarized in Table 5-13. Maximum concentrations of PFOS, PFOA, PFHpA,
PFNA, PFDeA, PFUA, PFDoA, PFBuS, and PFHxS measured in surface water collected from the Cape
Fear Basin, North Carolina were 287, 132, 329, 194, 120, 52.1, 4.46, 9.41, and 35.1 ng/L, respectively
(Nakayama et al. 2007). Much higher concentrations of PFOS (198–1,090 ng/L) have been measured in
Onondaga Lake in Syracuse, New York (Sinclair et al. 2006). Onondaga Lake is a Superfund site that has
become contaminated through industrial activity along its banks.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

572
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-12. Concentrations of PFOA and PFOS in Surface Water (ng/L)

Location

PFOA

Concentration
PFOS

Great Lakes

Reference
Boulanger et al. 2004

Lake Ontario (n=8)

15–70

6–121

Lake Erie (n=8)

21–47

11–39

New York State waters

Sinclair et al. 2006

Lake Ontario (n=13)

18–34

2.9–30

Niagara River (n=3)

18–22

3.3–6.7

Lake Erie (n=3)

13–27

2.8–5.5

Finger Lakes (n=13)

11–20

1.3–2.6

Onondaga Lake (n=3)

39–64

198–1,090
(median=756)

Oneida Lake (n=1)

19

3.5

Erie Canal (n=3)

25–59

5.7–13

Hudson River (n=8)

22–173
(median=35)

1.5–3.4

Lake Champlain (n=4)

10–46

0.8–7.7

Albany, New York

Kim and Kannan 2007

Lake water (n=11)

3.27–15.8
(median=7.20)

ND–9.30
(median=2.88)

Surface water runoff (n=14)

0.51–29.3
(median=3.80)

<0.25–14.6
(median=0.81)

New York/New Jersey/Rhode Island
metropolitan region

56 (maximum)
3.5 (median)

27.5 (maximum)
0.96 (median)

Detroit (n=10)

<8–16.14

<0.08–6.13

Flint (n=4)

<8–23.01

1.50–12.31

Saginaw Bay (n=5)

<8–24.08

3.10–12.69

Northeastern Michigan (n=2)

<8

0.87–6.34

Upper Peninsula (n=7)

<8–13.77

<0.8–3.09

Northwestern Michigan (n=2)

11.96

<0.8–4.48

Western Michigan (n=6)

<8–15.17

<0.8–5.32

Southwestern Michigan (n=5)

8.74–35.86

7.22–29.26

Lansing (n=3)

<8–13.37

1.04–4.96

Michigan water regions

Zhang et al. 2016
Sinclair et al. 2004

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

573
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-12. Concentrations of PFOA and PFOS in Surface Water (ng/L)

Location

PFOA

Concentration
PFOS

Minnesota Waters and Lake Michigan

Reference
Simcik and Dorweiler
2005

Remote (n=4)
Loiten

0.7

ND

Little Trout

0.3

1.2

Nipisiquit

0.1

ND

Tettegouche

0.5

0.2

Calhoun

20

47

Lake Harriet

3.5

21

Lake of the Isles

0.5

2.4

Minnesota River

1.2

9

<0.6–0.5

1–3.2

Urban (n=4)

Lake Michigan (n=4)
Cape Fear Basin, North Carolina

Nakayama et al. 2007

80 Sites (n=100)
Mean

43.4

31.2

Median

12.6

28.9

Minimum

ND

<1

287

132

7.6

0

Maximum
Percent not

detecteda

Raisin and St. Clair Rivers, Michigan

Kannan et al. 2005

Raisin River

14.7

3.5

St. Clair River (n=3)

4.0–5.0

1.9–3.9

Conasauga River, Georgia

253–1,150

192–318

Konwick et al. 2008

Dalton, Georgia

49.9–299

15.8–120

Konwick et al. 2008

2007

3.54–75.40

2.70–8.42

2008

3,99–48

3.53–11.7

2009

3.29–27.7

2.86–7.97

Several rivers in Japan

0.1–67,000

0.3–59

Lake Victoria Gulf, Kenya

0.4–96.4 (rivers)

<0.4–13.23 (rivers); Orata et al. 2009

0.4–11.6 (lakes)

<0.4–2.53 (lakes)

1–1,270

1–25

Tidal Delaware River (six locations)

River Po, Italy
aDetection

DRBC 2013

Harada and Koizumi
2009

Loos et al. 2008

limit is 0.05 ng/L

ND = not detected; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

574
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-13. Concentrations of Other Perfluoroalkyls in Surface Water
Concentration (ng/L)
Location
(reference)a

PFHpA PFNA

PFDeA PFUA

EtPFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH

Great Lakes (Boulanger et al. 2004)
Lake Ontario —
(n=8)

—

—

—

—

—

—

—

<0.3–10

Lake Erie
(n=8)

—

—

—

—

—

—

—

3–11

—

New York State waters (Sinclair et al. 2006)
Onondaga
Lake (n=3)

—

—

—

—

—

—

4.2–8.5

—

—

Erie Canal
(n=3)

—

—

—

—

—

—

2.5–5.6

—

—

Other lakes
and rivers

—

—

—

—

—

—

0.5–2.8

—

—

—

<0.25–
4.05

ND–
0.47

—

ND–13.5 ND–
2.14

—

6.2
(max);
0.4
(median)

43
—
(max);
0.7
(median)

—

Albany, New York (Kim and Kannan 2007)
Lake water
(n=11)

1.15–
12.7

ND–3.51 0.25–
3.58

Surface
water runoff
(n=14)

<0.25–
6.44

<0.25–
5.90

ND–1.45 ND–
<0.25

ND–8.39 ND–1.99 ND–1.60 —

New York/New Jersey/Rhode Island metropolitan region (Zhang et al. 2016)
48.2
(max);
0.9
(median)

14
(max);
0.4
(median)

5.8
(max);
0.2
(median)

1.9
(max);
0.1
(median)

2.6
(max);
0.0
(median)

Minnesota waters and Lake Michigan (Simcik and Dorweiler 2005)
Remote
(n=4)
Loiten

10

ND

ND

—

—

—

—

—

—

Little Trout

4.8

ND

ND

—

—

—

—

—

—

Nipisiquit

0.9

<0.3

ND

—

—

—

—

—

—

ND

ND

—

—

—

—

—

—

—

—

—

—

—

11

0.6

0.5

—

—

—

—

—

—

Lake Harriet 2.6

ND

ND

—

—

—

—

—

—

Lake of the 0.4
Isles

ND

ND

—

—

—

—

—

—

Minnesota
River

0.7

1.9

ND

—

—

—

—

—

—

Lake
Michigan
(n=4)

<0.6–4.1 <0.6–3.1 ND

—

—

—

—

—

—

Tettegouche 3.1
Urban (n=4)
Calhoun

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

575
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-13. Concentrations of Other Perfluoroalkyls in Surface Water
Concentration (ng/L)
Location
(reference)a

PFHpA PFNA

PFDeA PFUA

EtPFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH

Cape Fear Basin, North Carolina (Nakayama et al. 2007)
80 Sites
(n=100)
Mean

38.7

33.6

22.1

10.4

2.17

2.58

7.29

—

—

Median

14.8

5.70

13.2

5.67

1.95

2.46

5.66

—

—

Maximum

329

194

120

52.1

4.46

9.41

35.1

—

—

Percent not 32.9
detectedb

10.1

15.2

17.7

53.2

38.0

45.6

—

—

Raisin and St. Clair Rivers, Michigan (Kannan et al. 2005)
Raisin River —

—

—

—

—

—

<1

<10

—

St. Clair
River (n=3)

—

—

—

—

—

<1

<10

—

—

Delaware River (6 locations) (DRBC 2013)
2007

ND24.30

1.71-976 ND-9.97 ND-26

ND

—

ND-4.48 ND

—

2008

2.0316.3

3.24-650 ND-5.25 ND-11.8 ND

—

BD-3.55 ND

—

2009

1.1447.4

1.65-546 ND-1.75 ND-8.3

—

ND-3.62 ND

—

aSee

ND

Table 5-12 for numbers of samples collected at these locations.
limit = 0.05 ng/L.

bDetection

“—“ indicates no available data; Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; ND = not
detected; PFBuS = perfluorobutane sulfonic acid; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic
acid; PFHpA = perfluoroheptanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid

Higher levels perfluoroalkyl compounds are expected in surface water and groundwater surrounding
perfluorochemical industrial facilities. Paustenbach et al. (2007) estimated PFOA concentrations in
environmental media for communities located near the DuPont Washington Works chemical
manufacturing facility. From this analysis, the authors concluded that much of the PFOA detected in
groundwater near the facility was attributed to deposition to soil surfaces following atmospheric
emissions from the plant followed by subsequent leaching into groundwater. DuPont entered into an
agreement with the EPA to collect monitoring data for PFOA in the Ohio River around the DuPont
Washington Works facility. In its final phase III assessment, it was reported that levels of PFOA
downstream from the facility in 2011 were about an order of magnitude lower than when monitoring
began in 2002, while levels upstream essentially remained unchanged (URS 2012). The maximum

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

576
5. POTENTIAL FOR HUMAN EXPOSURE

concentration measured in 2011 was 200 ng/L and was obtained from a monitoring location adjacent to
the site.

Xiao et al. (2015) summarized soil and groundwater monitoring data of select perfluoroalkyls near the 3M
Cottage Grove facility in Minnesota. Groundwater samples for PFBA, PFHxA, PFOA, PFHxS, and
PFOS in wells surrounding the site obtained from 2009 to 2013 showed little or no change in
concentration over this time period, even though 3M had stopped producing perfluoroalkyls at this facility
in 2002. Levels in groundwater were shown to decrease exponentially with distance from the source. A
measured PFOA concentration of approximately 20,000 ng/L was observed from a well near the historical
unlined disposal site of this facility, but levels decreased to <100 ng/L measured at a distance of 1.4 km
away. Some other examples are provided in Table 5-14.

Table 5-14. Concentrations of Perfluoroalkyls in Surface Water and Groundwater
at Fluorochemical Industrial Facilities

Location

Percent detection and concentration (µg/L)
PFOA
PFBA PFOS
PFHxS PFBuS

Reference

DuPont Washington Works Facility, West Virginia
Groundwater
Borings (n=18)

Davis et al. 2007

Percent detected

89%a

—

—

—

—

Minimum

0.0912a

—

—

—

—

Maximum

78a

—

—

—

—

—

—

—

Wells (n=14)
Percent detected

100%a

—

—

—

—

Minimum

0.081a

—

—

—

—

Maximum

37.1a

—

—

—

—

—

—

—

Wells (n=3)
Percent detected

100%

—

—

—

—

Minimum

2.8

—

—

—

—

Maximum

100

—

—

—

—

Davis et al. 2007

EPA 2008d

Surface water
Outlets (n=4)

EPA 2008d

Percent detected

100%

—

—

—

—

Minimum

2.3

—

—

—

—

Maximum

61

—

—

—

—

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

577
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-14. Concentrations of Perfluoroalkyls in Surface Water and Groundwater
at Fluorochemical Industrial Facilities

Location

Percent detection and concentration (µg/L)
PFOA
PFBA PFOS
PFHxS PFBuS

Reference

3M Cottage Grove Facility, Minnesota
Groundwater
Wells (n=1–7)

3M 2007b

Percent detected

100%

100%

100%

100%

100%

Minimum

24.6

23.3

26.0

6.47

2.11

Maximum

619

318

26.0

40.0

26.1

Surface water
East and West Cove (n=3–9)

3M 2007b

Percent detected

100%

100%

100%

100%

78%

Minimum

0.172

0.803

0.227

0.0936

0.304

Maximum

2.79

1.01

3.12

4.58

9.69

Mississippi River Shoreline (n=52–80)

3M 2007b

Percent detected

60%

52%

43%

28%

56%

Maximum

0.760

6.92

0.539

1.04

3.05

Mississippi River Transect (n=34–44)

3M 2007b

Percent detected

14%

12%

0%

0%

0%

Maximum

0.0501

0.0530 ND

ND

ND

3M Decatur Facility, Alabama
Groundwater
Off-site groundwater (n=18)

3M 2008c

Percent detected

94%

—

—

—

—

Mean

1.87

—

—

—

—

Range

0.083–19.8 —

—

—

—

Off-site groundwater

3M 2010

Range

0.0396–
0.622

0.0615–
0.480

0.149–6.41 0.0104– 0.012–
1.26
0.151

0.0127–
0.0875

0.0101–
0.0766

Percent detected

100%

—

—

—

—

Median

2.66

—

—

—

—

Range

0.32–127

—

—

—

—

Groundwater range (n=51)

0.0404–
2.41

Lindstrom et al.
2011

Surface water
On-site surface water (n=7)

3M 2008c

Off-site surface water (n=60)

3M 2008c

Percent detected

98%

—

—

—

—

Range

0.026–27.7 —

—

—

—

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

578
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-14. Concentrations of Perfluoroalkyls in Surface Water and Groundwater
at Fluorochemical Industrial Facilities
Percent detection and concentration (µg/L)
PFOA
PFBA PFOS
PFHxS PFBuS

Location
Tennessee River

Hansen et al.
2002

Upstream of facility (n=19)
Percent detected

0%

—

100%

—

—

Range

<25

—

16.8–52.6 —

—

—

—

Downstream of facility (n=21)
Percent detected

0%

—

100%

—

—

Median

355

—

107

—

—

Range

<25–598

—

30.3–144

—

—

aAnalyte

Reference

was reported as APFO.

“—“ indicates no available data; APFO = ammonium perfluorooctanoate; ND = not detected; PFBA = perfluorobutyric
acid; PFBuS = perfluorobutane sulfonic acid; PFHxS = perfluorohexane sulfonic acid; PFOA = perfluorooctanoic
acid; PFOS = perfluorooctane sulfonic acid

Yamashita et al. (2005) measured PFOA, PFOS, PFNA, and PFHxS concentrations in ocean water
collected from locations in the Atlantic Ocean, Pacific Ocean, and areas near China, Korea, and Japan.
These concentrations are listed in Table 5-15. Wei et al. (2007a) measured perfluoroalkyl concentrations
in surface seawaters from the western Pacific Ocean, Indian Ocean, and near-Antarctic region. PFOS and
PFOA were detected in 60 and 40% of the samples, respectively, with maximum concentrations of
71.7 and 441.6 pg/L, respectively. Concentrations of other perfluoroalkyls (PFHxS, PFBuS, PFDoA,
PFDeA, PFNA, PFHpA) were generally below detection in most samples, with the exceptions being in
samples collected near Shanghai, the Philippines, and Indonesia. Maximum concentrations of these
perfluoroalkyls ranged from 3.1 to 70.2 pg/L near Shanghai. PFOA, PFOS, and other perfluoroalkyl
species were monitored in waste water effluents and 20 rivers located in Japan (Murakami et al. 2008).
Perfluoroalkyls were ubiquitous in the river water samples, with concentrations of PFOA as large as
0.054–0.192 µg/L in seven of the river samples with low waste water effluent sources.

Table 5-15. Concentrations of PFOA and PFOS in Ocean Water

Location

PFOA

Concentration (pg/L)
PFOS
PFNA

North Atlantic (n=9)

160–338

8.6–36

15–36

4.1–6.1

Mid Atlantic (n=7)

100–439

37–73

—

2.6–12

1.1–20

1.0–16

0.1–1.6

Central to Eastern Pacific (n=14) 15–62

***DRAFT FOR PUBLIC COMMENT***

PFHxS

 PERFLUOROALKYLS

579
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-15. Concentrations of PFOA and PFOS in Ocean Water

Location

PFOA

Concentration (pg/L)
PFOS
PFNA

Western Pacific (n=2)

136–142

54–78

—

2.2–2.8

Tokyo Bay (n=8)

1,800–192,000

338–57,700

163–71,000

17–5,600

Offshore Japan (n=4)

137–1,060

40–75

—

3.0–6.1

Coastal Hong Kong (n=12)

673–5,450

70–2,600

22–207

<5–311

Coastal China (n=14)

243–15,300

23–9,680

2.0–692

<5–1,360

Coastal Korea (n=10)

239–11,350

39–2,530

15–518

<5–1,390

Sulu Sea (n=5)

88–510

<17–109

—

<0.2

South China Sea (n=2)

160–420

8–113

—

<0.2

PFHxS

“—“ indicates no available data; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA =
perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid
Source: Yamashita et al. 2005

The presence of PFOA and PFOS in surface water in the United States indicates that drinking water taken
from these sources may contain detectable levels of these substances. The most vulnerable drinking water
systems are those in close proximity to sites that are contaminated with perfluoroalkyls. PFOA was
detected in 12 out of 13 samples collected from four municipal drinking water treatment plants that draw
water from the Tennessee River and are located downstream from the 3M Decatur Facility in Alabama.
Reported concentrations ranged from 0.025 to 0.16 μg/L (3M 2008c). PFOA was not detected in any
samples collected from a fifth plant located upstream of the 3M Decatur facility (3M 2008c).

Hu et al. (2016) presented geospatial monitoring data for six perfluoroalkyls (PFBuS, PFHxS, PFHpA,
PFOA, PFOS, and PFNA) in U.S. drinking water from information contained in the EPA third
Unregulated Contaminant Monitoring Rule (UCMR 3) program. Frequency of detection and
concentrations in drinking water was correlated with proximity to industrial facilities using
perfluoroalkyls, military fire training areas, and the number of waste water treatment plants. Under the
UCMR 3, the final data indicate that 63 out of 4,920 public water systems (PWSs) reported PFOA and
PFOS at concentrations that exceed the EPA Health Advisory of 0.07 µg/L (final UCMR 3 data posted
January 2017). These systems represent 1.3% of all of the PWSs that monitored under UCMR 3. Note
that UCMR 3 required all large water systems (4,120 PWSs, serving >10,000 people) and a representative
sample of 800 small water systems (serving ≤10,000 people) to monitor for PFOA and PFOS from 2013
through 2015. In addition, the results showed that one or more of the other four perfluoroalkyls

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

580
5. POTENTIAL FOR HUMAN EXPOSURE

monitored under UCMR 3 were detected above their respective minimum reporting levels at 118 out of
4,920 public water systems (EPA 2017).

It was also reported that 66 public drinking water systems that serve 6 million U.S. residents had at least
one sample that exceeded the current EPA health advisory level of 0.07 μg/L for PFOA and PFOS. The
dataset used by Hu et al. (2016) has since been updated by the EPA (EPA 2016c). UCMR 3 data are
added, removed, or updated following additional reviews by the analytical laboratory conducting the
testing, the individual states, and EPA staff. The most recent report dated July 2016 showed that the six
perfluoroalkyls were detected above their respective minimal reporting level in 365 out of 4,905 public
water systems and that PFOA and PFOS were identified above the health advisory level of 0.07 μg/L in
59 out of the 4,905 public water supplies (EPA 2016c).

Based on a memorandum of understanding with the EPA, DuPont began collecting water monitoring data
of both public and private wells near the Washington Works chemical plant. The quarterly reports and
monitoring data affiliated with these reports may be obtained from the regulations.gov portal
(http://www.regulations.gov). In samples of water collected at 17 public water facilities from 2002 to
2009 in West Virginia and Ohio, PFOA levels ranged from below the detection limit (0.0023 μg/L) to
nearly 100 μg/L in a few test wells in Little Hocking, Ohio (EPA 2010). In the final phase III summary
report, PFOA concentrations were reported to range from below the detection limit to 0.79 μg/L in
34 wells located approximately 3 miles upstream and 82 miles downstream from the facility (URS 2012).
Rumsby et al. (2009) have reviewed the presence of PFOS and PFOA in drinking waters worldwide and
discussed treatment methods for removing these substances from public water supplies. Conventional
waste water treatment does not always efficiently remove perfluoroalkyls, and effluent may contain
higher levels of some perfluoroalkyls than influent due to degradation of precursor substances during the
treatment process (Schultz et al. 2006a, 2006b). While granulated activated carbon and reverse osmosis
followed by nanofiltration have been shown to be effective methods of removing perfluoroalkyls,
conventional methods such as chlorination, ozonolysis, and slow sand filtration may not be as effective.
As a consequence, public drinking water systems impacted by effluent from waste water treatment plants
often contain higher levels of perfluoroalkyls than systems that are not impacted by waste water treatment
plant effluent. Quinones and Snyder (2009) analyzed raw and finished water at seven different public
water systems in the United States for the presence of perfluoroalkyls. Water systems that were heavily
impacted by waste water treatment plant effluents had greater frequency and higher levels of
perfluoroalkyls when compared to water systems that were not highly impacted by waste water treatment

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

581
5. POTENTIAL FOR HUMAN EXPOSURE

plants. For example, no perfluoroalkyls were detected in either influent or finished water from a public
water system in Aurora, Colorado with no impact from waste water treatment plant effluent; however,
PFHxA, PFOA, PFNA, PFDeA PFUA, PFHxS, and PFOS were detected in all samples of a Los Angeles,
California public water system that was highly impacted by waste water effluent. Perfluoroalkyls were
commonly detected in the influent and effluent of 10 waste water treatment plants across the United
States (Schultz et al. 2006a). PFBuS was detected in 100% of both influent and effluent samples of the
10 plants, while other perfluoroalkyls like PFOA, PFOS, PFNA, PFHxA, PFHpA, and PFHxS were
detected in 80% of the influent and effluent samples at the 10 plants. In a national study of
10 perfluoroalkyls in raw and treated drinking water of France, Boiteux et al. (2012) observed that several
perfluoroalkyl carboxylic acids had greater concentrations in treated water than the raw water. In eight
drinking water treatment plants, PFBA, PFHxA, and PFHpA were not detected in raw water, but were
detected in treated water, indicating that these substances were released from saturated activated carbon
used to treat raw waters or were formed by the degradation of precursor substances. Perfluoroalkyl
sulfonates appeared to be removed more efficiently than the carboxylates. PFHxS, PFBuS, and PFOS
were detected less frequently in treated water as compared to raw water influent. These three compounds
comprised 53% of the total concentration of perfluoroalkyls in the raw water samples, but only 37% of
the total concentration of the perfluoroalkyls in the treated water. The summed concentration of
10 perfluoroalkyls was analyzed in the raw water and treated water of two drinking water treatment plants
downstream from a fluoropolymer manufacturing facility located in France (Dauchy et al. 2012). The
total concentration of perfluoroalkyls in the raw water at four sampling locations of the first plant ranged
from 0.140 to 0.287 μg/L, while the summed concentration in the treated water was 0.179 μg/L. The total
concentration of the 10 perfluoroalkyls in raw water at the second plant was 0.132 μg/L, while the total
concentration in the treated water was 0.130 μg/L. Levels of PFHxA were greater in the treated water
than the raw water at three of the four raw water sampling points, and levels of PFNA were greater in
treated water than raw water at all sampling points of the first plant, but were slightly lower in the treated
water of the second drinking water plant even though both systems used simple chlorination to treat the
water.

According to ATSDR (2008), PFOA, PFOS, PFBA, PFHxS, and PFBuS have been detected in the
municipal drinking water of communities located near the 3M Cottage Grove fluorochemical facility.
According to Chang et al. (2008a), concentrations of PFBA in precipitation, surface waters, and water
treatment effluents were measured in the low ng/L range in effluent at these locations, but could be in the
μg/L range in public and private wells.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

582
5. POTENTIAL FOR HUMAN EXPOSURE

PFOA was detected in 65% of the public drinking water systems tested in New Jersey in 2006 at
concentrations ranging from 0.005 to 0.039 μg/L (Post et al. 2009). In a follow-up study conducted in
2009, PFOA was detected in 57% of raw water samples from 29 additional public drinking water systems
in New Jersey at a maximum concentration of 0.100 μg/L (Post et al. 2013). Nine other perfluoroalkyls
were also tested for, with PFOS and PFNA being the most frequently detected compounds (30% detection
frequency each) after PFOA. PFOA and PFOS were detected in tap water from 21 cities located in China
at concentrations of <0.0001–0.0459 and <0.0001–0.0148 μg/L, respectively (Jin et al. 2009). Mak et al.
(2009) published a study comparing detections of perfluoroalkyls including PFOA and PFOS in tap water
collected in China, Japan, India, Canada, and the United States. PFOA and PFOS were the predominant
species measured, accounting for 40–50% of the total perfluoroalkyls present in water, with the exception
of certain location of India where PFOS or PFOA may not have been present or were present at low
levels.

Gellrich et al. (2013) analyzed 119 samples of mineral water, 26 samples of tap water, 18 spring water
samples, and 14 raw water samples from Germany for the presence of perfluoroalkyls. Perfluoroalkyls
were detected in 58% of all of the samples tested, with the greatest summed total concentration observed
in tap water at 0.0427 μg/L. The maximum concentration of individual perfluoroalkyls occurring in
bottled water, spring water, untreated water, and tap water were observed for PFBuS (0.0133 μg/L),
PFOA (0.0074 μg/L), PFBuS (0.010 μg/L), and PFHxS (0.0121 μg/L), respectively. Perfluoroalkyls were
widely detected in drinking water samples collected in 2008 at 40 different locations of Catalonia, Spain
(Ericson et al. 2009). Median concentrations ranged from 0.00002 μg/L (PFOSA) to 0.00098 μg/L
(PFOA). The most frequently detected compounds were PFOS and PFHxS, which were detected in
35 and 31 samples, respectively. PFOS, PFOA, and PFHxS were detected in all samples collected in a
study of drinking water contamination of perfluoroalkyls in Rio de Janeiro, Brazil (Quinete et al. 2009).
Concentration ranges were 0.00058–0.00670 μg/L (PFOS), 0.00035–0.00282 μg/L (PFOA), and 0.00015–
0.001 μg/L (PFHxS) respectively.
5.5.3

Sediment and Soil

Concentrations of perfluoroalkyls in soils are expected to be greater in the vicinity of fluorochemical
plants that produced or used these substances as processing aids in the manufacture of fluoropolymers
than in the environment at large. Levels of some perfluoroalkyl compounds measured in soil and
sediment surrounding perfluorochemical industrial facilities are listed in Table 5-16. PFOA was detected
in most soil and sediment samples collected on- and off-site at the 3M Decatur facility in Alabama in

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

583
5. POTENTIAL FOR HUMAN EXPOSURE

monitoring studies conducted between October 2004 and December 2006. Maximum soil concentrations
were as high as 14,750 ng/g on-site and 7.85 ng/g off-site, and maximum sediment concentrations were as
high as 347 ng/g on-site and 2,385 ng/g off-site (3M 2008c). The highest levels of PFOA were measured
in soil from on-site fields formerly amended with PFOA-containing sludge. In its final project report for
this location, six on-site soil samples were analyzed in December 2012 for the presence of PFOA, PFOS,
PFBuS, and PFHxS. Average levels were 3.86–3,890 ng/g (PFOS), 3.56–270 ng/g (PFHxS), 0.423–
64.8 ng/g (PFBuS), and 17.0–1,410 ng/g (PFOA) (3M 2012).

Table 5-16. Concentrations of Perfluoroalkyls in Soil and Sediment at
Fluorochemical Industrial Facilities

Location

Percent detection and concentration (ng/g)
PFOA
PFBA
PFOS
PFHxS PFBuS Reference

DuPont Washington Works Facility, West Virginia
Soil
Boring samples (n=22)

Davis et al.
2007

Percent detected

36%a

—

—

—

—

Minimum

<0.17a

—

—

—

—

Maximum

170a

—

—

—

—

3M Cottage Grove Facility, Minnesota
Soil
Boring samples (n=50–108)

3M 2007b

Percent detected

100%

—

95%

90%

60%

Maximum

21,800

—

104,000

3,470

139

Fire training area (n=8–11)

3M 2007b

Percent detected

91%

82%

100%

100%

73%

Maximum

262

11.5

2,948

62.2

24.6

Sediment
East and West Cove (n=21–28)

3M 2007b

Percent detected

100%

93%

100%

96%

65%

Minimum

0.764

ND

40.0

ND

ND

Maximum

1,845

94.6

65,450

126

9.14

Mississippi River shoreline (n=84–92)

3M 2007b

Percent detected

70%

44%

80%

28%

29%

Maximum

341

124

79.0

11.5

29.4

Mississippi River transect (n=38–40)

3M 2007b

Percent detected

18%

0%

82%

0%

0%

Maximum

1.09

ND

3.16

ND

ND

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

584
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-16. Concentrations of Perfluoroalkyls in Soil and Sediment at
Fluorochemical Industrial Facilities

Location

Percent detection and concentration (ng/g)
PFOA
PFBA
PFOS
PFHxS PFBuS Reference

3M Decatur Facility, Alabama
Soil
On-site former sludge incorporation area (n=357)
Percent detected

99%

3M 2008c

—

—

—

—

Mean

885–929

Range

2.91–14,750 —

—

—

—

Percent detected

100%

—

—

—

—

Mean

3.53–4.1

Range

1.61–6.03

—

—

—

—

On-site background (n=18)

3M 2008c

Off-site soil (n=23)

3M 2008c

Percent detected

100%

Mean

3.68–4.6

Range

0.72–7.85

—

—

—

—

—

—

—

—

Sediment
On-site sediment (n=8)

3M 2008c

Percent detected

88%

Median

16.8

Range

1.64–347

—

—

—

—

—

—

—

—

Off-site sediment (n=30)

3M 2008c

Percent detected

93%

—

—

—

—

Range

0.39–2,385

—

—

—

—

3M Decatur Facility, Alabama December 2012
On-site former sludge
incorporation area (n=6)

3M 2012

Percent detected

100%

—

100%

86%

86%

Mean

17.0–1,410

—

3.86–
3,890

3.56–
270

0.423–
64.8

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

585
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-16. Concentrations of Perfluoroalkyls in Soil and Sediment at
Fluorochemical Industrial Facilities

Location

Percent detection and concentration (ng/g)
PFOA
PFBA
PFOS
PFHxS PFBuS Reference

3M Cottage Grove Facility, Minnesota
Surface soil along U.S.
Highway 10 near facility

aAnalyte

Xiao et al. 2015

Percent detected

100%

—

100%

Median

8.0

12.2

Range

5.5–125.7

0.2–28.2

—

—

was reported as APFO.

“—“ indicates no available data; APFO = ammonium perfluorooctanoate; ND = not detected; PFBA = perfluorobutyric
acid; PFBuS = perfluorobutane sulfonic acid; PFHxS = perfluorohexane sulfonic acid; PFOA = perfluorooctanoic
acid; PFOS = perfluorooctane sulfonic acid

PFOA, PFOS, and PFHxS were detected in 90–100% of soil samples collected from a former tar
neutralization area, a former sludge disposal area, a former solids burn pit area, a former waste water
treatment plant area, and a former fire training area at the 3M Cottage Grove facility in Minnesota (3M
2007b). PFBuS was detected in 60–73% of these samples. Maximum concentrations for these substances
were 21,800, 104,000, 3,470, and 139 ng/g, respectively. Levels of PFBA were only reported for soil in
the fire training area; it was detected in 9 out of 11 samples from this location at 0.306–11.5 ng/g. The
percent detection of these compounds in sediment from the East and West Cove sites was similar to that
in soil. Maximum concentrations of PFOA and PFOS were 1,845 and 65,450 ng/g, respectively. These
perfluoroalkyls were also analyzed in Mississippi River sediment near the Cottage Grove Facility. Levels
of these compounds were much greater along the facility shoreline compared to levels in transect samples
collected at points crossing the river. Maximum shoreline concentrations for PFOA, PFBA, PFOS,
PFHxS, and PFBuS were 341, 124, 79.0, 11.5, and 29.4 ng/g, respectively. PFHxS, PFBuS, and PFBA
were not detected in any of the transect samples, and PFOA was found in only 18%. Although the
maximum concentration of PFOS was 3.16 ng/g, it was still detected in 82% of the transect samples.

PFOA and PFOS were detected in all surface soils (top 10 cm) samples collected at 28 sites in September
and October of 2012 along U.S. Highway 10 running from Cottage Grove, Minnesota (where the former
3M perfluoroalkyl manufacturing facility was located) to Big Lake, Minnesota (Xiao et al. 2015).
Measured levels of PFOS and PFOA ranged from 0.2 to 28.2 and from 5.5 to 125.7 ng/g, respectively.
Subsurface soils up to a depth of 65 cm were collected at four sites as well. Levels of PFOA and PFOS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

586
5. POTENTIAL FOR HUMAN EXPOSURE

generally increased with increasing depth at each of the locations, suggesting a downward movement of
the contaminants and the potential to contaminate groundwater.

The use of aqueous firefighting foams at fire training areas of military installations has resulted in
widespread contamination of perfluoroalkyl compounds in the soil and groundwater at these facilities.
Monitoring data obtained from 40 sites at 10 U.S. military installations in the continental United States
and Alaska were collected for several perfluoroalkyls (Anderson et al. 2016). These data are summarized
in Table 5-17.

Table 5-17. Summary of Perfluoroalkyl Compounds Detected in Soil, Sediment,
Surface Water, and Groundwater at 10 Military Installationsa

Compound Parameter
PFBA

PFBuS

PFHxA

PFHxS

PFHpA

PFOA

PFOSA

PFOS

PFNA

Surface soil Subsurface
(µg/kg)
soil (µg/kg)

DF
Median
Maximum
DF
Median
Maximum
DF
Median
Maximum

38.46
1.00
31.0
35.16
0.775
52.0
70.33
1.75
51.0

29.81
0.960
14.0
34.62
1.30
79.0
65.38
1.04
140

DF
Median
Maximum
DF
Median
Maximum
DF
Median
Maximum
DF
Median
Maximum
DF
Median
Maximum
DF
Median
Maximum

76.92
5.70
1,300
59.34
0.705
11.4
79.12
1.45
58.0
64.84
1.20
620
98.90
52.5
9,700
71.43
1.30
23.0

59.62
4.40
520
45.19
0.660
17.0
48.08
1.55
140
29.81
0.470
160
78.85
11.5
1,700
14.42
1.50
6.49

Sediment
(µg/kg)
24.24
1.70
140
39.39
0.710
340
63.64
1.70
710
72.73
9.10
2,700
48.48
1.07
130
66.67
2.45
950
75.76
1.30
380
93.94
31.0
190,000
12.12
1.10
59.0

***DRAFT FOR PUBLIC COMMENT***

Surface water Groundwater
(µg/L)
(µg/L)
84.00
0.076
110
80.00
0.106
317
96.00
0.320
292

85.51
0.180
64.0
78.26
0.200
110
94.20
0.820
120

88.00
0.710
815
84.00
0.099
57.0
88.00
0.382
210
52.00
0.014
15.0
96.00
2.17
8,970
36.00
0.096
10.0

94.93
0.870
290
85.51
0.235
75.0
89.86
0.405
250
48.55
0.032
12.0
84.06
4.22
4,300
46.38
0.105
3.00

 PERFLUOROALKYLS

587
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-17. Summary of Perfluoroalkyl Compounds Detected in Soil, Sediment,
Surface Water, and Groundwater at 10 Military Installationsa

Compound Parameter
PFDeA

DF
Median
Maximum
DF
Median
Maximum
DF
Median
Maximum

PFUA

PFDoA

aWater

Surface soil Subsurface
(µg/kg)
soil (µg/kg)
67.03
0.980
15.0
45.05
0.798
10.0
21.98
1.95
18.0

12.50
1.40
9.40
9.62
1.15
2.00
6.73
2.40
5.10

Sediment
(µg/kg)
48.48
1.90
59.0
24.24
160
14.0
45.45
2.80
84.0

Surface water Groundwater
(µg/L)
(µg/L)
52.00
0.067
3.20
20.00
0.021
0.210
20.00
0.058
0.071

34.78
0.023
1.80
8.70
0.025
0.086
4.35
0.022
0.062

concentrations are ppb (µg/L); soil and sediment levels are ppb (µg/kg).

DF = detection frequency as a percentage; PFBA = perfluorobutyric acid; PFBuS = perfluorobutane sulfonic acid;
PFDoA = perfluorododecanoic acid; PFDeA = perfluoro-n-decanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxA = perfluorohexanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid;
PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid
Source: Anderson et al. (2016)

Perfluoroalkyls have been detected in soils that were amended with biosolids (Sepulvado et al. 2011).
Several perfluoroalkyls were detected in biosolid-amended soils, with PFOS being the predominant
compound with levels ranging from 5.5 to 483 ng/g, depending upon the loading rate.

5.5.4

Other Media

Levels of PFOS, PFOA, PFBuS, PFHxS, PFHxA, PFHpA, PFDeA, PFNA, and PFDoA were analyzed in
31 food items collected from 5 grocery stores located in Texas in 2009 (Schecter et al. 2010). PFOA was
the most frequently detected item (detected in 17 of 31 of the food samples), with levels ranging from
0.07 ng/g in potatoes to 1.80 ng/g in olive oil. PFOS, PFHxA, PFHpA, PFNA, PFDeA, and PFDoA were
not detected in any samples. PFBuS and PFHxS were detected in cod at 0.12 and 0.07 ng/g, respectively.
The data for PFOA are summarized in Table 5-18. Several studies have evaluated the levels of
perfluoroalkyls in fish from lakes and rivers in the United States; these data are summarized in
Table 5-19.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

588
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-18. Detections of PFOA in 31 U.S. Food Items
Food

PFOA concentration in ng/g (LOD for non-detects)

Hamburger
Bacon

0.15
0.24

Sliced turkey
Sausage
Ham
Sliced chicken breast
Roast beef
Canned chili

ND (0.02)
0.09
0.02
0.02
ND (0.02)
0.02

Salmon
Canned tuna
Fresh catfish fillet
Tilapia
Cod
Canned sardines

0.23
ND (0.05)
0.30
0.10
0.10
0.19

Frozen fish sticks
Butter
American cheese
Other cheese
Whole milk
Ice cream
Frozen yogurt

0.21
1.07
ND (0.04)
ND (0.04)
ND (0.02)
ND (0.03)
ND (0.02)

Whole milk yogurt
Cream cheese
Eggs
Olive oil
Canola oil
Margarine

ND (0.02)
ND (0.03)
ND (0.04)
1.80
ND (0.05)
0.19

Cereals
Apples
Potatoes
Peanut butter

ND (0.04)
ND (0.02)
0.07
0.10

LOD = limit of detection; ND = not detected; PFOA = perfluorooctanoic acid
Source: Schecter et al. 2010

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

589
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-19. Detections of Perfluoroalkyls in Fish from U.S. Lakes and Rivers
Type of seafood
(location)
PFOA
Lake trout
(Lake Superior)
Lake trout (Lake
Superior)
Lake trout
(Lake Michigan)
Lake trout
(Lake Huron)
Lake trout
(Lake Huron)
Lake trout
(Lake Erie)
Lake trout
(Eastern Lake Erie)
Walleye
(Western Lake
Erie)
Lake trout
(Lake Ontario)
Lake trout
(Lake Ontario)
Mixture of whole
fish
(Missouri River)
Mixture of whole
fish
(Mississippi River)
Mixture of whole
fish
(Ohio River)

Perfluoroalkyl concentration in ng/g

1.1

PFOS
4.8

<0.42

2.3

4.4

16

1.6

39

<0.42

17

1.6

121

<0.42

PFHpA
<0.02–
0.87

PFBuS

PFDoA
0.37

PFOSA
0.25

Reference

PFHxS
<0.01–
0.43
<0.10

PFNA
1.0

PFDeA
0.72

PFUA
0.90

0.70

0.39

1.1

<0.01–
0.87
<0.01–
6.2
<0.10

0.57

0.76

0.74

2.8

2.2

2.7

1.4

1.3

1.8

2.9

4.9

3.5

96

<0.01–
1.2
1.4

2.6

6.1

5.7

2.0

DeSilva et al. 2011

0.50

1.1

<0.10

1.2

3.6

3.1

1.1

DeSilva et al. 2011

1.5

46

0.65

1.1

1.8

1.6

0.88

2.5

0.70

0.90

1.4

2.

<0.02–
1.39
0.64

<1.00

84.7

1.89

0.43

0.25

<1.00

1.53

<0.20

83.1

0.42

0.78

1.24

3.38

<1.00

147

0.52

1.03

3.88

6.57

0.97
<0.02–
0.97
<0.02–
1.43

DeSilva et al. 2011

0.41

0.99

Furdui et al. 2007

0.88

1.6

Furdui et al. 2007

0.74
<0.02–
0.71

Furdui et al. 2007

0.97

0.70

DeSilva et al. 2011
2.1

0.82

Furdui et al. 2007

Furdui et al. 2007

0.32

DeSilva et al. 2011

<0.40

0.49

Ye et al. 2008

0.27

<0.20

<0.40

Ye et al. 2008

<4.00

<0.40

1.72

Ye et al. 2008

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

590
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-19. Detections of Perfluoroalkyls in Fish from U.S. Lakes and Rivers
Type of seafood
(location)
PFOA
Smallmouth bass
(Raisin River)
Smallmouth bass
(St Clair River)
Smallmouth bass
(Calumet River)
Carpa
(Saginaw Bay)
Carpa
(Saginaw Bay)
Lake whitefisha
(Michigan waters)
Chinook salmona
(Michigan waters)
Brown trouta
(Michigan waters)
Bluegill
(St Croix River,
Minnesota)
Bluegill
(Lake Calhoun,
Minnesota)
Bluegill
(Haw River, North
Carolina)
Mixture of fish fillet
(Zumbrol Lake,
Minnesota)
Mixture of fish fillet
(McCarrons Lake,
Minnesota)

Perfluoroalkyl concentration in ng/g
PFOSA
<1–4.1

Reference

<1

1.1–6.3

Kannan e al. 2005

2.5–7.6

<1

<1

Kannan e al. 2005

124

<34

<19

Kannan e al. 2005

<2

PFOS
PFHxS
2.0–41.3 <1

<2

<2–2.7

<2
<36

PFNA

PFDeA

PFUA

PFHpA

PFBuS

PFDoA

120
130
110
<6–46

Kannan e al. 2005

Giesy and Kannan
2001
Giesy and Kannan
2001
Giesy and Kannan
2001
Giesy and Kannan
2001
Delinsky et al.
2009

2.87

<LOQ

<LOQ

<LOQ

272

5.82

4.18

4.72

Delinsky et al.
2009

29.8

10.3

26.9

7.25

Delinsky et al.
2009

52.4

3.24

Delinsky et al.
2010

47.3

1.97

Delinsky et al.
2010

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

591
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-19. Detections of Perfluoroalkyls in Fish from U.S. Lakes and Rivers
Type of seafood
(location)
PFOA
Mixture of fish fillet
(Pickerel Lake,
Minnesota)
Mixture of fish fillet
(Carlos Lake,
Minnesota)
Mixture of fish fillet
(other Minnesota
lakes)
Mixture of fish fillet
(Mississippi River
sites, Minnesota)
aPerfluoroalkyl

Perfluoroalkyl concentration in ng/g
PFOS
10.0

PFHxS

PFNA

PFDeA
1.23

PFUA

PFHpA

PFBuS

PFDoA

PFOSA

Reference
Delinsky et al.
2010

12.3

Delinsky et al.
2010

1.08–
5.13

Delinsky et al.
2010

3.06–
20.00

0.47

2.94–
15.0

2.13–
6.72

3.74–
4.42

Delinsky et al.
2010

levels measured in muscle.

LOQ = limit of quantitation; PFBuS = perfluorobutane sulfonic acid; PFDoA = perfluorododecanoic acid; PFDeA = perfluorodecanoic acid;
PFHpA = perfluoroheptanoic acid; PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOS = perfluorooctane sulfonic acid;
PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

592
5. POTENTIAL FOR HUMAN EXPOSURE

Concentrations of perfluoroalkyls have been reported in foods sampled in Canada, the United Kingdom,
and Germany (Food Standards Agency 2006; Fromme et al. 2007b; Tittlemier et al. 2007).
Perfluoroalkyls were detected in only 9 out of 54 food composites collected during Canadian Total Diet
studies from 1992 to 2004 (Tittlemier et al. 2007). PFOS was detected in beef steak, ground beef,
luncheon meats, marine fish, freshwater fish, and microwave popcorn at concentrations ranging from
0.98 to 2.7 ng/g, wet weight. PFOA was detected in roast beef, pizza, and microwave popcorn at 0.74–
3.6 ng/g, wet weight. PFHpA was detected in pizza and microwave popcorn at 1.5–2.0 ng/g, wet weight.
PFNA was detected only in beef steak at 4.5 ng/g, wet weight. PFDeA, PFUA, and PFDoA were
analyzed for but not detected in any of the food composites. During the U.K. Food Standards Agency
Total Diet Study, PFOS was detected in eggs, sugars and preserves, potatoes, and canned vegetables at 1,
1, 10, and 2 μg/kg, respectively (Food Standards Agency 2006). PFOA was detected only in potatoes at
1 μg/kg. Neither substance was detected in the bread, miscellaneous cereals, carcass meats, offal, meat
products, poultry, fish, oils and fats, green vegetables, other vegetables, fresh fruit, fruit products,
beverages, milk, dairy products, or nuts categories. Fromme et al. (2007b) detected PFOS, PFOA, and
PFHxS in 33, 45, and 3%, respectively, of 214 daily duplicate food portions for 31 adults in the city of
Munich, Germany. Concentrations were 0.025–1.03 ng/g fresh weight for PFOS, 0.025–118.29 ng/g
fresh weight for PFOA, and 0.05–3.03 ng/g fresh weight for PFHxS. Reported 90th percentile values were
0.11 and 0.21 ng/g fresh weight for PFOS and PFOA, respectively (Fromme et al. 2007b).

The temporal trend of perfluoroalkyl residues in eggs, milk, and farmed rainbow trout from Sweden were
studied from 1999 to 2010 (Johansson et al. 2014). Over this period, the mean annual decreases in levels
of PFOS were 18 and 31% in rainbow trout and eggs, respectively. The mean annual decreases of PFOA
and PFHxS were 12 and 11%, respectively, in eggs. The detection frequency of PFOA and PFHxS was
too low in milk samples to assess changes in levels over the time period, and decreases in the levels of
PFOS were found to be not statistically significant over the temporal period. The mean annual decrease
in levels of PFHxS in rainbow trout was 4.3% annually.

Elevated levels of PFOS were measured in water and fish samples obtained from 2009 and 2012 at six
sampling locations along the Welland River and Lake Niapenco in Ontario, Canada (Gewurtz et al. 2014).
These locations were downstream from the Hamilton International airport where PFOS containing AFFF
was used until the mid-1990s at a firefighting training facility at the airport. PFOS concentrations were
generally highest in benthic feeding fish collected at the sampling locations nearest to the airport. The
maximum level of PFOS was observed in common carp collected at a site near the airport at a
concentration of 2,300 ng/g. Maximum levels of PFOS in smallmouth bass, largemouth bass, and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

593
5. POTENTIAL FOR HUMAN EXPOSURE

channel catfish were 640, 450, and 430 ng/g, respectively. Freshwater fish and seafood obtained from
China had detectable levels of nine perfluoroalkyls, with PFOS being detected in 62% of the samples at a
concentration range of <0.10–26.2 ng/g and PFOA detected in 70% of the samples at levels of <0.10–
1.99 ng/g (Zhang et al. 2011).
PFOA was detected in the packaging paper of two microwave popcorn bags at 0.3–4.7 ng/cm2 uncooked
and 0.5–4.3 ng/cm2 cooked (Sinclair et al. 2007). The mean mass of PFOA in the gas phase of popcorn
vapors following popping was 16–17 ng/cm2. PFHpA, PFNA, PFDeA, PFUA, and PFDoA were detected
in one of the bags at 0.4–3.2 ng/cm2 uncooked and 0.5–4.3 ng/cm2 cooked; however, these
perfluoroalkyls were not detected (<0.2 ng/cm2) in the second bag. Begley et al. (2005) measured PFOA
concentrations of 6–290 μg/kg in microwave popcorn bags. These authors also tested a hamburger
wrapper, sandwich wrapper, French fry box, and soak-proof paper plates and did not find PFOA above
the detection limit in these products. The concentration of PFOA measured in undiluted perfluoro paper
coating formulations ranged from 88,000 to 160,000 μg/kg (Begley et al. 2005).
A study of perfluorinated and polyfluorinated substances in food packaging from U.S. fast food
restaurants found perfluoroalkyl carboxylates (such as PFOA and PFHxA), perfluoroalkyl sulfonates
(such as PFBS), fluorotelomer sulfonates, and unknown polyfluorinated compounds (Schaider et al.
2017). PFOA was detected in 6 of the 20 samples collected in 2014–2015. Fluorotelomer alcohols that
can degrade or metabolize to perfluoroalkyl carboxylic acids such as PFOA continue to be detected in
food packaging materials. Yuan et al. (2016) analyzed 69 food contact materials (paper tableware, paper
cups, cupcake cups, paper boxes, paper bags, and microwave popcorn bags) produced in China and
25 materials (paper tableware, microwave popcorn bags, and paper cups) produced in the United States.
The median concentration of total fluorotelomer alcohols in food contact materials produced in China
(sum of 6:2, 8:2, 10:2, 12:2, 14:2, 16:2, and 18:2 fluorotelomer alcohol) ranged from 2 to 18,200 ng/g,
with the highest levels observed in microwave popcorn bags. The detection frequencies of 6:2, 8:2, 10:2,
12:2, 14:2, 16:2, and 18:2 fluorotelomer alcohol in all food contact materials were reported as 38, 65, 77,
70, 58, 35, and 30%, respectively. The only fluorotelomer detected in paper tableware produced in the
United States was 6:2 fluorotelomer alcohol, with a detection frequency of 11%; however, all
fluorotelomer alcohols, with the exception of 18:2 fluorotelomer alcohol, were detected in microwave
popcorn bags from the United States, although the levels were much lower than those produced in China.
Table 5-20 shows the median concentration of the fluorotelomer alcohols in microwave popcorn bags
produced in the United States versus those produced in China. In contrast to the products produced in
China, the predominant fluorotelomer alcohol currently detected in food contact materials produced in the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

594
5. POTENTIAL FOR HUMAN EXPOSURE

United States appears to be 6:2 fluorotelomer alcohol. The authors concluded that the results of the
PFOA Stewardship Program had effectively reduced the usage of long-chain fluorotelomer alcohols in the
United States; however, they were still widely present in products produced in China.

Table 5-20. Fluorotelomer Alcohols Detected in Microwaveable Popcorn Bags
Produced in China and the United States

Fluorotelomer
alcohol

Median concentration of FTOH in
microwave popcorn bag produced in
China (ng/g)

Median concentration of FTOH in
microwave popcorn bag produced in
the United States (ng/g)

6:2 FTOH
8:2 FTOH
10:2 FTOH
12:2 FTOH
14:2 FTOH
16:2 FTOH

80
4,810
6,700
5,650
384
61

485
1.36
0.73
0.55
0.12
<MQL

18:2 FTOH

7.5

Not detected

FTOH = fluorotelomer alcohol; MQL = method quantitation limit
Source: Yuan et al. 2016

Washburn et al. (2005) measured the concentration of the perfluorooctanoate anion in fluorotelomertreated consumer articles as well as the fluorotelomer formulations used for the treatments. PFOA was
detected in mill-treated carpeting (0.2–0.6 mg/kg), carpet-care solution-treated carpeting (0.2–2 mg/kg),
treated apparel (<0.02–1.4 mg/kg), treated home textiles (<0.02–1.4 mg/kg), industrial floor waxes and
wax removers (0.0005–0.06 mg/kg), latex paint (0.02–0.08 mg/kg), and home and office cleaners (0.005–
0.05 mg/kg). The concentrations of PFOA measured in the formulations used for these applications were
30–80, 1–50, <1–40, <1–40, 5–120, 50–150, and 50–150 mg/L, respectively. PFOA was not detected in
treated upholstery (<0.034 mg/kg), treated technical textiles (<0.034 mg/kg), treated nonwoven medical
garments (<0.034 mg/kg), or stone, tile, and wood sealants (<0.1 mg/kg).

Liu et al. (2014) measured levels of perfluoroalkyls in 35 consumer products that are typically used
indoors, such as treated home textiles, food contact paper, carpet care products, and floor waxes. All
products were obtained from retail stores in the United States between March 2007 and September 2011.
The general trend was that these products contained decreasing quantities of perfluorocarboxylic acids,
including PFOA, over the temporal period studied; however, there was an increase in use of PFBuS,
presumably as a replacement of PFOS in consumer products. Levels of perfluoroalkyls were analyzed in
115 random samples of consumer products obtained in Germany in 2010, including textiles, carpets,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

595
5. POTENTIAL FOR HUMAN EXPOSURE

cleaning and impregnating agents, leathers, food contact materials, baking and sandwich paper, and ski
waxes (Kotthoff et al. 2015). Cleaning agents and some baking and sandwich papers had nondetectable
or negligible amounts of perfluoroalkyls; however, PFOA and PFOS were frequently detected in outdoor
textiles, ski wax, leather products, gloves, awning cloths, nanosprays, impregnation sprays, and food
contact materials. Cleaning agents, nanosprays, and impregnation sprays tended to have the highest level
of fluorotelomer alcohols. EPA (2009b) analyzed 116 articles of commerce obtained from retail outlets in
the United States from March 2007 and May 2008 for the presence of perfluorocarboxylic acids ranging
in carbon number from C5 (perfluoropentanoic acid) to C12 (perfluorododecanoic acid). Total C5–C12
perfluorocarboxylic acids levels ranged from below the detection limit (1.5 ng/g) to 47,100 ng/g, with
levels of PFOA ranging from below the detection limit to 6,750 ng/g.

PTFE is a fluoropolymer used in applications such as nonstick cookware coatings and plumbing sealant
tape. In the past, PFOA had been used as a processing aid in the emulsion polymerization of PTFE
(DuPont 2008). PFOA was largely removed from the fluoropolymer material during the baking and
curing step of nonstick cookware coatings in a high temperature oven; however, residual PFOA could be
found in the final coatings (DuPont 2008). Begley et al. (2005) measured PFOA concentrations of 4–
75 μg/kg in PTFE cookware, 3 μg/kg in PTFE-based dental floss, 4 μg/kg in PTFE-based dental tape, and
1,800 μg/kg in PTFE film/sealant tape. PFOA was not detected in tubing made of a fluoro-ethylenepropene copolymer (Begley et al. 2005).

Studies have been conducted that investigated the release of PFOA from PTFE cookware when heated.
Sinclair et al. (2007) reported PFOA release concentrations ranging from 19 to 287 pg/cm2 measured
using four new nonstick frying pans. These concentrations were measured at normal cooking
temperatures within the range of 180–229°C. PFOA was detected in water (7 and 75 ng) boiled for
10 minutes in two out of five non-stick pans (Sinclair et al. 2007). PFOA was not found above the
detection limit (0.1 ng/cm2) during 40 extraction tests on PTFE cookware using an ethanol/water mixture
(Washburn et al. 2005). Likewise, Powley et al. (2005) conducted extraction texts on commercial
fluoropolymer-treated cookware using water and water/ethanol mixtures at 100 and 125°C. Under
simulated cooking conditions, PFOA was not identified above the detection limit of 100 pg/cm2. Begley
et al. (2005) reported that additional PFOA was not generated in the PTFE coating of three empty pans
heated to 320°C (DuPont 2008). According to DuPont, the non-stick coating on a pan may begin to
deteriorate if the pan is accidentally heated above 348°C, which is well above the maximum
recommended cooking temperature of 260°C (DuPont 2008). Although it is possible for an unattended

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

596
5. POTENTIAL FOR HUMAN EXPOSURE

empty pan to reach these high temperatures, overheating non-stick cookware is expected to be prevented
in most cases because food oils begin to generate smoke around 190°C (Begley et al. 2005).

Perfluoroalkyls in indoor and outdoor environments in urban, suburban, and rural locations near Toronto,
Ontario, Canada were assessed using window films for passive sampling. The sum of perfluoroalkyls
concentrations on outdoor window films ranged from 0.04 to 0.75 pg/cm2 in winter and from 0.04 to
0.92 pg/cm2 in summer, with higher values found in urban and suburban locations than in rural locations.
Indoors, concentrations on window film ranged from less than the detection limit (which ranged from
0.033 to 0.06 pg/cm2) to 2.1 pg/cm2 in winter and from 0.08 to 4.3 pg/cm2 in summer, although there were
no distinct trends between urban and rural for indoor concentrations (Gewurtz et al. 2009).

A comprehensive study that examined 116 articles of commerce (AOC) found perfluorocarboxylic acids,
including PFOA, in many commercially available substances, such as carpet care products and waxes
(EPA 2009b). Levels of PFOA ranged from nondetectable to 6,750 ng/g, and levels of total
perfluorocarboxylic acids (the sum of C5–C12 acids) ranged from nondetectable to 47,100 ng/g.
Perfluoroalkyl compounds, including PFOA, have been detected at low levels in personal care products
such as cosmetics and sunscreens (Fujii et al. 2013).
5.6

GENERAL POPULATION EXPOSURE

As a group of compounds, perfluoroalkyls appear to be ubiquitous in human blood based on the
widespread detection of these substances in human serum samples (Calafat et al. 2006b, 2007a, 2007b;
De Silva and Mabury 2006; Kuklenyik et al. 2004; Olsen et al. 2003b, 2003c, 2004b, 2004c, 2005,
2007a). Tables 5-21 and 5-22 list concentrations of perfluoroalkyl compounds measured in serum
samples collected from the general population in the United States. There is a clear trend in decreasing
serum levels of both PFOS and PFOA in the general population of the United States since 2000 as these
substances were phased out.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

597
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-21. Concentrations of PFOA and PFOS in Human Serum Collected in the
United States
Detection and concentration (ng/mL [ppb])a
Location

PFOA

PFOS

Reference

U.S. residents—NHANES
1999–2000 (n=1,562)
Percent >LOD

Calafat et al. 2007a
100%

100%

LOD

0.1

0.2

Geometric mean

5.2

30.4

11.9

75.6

95th percentile
2003–2004 (n=2,094)
Percent >LOD

Calafat et al. 2007b
99.7%

99.9%

LOD

0.1

0.4

Geometric mean

3.95

20.7

95th percentile

9.80

54.6

2005–2006 (n=2,120)
Percent >LOD

NR

NR

LOD

0.1

0.2

Geometric mean

3.92

17.1

95th percentile

11.3

47.5

2007–2008 (n=2,100)
Percent >LOD

CDC 2018
NR

NR

LOD

0.1

0.2

Geometric mean

4.12

13.2

95th percentile

9.60

40.5

2009–2010 (n=2,233)
Percent >LOD

CDC 2018
NR

NR

LOD

0.1

Geometric mean

3.07

95th percentile

7.50

0.2
9.32
32.0

2011–2012 (n=1,904)
Percent >LOD

CDC 2018
NR

NR

LOD

0.1

0.2

Geometric mean

2.08

6.31

95th percentile

5.68

21.7

2013–2014 (n=2,165)
Percent >LOD

CDC 2018

CDC 2018
NR

NR

LOD

0.1

0.2

Geometric mean

1.94

4.99

95th percentile

5.57

18.5

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

598
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-21. Concentrations of PFOA and PFOS in Human Serum Collected in the
United States
Detection and concentration (ng/mL [ppb])a
Location

PFOA

PFOS

Reference

U.S. residents
1990–2002 (n=23 pooled
samples)
Percent >LOD

Calafat et al. 2006b
100%

100%

LOD

0.2

0.4

Geometric mean

9.6

30.0

23.0

52.3

Percent >LLOQb

92%

99.8%

Geometric mean

4.6

34.9

12.1

88.5

52.3

1,656.0

95th percentile
U.S. blood donors
2000–2001 (n=645)

95th

percentilec

Maximum

Olsen et al. 2003b

2006 (n=600)

Olsen et al. 2017b

Geometric mean

3.44

14.5

95th percentile

7.9

31.5

2010 (n=600)

Olsen et al. 2017b

Geometric mean

2.44

8.3

95th

5.6

21.8

percentile

2015 (n=616)

Olsen et al. 2017b

Geometric mean

1.09

4.3

95th percentile

3.2

8.6

U.S. residents
(n=24)
Percent

Olsen et al. 2003c
>LLOQd

Geometric mean

NR

98%

2.5

14.7

Minimum

<3.0

<6.1

Maximum

7.0

58.3

Midwestern United States
2004–2005 (n=16)
Percent

detectedd

De Silva and Mabury 2006
100%

Mean

4.4

Maximum

8.6

Not detected

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

599
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-21. Concentrations of PFOA and PFOS in Human Serum Collected in the
United States
Detection and concentration (ng/mL [ppb])a
Location

PFOA

PFOS

Reference

Minneapolis-St. Paul blood donors (plasma)
2005 (n=40)
Percent >LLOQ
LLOQ

Olsen et al. 2007b
95%
NR

100%
3.4

Geometric mean

2.2

15.1

75th percentile

3.5

20.2

Maximum

4.7

36.9

100%

100%

LOD

0.1

0.4

Mean

4.9

55.8

Minimum

0.2

3.6

Maximum

10.4

164.0

Atlanta, Georgia
2003 (n=20)
Percent >LOD

Kuklenyik et al. 2004

Seattle, Washington elderly individuals
(n=238)
Percent

Olsen et al. 2004c
>LLOQb

Geometric mean
95th

percentileb

Maximum

99.2%

99.5%

4.2

31.0

9.7

84.1

16.7

175.0

Washington County, Maryland
1974 (n=178)
Percent >LLOQ

Olsen et al. 2005
71%

100%

LLOQ

1.9

3.9

Geometric mean

2.1

30.1

75th percentile

3.0

40.2

Percent >LLOQ

99%o

100%p

Geometric mean

5.5

33.3

LLOQ

1.9

3.9

75th percentile

6.7

44.0

1989 (n=178)

Olsen et al. 2005

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

600
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-21. Concentrations of PFOA and PFOS in Human Serum Collected in the
United States
Detection and concentration (ng/mL [ppb])a
Location

PFOA

PFOS

Reference

Boston, Massachusetts; Charlotte, North Carolina; Hagerstown, Maryland; Los Angeles, California;
Minneapolis-St. Paul, Minnesota; Portland, Oregon
2006 (n=600)

Olsen et al. 2008

Percent >LLOQ

99%

99%

Geometric mean

3.4

14.5

LLOQ

1.0

<2.5

95th percentile CI
geometric mean

3.3–3.6

13.9–15.2

a"Less

than" values indicate that the concentration was reported as below the LOD or LLOQ. For cases where
samples had concentrations below the limit of detection or lower limit of quantification, a value between zero and the
LOD or LLOQ was assigned when calculating the mean concentration.
bexperimental LLOQs not determined.
cReported as bias-corrected estimates.
dLLOQ, LOQ, or LOD not reported.
CI = confidence interval; LLOQ = lower limit of quantification; LOD = limit of detection; NR = not reported;
PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

Table 5-22. Concentrations of Other Perfluoroalkyls in Human Serum Collected
in the United States
Detection and concentration (ng/mL [ppb])a
Sample
population

PFHpA PFNA

PFDeA PFUA

MeEtPFOSA PFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH -AcOH

U.S. residents NHANES
1999–2000 (n=1,562) (Calafat et al. 2007a)
Percent >LOD 10%

95%

25%

12%

<1%

—

100%

100%

96%

91%

LOD

0.4

0.1

0.2

0.2

0.2

—

0.1

0.05

0.2

0.2

Geometric
mean

<0.4

0.5

<0.2

<0.2

<0.2

2.1

0.4

1.0

0.6

95th percentile

NR

1.7

0.5

NR

NR

8.7

1.4

3.2

2.2

2003–2004 (n=2,094) (Calafat et al. 2007b)
Percent >LOD 6.2%

98.8% 31.3% 9.7%

<0.1% <0.4% 98.3% 22.2% 27.5% 3.4%

LOD

0.3

0.1

0.3

0.3

1.0

0.4

0.3

0.2

0.6

0.4

Geometric
mean

<0.3

1.0

<0.3

<0.3

<1.0

<0.4

1.9

<0.2

<0.6

<0.4

3.2

0.8

0.6

<1.0

<0.4

8.3

0.2

1.3

<0.4

95th percentile 0.4

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

601
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-22. Concentrations of Other Perfluoroalkyls in Human Serum Collected
in the United States
Detection and concentration (ng/mL [ppb])a
Sample
population

PFHpA PFNA

PFDeA PFUA

MeEtPFOSA PFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH -AcOH

2005–2006 (n=2,120) (CDC 2018)
Percent >LOD NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

LOD

0.4

0.1

0.2

0.2

0.2

0.1

0.1

0.1

0.174

0.2

Geometric
mean

—

1.09

0.355

—

—

—

1.67

—

0.410

—

95th percentile

0.700

3.60

1.50

0.700

<LOD

0.1

8.30

0.300

1.57

0.300

2007–2008 (n=2,100) (CDC 2018)
Percent >LOD NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

LOD

0.4

0.082

0.2

0.2

0.2

0.1

0.1

0.1

0.174

0.2

Geometric
mean

—

1.22

0.286

—

—

—

1.95

—

0.303

—

95th percentile

0.500

3.28

0.900

0.600

<LOD

<LOD

9.80

<LOD

1.30

<LOD

2009–2010 (n=2,233) (CDC 2018)
Percent >LOD NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

LOD

0.1

0.082

0.1

0.1

0.1

0.1

0.1

0.1

0.087

0.1

Geometric
mean

—

1.26

0.279

0.172

—

—

1.66

—

0.198

—

95th percentile

0.200

3.77

0.900

0.900

<LOD

<LOD

6.90

<LOD

1.00

0.100

2011–2012 (n=1,904) (CDC 2018)
Percent >LOD NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

LOD

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

Geometric
mean

—

0.881

0.199

—

—

—

1.28

—

—

—

95th percentile

0.220

2.54

0.690

0.620

0.140

<LOD

5.44

<LOD

0.690

0.110
NR

2013–2014(n=2,168) (CDC 2018)
Percent >LOD NR

NR

NR

NR

NR

NR

NR

NR

NR

LOD

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

Geometric
mean

—

0.675

0.185

—

—

—

1.35

NR

—

NR

95th percentile

0.200

2.00

0.700

0.500

0.200

<LOD

5.60

NR

0.600

NR

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

602
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-22. Concentrations of Other Perfluoroalkyls in Human Serum Collected
in the United States
Detection and concentration (ng/mL [ppb])a
Sample
population

PFHpA PFNA

PFDeA PFUA

MeEtPFOSA PFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH -AcOH

U.S. residents (Calafat et al. 2006b)
1990–2002 (n=23
pooled samples)
Percent >LOD 0%

8.7%

0%

13%

0%

91.3% 26.1% 13%

56.5%

LOD

0.3

0.1

0.3

0.3

1

0.3

0.2

0.6

0.4

Geometric
mean

NA

<0.3

NA

<0.3

NA

1.6

<0.4

<0.6

<0.4

0.3

NA

1.3

NA

2.7

0.7

1.9

2.5

—

—

52%

2%

40%

42%

Geometric
mean

1.9

NR

1.3

2.0

95th percentileb

9.5

NR

5.0

7.6

Maximum

66.3

NR

16.4

60.1

95th percentile NA

—

U.S. blood donors (Olsen et al. 2003b)
2000–2001 (n=645)
Percent
>LLOQc

—

—

—

—

U.S. blood donors (Olsen et al. 2017a, 2017b)
2006 (n=600)
Percent>LLOQ 62%

100%

99.8% 99.8% 68%

1.2%

95.7%

62.4% 1.7%

LLOQd

<0.5

NR

<0.05

<0.05

<0.05

<0.5

<0.5

<1.3

<1.0

Geometric
mean

0.09

0.97

0.34

0.18

0.04

LLOQ

1.52

0.60

LLOQ

95th percentile

0.4

2.2

0.8

0.5

0.07

LLOQ

5.7

1.8

LLOQ

Percent>LLOQ 79.7% 100%

100%

99.8% 46.2% 22.5% 95.5%

LLOQd

<0.05

NR

NR

<0.025 <0.025 <1.0

<0.05

Geometric
mean

0.05

0.83

0.27

0.14

0.03

LLOQ

1.34

95th percentile

0.2

2.3

0.8

0.5

0.06

0.3

5.3

Percent>LLOQ 3.3%

100%

96.9% 47.9% 0.6%

8.4%

99.7% 0%

LLOQd

<0.093 <0.093 <0.466 <0.932 <0.047 <0.04

<0.093 <0.09

<0.23

0.43

0.15

1.1

0.49

2010 (n=600)

2015 (n=616)
<0.09

Geometric
mean
95th percentile

0.16

0.25

LLOQ

63.1% 8.2%

LLOQ

0.87

LLOQ

0.09

LLOQ

0.02

3.5

LLOQ

0.43

0.05

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

603
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-22. Concentrations of Other Perfluoroalkyls in Human Serum Collected
in the United States
Detection and concentration (ng/mL [ppb])a
Sample
population

PFHpA PFNA

PFDeA PFUA

MeEtPFOSA PFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH -AcOH

—

—

U.S. residents (Olsen et al. 2003c)
(n=24)
Geometrice
mean

—

—

—

—

1.8

3.0

—

—

Minimum

<1.2

<1.3

Maximum

5.9

22.1

—

—

—

—

—

—

100%

75%

100%

90%

0.2

0.6

0.4

Midwestern United States (De Silva and Mabury 2006)
2004–2005
(n=16)
Percent
detectede

—

100%

100%

13%

0%

Mean

0.77

0.17

NR

NA

Maximum

1.2

0.25

0.067

NA

85%

10%

Atlanta, Georgia (Kuklenyik et al. 2004)
2003 (n=20)
Percent >LOD 10%

100%

75%

LOD
Meanb

<0.3

2.6

0.7

0.8

<1

3.9

0.34

1.7

0.9

Maximum

8.5

3.9

1.2

1.4

1.6

11.2

0.7

5.2

1.4

—

—

76%

"Few"

65%

52%

Geometric
mean

2.2

NR

1.2

<1.6

95th percentileb

8.3

NR

3.8

7.8

Maximum

40.3

NR

6.6

21.1

63%

0%

4%

33%

Seattle, Washington (Olsen et al. 2004c)
(n=238)
Percent
>LLOQc

—

—

—

—

Washington County, Maryland (Olsen et al. 2005)
1974 (n=178)
Percent >LLOQ —

—

—

—

—

—

LLOQ

1.4

1.0

1.6

Geometric
mean

1.5

NA

0.5

1.2

75th percentile

2.5

NA

<1.0

1.8

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

604
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-22. Concentrations of Other Perfluoroalkyls in Human Serum Collected
in the United States
Detection and concentration (ng/mL [ppb])a
Sample
population

PFHpA PFNA

PFDeA PFUA

MeEtPFOSA PFOSA
PFDoA PFBuS PFHxS PFOSA -AcOH -AcOH

—

—

1989 (n=178)
Percent >LLOQ —

—

—

—

82%

0%

38%

93%

LLOQ

1.4

1.0

1.6

Geometric
mean

2.5

NA

0.8

3.6

75th percentile

1.6

NA

1.3

4.7

a"Less

than" values indicate that the concentration was reported as below the LOD or LLOQ. For cases where
samples had concentrations below the LOD or LLOQ, a value between zero and the LOD or LLOQ was assigned
when calculating the mean concentration.
bReported as bias-corrected estimates.
cExperimental LLOQs not determined.
dhighest LLOQ listed.
eLOD not reported.
“—“ indicates no available data; Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid;
LLOQ = lower limit of quantification; LOD = limit of detection; Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane
sulfonamide) acetic acid; NA = not applicable; NR = not reported; PFBuS = perfluorobutane sulfonic acid;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOSA = perfluorooctane sulfonamide;
PFUA = perfluoroundecanoic acid

Calafat et al. (2007a, 2007b) analyzed serum levels of different subsamples of the general population
during the 1999–2000 and 2003–2004 periods of the NHANES. The numbers of individuals included in
the analyses for each survey period were 1,562 and 2,094, respectively. PFOA, PFOS, PFNA, and
PFHxS were detected in 95–100% of serum samples collected during both survey periods. In an analysis
of NHANES data from 1999 to 2008, Kato et al. (2011) found that males had significantly higher levels
of PFOA, PFOS, and PFNA than females and that PFOS levels increased with age, especially in females.
An upward trend of PFNA levels and a downward trend for PFHxS levels were also noted. Mean
concentrations for PFOA, PFOS, and PFHxS declined by 10–30% in the 2003–2004 survey, while PFNA
values doubled from 0.5 to 1.0 ng/mL. NHANES survey data from 2005–2006, 2007–2008, 2009–2010,
2011–2012, and 2013–2014 have generally continued to show declining levels of PFOA and PFOS in
human serum samples (CDC 2015, 2018). A dramatic difference in detection frequency was observed for
PFOSA, Me-PFOSA-AcOH, and Et-PFOSA-AcOH, which were widely detected (91–100%) during the
1999–2000 survey period but were present in only 3.4–27.5% of samples collected during the 2003–2004
survey period. PFDoA and PFBuS were detected in <1% of the NHANES samples. The geometric mean
serum concentration of PFOS has declined nearly 80% from NHANES survey years 1999–2000 to 2013–

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

605
5. POTENTIAL FOR HUMAN EXPOSURE

2014 and the geometric mean serum levels of PFOA have declined nearly 70% over the same temporal
period. Olsen et al. (2008) reported a nearly 60% decline in PFOS blood levels when comparing data
from 2001 to 2006 American Red Cross surveys of participants.

Levels of perfluoroalkyl compounds have been measured in indoor air, outdoor air, dust, food, surface
water, and various consumer products. Possible exposure pathways have been proposed; however, the
relative importance of these pathways, including their association with the accumulation of
perfluoroalkyls in blood, remains unclear (Apelberg et al. 2007b; Begley et al. 2005; Calafat et al. 2006b;
Trudel et al. 2008; Washburn et al. 2005). For populations that have elevated levels of perfluoroalkyls in
water supplies, the primary route of exposure is expected to be ingestion of contaminated drinking water.
Using a stratified random sample of residents in the Little Hocking Water district in Ohio between July
2004 and February 2005, Emmett et al. (2006a) reported median serum PFOA levels of 329 ng/mL in
residents of households with a median drinking water concentration of 3.55 ng/mL. Median serum PFOA
levels were 371 ng/mL in residents for whom this was the only residential water source and 71 ng/mL in
those who used bottled, cistern, or spring water. Increased serum PFOA was associated with increasing
number of drinks of tap water daily and also with increasing use of water for making soups and stews and
in-home canning of fruits and vegetables. Use of a carbon water filter reduced PFOA levels by about
25%. In a follow-up study, 231 study participants in the Little Hocking Water District were evaluated
15 months later with 88% using bottled water exclusively; 8% had made other changes to their ingestion
of residential water including use of activated carbon water filters. PFOA levels had decreased an
average of 26% from the initial levels (Emmett et al. 2009).

A study conducted by the Minnesota Department of Health reported higher PFOA, PFHxS, and PFOS
serum levels in residents of two communities with contaminated water supplies as compared to the
general population (MDH 2009). Similar findings have been reported by Steenland et al. (2009a) in a
study of residents in six water districts in the mid-Ohio Valley located near the DuPont Washington
Works facility in Washington, West Virginia. The Minnesota Department of Health instituted a program
to reduce levels of perfluoroalkyls in drinking water that included using granulated activated carbon
(GAC) filters for home use in areas where private wells showed high levels of contamination and large
GAC filters for the municipal water supplies (MDH 2009). In two Mid-Ohio Valley locations with
PFOA-contaminated drinking water, blood serum levels of PFOA in residents declined significantly
following the implementation of GAC filtration of the public water supply (Bartell et al. 2010). The
Lubeck, West Virginia and Little Hocking, Ohio public water systems, which were contaminated with
PFOA from the DuPont Washington Works facility, began GAC treatment to remove PFOA from the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

606
5. POTENTIAL FOR HUMAN EXPOSURE

potable water supply in 2007. The average decrease in serum PFOA levels for Lubeck, West Virginia
residents primarily consuming public water at home (n=130) was 26% a year after treatment began.
Similar trends were reported for residents of Little Hocking, Ohio. The average decrease in PFOA serum
levels for residents primarily consuming public water (n=39) was about 11% 6 months after treatment
began (Bartell et al. 2010).

Trudel et al. (2008) provide a thorough analysis of general population exposure to PFOS and PFOA based
on the available information and have proposed the following possible exposure pathways: food and
water consumption, ingestion of house dust, hand-to-mouth transfer from treated carpets, migration into
food from PFOA-containing paper or cardboard, inhalation of indoor and ambient air, and inhalation of
impregnation spray aerosols. Other pathways proposed to be less significant included oral exposure from
hand-to-mouth contact with clothes and upholstery, migration into food prepared with PTFE-coated
cookware, dermal exposure from wearing treated clothes, deposition of spray droplets on skin while
impregnating, skin contact with treated carpet and upholstery, and deposition of dust onto skin (Trudel et
al. 2008). The strong correlation between PFOA and PFOS concentrations in human serum samples
indicates that common exposure pathways for these two substances are possible (Calafat et al. 2007a).

In order to estimate human uptake and the major pathways for human exposure to PFOS and PFOA,
reported levels of these compounds in various environmental media, including food and consumer
products, were analyzed with respect to product use patterns, personal activity patterns, and personal
intake rates (Trudel et al. 2008). For PFOS, the major exposure pathways in a high-exposure scenario
were proposed to be food and water ingestion, dust ingestion, and hand-to-mouth transfer from
mill-treated carpets. Relative contributions of these pathways to the total uptake of PFOS in adults were
estimated to be approximately 80, 15, and 5%, respectively (Trudel et al. 2008). For PFOA, the major
exposure pathways in a high-exposure scenario were proposed to be oral exposure resulting from
migration from paper packaging and wrapping into food, general food and water ingestion, inhalation
from impregnated clothes, and dust ingestion. Relative contributions of these pathways to the total uptake
of PFOA in adults were estimated to be approximately 60, 15, 15, and 10%, respectively (Trudel et al.
2008). Major exposure pathways for the intermediate and low exposure scenarios were proposed to be
through food and drinking water (PFOA and PFOS) and ingestion of house dust (PFOA only).

Based on these proposed exposure pathways, adult uptake doses estimated for low, medium, and high
exposure scenarios were approximately 7, 15, and 30 ng/kg body weight/day, respectively, for PFOS and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

607
5. POTENTIAL FOR HUMAN EXPOSURE

approximately 0.4, 2.5, and 41–47 ng/kg body weight/day, respectively, for PFOA (Trudel et al. 2008).
The estimated uptake values were similar for men and women.

Fromme et al. (2009) assessed human exposure to perfluoroalkyls for adults in the general population of
western countries. Based on measured in indoor and outdoor air, house dust, drinking water, and dietary
PFOS and PFOA levels, the investigators estimated average daily exposure levels of 1.6 ng/kg body
weight/day for PFOS and 2.9 ng/kg body weight/day for PFOA. Upper daily exposure levels were
determined to be 8.8 ng/kg body weight/day for PFOS and 12.6 ng/kg body weight/day for PFOA. The
investigators concluded that the oral route, especially diet, was the primary route of exposure to
perfluoroalkyls (Fromme et al. 2007a, 2007b, 2009). The geometric mean adult daily intakes for PFOA
and PFOS were estimated as 92.6 and 83.3 ng/day, respectively, for residents in Kansai, Japan and
53.7 and 63.8 ng/day, respectively, for residents in Tohoku, Japan (Harada and Koizumi 2009). The most
important exposure pathway for both compounds was food ingestion.

Limited monitoring data are available for PFBA. Monitoring efforts conducted in Washington County,
Minnesota near the 3M Cottage Grove Facility revealed widespread contamination of this substance in the
groundwater of that area in 2006. This compound has since also been detected along with PFOA, PFOS,
PFHxS, and PFBuS in municipal drinking water in Washington County (ATSDR 2008). Chang et al.
(2008a) measured concentrations of PFBA in the serum of 127 former employees and 50 current
employees of the 3M Cottage Grove Facility in Minnesota. PFBA serum concentrations were below the
detection limit in 73.2% of the former employees and 68.0% of the current employees. Only 4% of the
serum samples contained PFBA above 2 ng/mL, with maximum concentrations of 6.2 ng/mL for the
former employees and 2.2 ng/mL for the current employees. Landsteiner et al. (2014) conducted a
biomonitoring study of 196 residents of Washington County, Minnesota for the presence of
perfluoroalkyls. The residents of this area were suspected to have been exposed to perfluoroalkyls from
drinking water sources. PFOA, PFOS, and PFHxS were detected in all serum samples collected from the
196 residents, while PFBA and PFBuS were detected in 28 and 23% of the samples, respectively. The
geometric mean serum concentrations of PFOA, PFOS, and PFHxS were 15.4, 35.9, and 8.4 ng/mL,
respectively, which are several times greater than the levels of the U.S. general population when
compared to NHANES survey results shown in Tables 5-21 and 5-22. The authors also noted that the
geometric mean serum concentrations of these three substances were greater in residents obtaining
drinking water from municipal water supplies as compared to residents with private wells.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

608
5. POTENTIAL FOR HUMAN EXPOSURE

Another possible source for perfluoroalkyls in human blood is through uptake of precursor compounds
and then conversion of these within the human body (Trudel et al. 2008). For example, Me-PFOSAAcOH and Et-PFOSA-AcOH are the oxidation products of 2-(N-methyl-perfluorooctane sulfonamido)
ethanol and 2-(N-ethyl-perfluorooctane sulfonamido) ethanol, which were used in surface treatment
applications (Calafat et al. 2006a). Concentrations of Me- and Et-PFOSA-AcOH measured in human
serum may have resulted from exposure of individuals to these perfluoroalkyl sulfonamido ethanols and
then conversion of the ethanols to the perfluoroalkyl sulfonamido acetates within the body.

Levels of perfluoroalkyl compounds measured in the blood of occupationally exposed individuals are
listed in Table 5-23. 3M estimated doses for various on-site exposure scenarios based on monitoring
information collected at the Decatur Facility in Alabama (3M 2008c). Occupational exposure scenarios
included groundskeeper/maintenance worker and construction/utility worker exposed to on-site soils,
surface water, and sediment. According to 3M, estimated on-site exposure to PFOA ranged from
3.2x10-6 to 2.4 ng/kg/day, with the highest estimated exposure corresponding to construction/utility
workers engaged in projects involving contact with soil from an on-site field. Individuals who performed
jobs that require frequent contact with perfluoroalkyl containing products, such as firefighters, waste
handlers, and individuals who install and treat carpets, were also expected to have occupational exposure
to these substances (Emmett et al. 2006a). However, Emmett et al. (2006a) determined that levels of
PFOA in the serum of these types of individuals were only slightly higher than the non-occupational
exposure group (388 ng/mL compared to 329 ng/mL, respectively) while serum levels in workers at a
fluoropolymer manufacturing facility were much higher (775 ng/mL).

Table 5-23. Concentrations of PFOA, PFOS, and PFHxS in Human Serum for
Occupationally Exposed Individuals
Mean concentration (µg/mL [ppm])
PFOA
PFOS
PFHxS

Reference

1993 (n=111)

0.00–80.00 (range)

—

—

Olsen et al. 1998b

1995 (n=80)

0.00–114.10 (range)

—

—

Olsen et al. 1998b

1995 (n=90)

—

2.44

—

Olsen et al. 1999

1997 (n=84)

—

1.96

—

Olsen et al. 1999

2000 (n=215)
1999–2004 (n=26)a
Initial
Final

11.90

1.40

—

Olsen et al. 2003a
Olsen et al. 2007a

0.691
0.262

0.799
0.403

0.290
1.85

Location
Decatur, Alabama

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

609
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-23. Concentrations of PFOA, PFOS, and PFHxS in Human Serum for
Occupationally Exposed Individuals
Mean concentration (µg/mL [ppm])
PFOA
PFOS
PFHxS

Reference

1993 (n=111)

5.0

—

—

Olsen et al. 2000

1995 (n=80)

6.8

—

—

Olsen et al. 2000

1997 (n=74)

6.4

—

—

Olsen et al. 2000

2000 (n=122)

4.63

0.86

—

Olsen and Zobel
2007

Location
Cottage Grove, Minnesota

Washington Works, Little Hocking, Ohio
2004–2005
Substantial occupational
exposure (n=18)

Emmett et al. 2006a
0.824

—

—

1995 (n=88)

—

1.93

—

Olsen et al. 1999

1997 (n=65)

—

1.48

—

Olsen et al. 1999

2000 (n=206)

1.03

0.96

—

Olsen et al. 2003a

—

—

Costa et al. 2009

—

—

Costa et al. 2009

Antwerp, Belgium

Miteni, Trissino, Italy
2007
Current occupational exposure 5.71c (0.20–47.04)
(n=39)
Former occupational exposure 4.43c (0.53–18.66)
(n=11)
2000 (n=25)

18.8

—

—

Costa et al. 2009

2001 (n=42)

19.7

—

—

Costa et al. 2009

2002 (n=46)

19.3

—

—

Costa et al. 2009

2003 (n=41)

13.7

—

—

Costa et al. 2009

2004 (n=34)

11.4

—

—

Costa et al. 2009

2006 (n=49)

10.8

—

—

Costa et al. 2009

2007 (n=50)

11.6

—

—

Costa et al. 2009

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

610
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-23. Concentrations of PFOA, PFOS, and PFHxS in Human Serum for
Occupationally Exposed Individuals

Location

Mean concentration (µg/mL [ppm])
PFOA
PFOS
PFHxS

Reference

0.494 (median)

—

—

Sakr et al. 2007b

0.176 (median)

—

—

Sakr et al. 2007b

0.195 (median)

—

—

Sakr et al. 2007b

0.114 (median)

—

—

Sakr et al. 2007b

Washington Works
2004
Current occupational exposure
(n=259)
Intermittent current
occupational exposure (n=160)
Past occupational exposure
(n=264)
No occupational exposure
(n=342)
aData

include results from three retirees from the 3M plant in Cottage Grove, Minnesota.

“—“ indicates no available data; PFHxS = perfluorohexane sulfonic acid; PFOA = perfluorooctanoic acid; PFOS =
perfluorooctane sulfonic acid

Fu et al. (2014b) analyzed the effects of sex and age on levels of perfluoroalkyl compounds in a study of
133 (79 male, 54 female) participants. In general, higher levels of PFOA, PFOS, PFNA, and PFDeA
were observed in male subjects; however, differences were only statistically significant for PFOA and
PFDeA. Sex differences in other perfluoroalkyl compounds were not observed. For both male and
female subjects, increasing levels of PFOA, PFNA, and PFOS were positively correlated with increasing
age. Data from the 2005–2012 NHANES survey were used to derive a regression-based model to
estimate serum levels of total perfluoroalkyl residues in the human population using only age, sex, and
PFOA and PFOS levels (Jain 2015). Although the data used in the regression-based model were derived
from the U.S. population, the author expected that the data could be used to estimate exposures to
populations elsewhere in the world as well.

Christensen et al. (2016) conducted a biomonitoring study of perfluoroalkyls on male fishermen from
Wisconsin ≥50 years old with a history of sport fish consumption. Increasing age and lower BMI were
generally associated with higher levels of the perfluoroalkyls; however, there were only weak correlations
observed between amounts of fish consumption and perfluoroalkyl levels, with the exception of PFDeA.
Levels of PFOA, PFNA, and PFHxS in the blood of the male anglers were similar to the levels for a
subset of the NHANES 2011–2012 survey (non-Hispanic white males ≥50 years old); however, levels of
PFOS and PFDeA were approximately 2 times greater in the anglers as compared to the NHANES survey
subgroup. The median and 95th percentile concentrations of PFOS in the anglers were 19.00 and

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

611
5. POTENTIAL FOR HUMAN EXPOSURE

54.00 ng/mL, respectively, as compared to 10.33 and 25.83 ng/mL, respectively, in the NHANES study
group. The median and 95th percentile concentrations of PFDeA in the anglers were 0.52 and
1.90 ng/mL, respectively, as compared to 0.23 and 0.53 ng/mL in the NHANES study group.

Perfluoroalkyl compounds have been detected in childhood serum samples, human breast milk, and
umbilical cord blood; reported concentrations are listed in Tables 5-24 and 5-25. Measurements of
perfluoroalkyl compounds in amniotic fluid, meconium, neonatal blood, or other tissues have not been
located.

Table 5-24. Percent Detection and Levels of PFOA and PFOS in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL
[ppb])a
PFOA
PFOS

Location

Reference

Serum
U.S. adolescents—NHANES (ages 12–19)
1999–2000 (n=543)
Percent >LOD
LOD
Geometric mean
95th percentile

Calafat et al. 2007a
100%

100%

0.1

0.2

5.5

29.1

11.2

56.8

99.7%b

99.9%b

2003–2004 (n=640)
Percent >LOD

Calafat et al. 2007b

LOD

0.1

0.4

Geometric mean

3.9

19.3

95th percentile

8.6

42.2

2005–2006 (n=640)
Percent >LOD

CDC 2018
NR

NR

LOD

0.1

0.2

Geometric mean

3.59

15.0

95th percentile

8.40

38.5

2007–2009 (n=357)
Percent >LOD

CDC 2018
NR

NR

LOD

0.1

0.2

Geometric mean

3.91

11.3

95th percentile

7.30

28.0

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

612
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-24. Percent Detection and Levels of PFOA and PFOS in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL
[ppb])a
PFOA
PFOS

Location
2009–2010 (n=364)

Reference
CDC 2013

Percent >LOD

NR

NR

LOD

0.1

0.2

Geometric mean

2.74

6.84

95th percentile

5.00

18.1

2011–2012 (n=344)

CDC 2018

Percent >LOD

NR

NR

LOD

0.1

0.2

Geometric mean

1.80

4.16

95th percentile

3.59

10.8

2013–2014 (n=401)

CDC 2018

Percent >LOD

NR

NR

LOD

0.1

0.2

Geometric mean

1.66

3.54

95th percentile

3.47

9.30

U.S. children—NHANES 2013-2014
Ages 3-5 (n=181)
Percent >LOD

100%

100%

LOD

0.1

0.1

Geometric mean

2.00

3.38

95th percentile

5.58

8.82

Ye et al. 2018a; CDC
2018

Ages 6-11 (n=458)
Percent >LOD

100%

100%

LOD

0.1

0.1

Geometric mean

1.89

4.15

95th percentile

3.84

Ye et al. 2018a; CDC
2018

12.4

U.S. girls (ages 6–8)
Cincinnati, Ohio (n=353)
Percent >LODc

99.7%

99.7%

Geometric mean

7.8

13.2

Median

7.3

13.6

100%

100%

Geometric mean

5.7

13.2

Median

5.8

12.5

Pinney et al. 2014

U.S. girls (ages 6–8)
San Francisco, California (n=351)
Percent >LODc

***DRAFT FOR PUBLIC COMMENT***

Pinney et al. 2014

 PERFLUOROALKYLS

613
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-24. Percent Detection and Levels of PFOA and PFOS in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL
[ppb])a
PFOA
PFOS

Location

Reference

U.S. children (ages 2–12)
1994–1995 (n=598)
Percent >LLOQc
Geometric mean
95th

percentiled

Olsen et al. 2004b
96%

100%
4.9
10

37.5
89

U.S. children (West Virginia
and Ohio; ages 1–19)

Mondal et al. 2012

2005–2006 (n=4,943)
Geometric mean
90th percentile

31.2
201

19.2
36.8

U.S. children NHANES 2001–
2002 (ages 6–11)

Kato et al. 2009b

2001–2002 (n=936)
Least square mean

6.1–7.6

30.45–42.45

Umbilical cord blood
San Francisco, California
Percent >LODc

Morello-Frosch et al.
2016
56%

100%

Geometric mean

–

2.27

95th percentile

1.68

4.35

Baltimore THREE Study

Apelberg et al.
2007a, 2007b

Cord serum (n=299)
Percent >LOD

100%

99%

LOD

0.1–0.2

0.2

Geometric mean

1.6

4.9

Minimum

0.3

<0.2

Maximum

7.1

34.8

Maternal serum (n=293)
Median

1.4–1.6

4.1–5.0

Germany

Midasch et al. 2007

Cord plasma (n=11)
Percent detected

100%

100%

LOQ

0.5

0.5

Median

3.4

7.3

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

614
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-24. Percent Detection and Levels of PFOA and PFOS in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL
[ppb])a
PFOA
PFOS

Location

Reference

Maternal plasma (n=11)
Percent detected

100%

100%

LOQ

0.5

0.5

Median

2.6

13.0

Danish National Birth Cohort

Fei et al. 2007

Cord blood (n=50)
Percent >LLOQ

98%

100%

LLOQ

1.0

1.0

Mean

3.7

11.0

98%

100%

Maternal blood (n=200)
Percent >LLOQ
LLOQ

1.0

1.0

Mean

4.5

29.9

Japan

Inoue et al. 2004

Cord serum (n=15)
Percent >LOD

0%

100%

LOD

0.5

0.5

Range

1.6–5.3

Maternal serum (n=15)
Percent >LOD

20%

100%

LOD

0.5

0.5

Range

0.5–2.3

4.9–17.6

Breast milk
Massachusetts (n=45)

Tao et al. 2008

Milk
Percent >LOQc
Median

89%

96%

0.0361

0.106

Minimum

<0.0301

<0.032

Maximum

0.161

0.617

Sweden (n=12)

Kärrman et al. 2007

Milk
Percent >LOD

8%d

LOD

0.01

0.005

–

0.201

Mean
Range

100%

<0.209–0.492

***DRAFT FOR PUBLIC COMMENT***

0.060–0.470

 PERFLUOROALKYLS

615
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-24. Percent Detection and Levels of PFOA and PFOS in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL
[ppb])a
PFOA
PFOS

Location

Reference

Maternal serum
Percent >LOD

100%

LOD

0.01

Mean

3.8

Range

2.4–5.3

100%
0.005
20.7
8.2–48.0

China (n=19)
Percent >LOD

So et al. 2006b
100%

100%

LOD

0.021–0.027

0.001–0.0036

Range

0.047–0.210

0.045–0.360

Germany/Hungary (n=70)

Völkel et al. 2008

Percent >LOQ

16%

100%

Minimum

<0.200

0.028

Maximum

0.460

0.639

a"Less

than" values indicate that the concentration was reported as below the LOD or LLOQ. For cases where
samples had concentrations below the LOD or LLOQ, a value between zero and the LOD or LLOQ was assigned
when calculating the mean concentration.
bPercent detection for the adolescent age group was not specified for the 2003–2004 NHANES samples.
Percentages listed here are for the total sample population.
cLOD or LLOQ not reported.
dReported as bias-corrected estimates.
“—“ indicates no available data; LLOQ = lower limit of quantification; LOD = limit of detection; LOQ = limit of
quantification; NR = not reported; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

616
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-25. Percent Detection and Levels of Other Perfluoroalkyls in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL [ppb])a

Sample
population

PFHpA PFNA PFDeA

PFUA

PFDoA

MeEtPFOSA- PFOSAPFBuS PFHxS PFOSA AcOH AcOH

Serum
U.S. NHANES (ages 12–19)
1999–2000 (n=543) (Calafat et al. 2007a)
96%

15%

12%b

<1%b

—

100%

100%

100%

98%

Geometric —
mean

0.5

<0.2

—

—

—

2.7

0.4

1.3

0.8

95th
percentile

1.1

0.5

—

—

—

12.9

1.5

3.7

2.4

98.8%b 31.3%b

9.7%b

<0.1%b

<0.4%b

98.3%a 22.2%b

27.5%b

3.4%b

Geometric <0.3
mean

0.9

<0.3

<0.3

<1.0

<0.4

2.4

<0.2

<0.6

<0.4

95th
percentile

2.7

0.7

<0.3

<1.0

<0.4

13.1

0.3

1.4

<0.4

Percent
>LOD

10%b

—

2003–2004 (n=640) (Calafat et al. 2007b)
Percent
>LOD

6.2%b

0.5

2005–2006 (n=640) (CDC 2018)
Percent
>LOD

NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

Geometric —
mean

0.929

0.295

—

—

—

2.09

—

0.432

—

95th
percentile

2.70

0.800

0.500

<LOD

0.100

14.1

0.300

1.48

0.200

1.10

2007–2008 (n=357) (CDC 2018)
Percent
>LOD

NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

Geometric —
mean

1.16

0.231

—

—

—

2.40

—

0.340

—

95th
percentile

2.54

0.800

0.300

<LOD

<LOD

15.9

0.300

1.60

<LOD

0.600

2009–2010 (n=364) (CDC 2018)
Percent
>LOD

NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

Geometric —
mean

1.10

0.220

—

—

—

2.03

—

0.235

—

95th
percentile

2.62

0.600

0.400

<LOD

<LOD

12.3

<LOD

0.900

<LOD

0.400

2011–2012 (n=344) (CDC 2018)
Percent
>LOD

NR

Geometric —
mean

NR

NR

NR

NR

NR

NR

NR

NR

NR

0.741

0.146

—

—

—

1.28

—

0.134

—

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

617
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-25. Percent Detection and Levels of Other Perfluoroalkyls in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL [ppb])a

Sample
population
95th
percentile

PFHpA PFNA PFDeA

PFUA

PFDoA

MeEtPFOSA- PFOSAPFBuS PFHxS PFOSA AcOH AcOH

0.190

0.250

<LOD

<LOD

6.45

<LOD

0.510

<LOD

2.06

0.360

2013–2014 (n=402) (CDC 2018)
Percent
>LOD

NR

NR

NR

NR

NR

NR

NR

NR

NR

NR

Geometric —
mean

0.599

0.136

—

—

—

1.27

—

—

NR

95th
percentile

2.00

0.400

0.200

0.200

<LOD

6.30

<LOD

0.600

NR

0.200

2013-2014
Ages 3–5 (n=181) (Ye et al. 2018a, 2018b; CDC 2018)
Percent
>LOD

27%

100%

50%

27%

0%

1%

100%

7%

51%

4%

LOD

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

Geometric <LOD
mean

0.764

<LOD

<LOD

<LOD

<LOD

0.715

<LOD

<LOD

<LOD

95th
percentile

3.49

0.370

0.370

<LOD

<LOD

1.62

0.110

1.02

<LOD

0.310

Ages 6–11 (n=458) (Ye et al. 2018a, 2018b; CDC 2018)
Percent
>LOD

15%

100%

46%

28%

0%

7%

100%

2%

54%

3%

LOD

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.1

Geometric <LOD
mean

0.809

<LOD

<LOD

<LOD

<LOD

0.913

<LOD

<LOD

<LOD

95th
percentile

3.19

0.350

0.250

<LOD

0.130

4.14

<LOD

0.940

<LOD

0.170

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

618
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-25. Percent Detection and Levels of Other Perfluoroalkyls in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL [ppb])a

Sample
population

PFHpA PFNA PFDeA

PFUA

PFDoA

MeEtPFOSA- PFOSAPFBuS PFHxS PFOSA AcOH AcOH

—

—

—

U.S. children (ages 2–12)
1994–1995 (n=598) (Olsen et al. 2004b)
Percent
>LLOQ

—

—

—

LLOQ

78%

14%

67%

81%

NR

1.0

NR

NR

Geometric —
mean

—

—

—

—

—

4.5

<2.0

1.9

3.3

95th
—
percentileb

—

—

—

—

—

65

<2.0

12

10

U.S. girls (ages 6–8)
Cincinnati, Ohio (Pinney et al. 2014)
Percent
>LODc

—

99.9

75.8

—

—

—

99.7

19

94.9

14.4%

Geometric —
mean

1.4

0.3

—

—

—

5.1

<LOD

0.7

<LOD

Median

1.4

0.3

—

—

—

5.2

<LOD

0.7

<LOD

—

U.S. girls (ages 6–8)
San Francisco, California (Pinney et al. 2014)
Percent
>LODc

—

100

78.7

—

—

—

100

10

96.3

14.8

Geometric —
mean

1.7

0.3

—

—

—

3.0

<LOD

0.7

<LOD

Median

1.6

0.3

—

—

—

2.3

<LOD

0.6

<LOD

—

Umbilical cord
blood
San Francisco, California (Morello-Frosch et al. 2016) (n=65)
Percent
>LODc

56%

97%

9%

84%

0%

25%

—

91%

95%

63%

Geometric
mean

—

0.29

—

0.03

—

—

—

0.02

0.07

0.02

95th
percentile

0.23

0.93

0.49

0.16

—

0.03

—

0.10

0.38

0.05

—

26%

40%

1%

Baltimore THREE Study (Apelberg et al. 2007a, 2007b)
Cord serum (n=299)
Percent
>LOD

2%

—

LOD

0.4

Minimum

<0.4

—

Maximum

2.6

—

24%

34%

5%

3%

0.2

0.2

0.2

0.1

0.05

0.2

0.2

<0.2

<0.2

<0.2

<0.1

—

<0.05

<0.2

<0.2

1.1

1.9

1.7

0.2

—

0.8

1.8

0.5

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

619
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-25. Percent Detection and Levels of Other Perfluoroalkyls in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL [ppb])a

Sample
population

PFHpA PFNA PFDeA

PFUA

PFDoA

MeEtPFOSA- PFOSAPFBuS PFHxS PFOSA AcOH AcOH

—

—

—

Japan (Inoue et al. 2004)
Cord serum (n=15)
Percent
>LOD

—

—

—

—

LOD

0%

—

—

—

—

1.0

Maternal serum
(n=15)
Percent
>LOD

—

—

—

—

—

—

—

LOD

0%
1.0

Breast milk
Massachusetts (n=45) (Tao et al. 2008b)
Milk
Percent
>LOQc

<1%

64%

<1%

<1%

<1%

<1%

51%

—

—

—

Minimum

<0.010 <0.0052 <0.00772 <0.00499 <0.00440 <0.0100 <0.0120 —

Maximum

0.0234 0.0184 0.0111

—

—

0.00884

0.00974

0.0198

63.8

—

—

—

—

—

100%

67%

—

—

Sweden (n=12) (Kärrman et al. 2007a)
Milk
Percent
>LOD

—

LOD

17%

0%

0%

0.005

0.008

0.005

0.01

0.007

Mean

—

0.017

—

—

0.085

0.013

—

—

Range

—

<0.005–
0.020

—

—

0.031– <0.007– —
0.172 0.030

—

—

100%

100%

100%

—

—

100%

75%

—

—

0.005

0.008

0.005

0.01

0.007

—

—

0.80

0.53

0.40

4.7

0.24

0.43–
2.5

0.27–1.8

0.20–1.5

1.8–
11.8

0.16–
0.19

—

—

Maternal
serum
Percent
>LOD
LOD
Mean
Range

—

—

—

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

620
5. POTENTIAL FOR HUMAN EXPOSURE

Table 5-25. Percent Detection and Levels of Other Perfluoroalkyls in Children’s
Serum, Umbilical Cord Blood, and Breast Milk
Detection and concentration (ng/mL [ppb])a

Sample
population

PFHpA PFNA PFDeA

PFUA

PFDoA

MeEtPFOSA- PFOSAPFBuS PFHxS PFOSA AcOH AcOH

100%

—

11%

100%

0.001–
0.005

0.001–
0.010

China (n=19) (So et al. 2006b)
Percent
>LOD

37%

100%

100%

LOD

0.005– 0.001– 0.0011–
0.010 0.010 0.0025

0.0022–
0.0050

Range

<0.005– 0.01–
0.0067 0.062

0.0091–
0.056

0.0038–
0.011

—

—

—

—

<0.001– 0.004– —
0.0025 0.10

—

—

a"Less

than" values indicate that the concentration was reported as below the LOD or LLOQ. For cases where
samples had concentrations below the LOD or LLOQ, a value between zero and the LOD or LLOQ was assigned
when calculating the mean concentration.
bPercent detection for the adolescent age group was not specified for these samples. Percentages listed here are
for the total sample population.
cLOD or LLOQ not reported.
“—“ indicates no available data; Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid;
LLOQ = lower limit of quantification; LOD = limit of detection; Me-PFOSA-AcOH = 2-(N-methyl-perfluorooctane
sulfonamide) acetic acid; ND = no data; NR = not reported; PFBuS = perfluorobutane sulfonic acid;
PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOSA = perfluorooctane sulfonamide;
PFUA = perfluoroundecanoic acid

A few studies are available that reported serum levels of perfluoroalkyls measured in children. Calafat et
al. (2007a, 2007b) reported perfluoroalkyl serum concentrations measured in 543–640 adolescents who
made up the 12–19-year-old age subpopulation in the 1999–2000 and 2003–2004 NHANES surveys.
Olsen et al. (2003a) measured PFOA, PFOS, PFHxS, PFOSA, Me-PFOSA-AcOH, and Et-PFOSA-AcOH
in the serum of 598 children of ages 2–12 years from various locations in the United States who were
diagnosed with group A streptococcal infections. Mean serum concentrations of perfluoroalkyl
compounds measured in children from these studies are similar to mean concentrations reported for adults
(Calafat et al. 2007a, 2007b; Olsen et al. 2003a). For example, geometric mean concentrations of PFOA
and PFOS measured during the NHANES surveys were 3.9–5.5 and 19.3–29.1 ng/mL, respectively, in
adolescent serum and 3.9–5.2 and 20.7–30.4 ng/mL, respectively, in serum of the total population.
Emmett et al. (2006a) found that 2–5-year-old children had a higher serum PFOA (median 600 ng/mL) in
the Little Hocking Water Association district compared with residents in all other age groups (median
321 ng/mL) except for the group aged >60 years, whose levels were similar to those in young children.
Several factors may have contributed to the observed high levels in children: infants and young children

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

621
5. POTENTIAL FOR HUMAN EXPOSURE

proportionally drink more water per kg of body weight than adults; children (and also the elderly) tend to
spend more time at home with exclusive use of residential water than other age groups; and transplacental and breast milk exposures could also contribute to levels in children.

Kato et al. (2009b) reported serum levels in children aged 3–5 and 6–11 years from the 2001–2002
NHANES survey for PFNA, PFOA, PFOS, and PFOSA. The highest levels were typically observed for
PFOS. The least square mean (LSM is equivalent to arithmetic mean) serum concentrations for PFOS
ranged from 30.45 ng/mL for Mexican Americans to 42.45 ng/mL for non-Hispanic whites aged 6–
11 years (Kato et al. 2009b). The LSM for PFOA ranged from 6.1 ng/mL for Mexican Americans to
7.6 ng/mL for non-Hispanic whites aged 6–11 years. Among 3–5 year olds, specific data from pooled
samples were only presented for PFNA. The LSM serum PFNA serum concentrations for this age group
were 0.9, 1.2, and 0.6 ng/mL for non-Hispanic whites, non-Hispanic blacks, and Mexican Americans,
respectively (Kato et al. 2009b).

Blood serum levels of PFOA, PFOS, and PFHxS obtained in 2006–2007 from children residing in
Australia were reported by Toms et al. (2009). The highest levels tended to occur for PFOS. Mean PFOS
serum levels (combined male and female) ranged from 7.0 ng/mL for infants 0–0.5 years of age to
18.3 ng/mL for 6–9 year olds, while mean PFOA serum levels ranged from 4.5 ng/mL for infants 0–
0.5 years of age to 8.2 ng/mL for 6–9 year olds. Mean serum PFHxS levels ranged from 0.9 ng/mL for
infants 0–0.5 years of age to 5.8 ng/mL for 6–9 year olds (Toms et al. 2009).

Although mean serum concentrations of perfluoroalkyl compounds are reported to be similar for older
children (12–19 years of age) and adults, estimated 95th percentile values of PFHxS measured in
childhood serum were noted to be higher than values estimated for adults. Olsen et al. (2003a) reported
bias-corrected 95th percentile estimates of 65 ng/mL for PFHxS in the serum of children ages 2–12 years.
This value is higher than bias-corrected 95th percentile estimates of 9.5 and 8.3 ng/mL based on PFHxS
measurements in the serum of 645 U.S. adult blood donors and 238 elderly individuals from the Seattle,
Washington area, respectively (Olsen et al. 2003b, 2004b, 2004c). The difference is less extreme in the
NHANES data, with PFHxS 95th percentile values of 12.9–13.1 ng/mL reported for children compared to
values of 8.3–8.7 ng/mL reported for the total population. Olsen et al. (2004b) also noted statistically
higher levels of Me-PFOSA-AcOH measured in children citing estimated 95th percentile values of 12.0,
5.0, and 3.8 ng/mL for serum concentrations of this substance measured in children, adult donors, and
elderly individuals, respectively (Olsen et al. 2003b, 2004b, 2004c).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

622
5. POTENTIAL FOR HUMAN EXPOSURE

Reasons for the observed differences of PFHxS and Me-PFOSA-AcOH levels in childhood serum
samples compared to adult samples have not been determined. Olsen et al. (2004b) stated that different
exposure and activity patters between children and adults should be considered. For example, children
may have a higher exposure than adults to PFHxS, a substance that has been used in carpet treatment
applications, since they are lower to the ground and have increased contact with carpeted floors (Calafat
et al. 2007a; Olsen et al. 2004b).

When estimating PFOS and PFOA uptake doses for children, Trudel et al. (2008) assumed the same
exposure pathways for children as were proposed for adults, but considered exposure from hand-to-mouth
transfer from treated carpets to be much larger in children. This pathway was estimated to contribute 40–
60% of the total uptake of both PFOS and PFOA in infants (0–1 years), toddlers (1–4 years), and children
(5–11 years) in the high-exposure scenario. Exposure via human breast milk was included in the food
consumption pathway for infants. Exposure via mouthing of clothes, carpet, and upholstery was also
considered for children <12 years old; however, this was considered to be a minor pathway of exposure.
PFOS uptake doses estimated for the low-, medium-, and high-exposure scenarios were 18.1–219 ng/kg
body weight/day for infants, 14.8–201 ng/kg body weight/day for toddlers, and 9.7–101 ng/kg body
weight/day for children. PFOA uptake doses estimated for the low-, medium-, and high-exposure
scenarios were 2.2–121 ng/kg body weight/day for infants, 1.2–128 ng/kg body weight/day for toddlers,
and 0.8–65.2 ng/kg body weight/day for children. In contrast to the estimates for children under age 12,
relative exposure pathways and uptake doses estimated for teenagers (12–20 years old) were
approximately the same as for adults.

Tao et al. (2008b) measured perfluoroalkyl concentrations in 45 human breast milk samples collected
from Massachusetts. PFOS, PFOA, PFHxS, and PFNA were each detected in 96, 89, 51, and 64% of the
samples, respectively, with median concentrations of 106, 36.1, 12.1, and 6.97 pg/mL, respectively.
PFHpA, PFDeA, PFUA, PFDoA, and PFBuS were each detected in <1% of the samples. Perfluoroalkyls
have also been measured in the human breast milk of individuals from Sweden, China, and Germany/
Hungary (Kärrman et al. 2007; So et al. 2006b; Völkel et al. 2008). PFOS was detected in all samples,
while detection of PFOA ranged from 8 to 100% in these studies. The reported maximum concentrations
of PFOS and PFOA measured in human breast milk samples collected during these studies were 0.360–
0.639 and 0.210–0.490 ng/mL, respectively (Kärrman et al. 2007; So et al. 2006b; Völkel et al. 2008).
Other perfluoroalkyls detected in human breast milk included PFHpA, PFNA, PFDeA, PFUA, PFBuS,
PFHxS, and PFOSA. Maximum concentrations of these compounds were reported to be <0.18 ng/mL
(Kärrman et al. 2007).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

623
5. POTENTIAL FOR HUMAN EXPOSURE

The presence of perfluoroalkyl compounds in umbilical cord blood indicates that these substances can
cross the placental barrier resulting in the exposure of babies in utero (Apelberg et al. 2007a, 2007b; Fei
et al. 2007; Inoue et al. 2004; Midasch et al. 2007). In most studies, PFOS and PFOA have been detected
in 99–100% of umbilical cord blood samples with reported concentrations were 4.9–11.0 and 1.6–
3.7 ng/mL, respectively (Apelberg et al. 2007a, 2007b; Fei et al. 2007; Inoue et al. 2004; Midasch et al.
2007; Morello-Frosch et al. 2016). Inoue et al. (2004) did not detect PFOA in 15 cord blood samples
from Japan; however, this compound was only detected in the maternal serum of three mothers. Apelberg
et al. (2007a) also reported concentrations of other perfluoroalkyl compounds measured in 299 cord
serum samples collected during the Baltimore THREE Study. Of these compounds, PFDeA, PFUA,
PFOSA, and Me-PFOSA-AcOH were detected most frequently (24, 34, 26, and 40%, respectively).
Maximum concentrations in these samples ranged from 1.1 to 1.9 ng/mL. PFHpA, PFDoA, PFBuS, and
Et-PFOSA-AcOH were each detected in <6% of the samples, with maximum concentrations ranging from
0.2 to 2.6 ng/mL. Manzano-Salgado et al. (2015) studied the potential transfer of perfluoroalkyls from
mothers to their children during pregnancy. Maternal blood and cord serum were collected from
66 mother-child pairs and analyzed for the presence of perfluoroalkyls. A positive correlation was found
between maternal plasma and maternal serum with cord serum levels, and the authors concluded that
either maternal plasma or maternal serum could be used as a method to estimate fetal exposure to
perfluoroalkyls. Median concentrations of PFOS and PFOA were 6.18 and 2.85 ng/mL, respectively, in
maternal plasma and 6.99 and 2.97 ng/mL, respectively, in maternal serum. PFOS and PFOA levels in
cord serum were 1.86 and 1.90 ng/mL, respectively. A biomonitoring survey of 1,533 pregnant females
in Denmark from 2008 to 2013 showed decreasing levels of most perfluoroalkyls in the females’ blood
during this time period (Bjerregaard-Olesen et al. 2016). The results of this study showed that serum
levels of PFHxS, PFOS, PFOA, PFNA, and PFDeA decreased at a rate of 7.0, 9.3, 9.1, 6.2, and 6.3 per
year, respectively. Morello-Frosch et al. (2016) measured the levels of perfluoroalkyls in 77 maternal and
65 paired umbilical cord blood samples from pregnant females and newborn children in San Francisco,
California. Perfluoroalkyls, including PFOA and PFOS, were widely detected; however, concentrations
in cord blood or serum were typically equal to or lower than maternal blood levels.
5.7

POPULATIONS WITH POTENTIALLY HIGH EXPOSURES

Potentially high exposures to perfluoroalkyls can occur in the following population categories:
perfluoroalkyl production and manufacturing workers, communities located near fluorochemical
facilities, and individuals with prolonged use of perfluoroalkyl-containing products. These populations

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

624
5. POTENTIAL FOR HUMAN EXPOSURE

may have higher exposure to perfluoroalkyl compounds than the general population based on elevated
concentrations of these substances measured in air, soil, sediment, surface water, groundwater, and
vegetation surrounding these facilities (3M 2007b, 2008b, 2008c; Barton et al. 2006; Davis et al. 2007;
Olsen 2015).

A number of studies have evaluated serum perfluoroalkyl levels in former workers at 3M and DuPont
PFOA and/or PFOS facilities in the United States. At the 3M Cottage Grove facility, the geometric mean
serum PFOA level was 5,200 ng/mL in PFOA production workers and the PFOS level was 1,760 ng/mL
in PFOS production workers in 2000 (Olsen 2015). At another 3M facility in Decatur, geometric mean
serum PFOA, PFOS, and PFHxS levels were 910, 1,130, and 180 ng/mL, respectively (Olsen 2015). The
estimated mean serum PFOA level at DuPont’s Washington Works facility was 2,050 ng/mL for all
workers (Woskie et al. 2012)

Family members of occupationally exposed workers have been shown to have higher exposure to
perfluoroalkyls via dust transfer as compared to family members of nonoccupationally exposed workers
(Fu et al. 2015). PFOA, PFOS, PFBA, PFBuS, PFNA, and PFHxS have been detected in the municipal
drinking water and private wells of some communities located near fluorochemical facilities (3M 2008c;
ATSDR 2008; Emmett et al. 2006a; Hoffman et al. 2011; Hölzer et al. 2008; Post et al. 2013; Steenland et
al. 2009a; Wilhelm et al. 2009). Emmett et al. (2006a) compared PFOA serum levels to various types of
exposure for individuals living in the Little Hocking community (near DuPont's Washington Works
facility) and concluded that residential water source was the primary determinant of serum PFOA at this
location. These authors reported that the mean human serum PFOA level was 105 times higher than the
residential drinking water level. In residents with residential drinking water but without occupational
exposure, the model of best-fit for serum PFOA also varied significantly by age (highest in children
≤5 years old and adults >60 years old), use of carbon home water filters (negative effect), number of
servings of home-grown fruits and vegetables (positive effect), and number of tap water-based drinks per
day (positive effect). Median PFOA serum levels for residents currently residing in six water districts
located in the mid-Ohio Valley near the Washington Works facility ranged from 12.1 to 224.1 ng/mL,
while the median concentration ranged from 10.5 to 33.7 ng/mL for residents who previously worked or
resided in these districts (Steenland et al. 2009a). PFOA serum levels tended to be highest for children
aged 0–9 years and persons >50 years old. These authors also reported that former employees at the
chemical plant had much higher levels (median=75 ng/mL) than people who had not worked at the plant
(median=24 ng/mL), but lower levels than those who continued to be employed at the plant during the
monitoring period (median=148 ng/mL). The serum levels of the 69,030 residents participating in this

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

625
5. POTENTIAL FOR HUMAN EXPOSURE

study categorized by age are provided in Table 5-26. Additional blood serum levels of PFOA and PFOS
for residents in selected areas of Ohio, West Virginia, New Jersey, and Minnesota whose residential
source of drinking water may have been contaminated are available from the EPA docket on PFOA and
related perfluoroalkyls (EPA-HQ-OPPT-2003-0012) (Bilott 2004, 2005a, 2005b, 2007).

Table 5-26. Blood Serum Levels for 69,030 Current and Former Residents of Six
Water Districts in the Mid-Ohio Valley (2005–2006)
Age (years)

Number (percentage of total)

Median perfluorooctanoic acid
(PFOA) level (ng/mL)

0–9
10–19
20–29
30–39
40–49
50–59

4,915 (7.1)
9,658 (14.0)
10,073 (14.6)
10,547 (15.3)
12,113 (17.6)
10,515 (15.2)

32.8
26.6
21.0
22.7
28.0
33.6

60–69
≥70

6,881 (10)
4,328 (6.3)

42.9
40.1

Source: Steenland et al. 2009a

ATSDR performed an exposure investigation for residents of Decatur, Alabama following an accidental
release of perfluoroalkyls into the Decatur waste water treatment plant (ATSDR 2013). A group of
155 residents had their blood tested for levels of eight perfluoroalkyls. Serum levels of five of the
substances were lower or similar to levels of the general population when compared to the NHANES
results. Residents who used the West Morgan/East Lawrence public water supply had significantly
higher geometric mean serum levels of two substances (PFOA and PFHxS) as compared to the geometric
mean for a similar demographic group from the NHANES survey. Serum levels in the residents for all
eight compounds were shown to be much lower than levels found in occupationally exposed individuals
who regularly worked with these substances (ATSDR 2013).

Residents of Arnsberg, Germany who were exposed to perfluoroalkyls from contaminated drinking water
had serum PFOA levels that were 4.5–8.3 times greater than those from a non-exposed reference
population depending upon the age and sex of the groups (Hölzer et al. 2008).

Shin et al. (2011) provided PFOA exposure data on 45,276 non-occupationally exposed individuals
residing in eastern Ohio and West Virginia who were exposed to PFOA via drinking water. The median
serum concentration for all subjects was 24.3 ng/mL (ppb) measured in 2005–2006. This was about

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

626
5. POTENTIAL FOR HUMAN EXPOSURE

6 times greater than the median concentration (3.20 ng/mL) of 2,120 residents of the general population
taken from the NHANES data for 2005–2006 (CDC 2018).

Individuals involved in activities with prolonged use of perfluoroalkyl-containing products, such as the
application of protective coatings for fabrics and carpet and the use of paper coatings or ski wax, may
have higher levels of exposure to perfluoroalkyl compounds than the general population (Calafat et al.
2006a; Olsen 2015). Elevated serum levels of PFOA, PFNA, PFHxA, PFHpA, PFDeA, and PFUA have
been found in professional ski waxers using products containing fluorotelomers; the perfluoroalkyls were
likely formed via fluorotelomer metabolism (Olsen 2015). Some firefighting foams contain
perfluoroalkyls, and firefighters who use these products have been shown to have greater exposures as
compared to the general population. Dobraca et al. (2015) compared perfluoroalkyl serum levels of a
group of firefighters in California to an adult population from the NHANES survey. Levels of PFOA and
PFOS were only slightly higher in the firefighter group (geometric means 3.75 and 12.50 ng/mL,
respectively) when compared to adult males in the 2009–2010 NHANES general population survey
(3.61 and 12.13 ng/mL, respectively); however, PFDeA serum concentrations of firefighters were up to
3 times greater than the NHANES comparison group for the 25th–95th percentiles (50th percentile in
firefighters; 0.72 ng/mL compared to 0.30 ng/mL) and the geometric mean (0.90 ng/mL compared to
0.30 ng/mL). In a small-scale study of 37 firefighters participating in the C8 Health Project, significantly
(adjusted for age, water district, household income, and smoking) higher levels of PFOA and PFHxS
were found in the firefighters compared to 5,373 male participants with other jobs (Jin et al. 2011).
Geometric mean PFOA and PFHxS levels were 37.59 and 4.77 ng/mL, respectively, in the firefighters
and 31.59 and 3.62 ng/mL, respectively, in the other participants. No significant differences in PFOS or
PFNA levels were found between the groups. A biomonitoring study of 149 firefighters in Australia
showed that 100% of serum samples collected had detectable levels of PFOA, PFOS, PFHxS, and PFNA
(Rotander et al. 2015). Serum levels of PFHxS were found to be approximately 10–15 times higher than
levels found in the general population of Australia and Canada, while PFOS levels in the firefighters were
approximately 6–10 times greater than the general population of these nations.

3M estimated doses for various off-site exposure scenarios based on monitoring information collected at
the Decatur Facility in Alabama (3M 2008c). Exposure scenarios include local children and adult
residents exposed to PFOA in off-site soils, groundwater, municipal water, fish from the Tennessee River,
and surface water and sediments in the Tennessee River. According to 3M, estimated off-site exposure of
local residents to PFOA ranged from 0.011 to 260 ng/kg/day, with the highest estimated exposure

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

627
5. POTENTIAL FOR HUMAN EXPOSURE

corresponding to children whose source of drinking water was groundwater adjacent to the southern side
of the facility.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

628

CHAPTER 6. ADEQUACY OF THE DATABASE
Section 104(i)(5) of CERCLA, as amended, directs the Administrator of ATSDR (in consultation with the
Administrator of EPA and agencies and programs of the Public Health Service) to assess whether
adequate information on the health effects of perfluoroalkyls is available. Where adequate information is
not available, ATSDR, in conjunction with NTP, is required to assure the initiation of a program of
research designed to determine the adverse health effects (and techniques for developing methods to
determine such health effects) of perfluoroalkyls.

Data needs are defined as substance-specific informational needs that, if met, would reduce the
uncertainties of human health risk assessment. This definition should not be interpreted to mean that all
data needs discussed in this section must be filled. In the future, the identified data needs will be
evaluated and prioritized, and a substance-specific research agenda will be proposed.
6.1

Existing Information on Health Effects

Studies evaluating the health effects of inhalation, oral, and dermal exposure of humans and animals to
PFOA, PFOS, and other perfluoroalkyls that are discussed in Chapter 2 are summarized in Figures 6-1,
6-2, and 6-3, respectively. The purpose of these figures is to illustrate the information concerning the
health effects of perfluoroalkyls. The number of human and animal studies examining each endpoint is
indicated regardless of whether an effect was found and the quality of the study or studies.

As illustrated in Figures 6-1, 6-2, and 6-3, most of the data on the toxicity of PFOA, PFOS, and other
perfluoroalkyls come from epidemiology studies in humans; oral exposure is the assumed route of
exposure for the epidemiology studies. The epidemiology database consists of health evaluations of
subjects exposed in occupational settings (primarily PFOA and PFOS), highly exposed residents living
near a PFOA facility, and studies of the general population. The most commonly examined endpoints
examined in the epidemiology studies were developmental, hepatic, reproductive, and immunological
effects.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

629
6. ADEQUACY OF THE DATABASE

Figure 6-1. Summary of Existing Health Effects Studies on PFOA By Route and
Endpoint*
Potential body weight, hepatic, and developmental effects were the most studied
endpoints
The majority of the studies examined oral exposure in humans (versus animals)

*In this figure, the number of human studies is referring to the number of publications; a “—“ indicates that no studies
are available.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

630
6. ADEQUACY OF THE DATABASE

Figure 6-2. Summary of Existing Health Effects Studies on PFOS By Route and
Endpoint*
Potential developmental, hepatic, reproductive, and body weight effects were the most
studied endpoints
The studies examined oral exposure in humans (versus animals)

*Oral exposure was the presumed route of exposure for epidemiology studies; in this figure, the number of human
studies is referring to the number of publications.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

631
6. ADEQUACY OF THE DATABASE

Figure 6-3. Summary of Existing Health Effects Studies on Other Perfluoroalkyls
By Route and Endpoint*
Potential body weight, hepatic, and developmental effects were the most studied
endpoints
The majority of the studies examined oral exposure in humans (versus animals)

*In this figure, the number of human studies is referring to the number of publications; a “—“ indicates that no studies
are available. Includes data for PFBA, PFHxA, PFHpA, PFNA, PFDeA, PFUA, PFBuS, PFHxS, PFDoA, PFOSA, MePFOSA-AcOH, and Et-PFOSA-AcOH.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

632
6. ADEQUACY OF THE DATABASE

Studies in laboratory animals have primarily been conducted using oral exposure. Most of the
information regarding the effects of perfluoroalkyl compounds in animals has been derived from oral
studies; considerably less information is available from inhalation and dermal exposure studies. PFOA
and PFOS have been the most extensively studied members of this class of chemicals, and oral
administration has been the preferred route of exposure in animal studies. Information regarding other
perfluoroalkyls covered in this profile is limited to acute-duration oral studies with PFNA, PFDeA,
PFBA, PFDoA, and PFOSA; intermediate-duration oral studies with PFHxS, PFNA, PFUA, PFBuS,
PFBA, PFDoA, and PFHxA; and a chronic-duration oral study with PFHxA. An acute-durationinhalation study with PFNA is also available. The most commonly examined endpoints were hepatic,
body weight, developmental, reproductive, and immunological effects.
6.2

Identification of Data Needs

Missing information in Figures 6-1, 6-2, and 6-3 should not be interpreted as a “data need.” A data need,
as defined in ATSDR’s Decision Guide for Identifying Substance-Specific Data Needs Related to
Toxicological Profiles (ATSDR 1989), is substance-specific information necessary to conduct
comprehensive public health assessments. Generally, ATSDR defines a data gap more broadly as any
substance-specific information missing from the scientific literature.
Acute-Duration MRLs. The available acute inhalation database for PFOA was considered inadequate
for derivation of an MRL due to lack of measured serum PFOA levels in the available animal studies and
the lack of PBPK model parameters that could be used to predict serum levels. The inhalation database
for PFNA was not considered adequate due to the limited endpoints examined and the short exposure
duration of the only available study. No inhalation data were available for PFOS or the other
perfluoroalkyl compounds. A number of studies have evaluated the acute toxicity of PFOA and PFOS
following oral exposure and have identified several sensitive targets of toxicity. Smaller numbers of
studies evaluated potential sensitive targets of acute toxicity for PFNA and PFDeA. However,
toxicokinetic differences between humans and laboratory animals, particularly the relative short half-life
in rodents compared to humans, preclude derivation of an acute MRL for these compounds. For other
perfluoroalkyls (PFHxS, PFBA, PFDoA, PFOSA), the available studies were not considered adequate for
identification of critical targets; did not examine sensitive targets that were identified for other
perfluoroalkyl compounds, such as developmental and immunological endpoints; or involved a single
exposure. No acute oral data were identified for PFUA, PFHpA, or PFBuS. Research is needed to
develop a PBPK model that would allow for extrapolation from rodents to humans. Additionally, toxicity

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

633
6. ADEQUACY OF THE DATABASE

studies are needed for most perfluoroalkyl compounds to identify critical targets of toxicity and/or
establish dose-response relationships. These studies should examine developmental and reproductive
endpoints that have been established as the most sensitive targets of toxicity for PFOA and PFOS.
Intermediate-Duration MRLs. No intermediate-duration inhalation studies were identified for
perfluoroalkyl compounds. Oral studies suggest that developmental and immune effects are the most
sensitive targets of toxicity; similar effects are likely to occur following inhalation exposure because
perfluoroalkyl compounds are not metabolized. Inhalation studies are needed to establish dose-response
relationships and to establish whether the respiratory tract is a sensitive target of toxicity. The
intermediate-duration oral databases were considered adequate for derivation of provisional MRLs for
PFOA and PFOS. A modifying factor was used for PFOS due to the lack of PBPK modeling parameters
that would allow for predicting steady-state serum PFOS levels for immunotoxicity studies in laboratory
animals. Provisional intermediate-duration MRLs were also derived for PFHxS and PFNA; however,
these were based on marginal databases and additional dose-response studies are needed to support the
basis of the MRL. The databases were not considered adequate for PFUA, PFBuS, PFBA, or PFDoA due
to the lack of studies examining potential sensitive targets (developmental and/or immune effects). No
intermediate-duration oral studies are available for PFDeA, PFHpA, or PFOSA. Intermediate-duration
oral studies are needed for these seven perfluoroalkyls to provide information on sensitive targets and
establish dose-response relationships. These studies should include measurement of serum perfluoroalkyl
levels, which would allow for estimating human equivalent doses (HEDs).
Chronic-Duration MRLs. The lack of chronic-duration inhalation studies for perfluoroalkyl
compounds precluded derivation of chronic MRLs. Chronic toxicity studies examining a wide range of
endpoints are needed to identify the most sensitive target and establish concentration-response
relationships. A small number of chronic duration oral studies have been identified in laboratory animals.
Four studies examined the chronic toxicity of PFOA, PFOS, or PFHxA. These studies were not
considered suitable for derivation of MRLs because they did not evaluate immunotoxicity which was a
sensitive target following shorter term exposures. Studies examining this potentially sensitive endpoint
are needed to identify the most sensitive target following chronic exposure.
Health Effects. Over 600 studies have evaluated the toxicity of perfluoroalkyl compounds;
epidemiology studies account for over 400 of the toxicity studies. Evidence from epidemiology studies
suggest links between perfluoroalkyl exposure and several health outcomes including liver damage,
increases in serum lipids, thyroid disease, immune effects, reproductive toxicity, and developmental

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

634
6. ADEQUACY OF THE DATABASE

toxicity. The primary health effects observed in laboratory animals are liver, developmental, and immune
toxicity. Although a large number of studies evaluating health effects are available, there is a need for
additional studies to address data gaps. Future laboratory animal studies should include measurement of
serum perfluoroalkyl levels, as this would provide valuable information for comparing effects observed in
laboratory animals to effects observed in humans.

Hepatic Effects. Evidence from acute, intermediate, and/or chronic oral studies in rats, mice, and
monkeys indicates that the liver is a sensitive target of PFOA, PFOS, PFHxS, PFNA, PFDeA, PFUA,
PFBA, PFBuS, PFDoA, and PFHpA toxicity. The effects observed in rodents differ from those
observed in humans. In humans, exposure to PFOA, PFOS, PFNA, and PFDeA appear to result in
increases in serum lipid levels, particularly total cholesterol levels. However, animal studies have
found decreases in serum lipid levels associated with exposure to most perfluoroalkyls. It is not
known if the species differences are due to different mechanisms of toxicity or differences in
exposure levels (serum levels observed in animal studies are orders of magnitude higher than those in
human studies).
Immune Effects: Epidemiology data suggest a link between PFOA, PFOS, PFHxS, and PFDeA
and decreased antibody response to vaccines. This is supported by acute- and intermediate-duration
studies of PFOA and PFOS in laboratory animals. There is also evidence of immunotoxicity
following a single injection of PFNA; some of the immune effects persisted 4 weeks post-exposure.
Shorter-term studies are needed for other perfluoroalkyl compounds. In addition, chronic-duration
studies evaluating immune endpoints, particularly immunosuppression, for all perfluoroalkyls would
allow for identification of the critical targets of toxicity.
Reproductive Effects. Decreases in mammary gland development have been demonstrated in
several PFOA mouse studies. The effect levels observed in these studies are very low, although there
is some indication that at lower doses, the changes in mammary gland development do not affect
lactation. Additional studies are needed to evaluate the adversity of these alterations. This endpoint
has not been evaluated for other perfluoroalkyl compounds and studies are needed to determine
whether it is also a sensitive effect for these compounds.

Developmental Effects. Based on the results of laboratory animal studies, developmental
endpoints are targets of PFOA, PFOS, PFHxS, PFNA, PFDeA, PFUA, and PFBA toxicity following
acute- and/or intermediate-duration oral exposure. Studies are needed to evaluate potential

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

635
6. ADEQUACY OF THE DATABASE

developmental effects for PFHxS following intermediate-duration oral exposure. Epidemiology
studies in children suggest altered responses to vaccination; two animal studies have evaluated
immune effects following perinatal exposure to PFOA and PFOS, but data are lacking for other
perfluoroalkyls.

Potential Interactions Between Perfluoroalkyls. A common limitation of the epidemiology
data is co-exposure to multiple perfluoroalkyl compounds. There are limited data on possible
interactions between perfluoroalkyl compounds and possible effects on toxicity and toxicokinetics.
Animal studies examining the possible interactions between perfluoroalkyl compounds would be
useful for interpreting the epidemiology study results; this is especially important since humans are
typically exposed to multiple perfluoroalkyls and many of them are likely to have similar mechanisms
of action.
Mechanisms of Toxicity. Many of the effects observed in rodents, particularly liver and
developmental effects, involve the activation of PPARα; humans and nonhuman primates are less
responsive to PPARα agonists than rats and mice. However, the results of studies in PPARα-null
mice suggest that PPARα-independent mechanisms also play a role in the liver, immunological, and
developmental toxicity. Additional studies are needed on the mechanisms of toxicity to assess
whether the effects observed in laboratory animal are relevant to humans.
Epidemiology and Human Dosimetry Studies. As previously mentioned, information is available
regarding the effects of exposure to perfluoroalkyl compounds in humans derived from health evaluations
of subjects exposed in occupational settings, residents living near a PFOA manufacturing facility with
high levels of PFOA in the drinking water, and the general population. Although many studies found
statistically significant associations between serum perfluoroalkyl levels and the occurrence of an adverse
health effect, the findings were not consistent across studies. Interpretation of the human data is limited
by the reliance of cross-sectional studies, which do not establish causality, and the lack of exposure data.
Studies on serum lipids suggest that the dose-response curve is steeper at lower concentrations and
flattens out at higher serum perfluoroalkyl concentrations (Steenland et al. 2010a); additional studies that
could be used to establish dose-response relationships would be valuable. Mechanistic studies examining
the association between perfluoroalkyl exposure and serum lipid levels would also provide valuable
insight. Clarification of the significance and dose-response relationships for other observed effects is also
needed. Longitudinal studies examining a wide range of endpoints would be useful for identifying critical
targets of toxicity in humans exposed to perfluoroalkyls. The available human studies have identified

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

636
6. ADEQUACY OF THE DATABASE

some potential targets of toxicity; however, cause-and-effect relationships have not been established for
any of the effects, and the effects have not been consistently found in all studies. Mechanistic studies
would be useful for establishing causality. When possible, health assessments should include subjects of
different race/ethnicity and age to determine potential race/ethnicity- and age-based susceptibilities.
Another limitation of the epidemiology studies is co-exposure to other perfluoroalkyl compounds; studies
that statistically controlled for co-exposure to other pollutants would decrease this uncertainty. As noted
previously, there is a need for studies evaluating potential interactions between perfluoroalkyl
compounds.
Biomarkers of Exposure and Effect. Data are available regarding levels of perfluoroalkyl
compounds in serum from the general population, highly exposed residents, and perfluoroalkyl workers.
Information is needed regarding the toxicokinetics (see also below) of perfluoroalkyl compounds in
humans to be able to relate levels of these compounds in serum to exposure to specific perfluoroalkyls;
data on matched serum and urine samples would be valuable. Also needed is further information on the
relationship between serum and liver concentrations of perfluoroalkyl compounds in humans.
Absorption, Distribution, Metabolism, and Excretion. Several epidemiology studies have
examined the kinetics of serum perfluoroalkyl concentrations following a change in environmental or
occupational exposure, from which estimates of terminal elimination half-lives in adults are available for
PFOA, PFOS, PFHxS, PFBA, and PFBuS. Other studies provide data on the renal clearances of PFOA
and PFOS, binding of PFOA, PFOS, and PFHxS to human plasma protein, tissue levels (primarily blood,
maternal and fetal cord serum, and breast milk. Data on other aspects of the toxicokinetics of
perfluoroalkyls in humans are not available and could serve to improve predictions of internal dosimetry
associated with exposures to perfluoroalkyls (bioavailability, kinetics of tissue distribution and
elimination, binding in tissues, external-internal dose relationships, all aspects of toxicokinetics in
children and aging populations).

Toxicokinetics of perfluoroalkyls have been studied much more extensively in rodents (rats and mice) and
less extensively in Cynomolgus monkeys; however, a number of data gaps have been identified:
•

Absorption studies; oral absorption data are available for PFOA, PFOS, and PFBA, but are more
limited for other perfluoroalkyl compounds and for other exposure routes. Studies elucidating
the mechanisms of pulmonary and gastrointestinal absorption are also needed.

•

Studies have shown that elimination kinetics, and therefore, internal dose-external dose
relationships, are dependent on structure, including the terminal acid group (carboxylate or

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

637
6. ADEQUACY OF THE DATABASE

sulfonate), carbon chain length, and carbon chain branching. These structural features affect
plasma and tissue protein binding, renal and biliary clearances, tissue levels, maternal-fetal
transfer, and lactational transfer of perfluoroalkyls. Studies examining differences between
perfluoroalkyl compounds would be useful for extrapolating health effects and toxicokinetic data
across compounds.
•

Toxicokinetic studies have found sex- and dose-dependent subcellular distribution of PFOA in
rats. Further studies on the mechanisms for dose-dependency, characterization of subcellular
binding proteins, and mechanistic linkages between subcellular distribution and toxicity of
perfluoroalkyls are needed.

•

The distribution and elimination of PFOA and PFOS are greatly influenced by binding
interactions with albumin and other high molecular weight plasma proteins; available data
suggest that binding to plasma proteins, as well as the volume of distribution, may be sex- and
species-specific. Interactions with albumin have been partially characterized to the extent that
binding capacity and affinity constants have been estimated, but the rates of association and
dissociation have not been reported.

•

Liver uptake and renal clearance of PFOS also appeared to be time-dependent in a PBPK model
used to predict plasma and liver in concentrations of PFOS in a chronic rat study (Harris and
Barton et al. 2008). Mechanisms underlying these time dependencies have not been elucidated.

Comparative Toxicokinetics. Toxicokinetic studies conducted in various rodent species (mice, rats,
hamsters, rabbits) and in Cynomolgus monkeys have revealed profound species and sex differences as
well as dose dependencies in the tissue distribution and elimination kinetics of PFOA and PFOS. Studies
conducted in rats have revealed contributing mechanisms for sex differences in elimination of PFOA;
slower elimination of PFOA in male rats compared to female rats has been attributed to sex hormonemodulated renal tubular transport of PFOA that results in markedly lower renal clearance of PFOA in the
sexually mature male rat (see Section 3.5.1, Excretion). Sex differences in elimination of PFOA have
also been observed in hamsters; unlike the rat, male hamsters excreted absorbed PFOA more rapidly than
female hamsters. Sex differences in elimination of PFOA have not been observed in other rodent species,
in Cynomolgus monkeys, or in limited observations made in humans. Sex differences in elimination rates
of perfluoroalkyls in humans have not been demonstrated in population studies of serum elimination
kinetics or renal clearance. Although the few studies that estimated elimination half-lives or renal
clearances in male and female humans have not found significant sex differences, these outcomes may
reflect the relatively low serum concentrations in these subjects compared with studies that were
conducted in nonhuman primates and rodents (i.e., sex differences in elimination may vary with dose
and/or plasma concentration). Additionally, the failure to account for the influence of reduced estrogen
levels (in postmenopausal women) and reduced testosterone levels (in older males) in occupational and/or
site-related epidemiology studies may also account for the lack of finding of sex-related differences.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

638
6. ADEQUACY OF THE DATABASE

Children’s Susceptibility. Data needs relating to both prenatal and childhood exposures, and
developmental effects expressed either prenatally or during childhood, are discussed in the Health Effects
subsection above. It is not known whether children are more or less susceptible than adults to the effects
of exposure to perfluoroalkyl compounds because there are no studies that specifically addressed this
question. Several studies have examined the possible associations between perfluoroalkyl exposure and
health outcome in children living in an area with high PFOA contamination and in the general population.
Although some studies have found statistically significant associations, they are not adequate for
establishing causality. Follow-up studies of the C8 population could allow for a longitudinal assessment
of health effects in children and would be useful in determining whether the observed effects are due to
perfluoroalkyl exposure. Toxicokinetics information in children is needed. Half-life studies have been
conducted in adults; there is the need to understand if these are applicable to children. There are no
studies that have examined whether young animals are more or less susceptible than adults to
perfluoroalkyls toxicity. Additional information on this issue would be useful.
Physical and Chemical Properties.

Perfluoroalkyl compounds have unique and complex physical

and chemical properties (Kissa 2001; Schultz et al. 2003). Sources are available that provide helpful
insights into the structural aspects and surfactant nature of these substances; however, many of the
properties are still not well understood (CEMN 2008; Kissa 2001; Schultz et al. 2003). In general,
specific properties such as physical state, melting point, boiling point, density, solubility, vapor pressure,
micelle formation, and acid dissociation in water have not been determined or are not well described for
these compounds. Measurements of these endpoints are needed. Information regarding the potential
association of these species in water would be useful. Where determination of a particular endpoint is not
possible, a thorough description of the physical and chemical properties as they relate to that endpoint
would be helpful. Perfluoroalkyls discussed in this profile exist as a mixture of linear and branched
isomers. Isomer-specific data would also be useful for the various physical-chemical properties. Wang et
al. (2013b) identified several of the fluorinated compounds that are currently being used by major
manufacturers as alternatives to PFOA and PFOS. These compounds are being used as processing aids in
the emulsion polymerization of PTFE and other polymers as well as surface treatment uses, metal plating
uses, firefighting foams, and other miscellaneous uses such as food contact materials. A data need exists
to determine the physical and chemical properties of these replacement substances.

The production, use, import, and export of perfluoroalkyl compounds have changed dramatically since
2000. Most nations no longer produce or use PFOS or PFOA (China is a notable exception). Major
fluoropolymer manufacturers in the United States have altered their chemical processes to use alternative

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

639
6. ADEQUACY OF THE DATABASE

fluorinated substances in their production processes. Information regarding the production, import, and
export volumes of these substances is needed.

Recommended methods for the disposal of perfluoroalkyl compounds have not been located. In the past,
perfluoroalkyl-containing waste has been disposed of in on- and off-site landfills, through sludge
incorporation, and through incineration (3M 2007b, 2008b; ATSDR 2005). New disposal methods that
avoid release of these substances into the open environment and prevent contamination of nearby soil,
sediment, and groundwater should be developed. The eventual breakdown of fluorotelomer-based
polymers with the eventual release of substances such as PFOA is not well understood. Early researchers
have concluded that the half-life for this process is >1,000 years; however, more recent data suggest much
shorter time scale of 1–2 decades (Rankin et al. 2014; Washington and Jenkins 2015; Washington et al.
2009, 2014, 2015). Additional studies on the potential release of perfluoroalkyls from the eventual
degradation of fluoropolymers in landfills would be useful.

Environmental Fate.

Perfluoroalkyls are very stable compounds and are resistant to biodegradation,

direct photolysis, atmospheric photooxidation, and hydrolysis (3M 2000; EPA 2008a; OECD 2002, 2007;
Schultz et al. 2003). The chemical stability of perfluoroalkyls and the low volatility of these substances
in ionic form indicate that perfluoroalkyls will be persistent in water and soil (3M 2000; Prevedouros et
al. 2006). Koc values ranging from 17 to 230 indicate that PFOA will be mobile in soil and can leach into
groundwater (Davis et al. 2007; Prevedouros et al. 2006). Environmental fate and potential pathways of
PFOA exposure at and near the DuPont Washington Works site have been discussed (Small 2009). Wang
et al. (2013b) identified several of the fluorinated compounds that are currently being used by major
manufacturers as alternatives to PFOA and PFOS. A data need exists to determine the fate properties of
these replacement substances.
Bioavailability from Environmental Media.

Perfluoroalkyls are widely detected in humans and

animals, indicating that several of these substances are bioavailable. The bioaccumulation potential of
perfluoroalkyls is reported to increase with increasing chain length (de Vos et al. 2008; Furdui et al. 2007;
Martin et al. 2004b). In living organisms, perfluoroalkyls bind to protein albumin in blood, liver, and
eggs and do not accumulate in fat tissue (de Vos et al. 2008; Kissa 2001). The mechanism of
perfluoroalkyl uptake in animals is not fully understood; additional studies would be helpful (de Vos et al.
2008). Perfluoroalkyls discussed in this profile exist as a mixture of linear and branched isomers. Data
regarding the bioavailability of branched versus linear substances would be useful. A data need exists to

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

640
6. ADEQUACY OF THE DATABASE

determine the bioavailability of the replacement substances identified in Wang et al. (2013b) used in place
of PFOA and PFOS.
Food Chain Bioaccumulation.

High levels of certain perfluoroalkyls in animals have been

measured in apex predators, such as polar bears, which indicates that some perfluoroalkyls possess the
ability to bioaccumulate (de Vos et al. 2008; Houde et al. 2006a; Kannan et al. 2005; Smithwick et al.
2005a, 2005b, 2006). Perfluoroalkyl sulfonates with carbon chain length lower than 8 tend to
bioaccumulate less than PFOS. Ongoing monitoring of perfluoroalkyl levels in animals may help to
determine whether efforts to phase out these substances will have had an effect on their biomagnification.
A data need exists to determine the bioaccumulation potential of the new replacement substances used in
place of PFOA and PFOS.
Exposure Levels in Environmental Media.

Reliable monitoring data for the levels of

perfluoroalkyls in contaminated media at hazardous waste sites are needed so that the information
obtained on levels of perfluoroalkyls in the environment can be used in combination with the known body
burden of perfluoroalkyls to assess the potential risk of adverse health effects in populations living in the
vicinity of hazardous waste sites.

Concentrations of perfluoroalkyls have been measured in surface water from several locations across the
United States (Boulanger et al. 2004; Kannan et al. 2005; Kim and Kannan 2007; Nakayama et al. 2007;
Simcik and Dorweiler 2005; Sinclair et al. 2004, 2006). Continued monitoring for perfluoroalkyls in
surface water would be useful. Data are available regarding levels of perfluoroalkyls in outdoor air,
indoor air, indoor dust, food, food packaging, and consumer products (3M 2001; Barber et al. 2007;
Begley et al. 2005; Food Standards Agency 2006; Fromme et al. 2007b; Harada et al. 2005b, 2006;
Jogsten et al. 2009; Kim and Kannan 2007; Kubwabo et al. 2005; Moriwaki et al. 2003; Tittlemier et al.
2007; Washburn et al. 2005). Comprehensive studies monitoring for perfluoroalkyls in these matrices
within the United States are needed. Elevated concentrations of perfluoroalkyls have been measured in
air, water, soil, and sediment near fluorochemical industrial facilities (3M 2007b, 2008b, 2008c; Barton et
al. 2006; Davis et al. 2007; Hansen et al. 2002). Continued monitoring for perfluoroalkyls in these
matrices are needed to assess exposure of individuals working at these locations and individuals who live
near these facilities. A data need also exists to perform environmental monitoring of the replacement
substances identified in Wang et al. (2013b) used in place of PFOA and PFOS, particularly near
manufacturing locations.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

641
6. ADEQUACY OF THE DATABASE

Exposure Levels in Humans.

Trudel et al. (2008) provided a thorough assessment of the exposure

of the general population to PFOS and PFOA. 3M (2008b) provided an assessment of exposure of
individuals to PFOA on-site at a fluoropolymer facility. Uptake values and exposure pathways
determined in these studies should be examined further. Conclusions made in these assessments are
expected to be adjusted as future monitoring data are made available. Large-scale monitoring of
perfluoroalkyls in human serum in the United States is ongoing (Calafat et al. 2006a). Future results of
human monitoring studies would be useful for assessing human exposure to these substances over time.
The results of these studies can be examined for correlations between human perfluoroalkyl levels and the
phasing out of perfluoroalkyl compounds by companies of the fluorochemical industry. Levels of
perfluoroalkyl compounds in human urine have been reported (Jurado-Sanchez et al. 2014). Higher
exposure levels for individuals who reside in areas where substances such as PFOA contaminated both
public and private water supplies have been documented (Emmett et al. 2006a, 2009). Continued
biomonitoring of legacy compounds such as PFOA and PFOS as well as replacement substances is
needed.

This information is necessary for assessing the need to conduct health studies on these populations.
Exposures of Children. Trudel et al. (2008) provided a thorough assessment of the exposure of
children to PFOS and PFOA. These conclusions should be reexamined with respect to future
biomonitoring data when they become available. Data are available regarding the levels of
perfluoroalkyls in young children (Kato et al. 2009b; Olsen et al. 2004b; Toms et al. 2009). NHANES
monitoring data for 2013–2014 for children of ages 3–11 years have recently been released (CDC 2018;
Ye et al. 2018a). Data provided from these efforts will be useful in assessing the exposure of young
children to perfluoroalkyls.

Concentrations of perfluoroalkyls have been measured in human breast milk and cord blood (Apelberg et
al. 2007a, 2007b; Fei et al. 2007; Inoue et al. 2004; Kärrman et al. 2007; Midasch et al. 2007; So et al.
2006b; Völkel et al. 2008). Additional monitoring for perfluoroalkyls in these media would be useful.
Continued biomonitoring of legacy compounds such as PFOA and PFOS as well as replacement
substances is needed.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

642
6. ADEQUACY OF THE DATABASE

6.3

Ongoing Studies

A number of ongoing studies were identified in NIH Reporter (2017), these studies are summarized in
Table 6-1.

Table 6-1. Ongoing Studies on Perfluoroalkyls
Investigator Affiliation

Research description

Sponsor

Bruan J

Brown University

Early life exposure to PFOA, PFOS, PFNA, and
PFHxS on obesity mechanisms and phenotyping

NIEHS

Chen, A

University of
Cincinnati

Longitudinal study of exposure to PBDEs and PFCs
and child neurobehavior

NIEHS

Fenton S

NIEHS

Examination of the mammary gland as a sensitive
endpoint to effects of endocrine disruptors

NIEHS

Frisbee S

West Virginia
University

Study of the effects of perfluoroalkyl chemicals on
stroke incidence, mortality and morbidity in humans

NIGMS

Hijtmancik M

Battelle Centers

Chronic toxicity and carcinogenicity study of PFOA in NIEHS
rats

Hocevar B

Indiana University,
Bloomington

Study of the promotion of pancreatic cancer by PFOA NIEHS

Kissling G

NIEHS

PFOA-induced liver toxicity in wild and PPARα-null
rats and mice

NIEHS

Sagiv S

University of
California, Berkeley

Prenatal exposure to perfluoroalkyls (PFOA, PFOS,
PFHxS, PFNA) effect on growth and development in
children

NIEHS

Shankar A

West Virginia
University

Perfluoroalkyls (PFOA, PFOS, PFHxS, PFNA,
NIEHS
PFHpA, PFDeA, PFUA, PFDoA, PFOSA, Et-PFOSAAcOH, Me-PFOSA-AcOH) and the risk of developing
clinical and subclinical cardiovascular disease in
humans

Stein C

Icahn School of
Medicine at Mount
Sinai

Examination of possible associations between
environmental toxicants (PFOA and other
perfluoroalkyls) and child neurobehavioral
development

NIEHS

Sun Q

Harvard School of
Public Health

Examination of the relationship between
perfluoroalkyls and other pollutants and the risk of
diabetes in the Nurses’ Health Study

NIEHS

Sun Q

Harvard School of
Public Health

Perspective study of possible effects of
perfluoroalkyls and other pollutants on body weight
regulation

NIEHS

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

643
6. ADEQUACY OF THE DATABASE

Table 6-1. Ongoing Studies on Perfluoroalkyls
Investigator Affiliation

Research description

Sponsor

Taylor L

Toxicity study of perfluorinated compounds using
microarray analysis of rat liver samples

NIEHS

Battelle Centers

Source: NIH Reporter 2017
Et-PFOSA-AcOH = 2-(N-ethyl-perfluorooctane sulfonamide) acetic acid; Me-PFOSA-AcOH = 2-(N-methylperfluorooctane sulfonamide) acetic acid; NIEHS = National Institute of Environmental Health Sciences;
NIGMS = National Institute of General Medical Sciences; PBDE = polybrominated diphenyl ether; PFC = perfluoro
compound; PFDeA = perfluorodecanoic acid; PFDoA = perfluorododecanoic acid; PFHpA = perfluoroheptanoic acid;
PFHxS = perfluorohexane sulfonic acid; PFNA = perfluorononanoic acid; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; PFOSA = perfluorooctane sulfonamide; PFUA = perfluoroundecanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

644

CHAPTER 7. REGULATIONS AND GUIDELINES
Pertinent international and national regulations, advisories, and guidelines regarding perfluoroalkyls in
air, water, and other media are summarized in Table 7-1. This table is not an exhaustive list, and current
regulations should be verified by the appropriate regulatory agency. A list of some select state drinking
water regulations/guidelines or health based values are summarized in Table 7-2.

ATSDR develops MRLs, which are substance-specific guidelines intended to serve as screening levels by
ATSDR health assessors and other responders to identify contaminants and potential health effects that
may be of concern at hazardous waste sites. See Section 1.3 and Appendix A for detailed information on
the provisional MRLs for perfluoroalkyls.

Table 7-1. Regulations and Guidelines Applicable to Perfluoroalkyls
Agency

Description

Information

Reference

Air
EPA
WHO
EPA

WHO
FDA
ACGIH
HHS
EPA
IARC

ACGIH
OSHA

RfC
Air quality guidelines

No data
No data
Water & Food
Drinking water standards and health advisories
Lifetime Health Advisory
PFOA
0.07 μg/L
PFOS
0.07 μg/L
National primary drinking water regulations No data
RfD
No data
PFOA
2x10-5 mg/kg/day
PFOS
2x10-5 mg/kg/day
Drinking water quality guidelines
No data
EAFUS
No dataa
Cancer
Carcinogenicity classification
No data
Carcinogenicity classification
No data
Carcinogenicity classification
Carcinogenicity classification
PFOA

No data

Group 2Bb
Occupational
TLV
No data
PEL (8-hour TWA) for general industry,
No data
shipyards and construction

***DRAFT FOR PUBLIC COMMENT***

IRIS 2017
WHO 2010

EPA 2016e
EPA 2016f
EPA 2009d
IRIS 2017
EPA 2016e
EPA 2016f
WHO 2017
FDA 2013
ACGIH 2016
NTP 2016a
IRIS 2017
IARC 2017

ACGIH 2016
OSHA 2013
29 CFR 1910.1000,
Table Z-1

 PERFLUOROALKYLS

645
7. REGULATIONS AND GUIDELINES

Table 7-1. Regulations and Guidelines Applicable to Perfluoroalkyls
Agency

Description

Information

Reference

PEL (8-hour TWA) for shipyards and
construction

No data

PEL (8-hour TWA) for construction

No data

OSHA 2014b
29 CFR 1915.1000,
Table Z
OSHA 2014a
29 CFR 1926.55,
Appendix A
NIOSH 2016

NIOSH

REL (up to 10-hour TWA)

EPA
AIHA

AEGLs-air
ERPGs

DOE

PACs-air
PFOA
PAC-1c
PAC-2c
PAC-3c
PFBA
PAC-1c
PAC-2c
PAC-3c

No data
Emergency Criteria
No data
No data

EPA 2016b
AIHA 2015
DOE 2016a

1.1 mg/m3
12 mg/m3
75 mg/m3
0.5 mg/m3
5.5 mg/m3
33 mg/m3

aThe

EAFUS list of substances contains ingredients added directly to food that FDA has either approved as food
additives or listed or affirmed as GRAS.
bGroup 2B: possibly carcinogenic to humans.
cDefinitions of PAC terminology are available from DOE (2016b).
ACGIH = American Conference of Governmental Industrial Hygienists; AEGL = acute exposure guideline level;
AIHA = American Industrial Hygiene Association; CFR = Code of Federal Regulations; DOE = Department of
Energy; EAFUS = Everything Added to Food in the United States; EPA = Environmental Protection Agency;
ERPG = emergency response planning guidelines; FDA = Food and Drug Administration; GRAS = generally
recognized as safe; HHS = Department of Health and Human Services; IARC = International Agency for Research
on Cancer; IRIS = Integrated Risk Information System; NIOSH = National Institute for Occupational Safety and
Health; NTP = National Toxicology Program; OSHA = Occupational Safety and Health Administration;
PAC = Protective Action Criteria; PEL = permissible exposure limit; PFOA = perfluorooctanoic acid;
PFOS = perfluorooctane sulfonic acid; REL = recommended exposure limit; RfC = inhalation reference
concentration; RfD = oral reference dose; TLV = threshold limit values; TWA = time-weighted average;
WHO = World Health Organization

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

646
7. REGULATIONS AND GUIDELINES

Table 7-2. Select State Drinking Water and Daily Intake Levels for Perfluoroalkyls
Value (ppb or µg/L)
Value type
PFOA PFOS PFBuS PFBA PFNA
Maine
Maximum exposure guideline for
0.07a
0.07a ND
ND
ND
drinking water
Michigan
Human noncancer drinking water
0.42
0.011 ND
ND
ND
value
Minnesota
Health risk limit
Short-term
ND
ND
ND
7
ND
Subchronic
ND
ND
9
7
ND
Chronic
0.3
0.3
7
7
ND
Health-based value
Short-term, subchronic or chronic 0.035 0.027 ND
7
ND
Nevada
Basic comparison level
New Jersey
Health-based chronic maximum
contaminant level

North Carolina
Interim maximum allowable
concentration in groundwater
Vermont
Drinking water health advisory

ND

MECDC
2016

ND

Michigan
DEQ 2016

ND
ND
ND

MDH
2017a,
2017b

0.027b

0.667

0.667

667

ND

0.014

0.013
(draft)

ND

ND

ND

ND

ND

ND

NCDENR
2012, NC
DEQ 2013

ND

ND

ND

ND

Vermont
DOH 2017

1.1–1.6 ND

0.02c

0.02c

ND

PFHxS Reference

ND

0.013 ND

NDEP 2017
DWQI
2015,
2017a,
2017b

aMECDC

notes that according to the U.S. EPA lifetime health advisory for PFOA and PFOS, when both PFOS and
PFOA are present in drinking water combined levels are not to exceed 0.07 ppb (EPA 2016j).
bMDH recommends using the health based value for PFOS (0.027 ppb) as a surrogate for PFHxS until more
toxicological research on PFHxS is available (MDH 2017b).
cSum of PFOS and PFOA not to exceed 0.02 µg/L.
DEC = Department of Environmental Conservation; DEQ = Department of Environmental Quality;
DOH = Department of Health; DTSC = Department of Toxic Substances Control; DWQI = Drinking Water Quality
Institute; MECDC = Maine Center for Disease Control & Prevention; MDH = Minnesota Department of Health;
NCDENR = North Carolina Department of Environment and Natural Resources; ND = no data; NDEP = Nevada
Division of Environmental Protection

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

647

CHAPTER 8. REFERENCES
+3M. 1983. Two year oral (diet) toxicity/carcinogenicity study of fluorochemical FC-143 in rats.
Washington, DC: U.S. Environmental Protection Agency. Submitted to the U.S. Environmental
Protection Agency under TSCA Section 8E. OTS0204926-1.
3M. 1999. The science of organic fluorochemistry. U.S. Environmental Protection Agency. OPPT2002-0043-0006. http://www.fluoridealert.org/pesticides/pfos.fr.final.docket.0006.pdf. July 08,
2008.
3M. 2000. Sulfonated perfluorochemicals in the environment: Sources; dispersion, fate and effects. 3M
Company submission to the U.S. Environmental Protection Agency's Administrative Record.
OPPT-2002-0043-0005.
+3M. 2001. A 28-day oral (gavage) toxicity study of T-7485 in Sprague-Dawley rats. St Paul, MN: 3M
Corporate Toxicology.
+3M. 2007a. A 5-day repeat dose oral toxicity screening study in rats with a 7-day recovery period with
MTDID. St. Paul, MN: 3M Corporate Toxicology.
3M. 2007b. Remedial investigation report. Phase 2. Fluorochemical (FC) data assessment report for the
Cottage Grove, MN site. St. Paul, MN: 3M Corporate Toxicology.
3M. 2008a. Information about PFOS and PFOA. 3M Company.
http://solutions.3m.com/wps/portal/3M/en_US/PFOS/PFOA/. April 01, 2008.
3M. 2008b. Data assessment report. 3M Decatur, Alabama facility PFOA site-related environmental
monitoring program. St. Paul, MN: 3M Company.
3M. 2008c. Screening level human exposure assessment report. 3M Decatur, Alabama facility PFOA
site-related environmental monitoring program. St. Paul, MN: 3M Company.
3M. 2010. GLP10-01-01, Interim report 10: Analysis of PFBS, PFHS, and PFOS in groundwater
samples collected at off-site wells located in Decatur, AL in October 2010. In: Docket EPA-HQ2004-0112-0155. Letter to USEPA, Docket Center, from Gary A. Hohenstein, Manager of
Environmental and Regulatory Affairs, 3M Company. 3M.
https://www.regulations.gov/contentStreamer?documentId=EPA-HQ-OPPT-2004-01120155&attachmentNumber=9&contentType=pdf February 28, 2017.
3M. 2012. E12-0849 Revised final report- Decatur soils- December 2012.
https://www.regulations.gov/contentStreamer?documentId=EPA-HQ-OPPT-2004-01120155&attachmentNumber=36&contentType=pdf February 28, 2017.
Abbott BD. 2009. Review of the expression of peroxisome proliferator-activated receptors alpha
(PPARα), beta (PPARβa), and gamma (PPARγ) in rodent and human development. Reprod Toxicol
27(3-4, Sp. Iss. SI):246-257. 10.1016/j.reprotox.2008.10.001.
+Abbott BD, Wolf CJ, Das KP, et al. 2009. Developmental toxicity of perfluorooctane sulfonate (PFOS)
is not dependent on expression of peroxisome proliferator activated receptor-alpha (PPARα) in the
mouse. Reprod Toxicol 27(3-4):258-265.
+Abbott BD, Wolf CJ, Schmid JE, et al. 2007. Perfluorooctanoic acid (PFOA)-induced developmental
toxicity in the mouse is dependent on expression of peroxisome proliferator activated receptor-alpha.
Toxicol Sci 98(2):571-581.
Abbott BD, Wood CR, Watkins AM, et al. 2010. Peroxisome proliferator-activated receptors alpha, beta,
and gamma mRNA and protein expression in human fetal tissues. PPAR Res 10.1155/2010/690907.
Abbott BD, Wood CR, Watkins AM, et al. 2012. Effects of perfluorooctanoic acid (PFOA) on
expression of peroxisome proliferator-activated receptors (PPAR) and nuclear receptor-regulated
genes in fetal and postnatal CD-1 mouse tissues. Reprod Toxicol 33(4):491-505.
_______________________
+ Cited in supplemental document

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

648
8. REFERENCES

Abdellatif A, Al-Tonsy AH, Awad ME, et al. 2004. Peroxisomal enzymes and
δ-hydroxydeoxyguanosine in rat liver treated with perfluorooctanoic acid. Dis Markers 19(1):19-25.
Abdellatif AG, Preat V, Taper HS, et al. 1991. The modulation of rat liver carcinogenesis by
perfluorooctanoic acid, a peroxisome proliferator. Toxicol Appl Pharmacol 111:530-537.
ACGIH. 2016. CAS Number index. TLVs and BEIs based on the documentation of the threshold limit
values for chemical substances and physical agents and biological exposure indices. Cincinnati, OH:
American Conference of Governmental Industrial Hygienists, 233-250.
Ahmed DY, Abd Ellah MR. 2012. Effect of exposure to perfluorooctanoic acid on hepatic antioxidants
in mice. Comp Clin Pathol 21(6):1643-1645.
AIHA. 2015. Current ERPG Values (2015). Fairfax, VA: American Industrial Hygiene Association.
https://www.aiha.org/getinvolved/AIHAGuidelineFoundation/EmergencyResponsePlanningGuidelines/Documents/2015%20
ERPG%20Levels.pdf. March 22, 2016.
+Albrecht PP, Torsell NE, Krishnan P, et al. 2013. A species difference in the peroxisome proliferatoractivated receptor α-dependent response to the developmental effects of perfluorooctanoic acid.
Toxicol Sci 131(2):568-582.
Alexander BH, Olsen GW. 2007. Bladder cancer in perfluorooctanesulfonyl fluoride manufacturing
workers. Ann Epidemiol 17(6):471-478.
Alexander BH, Olsen GW, Burris JM, et al. 2003. Mortality of employees of a perfluorooctanesulphonyl
fluoride manufacturing facility. Occup Environ Med 60:722-729.
Allen BC, Covington TR, Clewell HJ. 1996. Investigation of the impact of pharmacokinetic variability
and uncertainty on risks predicted with a pharmacokinetic model for chloroform. Toxicology 111(13):289-303.
Alves A, Jacobs G, Vanermen G, et al. 2015. New approach for assessing human perfluoroalkyl
exposure via hair. Talanta 144:574-583. 10.1016/j.talanta.2015.07.009.
Andersen CS, Fei C, Gamborg M, et al. 2010. Prenatal exposures to perfluorinated chemicals and
anthropometric measures in infancy. Am J Epidemiol 172(11):1230-1237.
Andersen CS, Fei C, Gamborg M, et al. 2013. Prenatal exposures to perfluorinated chemicals and
anthropometry at 7 years of age. Am J Epidemiol 178(6):921-927.
Andersen ME, Butenhoff JL, Chang SC, et al. 2008. Perfluoroalkyl acids and related chemistriestoxicokinetics and modes of action. Toxicol Sci 102(1):3-14.
Andersen ME, Clewell HJ, Tan YM, et al. 2006. Pharmacokinetic modeling of saturable, renal
resorption of perfluoroalkylacids in monkeys - probing the determinants of long plasma half-lives.
Toxicology 227(1-2):156-164.
Anderson RH, Long GC, Porter RC, et al. 2016. Occurrence of select perfluoroalkyl substances at U.S.
Air Force aqueous film-forming foam release sites other than fire-training areas: Field-validation of
critical fate and transport properties. Chemosphere 150:678-685.
10.1016/j.chemosphere.2016.01.014.
Anderson-Mahoney P, Kotlerman J, Takhar H, et al. 2008. Self-reported health effects among
community residents exposed to perfluorooctanoate. New Solut 18(2):129-143.
Apelberg BJ, Goldman LR, Calafat AM, et al. 2007a. Determinants of fetal exposure to polyfluoroalkyl
compounds in Baltimore, Maryland. Environ Sci Technol 41:3891-3897.
Apelberg FJ, Witter FR, Herbstman JB, et al. 2007b. Cord serum concentrations of perfluorooctane
sulfonate (PFOS) and perfluorooctanoate (PFOA) in relation to weight and size at birth. Environ
Health Perspect 115:1670-1676.
Armitage J, Cousins I, Buck RC, et al. 2006. Modeling global-scale fate and transport of
perfluorooctanoate emitted from direct sources. Environ Sci Technol 40:6969-6975.
Ashley-Martin J, Dodds L, Arbuckle TE, et al. 2016. Maternal and neonatal levels of perfluoroalkyl
substances in relation to gestational weight gain. Int J Environ Res Public Health 13(1):146.
10.3390/ijerph13010146.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

649
8. REFERENCES

Ashley-Martin J, Dodds L, Levy AR. 2015. Prenatal exposure to phthalates, bisphenol A and
perfluoroalkyl substances and cord blood levels of IgE, TSLP and IL-33. Environ Res 140:360-368.
ATSDR. 1989. Decision guide for identifying substance-specific data needs related to toxicological
profiles; Notice. Agency for Toxic Substances and Disease Registry, Division of Toxicology. Fed
Regist 54(174):37618-37634.
ATSDR. 2003. Guidance for the preparation of toxicological profiles. Agency for Toxic Substances and
Disease Registry. https://www.atsdr.cdc.gov/toxprofiles/index.asp#profiledevelopment. July 6,
2017.
ATSDR. 2005. Health consultation. 3M chemolite. Perfluorochemical releases at the 3M-Cottage
Grove facility. Atlanta, GA: Agency for Toxic Substances and Disease Registry.
http://www.atsdr.cdc.gov/HAC/pha/3M-CGF021805-MN/3M-CGF021805-MN_pt1.pdf. June 27,
2008.
ATSDR. 2008. Public health Assessment for perfluorochemical contamination in Lake Elmo and
Oakdale, Washington County, Minnesota. EPA facility ID: MND980704738 and MND980609515
August 29, 2008. Agency for Toxic Substances and Disease Registry.
http://www.health.state.mn.us/divs/eh/hazardous/sites/washington/lakeelmo/phaelmooakdale.pdf.
November 13, 2008.
ATSDR. 2013. Health consultation. Exposure investigation report. Perfluorochemical serum sampling
in the vicinity of Decatur, Alabama, Morgan, Lawrence, and Limestone counties. Atlanta, GA:
Agency for Toxic Substances and Disease Registry, U.S. Department of Health and Human
Services.
ATSDR. 2017. Perfluoroalkyls. Full SPL data. Substance priority list (SPL) resource page. Agency for
Toxic Substances and Disease Registry, Centers for Disease Control and Prevention.
http://www.atsdr.cdc.gov/SPL/resources/index.html. October 6, 2017.
Bach CC, Bech BH, Nohr EA, et al. 2016. Perfluoroalkyl acids in maternal serum and indices of fetal
growth: The Aarhus Birth Cohort. Environ Health Perspect 10.1289/ehp.1510046.
Bach CC, Bech BH, Nohr EA, et al. 2015a. Serum perfluoroalkyl acids and time to pregnancy in
nulliparous women. Environ Res 142:535-541. 10.1016/j.envres.2015.08.007.
Bach CC, Liew Z, Bech BH, et al. 2015b. Perfluoroalkyl acids and time to pregnancy revisited: An
update from the Danish National Birth Cohort. Environ Health 14:59. 10.1186/s12940-015-0040-9.
Bach CC, Liew Z, Bech BH, et al. 2015c. Additional file to perfluoroalkyl acids and time to pregnancy
revisited: An update from the Danish National Birth Cohort [Environ Health 14:59]. Environ
Health 10.1186/s12940-015-0040-9.
Bae J, Kim S, Schisterman EF, et al. 2015. Maternal and paternal serum concentrations of perfluoroalkyl
and polyfluoroalkyl substances and the secondary sex ratio. Chemosphere 133:31-40.
10.1016/j.chemosphere.2015.03.015.
Barbarossa A, Masetti R, Gazzotti T, et al. 2013. Perfluoroalkyl substances in human milk: A first
survey in Italy. Environ Int 51:27-30.
Barber JL, Berger U, Chaemfa C, et al. 2007. Analysis of per- and polyfluorinated alkyl substances in air
samples from Northwest Europe. J Environ Monit 9:530-541.
Barnes DG, Dourson M. 1988. Reference dose (RfD): Description and use in health risk assessments.
Regul Toxicol Pharmacol 8(4):471-486.
Barrett ES, Chen C, Thurston SW, et al. 2015. Perfluoroalkyl substances and ovarian hormone
concentrations in naturally cycling women. Fertil Steril 103(5):1261-1270 e1263.
10.1016/j.fertnstert.2015.02.001.
Barry V, Darrow LA, Klein M, et al. 2014. Early life perfluorooctanoic acid (PFOA) exposure and
overweight and obestity risk in adulthood in a community with elevated exposure. Environ Res
132:62-69.
Barry V, Winquist A, Steenland K. 2013. Perfluorooctanoic acid (PFOA) exposures and incident cancers
among adults living near a chemical plant. Environ Health Perspect 121(11-12):1313-1318.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

650
8. REFERENCES

Bartell SM, Calafat AM, Lyu C, et al. 2010. Rate of decline in serum PFOA concentrations after
granular activated carbon filtration at two public water systems in Ohio and West Virginia. Environ
Health Perspect 118(2):222-228.
Barton CA, Butler LE, Zarzecki CJ, et al. 2006. Characterizing perfluorooctanoate in ambient air near
the fence line of a manufacturing facility: Comparing modeled and monitored values. J Air Waste
Manag Assoc 56:48-55.
Barton CA, Kaiser MA, Russell MH. 2007. Partitioning and removal of perfluorooctanoate during rain
events: The importance of physical-chemical properties. J Environ Monit 9:839-846.
Beesoon S, Martin JW. 2015. Isomer-specific binding affinity of perfluorooctanesulfonate (PFOS) and
perfluorooctanoate (PFOA) to serum proteins. Environ Sci Technol 49(9):5722-5731.
Begley TH, White K, Honigfort P, et al. 2005. Perfluorochemicals: Potential sources of and migration
from food packaging. Food Addit Contam 22(10):1023-1031.
Benninghoff AD, Bisson WH, Koch DC, et al. 2011. Estrogen-like activity of perfluoroalky acids in
vivo and interaction with human and rainbow trout estrogen receptors in vitro. Toxicol Sci
120(1):42-58.
Benninghoff AD, Orner GA, Buchner CH, et al. 2012. Promotion of hepatocarcinogenesis by
perfluoroalkyl acids in rainbow trout. Toxicol Sci 125(1):69-78. 10.1093/toxsci/kfr267.
Benskin JP, De Silva AO, Martin LJ, et al. 2009. Disposition of perfluorinated acid isomers in SpragueDawley rats: Part 1: Single dose. Environ Toxicol Chem 28(3):542-554.
Berg V, Nost TH, Hansen S, et al. 2015. Assessing the relationship between perfluoroalkyl substances,
thyroid hormones and binding proteins in pregnant women; a longitudinal mixed effects approach.
Environ Int 77:63-69. 10.1016/j.envint.2015.01.007.
+Biegel LB, Hurtt ME, Frame SR, et al. 2001. Mechanisms of extrahepatic tumor induction by
peroxisome proliferators in male CD rats. Toxicol Sci 60(1):44-55.
+Biegel LB, Liu RC, Hurtt ME, et al. 1995. Effects of ammonium perfluorooctanoate on Leydig cell
function: In vitro, in vivo, and ex vivo studies. Toxicol Appl Pharmacol 134(1):18-25.
+Bijland S, Rensen PC, Pieterman EJ, et al. 2011. Perfluoroalkyl sulfonates cause alkyl chain lengthdependent hepatic steatosis and hypolipidemia mainly by impairing lipoprotein production in
APOE*3-Leiden CETP mice. Toxicol Sci 123(1):290-303. 10.1093/toxsci/kfr142.
Bilott RA. 2004. PFOA-exposed community blood sample results (for AR-226 and OPPT-2003-0012).
Submitted to the U.S. Environmental Protection Agency under TSCA Section FYI.
http://www.epa.gov/oppt/tsca8e/pubs/8ehq/2004/oct04/fyi_1004_01480a.pdf. May 5, 2014.
Bilott RA. 2005a. Perfluorochemical residential exposure data for Washington County, Minnesota.
Submitted to the U.S. Environmental Protection Agency.
Bilott RA. 2005b. Perfluorochemical residential exposure data for Washington County, Minnesota.
Submitted to the U.S. Environmental Protection Agency.
http://yosemite.epa.gov/sab/sabproduct.nsf/1BF9ABCD791BDDE38525711F000156BE/$File/pfoa_
sab_lett-bilott_05-20-05.pdf. May 5, 2014.
Bilott RA. 2007. Perfluorochemical residential exposure data for Washington County, Minnesota. EPAHQ-OPPT-2003-0012-1230. https://www.regulations.gov/document?D=EPA-HQ-OPPT-20030012-1230. February 28, 2017.
Bischel HN, MacManus-Spencer LA, Zhang C, et al. 2011. Strong associations of short-chain
perfluoroalkyl acids with serum albumin and investigation of binding mechanisms. Environ Toxicol
Chem 30(11):2423-2430.
Bjermo H, Darnerud PO, Pearson M, et al. 2013. Serum concentrations of perfluorinated alkyl acids and
their associations with diet and personal characteristics among Swedish adults. Mol Nutr Food Res
57(12):2206-2215.
Bjerregaard-Olesen C, Bach CC, Long M, et al. 2016. Time trends of perfluorinated alkyl acids in serum
from Danish pregnant women 2008-2013. Environ Int 91:14-21. 10.1016/j.envint.2016.02.010.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

651
8. REFERENCES

Bjork JA, Wallace KB. 2009. Structure-activity relationships and human relevance for perfluoroalkyl
acid-induced transcriptional activation of peroxisome proliferation in liver cell cultures. Toxicol Sci
111(1):89-99.
Bjork JA, Butenhoff JL, Wallace KB. 2011. Multiplicity of nuclear receptor activation by PFOA and
PFOS in primary human and rodent hepatocytes. Toxicology 288(1-3):8-17.
10.1016/j.tox.2011.06.012.
Björklund JA, Thuresson K, de Wit CA. 2009. Perfluoroalkyl compounds (PFCs) in indoor dust:
Concentrations, human exposure estimates, and sources. Environ Sci Technol 43(7):2276-2281.
Blaine AC, Rich CD, Hundal LS, et al. 2013. Uptake of perfluoroalkyl acids into edible crops via land
applied biosolids: Field and greenhouse studies. Environ Sci Technol 47:14062-14069.
Blaine AC, Rich CD, Sedlacko EM, et al. 2014a. Perfluoroalkyl acid distribution in various plant
compartments of edible crops grown in biosolids-amended soils. Environ Sci Technol 48(14):78587865. 10.1021/es500016s.
Blaine AC, Rich CD, Sedlacko EM, et al. 2014b. Perfluoroalkyl acid uptake in lettuce (Lactuca sativa)
and strawberry (Fragaria ananassa) irrigated with reclaimed water. Environ Sci Technol 48:1436114368.
Bloom MS, Kannan K, Spliethoff HM, et al. 2010. Exploratory assessment of perfluorinated compounds
and human thyroid function. Physiol Behav 99(2):240-245.
Bogdanska J, Borg D, Sundstrom M, et al. 2011. Tissue distribution of 35S-labelled perfluorooctane
sulfonate in adult mice after oral exposure to a low environmentally relevant dose or a high
experimental dose. Toxicology 284(1-3):54-62.
Bogdanska J, Sundstrom M, Bergstrom U, et al. 2014. Tissue distribution of 35S-labelled
perfluorobutanesulfonic acid in adult mice following dietary exposure for 1-5 days. Chemosphere
98:28-36.
Boiteux V, Dauchy X, Rosin C, et al. 2012. National screening study on 10 perfluorinated compounds in
raw and treated tap water in France. Arch Environ Contam Toxicol 63(1):1-12. 10.1007/s00244012-9754-7.
Bonefeld-Jorgensen EC, Long M, Bossi R, et al. 2011. Perfluorinated compounds are related to breast
cancer risk in Greenlandic Inuit: A case control study. Environ Health 10:88. 10.1186/1476-069X10-88.
Bonefeld-Jorgensen EC, Long M, Fredslund SO, et al. 2014. Breast cancer risk after exposure to
perfluorinated compounds in Danish women: A case-control study nested in the Danish National
Birth Cohort. Cancer Causes Control 25(11):1439-1448. 10.1007/s10552-014-0446-7.
Borg D, Bogdanska J, Sundstrom M, et al. 2010. Tissue distribution of 35S-labelled perfluorooctane
sulfonate (PFOS) in C57Bl/6 mice following late gestational exposure. Reprod Toxicol 30(4):558565.
Bossi R, Riget FF, Dietz R. 2005. Temporal and spatial trends of perfluorinated compounds in ringed
seal (Phoca hispida) from Greenland. Environ Sci Technol 39:7416-7422.
Boulanger B, Peck AM, Schnoor JL, et al. 2005. Mass budget of perfluorooctane surfactants in Lake
Ontario. Environ Sci Technol 39:74-79.
Boulanger B, Vargo J, Schnoor JL, et al. 2004. Detection of perfluorooctane surfactants in Great Lakes
water. Environ Sci Technol 38:4064-4070.
Brantsaeter AL, Whitworth KW, Ydersbond TA, et al. 2013. Determinants of plasma concentration of
perfluoroalkyl substances in pregnant Norwegian women. Environ Int 54:74-84.
Braun JM, Chen A, Romano ME, et al. 2016a. Prenatal perfluoroalkyl substance exposure and child
adiposity at 8 years of age: The HOME study. Obesity (Silver Spring, Md.) 24(1):231-237.
10.1002/oby.21258.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

652
8. REFERENCES

Braun JM, Chen A, Romano ME, et al. 2016b. Supplemental material for: Prenatal perfluoroalkyl
substance exposure and child adiposity at 8 years of age: The HOME study. [Obesity 24(1):231237]. Obesity (Silver Spring, Md.) 24(1):231-237.
http://onlinelibrary.wiley.com/store/10.1002/oby.21258/asset/supinfo/oby21258-sup-0001suppinfo01.docx?v=1&s=cac722d8275d428f198a62d78041881f25813e38.
Braun JM, Kalkbrenner AE, Just AC, et al. 2014. Gestational exposure to endocrine-disrupting
chemicals and reciprocal social, repetitive, and stereotypic behaviors in 4- and 5-year-old children:
The HOME study. Environ Health Perspect 122(5):513-520. 10.1289/ehp.1307261.
+Brewster DW, Birnbaum LS. 1989. The biochemical toxicity of perfluorodecanoic acid in the mouse is
different from that of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol 99:544-554.
Brochot C, Smith TJ, Bois FY. 2007. Development of a physiologically based toxicokinetic model for
butadiene and four major metabolites in humans: Global sensitivity analysis for experimental design
issues. Chem Biol Interact 167(3):168-183. 10.1016/j.cbi.2007.02.010.
Buck RC, Franklin J, Berger U, et al. 2011. Perfluoroalkyl and polyfluoroalkyl substances in the
environment: Terminology, classification, and origins. Integr Environ Assess Manag 7(4):513-541.
Buck Louis GM, Chen Z, Schisterman EF, et al. 2015. Perfluorochemicals and human semen quality:
The LIFE study. Environ Health Perspect 123(1):57-63. 10.1289/ehp.1307621.
Buck Louis GM, Peterson CM, Chen Z, et al. 2012. Perfluorochemicals and endometriosis. The ENDO
study. Epidemiology 23(6):799-805.
Buist SCN, Klaassen CD. 2004. Rat and mouse differences in gender-predominant expression of organic
anion transporter (OAT1-3; SLC22A6-9) mRNA levels. Drug Metab Dispos 32(6):620-625.
Burns DC, Ellis DA, Li H, et al. 2008. Experimental pΚa determination for perfluorooctanoic acid
(PFOA) and the potential impact of pKa concentration dependence on laboratory-measured
phenomena and environmental modeling. Environ Sci Technol 42(24):9283-9288.
Buser MC, Scinicariello F. 2016. Perfluoroalkyl substances and food allergies in adolescents. Environ
Int 88:74-79. 10.1016/j.envint.2015.12.020.
+Butenhoff JL, Bjork JA, Chang SC, et al. 2012a. Toxicological evaluation of ammonium
perfluorobutyrate in rats: Twenty-eight-day and ninety-day oral gavage studies. Reprod Toxicol
33(4):513-530.
+Butenhoff JL, Chang S, Ehresman DJ, et al. 2009a. Evaluation of potential reproductive and
developmental toxicity of potassium perfluorohexanesulfonate in Sprague Dawley rats. Reprod
Toxicol 27:331-341.
+Butenhoff JL, Chang SC, Olsen GW, et al. 2012b. Chronic dietary toxicity and carcinogenicity study
with potassium perfluorooctanesulfonate in Sprague Dawley rats. Toxicology 293(1-3):1-15.
+Butenhoff J, Costa G, Elcombe C, et al. 2002. Toxicity of ammonium perfluorooctanoate in male
Cynomolgus monkeys after oral dosing for 6 months. Toxicol Sci 69(1):244-257.
+Butenhoff JL, Ehresman DJ, Chang SC, et al. 2009b. Gestational and lactational exposure to potassium
perfluorooctanesulfonate (K+PFOS) in rats: Developmental neurotoxicity. Reprod Toxicol 27(34):319-330.
Butenhoff JL, Gaylor DW, Moore JA, et al. 2004a. Characterization of risk for general population
exposure to perfluorooctanoate. Regul Toxicol Pharmacol 39(3):363-380.
+Butenhoff JL, Kennedy GL, Frame SR, et al. 2004b. The reproductive toxicology of ammonium
perfluorooctanoate (APFO) in the rat. Toxicology 196(1-2):95-116.
Butenhoff JL, Kennedy GL, Hinderliter PM, et al. 2004c. Pharmacokinetics of perfluorooctanoate in
Cynomolgus monkeys. Toxicol Sci 82:394-406.
Butenhoff JL, Kennedy GL, Jung R, et al. 2014. Evaluation of perfluorooctanoate for potential
genotoxicity. Toxicol Rep 1:252-270.
Butenhoff JL, Pieterman E, Ehresman DJ, et al. 2012c. Distribution of perfluorooctanesulfonate and
perfluorooctanoate into human plasma lipoprotein fractions. Toxicol Lett 210(3):360-365.
10.1016/j.toxlet.2012.02.013.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

653
8. REFERENCES

Butt CM, Mabury SA, Kwan M, et al. 2008. Spatial trends of perfluoroalkyl compounds in ringed seals
(Phoca hispida) from the Canadian Arctic. Environ Toxicol Chem 27(3):542-553.
Butt CM, Mabury SA, Muir DCG, et al. 2007a. Prevalence of long-chained perfluorinated carboxylates
in seabirds from the Canadian Arctic between 1975 and 2004. Environ Sci Technol 41:3521-3528.
Butt CM, Muir DCG, Stirling I, et al. 2007b. Rapid response of arctic ringed seals to changes in
perfluoroalkyl production. Environ Sci Technol 41(1):42-49.
Calafat AM, Kuklenyik Z, Caudill SP, et al. 2006a. Perfluorochemicals in pooled serum samples from
United States residents in 2001 and 2002. Environ Sci Technol 40:2128-2134.
Calafat AM, Kuklenyik Z, Reidy JA, et al. 2007a. Serum concentrations of 11 polyfluoroalkyl
compounds in the U.S. population: Data from the National Health and Nutrition Examination
Survey (NHANES) 1999-2000. Environ Sci Technol 41:2237-2242.
Calafat AM, Needham LL, Kuklenyik Z, et al. 2006b. Perfluorinated chemicals in selected residents of
the American continent. Chemosphere 63:490-496.
Calafat AM, Wong L, Kuklenyik Z, et al. 2007b. Polyfluoroalkyl chemicals in the U.S. population:
Data from the National Health and Nutrition Examination Survey (NHANES) 2003-2004 and
comparisons with NHANES 1999-2000. Environ Health Perspect 115:1596-1602.
Cariou R, Veyrand B, Yamada A, et al. 2015. Perfluoroalkyl acid (PFAA) levels and profiles in breast
milk, maternal and cord serum of French women and their newborns. Environ Int 84:71-81.
Carr CK, Watkins AM, Wolf CJ, et al. 2013. Testing for departures from additivity in mixtures of
perfluoroalkyl acids (PFAAs). Toxicology 306:169-175. 10.1016/j.tox.2013.02.016.
CAS. 2008. Registry. Columbus, OH: Chemical Abstracts Service. http://stnweb.cas.org/. March 14,
2008.
+Case MT, York RG, Christian MS. 2001. Rat and rabbit oral developmental toxicology studies with
two perfluorinated compounds. Int J Toxicol 20(2):101-109.
Cattley RC, DeLuca J, Elcombe C, et al. 1998. Do peroxisome proliferating compounds pose a
hepatocarcinogenic hazard to humans? Regul Toxicol Pharmacol 27:47-60.
CDC. 2013. Fourth national report on human exposure to environmental chemicals. Updated tables,
September 2013. U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention.
CDC. 2015. Fourth national report on human exposure to environmental chemicals. February 2015.
U.S. Department of Health and Human Services, Centers for Disease Control and Prevention.
https://www.cdc.gov/biomonitoring/pdf/fourthreport_updatedtables_feb2015.pdf. May 25, 2017.
CDC. 2018. Fourth national report on human exposure to environmental chemicals. Updated Tables,
March 2018. Atlanta, GA: Centers for Disease Control and Prevention, Department of Health and
Human Services.
https://www.cdc.gov/exposurereport/pdf/FourthReport_UpdatedTables_Volume1_Mar2018.pdf.
April 23, 2018.
Celik A, Eke D, Ekinci SY, et al. 2013. The protective role of curcumin on perfluorooctane sulfonateinduced genotoxicity: Single cell gel electrophoresis and micronucleus test. Food Chem Toxicol
53:249-255.
CEMN. 2008. CEMC focussing on ionizing surfactants. Canadian Environmental Modelling Network.
http://www.trentu.ca/academic/aminss/envmodel/cemn/NewsReports/CEMNnews200804.pdf. June
24, 2008.
Chan E, Burstyn I, Cherry N, et al. 2011. Perfluorinated acids and hypothyroxinemia in pregnant
women. Environ Res 111(4):559-564.
Chang ET, Adami H, Boffetta P, et al. 2014. A critical review of perfluorooctanoate and
perfluorooctanesulfonate exposure and cancer risk in humans. Crit Rev Toxicol 44:1-81.
Chang S, Das K, Ehresman DJ, et al. 2008a. Comparative pharmacokinetics of perfluorobutyrate
(PFBA) in rats, mice, monkeys, and humans and relevance to human exposure via drinking water.
Toxicol Sci 104(1):40-53.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

654
8. REFERENCES

+Chang S, Ehresman DJ, Bjork JA, et al. 2009. Gestational and lactational exposure to potassium
perfluorooctanesulfonate (K+PFOS) in rats: Toxicokinetics, thyroid hormone status, and related
gene expression. Reprod Toxicol 27(3-4):387-399.
Chang S, Noker PE, Gorman GS, et al. 2012. Comparative pharmacokinetics of
perfluorooctanesulfonate (PFOS) in rats, mice, and monkeys. Reprod Toxicol 33(4):428-440.
+Chang S, Thibodeaux JR, Eastvold ML, et al. 2008b. Thyroid hormone status and pituitary function in
adult rats given oral doses of perfluorooctanesulfonate (PFOS). Toxicology 243:330-339.
Château-Degat ML, Pereg D, Dallaire R, et al. 2010. Effects of perfluorooctanesulfonate exposure on
plasma lipid levels in the Inuit population of Nunavik (Northern Quebec). Environ Res 110(7):710717.
ChemIDplus. 2008. Perfluoroalkyls. ChemIDplus. Bethesda, MD: U.S. National Library of Medicine.
http://sis.nlm.nih.gov/chemical.html. July 10, 2008.
ChemIDplus. 2017. Ammonium perfluorooctanoate. Bethesda, MD: ChemIDplus. U.S. National
Library of Medicine. http://sis.nlm.nih.gov/chemical.html. December 22, 2017.
Chen MH, Ha EH, Liao HF, et al. 2013. Perfluorinated compound levels in cord blood and
neurodevelopment at 2 years of age. Epidemiology 24(6):800-808.
10.1097/EDE.0b013e3182a6dd46.
Chen MH, Ha EH, Wen TW, et al. 2012a. Perfluorinated compounds in umbilical cord blood and
adverse birth outcomes. PLoS ONE 7(8):e42474.
+Chen T, Zhang L, Yue JQ, et al. 2012b. Prenatal PFOS exposure induces oxidative stress and apoptosis
in the lung of rat off-spring. Reprod Toxicol 33(4):538-545.
Chen YM, Guo LH. 2009. Fluorescence study on site-specific binding of perfluoroalkyl acids to human
serum albumin. Arch Toxicol 83(3):255-261.
+Cheng J, Fujimura M, Zhao W, et al. 2013. Neurobehavioral effects, c-Fos/Jun expression and tissue
distribution in rat offspring prenatally co-exposed to MeHg and PFOA: PFOA impairs Hg retention.
Chemosphere 91(6):758-764.
Cheng W, Ng CA. 2017. A permeability-limited physiologically based pharmacokinetic (PBPK) model
for perfluorooctanoic acid (PFOA) in male rats. Environ Sci Technol 51(17):9930-9939.
10.1021/acs.est.7b02602.
Cheng X, Klaassen CD. 2008a. Critical role of PPAR-α in perfluorooctanoic acid- and
perfluorodecanoic acid-induced downregulation of oatp uptake transporters in mouse livers. Toxicol
Sci 106(1):37-45. 10.1093/toxsci/kfn161.
Cheng X, Klaassen CD. 2008b. Perfluorocarboxylic acids induce cytochrome P450 enzymes in mouse
liver through activation of PPAR-alpha and CAR transcription factors. Toxicol Sci 106(1):29-36.
10.1093/toxsci/kfn147.
Chengelis CP, Kirkpatrick JB, Myers NR, et al. 2009. Comparison of the toxicokinetic behavior of
perfluorohexanoic acid (PFHxA) and nonafluorobutane-1-sulfonic acid (PFBS) in Cynomolgus
monkeys and rats. Reprod Toxicol 27:400-406.
Christensen KY, Maisonet M, Rubin C, et al. 2011. Exposure to polyfluoroalkyl chemicals during
pregnancy is not associated with offspring age at menarche in a contemporary British cohort.
Environ Int 37(1):129-135.
Christensen KY, Raymond M, Thompson BA, et al. 2016. Perfluoroalkyl substances in older male
anglers in Wisconsin. Environ Int 91:312-318. 10.1016/j.envint.2016.03.012.
Clara M, Gans O, Weiss S, et al. 2009. Perfluorinated alkylated substances in the aquatic environment:
An Austrian case study. Water Res 43(18):4760-4768. 10.1016/j.watres.2009.08.004.
Clarke BO, Smith SR. 2011. Review of 'emerging' organic contaminants in biosolids and assessment of
international research priorities for the agricultural use of biosolids. Environ Int 37(1):226-247.
10.1016/j.envint.2010.06.004.
Clewell HJ, Andersen ME. 1985. Risk assessment extrapolations and physiological modeling. Toxicol
Ind Health 1(4):111-131.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

655
8. REFERENCES

Conder JM, Hoke RA, De Wolf W, et al. 2008. Are PFCAs bioaccumulative? A critical review and
comparison with regulatory criteria and persistent lipophilic compounds. Environ Sci Technol
42(4):995-1003.
Consonni D, Straif K, Symons JM, et al. 2013. Cancer risk among tetrafluoroethylene synthesis and
polymerization workers. Am J Epidemiol 178(3):350-358.
+Cook JC, Murray SM, Frame SR, et al. 1992. Induction of Leydig cell adenomas by ammonium
perfluorooctanoate: A possible endocrine-related mechanism. Toxicol Appl Pharmacol 113(2):209217.
Corsini E, Luebke RW, Germolec DR, et al. 2014. Perfluorinated compounds: Emerging POPs with
potential immunotoxicity. Toxicol Lett 230(2):263-270. 10.1016/j.toxlet.2014.01.038.
Corton JC, Cunningham ML, Hummer BT, et al. 2014. Mode of action framework analysis for receptormediated toxicity: The peroxisome proliferator-activated receptor alpha (PPARα) as a case study.
Crit Rev Toxicol 4444(1):1-49. 10.3109/10408444.2013.835784.
Costa G. 2004. Report on the meeting held on Friday 20th and Saturday 21st 2004 at the Inn at
Montchanin Village (Wilmington, USA) with 3M and DuPont delegations. DuPont. Submitted to
the U.S. Environmental Protection Agency. AR226-1866.
Costa G, Sartori S, Consonni D. 2009. Thirty years of medical surveillance in perfluooctanoic acid
production workers. J Occup Environ Med 51(3):364-372.
+Cui L, Zhou QF, Liao CY, et al. 2009. Studies on the toxicological effects of PFOA and PFOS on rats
using histological observation and chemical analysis. Arch Environ Contam Toxicol 56(2):338-349.
+Curran I, Hierlihy SL, Liston V, et al. 2008. Altered fatty acid homeostasis and related toxicologic
sequelae in rats exposed to dietary potassium perfluorooctanesulfonate (PFOS). J Toxicol Environ
Health A 71(23):1526-1541.
Dai J, Li M, Jin Y, et al. 2006. Perfluorooctanesulfonate and perfluorooctanoate in red panda and giant
panda from China. Environ Sci Technol 40:5647-5652.
Dallaire R, Ayotte P, Pereg D, et al. 2009. Determinants of plasma concentrations of
perfluorooctanesulfonate and brominated organic compounds in Nunavik Inuit adults (Canada).
Environ Sci Technol 43(13):5130-5136.
Dalsager L, Christensen N, Husby S, et al. 2016. Association between prenatal exposure to
perfluorinated compounds and symptoms of infections at age 1-4 years among 359 children in the
Odense Child Cohort. Environ Int 96:58-64.
Darrow LA, Groth AC, Winquist A, et al. 2016. Modeled perfluorooctanoic acid (PFOA) exposure and
liver function in a mid-Ohio Valley community. Environ Health Perspect [Advance publication].
10.1289/ehp.1510391.
Darrow LA, Howards PP, Winquist A, et al. 2014. PFOA and PFOS serum levels and miscarriage risk.
Epidemiology 25(4):505-512. 10.1097/ede.0000000000000103.
Darrow LA, Stein CR, Steenland K. 2013. Serum perfluorooctanoic acid and perfluorooctane sulfonate
concentrations in relation to birth outcomes in the Mid-Ohio Valley, 2005-2010. Environ Health
Perspect 121(10):1207-1213.
+Das KP, Grey BE, Zehr RD, et al. 2008. Effects of perfluorobutyrate exposure during pregnancy in the
mouse. Toxicol Sci 105(1):173-181.
+Das KP, Grey BE, Rosen MB, et al. 2015. Developmental toxicity of perfluorononanoic acid in mice.
Reprod Toxicol 51:133-144. 10.1016/j.reprotox.2014.12.012.
Das KP, Wood CR, Lin MT, et al. 2017. Perfluoroalkyl acids-induced liver steatosis: Effects on genes
controlling lipid homeostasis. Toxicology 378:37-52. 10.1016/j.tox.2016.12.007.
Dauchy X, Boiteux V, Rosin C, et al. 2012. Relationship between industrial discharges and
contamination of raw water resources by perfluorinated compounds. Part I: Case study of a
fluoropolymer manufacturing plant. Bull Environ Contam Toxicol 89(3):525-530. 10.1007/s00128012-0704-x.
Davis KL, Aucoin MD, Larsen BS, et al. 2007. Transport of ammonium perfluorooctanoate in
environmental media near a fluoropolymer manufacturing facility. Chemosphere 67:2011-2019.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

656
8. REFERENCES

de Cock M, de Boer MR, Lamoree M, et al. 2014. First year growth in relation to prenatal exposure to
endocrine disruptors - a Dutch prospective cohort study. Int J Environ Res Public Health
11(7):7001-7021. 10.3390/ijerph110707001.
Delinsky AD, Strynar MJ, McCann PJ, et al. 2010. Geographical distribution of perfluorinated
compounds in fish from Minnesota lakes and rivers. Environ Sci Technol 44(7):2549-2554.
10.1021/es903777s.
Delinsky AD, Strynar MJ, Nakayama SF, et al. 2009. Determination of ten perluorinated compounds in
bluegill sunfish (Lepomis macrochirus) fillets. Environ Res 109:975-984.
D'eon JC, Mabury SA. 2007. Production of perfluorinated carboxylic acids (PFCAs) from the
biotransformation of polyfluoroalkyl phosphate surfactants (PAPS): Exploring routes of human
contamination. Environ Sci Technol 41(13):4799-4805.
D'eon JC, Crozier PW, Furdui VI, et al. 2009. Observation of a commercial fluorinated material, the
polyfluoroalkyl phosphoric acid diesters, in human sera, wastewater treatment plant sludge, and
paper fibers. Environ Sci Technol 43(12):4589-4594.
D'eon JC, Hurley MD, Wallington TJ, et al. 2006. Atmospheric chemistry of n-methyl perfluorobutane
sulfonamidoethanol, C4F9SO2N (Ch3)CH2CH2OH: Kinetics and mechanism of reaction with OH.
Environ Sci Technol 40:1862-1868.
De Silva AO, Benskin JP, Martin LJ, et al. 2009. Disposition of perfluorinated acid isomers in SpragueDawley rats. Part 2: Subchronic dose. Environ Toxicol Chem 28(3):555-567.
De Silva AO, Mabury SA. 2006. Isomer distribution of perfluorocarboxylates in human blood: Potential
correlation to source. Environ Sci Technol 40:2903-2909.
De Silva AO, Spencer C, Scott BF, et al. 2011. Detection of a cyclic perfluorinated acid,
perfluoroethylcyclohexane sulfonate, in the Great Lakes of North America. Environ Sci Technol
45:8060-8066.
de Vos MG, Huijbregts MAJ, van den Heuvel-Greve MJ, et al. 2008. Accumulation of perfluorooctane
sulfonate (PFOS) in the food chain of the Western Scheldt estuary: Comparing field measurements
with kinetic modeling. Chemosphere 70:1766-1773.
+DeWitt JC, Copeland CB, Luebke RW. 2009. Suppression of humoral immunity by perfluorooctanoic
acid is independent of elevated serum corticosterone concentration in mice. Toxicol Sci 109:106112.
+DeWitt JC, Copeland CB, Strynar MJ, et al. 2008. Perfluorooctanoic acid-induced immunomodulation
in adult C57BL/6J or C57BL/6N female mice. Environ Health Perspect 116(5):644-650.
DeWitt JC, Peden-Adams MM, Keller JM, et al. 2012. Immunotoxicity of perfluorinated compounds:
Recent developments. Toxicol Pathol 40(2):300-311. 10.1177/0192623311428473.
+DeWitt JC, Williams WC, Creech NJ, et al. 2016. Suppression of antigen-specific antibody responses
in mice exposed to perfluorooctanoic acid: Role of PPARalpha and T- and B-cell targeting. J
Immunotoxicol 13(1):38-45. 10.3109/1547691x.2014.996682.
Dhingra R, Darrow LA, Klein M, et al. 2016a. Perfluorooctanoic acid exposure and natural menopause:
A longitudinal study in a community cohort. Environ Res 146:323-330.
10.1016/j.envres.2015.12.037.
Dhingra R, Lally C, Darrow LA, et al. 2016b. Perfluorooctanoic acid and chronic kidney disease:
Longitudinal analysis of a Mid-Ohio Valley community. Environ Res 145:85-92.
10.1016/j.envres.2015.11.018.
Dinglasan-Panlilio MJA, Mabury SA. 2006. Significant residual fluorinated alcohols present in various
fluorinated materials. Environ Sci Technol 40:1447-1453.
Dixon D, Reed CE, Moore AB, et al. 2012. Histopathologic changes in the uterus, cervix and vagina of
immature CD-1 mice exposed to low doses of perfluorooctanoic acid (PFOA) in a uterotrophic
assay. Reprod Toxicol 33(4):506-512. 10.1016/j.reprotox.2011.10.011.
Dobraca D, Israel L, McNeel S, et al. 2015. Biomonitoring in California firefighters: Metals and
perfluorinated chemicals. J Occup Environ Med 57(1):88-97. 10.1097/jom.0000000000000307.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

657
8. REFERENCES

DOE. 2016a. Table 3: Protective Action Criteria (PAC) Rev. 29 based on applicable 60-minute AEGLs,
ERPGs, or TEELs. The chemicals are listed by CASRN. May 2016. Oak Ridge, TN: U.S.
Department of Energy. https://sp.eota.energy.gov/pac/teel/Revision_29_Table3.pdf. February 28,
2017.
DOE. 2016b. Protective Action Criteria (PAC) with AEGLs, ERPGs, & TEELs: Rev. 29 for Chemicals
of Concern - May 2016. Oak Ridge, TN: U.S. Department of Energy.
https://energy.gov/ehss/protective-action-criteria-pac-aegls-erpgs-teels-rev-29-chemicals-concernmay-2016. February 28, 2017.
Domingo JL, Ericson-Jogsten I, Perello G, et al. 2012a. Human exposure to perfluorinated compounds
in Catalonia, Spain: Contribution of drinking water and fish and shellfish. J Agric Food Chem
60(17):4408-4415.
Domingo JL, Jogsten IE, Eriksson U, et al. 2012b. Human dietary exposure to perfluoroalkyl substances
in Catalonia, Spain. Temporal trend. Food Chem 135:1575-1582.
Donauer S, Chen A, Xu Y, et al. 2015. Prenatal exposure to polybrominated diphenyl ethers and
polyfluoroalkyl chemicals and infant neurobehavior. J Pediatr 166(3):736-742.
10.1016/j.jpeds.2014.11.021.
+Dong GH, Liu MM, Wang D, et al. 2011. Sub-chronic effect of perfluorooctanesulfonate (PFOS) on
the balance of type 1 and type 2 cytokine in adult C57BL6 mice. Arch Toxicol 85(10):1235-1244.
Dong GH, Tung KY, Tsai CH, et al. 2013. Serum polyfluoroalkyl concentrations, asthma outcomes, and
immunological markers in a case-control study of Taiwanese children. Environ Health Perspect
121(4):507-513.
+Dong GH, Zhang YH, Zheng L, et al. 2009. Chronic effects of perfluorooctanesulfonate exposure on
immunotoxicity in adult male C57BL/6 mice. Arch Toxicol 83(9):805-815.
DRBC. 2013. Contaminants of emerging concern in the tidal Delaware River: Pilot monitoring survey
2007-2009. Delaware River Basin Commission.
http://www.nj.gov/drbc/library/documents/contaminants-of-emerging-concernAug2013rev.pdf. July
6, 2017.
Ducatman AM, Zhang J, Fan H. 2015a. Letter to the editor. (Prostate-specific antigen and
perfluoroalkyl acids in the C8 health study population. J Occup Environ Med 57(1):111-114). J
Occup Environ Med 57(6):e61. 10.1097/jom.0000000000000470.
Ducatman A, Zhang J, Fan H. 2015b. Prostate-specific antigen and perfluoroalkyl acids in the C8 health
study population. J Occup Environ Med 57(1):111-114. 10.1097/jom.0000000000000319.
DuPont. 2008. Information on PFOA. http://repanet.de/PFOA2/en_US/index.html. April 07, 2008.
DWQI. 2015. Maximum contaminant level recommendation for perfluorononanoic acid in drinking
water. Trenton, NJ: New Jersey Drinking Water Quality Institute.
http://www.nj.gov/dep/watersupply/pdf/pfna-recommend-final.pdf. December 18, 2017.
DWQI. 2017a. Maximum contaminant level recommendation for perfluorooctanoic acid in drinking
water. Trenton, NJ: New Jersey Drinking Water Quality Institute.
http://www.nj.gov/dep/watersupply/pdf/pfoa-recommend.pdf. December 18, 2017.
DWQI. 2017b. Public review draft: Perfluorooctane sulfonate (PFOS) (CAS #: 1763-23-1; Chemical
Formula: C8HF17O3S). Health-based maximum contaminant level support document. Trenton, NJ:
Health Effects Subcommitttee, New Jersey Drinking Water Quality Institute.
http://www.nj.gov/dep/watersupply/pdf/dwqi-pfos-mcl-draft.pdf. December 18, 2017.
Eggen T, Moeder M, Arukwe A. 2010. Municipal landfill leachates: A significant source for new and
emerging pollutants. Sci Total Environ 408:5147-5157.
Ehresman DJ, Froehlich JW, Olsen GW, et al. 2007. Comparison of human whole blood, plasma, and
serum matrices for the determination of perfluorooctanesulfonate (PFOS), perfluorooctanoate
(PFOA), and other fluorochemicals. Environ Res 103:176-184.
Eke D, Celik A. 2016. Curcumin prevents perfluorooctane sulfonate-induced genotoxicity and oxidative
DNA damage in rat peripheral blood. Drug Chem Toxicol 39(1):97-103.
10.3109/01480545.2015.1041601.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

658
8. REFERENCES

+Elcombe CR, Elcombe BM, Foster JR, et al. 2010. Hepatocellular hypertrophy and cell proliferation in
Sprague-Dawley rats following dietary exposure to ammonium perfluorooctanoate occurs through
increased activation of the xenosensor nuclear receptors PPARα and CAR/PXR. Arch Toxicol
84(10):787-798.
+Elcombe CR, Elcombe BM, Foster JR, et al. 2012b. Evaluation of hepatic and thyroid responses in
male Sprague Dawley rats for up to eighty-four days following seven days of dietary exposure to
potassium perfluorooctanesulfonate. Toxicology 293(1-3):30-40.
+Elcombe CR, Elcombe BM, Foster JR, et al. 2012a. Hepatocellular hypertrophy and cell proliferation
in Sprague-Dawley rats from dietary exposure to potassium perfluorooctanesulfonate results from
increased expression of xenosensor nuclear receptors PPARα and CAR/PXR. Toxicology 293(13):16-29.
Eldasher LM, Wen X, Little MS, et al. 2013. Hepatic and renal Bcrp transporter expression in mice
treated with perfluorooctanoic acid. Toxicology 306:108-113.
Ellis DA, Martin JW, De Silva AO, et al. 2004. Degradation of fluorotelomer alcohols: A likely
atmospheric source of perfluorinated carboxylic acids. Environ Sci Technol 27:3316-3321.
Emmett EA, Shofer FS, Zhang H, et al. 2006a. Community exposure to perfluorooctanoate:
Relationships between serum concentrations and exposure sources. J Occup Environ Med 48:759770.
Emmett EA, Zhang H, Shofer FS, et al. 2006b. Community exposure to perfluorooctanoate:
Relationships between serum levels and certain health parameters. J Occup Environ Med 48(8):771779.
Emmett EA, Zhang H, Shofer FS, et al. 2009. Development and successful application of a "communityfirst" communication model for community-based environmental health research. J Occup Environ
Med 51(2):146-156.
EPA. 1988. Recommedations for and documentation of biological values for use in risk assessment.
Washington, DC: U.S. Environmental Protection Agency. PB88179874.
EPA. 2002. Perfluoroalkyl sulfonates; significant new use rule. U.S. Environmental Protection Agency.
Fed Regist 67(236):72854-72867.
EPA. 2005a. Draft risk assessment of the potential human health effects associated with exposure to
perfluorooctanoic acid and its salts. U.S. Environmental Protection Agency.
http://www.epa.gov/opptintr/pfoa/pubs/pfoarisk.pdf. June 26, 2007.
EPA. 2005b. Toxic chemical release inventory reporting forms and instructions: Revised 2004 version.
Section 313 of the Emergency Planning and Community Right-to-Know Act (Title III of the
Superfund Amendments and Reauthorization Act of 1986). U.S. Environmental Protection Agency.
Office of Environmental Information. EPA260B05001.
EPA. 2005c. Guidelines for carcinogen risk assessment. Washington, DC: U.S. Environmental
Protection Agency. EPA630P03001F.
http://www.epa.gov/raf/publications/pdfs/CANCER_GUIDELINES_FINAL_3-25-05.PDF.
December 10, 2014.
EPA. 2006. SAB review of EPA's draft risk assessment of potential human health effects associated with
PFOA and its salts. U.S. Environmental Protection Agency.
http://yosemite.epa.gov/sab/sabproduct.nsf/A3C83648E77252828525717F004B9099/$File/sab_06_
006.pdf. April 24, 2008.
EPA. 2007a. Perfluoroalkyl sulfonates; significant new use rule. U.S. Environmental Protection
Agency. Fed Regist 72:57222-57235.
EPA. 2007b. Quarterly MOU status report #6: Phase II monitoring/sampling work plan, DuPont
Washington Works (PFOA site-related environmental assessment program), Appendices 3.1 through
5.1. U.S. Environmental Protection Agency. EPA-HQ-OPPT-2004-0113-0242.1.
http://www.regulations.gov/fdmspublic/component/main?main=DocketDetail&d=EPA-HQ-OPPT2004-0113. May 18, 2009.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

659
8. REFERENCES

EPA. 2008a. Perfluorooctanoic acid (PFOA) and fluorinated telomers. U.S. Environmental Protection
Agency. http://www.epa.gov/oppt/pfoa/. May 29, 2008.
EPA. 2008b. Non-confidential IUR production volume information. Inventory updating reporting. U.S.
Environmental Protection Agency. http://www.epa.gov/opptintr/iur/tools/data/2002-vol.htm. July
09, 2008.
EPA. 2008c. Perfluoroalkyls. Substance registry system. U.S. Environmental Protection Agency.
http://www.epa.gov/srs/. July 10, 2008.
EPA. 2008d. Quarterly MOU status report #10: Phase II monitoring/sampling work plan, DuPont
Washington Works (PFOA site-related environmental assessment program), text, tables 4.1 through
5.10. U.S. Environmental Protection Agency. EPA-HQ-OPPT-2004-0113-0430.
http://www.regulations.gov/fdmspublic/component/main?main=DocketDetail&d=EPA-HQ-OPPT2004-0113. May 18, 2009.
EPA. 2009a. PFOS chromium electroplater study. Cleveland, OH: U.S. Environmental Protection
Agency- Region 5.
EPA. 2009b. Perfluorocarboxylic acid content in 116 articles of commerce. Research Triangle Park,
NC: U.S. Environmental Protection Agency, National Risk Management Research Laboratory,
Office of Research and Development. EPA600R09033.
EPA. 2009c. Method 537. Determination of selected perfluorinated alkyl acids in drinking water by
solid phase extraction and liquid chromatography/tandem mass spectrometry (LC/MS/MS). U.S.
Environmental Protection Agency.
http://www.epa.gov/microbes/documents/Method%20537_FINAL_rev1.1.pdf. May 5, 2014.
EPA. 2009d. National primary drinking water regulations. Washington, DC: Office of Ground Water
and Drinking Water, U.S. Environmental Protection Agency. EPA816F090004.
https://www.epa.gov/sites/production/files/2016-06/documents/npwdr_complete_table.pdf.
February 28, 2017.
EPA. 2010. Quarterly MOU Status Report #17. Phase II. Monitoring/sampling work plan. DuPont
Washington Works (OPPT-2004-0113 PFOA site-related environmental asssessment program).
U.S. Environmental Protection Agency. Project No: 18984356.05013.
EPA. 2014. Health effects document for perfluorooctanoic acid (PFOA). U.S. Environmental Protection
Agency. EPA822R14001. https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=P100IRZ1.txt.
December 21, 2017.
EPA. 2015. Long-chain perfluoroalkyl carboxylate and perfluoroalkyl sulfonate chemical substances:
Significan new use rule. Fed Regist 80(13):2885-2898.
EPA. 2016a. 2010/2015 PFOA Stewardship Program- 2014 annual progress reports. U.S.
Environmental Protection Agency. https://www.epa.gov/assessing-and-managing-chemicals-undertsca/20102015-pfoa-stewardshiip-program-2014-annual-progress#summary. August 08, 2016.
EPA. 2016b. Compiled AEGL values. Acute Exposure Guideline Levels (AEGLs) Values. U.S.
Environmental Protection Agency. https://www.epa.gov/sites/production/files/201603/documents/compiled_aegl_update_.pdf. February 28, 2017.
EPA. 2016c. The Third Unregulated Contaminant Monitoring Rule (UCMR 3): Data summary, July
2016. U.S. Environmental Protection Agency, Office of Water.
EPA. 2016d. Chemical Data Access Tool (CDAT). U.S. Environmental Protection Agency.
https://java.epa.gov/oppt_chemical_search/. December 09, 2016.
EPA. 2016e. Drinking water health advisory for perfluorooctanoic acid (PFOA). U.S. Environmental
Protection Agency. https://www.epa.gov/sites/production/files/201605/documents/pfoa_health_advisory_final-plain.pdf. March 3, 2017.
EPA. 2016f. Drinking water health advisory for perfluorooctane sulfonate (PFOS). U.S. Environmental
Protection Agency. https://www.epa.gov/sites/production/files/201605/documents/pfos_health_advisory_final-plain.pdf. March 3, 2017.
EPA. 2016g. Toxic chemical release inventory reporting forms and instructions: Revised 2016 version.
Section 313 of the Emergency Planning and Community Right-to-Know Act (Title III of the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

660
8. REFERENCES

Superfund Amendments and Reauthorization Act of 1986). U.S. Environmental Protection Agency,
Office of Environmental Information. EPA740B16001.
https://ofmpub.epa.gov/apex/guideme_ext/guideme_ext/r/files/static/v3321/rfi/RY_2016_RFI.pdf.
July 5, 2017.
EPA. 2016h. Health effects support document for perfluorooctanoic acid (PFOA). U.S. Environmental
Protection Agency. EPA822R16003. https://www.epa.gov/sites/production/files/201605/documents/pfoa_hesd_final-plain.pdf. July 6, 2017.
EPA. 2016i. Health effects support document for perfluorooctane sulfate (PFOS). U.S. Environmental
Protection Agency. EPA822R16002. https://www.epa.gov/sites/production/files/201605/documents/pfos_hesd_final_508.pdf. July 6, 2017.
EPA. 2016j. Lifetime health advisories and health effects support documents for perfluorooctanoic acid
and perfluoroactoane sulfonates. Federal Register. U.S. Environmental Protection Agency. Vol. 81
FR 33250 (May 25, 2016). https://www.epa.gov/sites/production/files/2016-05/documents/201612361.pdf. December 18, 2017.
+Era S, Harada KH, Toyoshima M, et al. 2009. Cleft palate caused by perfluorooctane sulfonate is
caused mainly by extrinsic factors. Toxicology 256(1-2):42-47.
EPA. 2017. The third Unregulated Contaminant Monitoring Rule (UCMR 3): Data summary, January
2017. U.S. Environmental Protection Agency. https://www.epa.gov/sites/production/files/201702/documents/ucmr3-data-summary-january-2017.pdf. July 6, 2017.
Ericson I, Domingo JL, Nadal M, et al. 2009. Levels of perfluorinated chemicals in municipal drinking
water from Catalonia, Spain: Public health implications. Arch Environ Contam Toxicol 57(4):631638. 10.1007/s00244-009-9375-y.
Ericson I, Gómez M, Nadal M, et al. 2007. Perfluorinated chemicals in blood of residents in Catalonia
(Spain) in relation to age and gender: A pilot study. Environ Int 33:616-623.
Eriksen KT, Raaschou-Nielsen O, McLaughlin JK, et al. 2013. Association between plasma PFOA and
PFOS levels and total cholesterol in a middle-aged Danish population. PLoS ONE 8(2):e56969.
Eriksen KT, Raaschou-Nielsen O, Sorensen M, et al. 2010. Genotoxic potential of the perfluorinated
chemicals PFOA, PFOS, PFBS, PFNA and PFHxA in human HepG2 cells. Mutat Res 700(1-2):3943. 10.1016/j.mrgentox.2010.04.024.
Eriksen KT, Sorensen M, McLaughlin JK, et al. 2009. Perfluorooctanoate and perfluorooctanesulfonate
plasma levels and risk of cancer in the general Danish population. J Natl Cancer Inst 101(8):605609.
Fábrega F, Kumar V, Schuhmacher M, et al. 2014. PBPK modeling for PFOS and PFOA: Validation
with human experimental data. Toxicol Lett 230(2):244-251.
Fábrega F, Nadal M, Schuhmacher M, et al. 2016. Influence of the uncertainty in the validation of PBPK
models: A case-study for PFOS and PFOA. Regul Toxicol Pharmacol 77:230-239.
10.1016/j.yrtph.2016.03.009
+Fairley KJ, Purdy R, Kearns S, et al. 2007. Exposure to the immunosuppresant, perfluorooctanoic acid,
enhances the murine IgE and airway hyperreactivity response to ovalbumin. Toxicol Sci 97(2):375383.
+Fang X, Feng Y, Shi Z, et al. 2009. Alterations of cytokines and MAPK signaling pathways are related
to the immunotoxic effect of perfluorononanoic acid. Toxicol Sci 108(2):367-376.
10.1093/toxsci/kfp019.
+Fang X, Feng Y, Wang J, et al. 2010. Perfluorononanoic acid-induced apoptosis in rat spleen involves
oxidative stress and the activation of caspase-independent death pathway. Toxicology 267(1-3):5459. 10.1016/j.tox.2009.10.020.
+Fang X, Gao G, Xue H, et al. 2012a. Exposure of perfluorononanoic acid suppresses the hepatic insulin
signal pathway and increases serum glucose in rats. Toxicology 294(2-3):109-115.
+Fang X, Gao G, Xue H, et al. 2012b. In vitro and in vivo studies of the toxic effects of
perfluorononanoic acid on rat hepatocytes and Kupffer cells. Environ Toxicol Pharmacol 34(2):484494.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

661
8. REFERENCES

+Fang X, Zhang L, Feng Y, et al. 2008. Immunotoxic effects of perfluorononanoic acid on BALB/c
mice. Toxicol Sci 105(2):312-321.
Fasano WJ, Carpenter SC, Gannon SA, et al. 2006. Absorption, distribution, metabolism, and
elimination of 8-2 fluorotelomer alcohol in the rat. Toxicol Sci 91(2):341-355.
Fasano WJ, Kennedy GL, Szostek B, et al. 2005. Penetration of ammonium perfluorooctanoate through
rat and human skin in vitro. Drug Chem Toxicol 28(1):79-90.
FDA. 2013. Everything added to food in the United States (EAFUS). Washington, DC: U.S. Food and
Drug Administration. http://www.accessdata.fda.gov/scripts/fcn/fcnnavigation.cfm?rpt=eafuslisting.
February 28, 2017.
Fei C, Olsen J. 2011. Prenatal exposure to perfluorinated chemicals and behavioral or coordination
problems at age 7 years. Environ Health Perspect 119(4):573-578.
Fei C, McLaughlin JK, Lipworth L, et al. 2009. Maternal levels of perfluorinated chemicals and
subfecundity. Hum Reprod 24(5):1200-1205.
Fei C, McLaughlin JK, Lipworth L, et al. 2008b. Prenatal exposure to perfluorooctanoate (PFOA) and
perfluorooctanesulfonate (PFOS) and maternally reported developmental milestones in infancy.
Environ Health Perspect 116(10):1391-1395.
Fei C, McLaughlin JK, Tarone RE, et al. 2007. Perfluorinated chemicals and fetal growth: A study
within the Danish National Birth Cohort. Environ Health Perspect 115:1677-1682.
Fei C, McLaughlin JK, Tarone RE, et al. 2008a. Fetal growth indicators and perfluorinated chemicals:
A study in the Danish National Birth Cohort. Am J Epidemiol 168(1):66-72.
Fei C, McLaughlin JK, Lipworth L, et al. 2010. Prenatal exposure to PFOA and PFOS and risk of
hospitalization for infectious diseases in early childhood. Environ Res 110(8):773-777.
10.1016/j.envres.2010.08.004.
Fei C, Weinberg CR, Olsen J. 2012. Commentary: Perfluorinated chemicals and time to pregnancy: A
link based on reverse causation? Epidemiology 23(2):264-266.
Feng Y, Fang X, Shi Z, et al. 2010. Effects of PFNA exposure on expression of junction-associated
molecules and secretory function in rat Sertoli cells. Reprod Toxicol 30(3):429-437.
10.1016/j.reprotox.2010.05.010.
Feng Y, Shi Z, Fang X, et al. 2009. Perfluorononanoic acid induces apoptosis involving the Fas death
receptor signaling pathway in rat testis. Toxicol Lett 190(2):224-230. 10.1016/j.toxlet.2009.07.020.
Fenton SE, Reiner JL, Nakayama SF, et al. 2009. Analysis of PFOA in dosed CD-1 mice. Part 2.
Disposition of PFOA in tissues and fluids from pregnant and lactating mice and their pups. Reprod
Toxicol 27(3-4):365-372.
Fernández Freire P, Pérez Martin JM, Herrero O, et al. 2008. In vitro assessment of the cytotoxic and
mutagenic potential of perfluorooctanoic acid. Toxicol in Vitro 22:1228-1233.
Filgo AJ, Quist EM, Hoenerhoff MJ, et al. 2015a. Perfluorooctanoic Acid (PFOA)-induced liver lesions
in two strains of mice following developmental exposures: PPARalpha is not required. Toxicol
Pathol 43(4):558-568. 10.1177/0192623314558463.
Filgo AJ, Quist EM, Hoenerhoff MJ, et al. 2015b. Supplemental data to perfluorooctanoic acid (PFOA)induced liver lesions in two strains of mice following developmental exposures: PPARalpha is not
required. [Toxicol Pathol 43(4):558-568]. Toxicol Pathol 10.1177/0192623314558463.
Fisher M, Arbuckle TE, Wade M, et al. 2013. Do perfluoroalkyl substances affect metabolic function
and plasma lipids? Analysis of the 2007-2009, Canadian Health Measures Survey (CHMS) Cycle 1.
Environ Res 121:95-103.
Fisher M, Arbuckle TE, Liang CL, et al. 2016. Concentrations of persistent organic pollutants in
maternal and cord blood from the maternal-infant research on environmental chemicals (MIREC)
cohort study. Environ Health 15(1):59.
Fitz-Simon N, Fletcher T, Luster MI, et al. 2013. Reductions in serum lipids with a 4-year decline in
serum perfluorooctanoic acid and perfluorooctanesulfonic acid. Epidemiology 24(4):569-576.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

662
8. REFERENCES

Florentin A, Deblonde T, Diguio N, et al. 2011. Impacts of two perfluorinated compounds (PFOS and
PFOA) on human hepatoma cells: Cytotoxicity but no genotoxicity? Int J Hyg Environ Health
214(6):493-499.
Food Standards Agency. 2006. FSIS 11/06. Fluorinated chemicals: UK dietary intakes. Food
Standards Agency. http://www.food.gov.uk/multimedia/pdfs/fsis1106.pdf. June 28, 2008.
Foreman JE, Chang SC, Ehresman DJ, et al. 2009. Differential hepatic effects of perfluorobutyrate
mediated by mouse and human PPAR-α. Toxicol Sci 110(1):204-211. 10.1093/toxsci/kfp077.
Forns J, Iszatt N, White RA, et al. 2015. Perfluoroalkyl substances measured in breast milk and child
neuropsychological development in a Norwegian birth cohort study. Environ Int 83:176-182.
Franko J, Meade BJ, Frasch HF, et al. 2012. Dermal penetration potential of perfluorooctanoic acid
(PFOA) in human and mouse skin. J Toxicol Environ Health A 75(1):50-62.
Frisbee SJ, Brooks AP, Jr., Maher A, et al. 2009. The C8 health project: Design, methods, and
participants. Environ Health Perspect 117(12):1873-1882.
Frisbee SJ, Shankar A, Knox SS, et al. 2010. Perfluorooctanoic acid, perfluorooctanesulfonate, and
serum lipids in children and adolescents: Results from the C8 Health Project. Arch Pediatr Adolesc
Med 164(9):860-869.
Fromme H, Midasch O, Twardella D, et al. 2007a. Occurrence of perfluorinated substances in an adult
German population in southern Bavaria. Int Arch Occup Environ Health 80:313-319.
Fromme H, Mosch C, Morovitz M, et al. 2010. Pre- and postnatal exposure to perfluorinated compounds
(PFCs). Environ Sci Technol 44(18):7123-7129.
Fromme H, Schlummer M, Moller A, et al. 2007b. Exposure of an adult population to perfluorinated
substances using duplicate diet portions and biomonitoring data. Environ Sci Technol 41:79287933.
Fromme H, Tittlemier SA, Volkel W, et al. 2009. Perfluorinated compounds-exposure assessment for
the general population in western countries. Int J Hyg Environ Health 212(3):239-270.
Fu J, Gao Y, Wang T, et al. 2015. Elevated levels of perfluoroalkyl acids in family members of
occupationally exposed workers: The importance of dust transfer. Scientific reports 5:9313.
10.1038/srep09313.
Fu Y, Wang T, Fu Q, et al. 2014a. Associations between serum concentrations of perfluoroalkyl acids
and serum lipid levels in a Chinese population. Ecotoxicol Environ Saf 106:246-252.
10.1016/j.ecoenv.2014.04.039.
Fu Y, Wang T, Wang P, et al. 2014b. Effects of age, gender and region on serum concentrations of
perfluorinated compounds in general population of Henan, China. Chemosphere 110:104-110.
10.1016/j.chemosphere.2014.02.020.
+Fuentes S, Colomina MT, Rodriguez J, et al. 2006. Interactions in developmental toxicology:
Concurrent exposure to perfluorooctane sulfonate (PFOS) and stress in pregnant mice. Toxicol Lett
164(1):81-89.
Fuentes S, Colomina MT, Vicens P, et al. 2007a. Influence of maternal restraint stress on the longlasting effects induced by prenatal exposure to perfluorooctane sulfonate (PFOS) in mice. Toxicol
Lett 171:162-170.
Fuentes S, Colomina MT, Vicens P, et al. 2007b. Concurrent exposure to perfluoroactane sulfonate and
restraint stress during pregnancy in mice: Effects on postnatal development and behavior of the
offspring. Toxicol Sci 98:589-598.
+Fuentes S, Vicens P, Colomina MT, et al. 2007c. Behavioral effects in adult mice exposed to
perfluorooctane sulfonate (PFOS). Toxicology 242:123-129.
Fujii Y, Harada KH, Koizumi A. 2013. Occurrence of perfluorinated carboxylic acids (PFCAs) in
personal care products and compounding agents. Chemosphere 93(3):538-544.
Fujii Y, Niisoe T, Harada KH, et al. 2015a. Toxicokinetics of perfluoroalkyl carboxylic acids with
different carbon chain lengths in mice and humans. J Occup Health 57(1):1-12. 10.1539/joh.140136-OA.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

663
8. REFERENCES

Fujii Y, Niisoe T, Harada KH, et al. 2015b. Supplemental materials: Toxicokinetics of perfluoroalkyl
carboxylic acids with different carbon chain lengths in mice and humans. J Occup Health 57(1):112.
Fujii Y, Yan J, Harada KH, et al. 2012. Levels and profiles of long-chain perfluorinated carboxylic acids
in human breast milk and infant formulas in East Asia. Chemosphere 86(3):315-321.
Furdui F, Stock N, Ellis DA, et al. 2007. Spatial distribution of perfluoroalkyl contaminants in lake trout
from the Great Lakes. Environ Sci Technol 41:1554-1559.
Gallo V, Leonardi G, Brayne C, et al. 2013. Serum perfluoroalkyl acids concentrations and memory
impairment in a large cross-sectional study. BMJ Open 3(6):e002414.
Gallo V, Leonardi G, Genser B, et al. 2012. Serum perfluorooctanoate (PFOA) and perfluorooctane
sulfonate (PFOS) concentrations and liver function biomarkers in a population with elevated PFOA
exposure. Environ Health Perspect 120(5):655-660.
Gao B, He X, Liu W, et al. 2015. Distribution of perfluoroalkyl compounds in rats: Indication for using
hair as bioindicator of exposure. J Expo Sci Environ Epidemiol 25(6):632-638.
Gauthier SA, Mabury SA. 2005. Aqueous photolysis of 8:2 fluorotelomer alcohol. Environ Toxicol
Chem 24(8):1837-1846.
Geiger SD, Xiao J, Shankar A. 2014a. No association between perfluoroalkyl chemicals and
hypertension in children. Integrated blood pressure control 7:1-7. 10.2147/ibpc.s47660.
Geiger SD, Xiao J, Ducatman A, et al. 2014b. The association between PFOA, PFOS and serum lipid
levels in adolescents. Chemosphere 98:78-83. 10.1016/j.chemosphere.2013.10.005.
Geiger SD, Xiao J, Shankar A. 2013. Positive association between perfluoroalkyl chemicals and
hyperuricemia in children. Am J Epidemiol 177(11):1255-1262.
Gellrich V, Brunn H, Stahl T. 2013. Perfluoroalkyl and polyfluoroalkyl substances (PFASs) in mineral
water and tap water. J Environ Sci Health A Tox Hazard Subst Environ Eng 48(2):129-135.
10.1080/10934529.2013.719431.
Genuis SJ, Beesoon S, Birkholz D. 2013. Biomonitoring and elimination of perfluorinated compounds
and polychlorinated biphenyls through perspiration: Blood, urine, and sweat study. ISRN
toxicology 2013:483832.
George ME, Andersen ME. 1986. Toxic effects of nonadecafluoro-n-decanoic acid in rats. Toxicol Appl
Pharmacol 85:169-180.
Gewurtz SB, Bhavsar SP, Crozier PW, et al. 2009. Perfluoroalkyl contaminants in window film:
Indoor/outdoor, urban/rural, and winter/summer contamination and assessment of carpet as a
possible source. Environ Sci Technol 43(19):7317-7323.
Gewurtz SB, Bhavsar SP, Petro S, et al. 2014. High levels of perfluoroalkyl acids in sport fish species
downstream of a firefighting training facility at Hamilton International Airport, Ontario, Canada.
Environ Int 67:1-11. 10.1016/j.envint.2014.02.005.
Giesy JP, Kannan K. 2001. Global distribution of perfluorooctane sulfonate in wildlife. Environ Sci
Technol 35(7):1339-1342.
Gilliland FD, Mandel JS. 1993. Mortality among employees of a perfluorooctanoic acid production
plant. J Occup Med 35(9):950-954.
Gilliland FD, Mandel JS. 1996. Serum perfluorooctanoic acid and hepatic enzymes, lipoproteins, and
cholesterol: A study of occupationally exposed men. Am J Ind Med 29(5):560-568.
Gleason JA, Post GB, Fagliano JA. 2015. Associations of perfluorinated chemical serum concentrations
and biomarkers of liver function and uric acid in the US population (NHANES), 2007-2010.
Environ Res 136:8-14. 10.1016/j.envres.2014.10.004.
Glynn A, Berger U, Bignert A, et al. 2012. Perfluorinated alkyl acids in blood serum from primiparous
women in Sweden: Serial sampling during pregnancy and nursing, and temporal trends 1996-2010.
Environ Sci Technol 46:9071-9079.
Goecke CM, Jarnot BM, Reo NV. 1992. A comparative toxicological investigation of
perfluorocarboxylic acids in rats by fluorine-19 NMR spectroscopy. Chem Res Toxicol 5(4):512519.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

664
8. REFERENCES

Goecke-Flora CM, Reo NV. 1996. Influence of carbon chain length on the hepatic effects of
perfluorinated fatty acids. A 19F- and 31P-NMR investigation. Chem Res Toxicol 9(4):689-695.
+Gortner EG, Lamprecht EG, Case MT. 1982. Oral teratology study of T-3141CoC in rabbits. St. Paul,
MN: Riker Laboratories, Inc.
Gotoh Y, Kato Y, Stieger B, et al. 2002. Gender difference in the Oatp1-mediated tubular reabsorption
of estradiol 17β-D-glucuronide in rats. Am J Physiol Endocrinol Metab 282:E1245-E1254.
Goudarzi H, Nakajima S, Ikeno T, et al. 2016. Prenatal exposure to perfluorinated chemicals and
neurodevelopment in early infancy: The Hokkaido Study. Sci Total Environ 541:1002-1010.
10.1016/j.scitotenv.2015.10.017.
Govarts E, Remy S, Bruckers L, et al. 2016. Combined effects of prenatal exposures to environmental
chemicals on birth weight. Int J Environ Res Public Health 13(5) 10.3390/ijerph13050495.
Grandjean P, Budtz-Jorgensen E. 2013. Immunotoxicity of perfluorinated alkylates: Calculation of
benchmark doses based on serum concentrations in children. Environ Health 12(1):35.
Grandjean P, Andersen EW, Budtz-Jorgensen E, et al. 2012. Serum vaccine antibody concentrations in
children exposed to perfluorinated compounds. J Am Med Assoc 307(4):391-397.
Grandjean P, Heilmann C, Weihe P, et al. 2017. Serum vaccine antibody concentrations in adolescents
exposed to perfluorinated compounds. Environ Health Perspect. 125(7):077018. 10.1289/EHP275.
Granum B, Haug LS, Namork E, et al. 2013. Pre-natal exposure to perfluoroalkyl substances may be
associated with altered vaccine antibody levels and immune-related health outcomes in early
childhood. J Immunotoxicol 10(4):373-379.
+Grasty RC, Bjork JA, Wallace KB, et al. 2005. Effects of prenatal perfluorooctane sulfonate (PFOS)
exposure on lung maturation in the perinatal rat. Birth Defects Res B Dev Reprod Toxicol 74:405416.
+Grasty RC, Wolf DC, Grey BE, et al. 2003. Prenatal window of susceptibility to perfluorooctane
sulfonate-induced neonatal mortality in the Sprague-Dawley rat (corrigendum in Birth Defects Res B
77(1):86). Birth Defects Res B 68(6):465-471.
Grice MM, Alexander BH, Hoffbeck R, et al. 2007. Self-reported medical conditions in
perfluorooctanesulfonyl fluoride manufacturing workers. J Occup Environ Med 49(7):722-729.
+Griffith FD, Long JE. 1980. Animal toxicity studies with ammonium perfluorooctanoate. Am Ind Hyg
Assoc J 41(8):576-583.
Gump BB, Wu Q, Dumas AK, et al. 2011. Perfluorochemical (PFC) exposure in children: Associations
with impaired response inhibition. Environ Sci Technol 45(19):8151-8159.
Guo R, Sim WJ, Lee ES, et al. 2010. Evaluation of the fate of perfluoroalkyl compounds in wastewater
treatment plants. Water Res 44(11):3476-3486. 10.1016/j.watres.2010.03.028.
+Guruge KS, Hikono H, Shimada N, et al. 2009. Effect of perflluorooctane sulfonate (PFOS) on
influenza A virus-induced mortality in female B6C3F1 mice. J Toxicol Sci 34(6):687-691.
Guruge KS, Yeung LW, Yamanaka N, et al. 2006. Gene expression profiles in rat liver treated with
perfluorooctanoic acid (PFOA). Toxicol Sci 89(1):93-107.
Gützkow KB, Haug LS, Thomsen C, et al. 2012. Placental transfer of perfluorinated compounds is
selective. A Norwegian mother and child sub-cohort study. Int J Hyg Environ Health 215(2):216219.
+Hadrup N, Pedersen M, Skov K, et al. 2016. Perfluorononanoic acid in combination with 14 chemicals
exerts low-dose mixture effects in rats. Arch Toxicol 90(3):661-675. 10.1007/s00204-015-1452-6.
HAES. 2017. GEF-Reduction and phase-out of PFOS in priority sectors in China. Environmenstal &
Social Management Framework. Hubei Academy of Environmental Sciences, Foreign Economic
Cooperation Office, Ministry of Environmental Protection (FECO).
http://documents.worldbank.org/curated/en/2752514607008/Environmental-and-social-managementframework. February 28, 2017.
Hall AP, Elcombe CR, Foster JR, et al. 2012. Liver hypertrophy: A review of adaptive (adverse and
non-adverse) changes- conclusions from the 3rd International ESTP Expert Workshop. Toxicol
Pathol 40:971-994.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

665
8. REFERENCES

Halldorsson TI, Rytter D, Haug LS, et al. 2012. Prenatal exposure to perfluorooctanoate and risk of
overweight at 20 years of age: A prospective cohort study. Environ Health Perspect 120(5):668673.
+Hallgren S, Fredriksson A, Viberg H. 2015. More signs of neurotoxicity of surfactants and flame
retardants - neonatal PFOS and PBDE 99 cause transcriptional alterations in cholinergic genes in the
mouse CNS. Environ Toxicol Pharmacol 40(2):409-416. 10.1016/j.etap.2015.06.014.
Hamm MP, Cherry NM, Chan E, et al. 2010. Maternal exposure to perfluorinated acids and fetal growth.
J Expo Sci Environ Epidemiol 20(7):589-597.
Hamm MP, Martin JW, Chan E, et al. 2009. Maternal exposure to perfluorinated acids and fetal growth
restriction. Am J Epidemiol 169(Suppl. 11):S43.
Han X, Hinderliter PM, Snow TA, et al. 2004. Binding of perfluorooctanoic acid to rat liver-form and
kidney-form α2u-globulins. Drug Chem Toxicol 27(4):341-360.
Han X, Kemper RA, Jepson GW. 2005. Subcellular distribution and protein binding of
perfluorooctanoic acid in rat liver and kidney. Drug Chem Toxicol 28(2):197-209.
Han X, Snow TA, Kemper RA, et al. 2003. Binding of perfluorooctanoic acid to rat and human plasma
proteins. Chem Res Toxicol 16:775-781.
Hanhijarvi H, Ophaug RH, Singer L. 1982. The sex-related difference in perfluorooctanoate excretion in
the rat. Proc Soc Exp Biol Med 171:50-55.
Hanhijarvi H, Ylinen M, Kojo A, et al. 1987. Elimination and toxicity of perfluorooctanoic acid during
subchronic administration in the Wistar rat. Pharmacol Toxicol 61(1):66-68.
Hansen KJ, Johnson HO, Eldridge JS, et al. 2002. Quantitative characterization of trace levels of PFOS
and PFOA in the Tennessee River. Environ Sci Technol 36:1681-1685.
Hanssen L, Dudarev AA, Huber S, et al. 2013. Partition of perfluoroalkyl substances (PFASs) in whole
blood and plasma, assessed in maternal and umbilical cord samples from inhabitants of arctic Russia
and Uzbekistan. Sci Total Environ 447:430-437.
Hanssen L, Rollin H, Odland JO, et al. 2010. Perfluorinated compounds in maternal serum and cord
blood from selected areas of South Africa: Results of a pilot study. J Environ Monit 12(6):13551361.
Harada K, Inoue K, Morikawa A, et al. 2005a. Renal clearance of perfluorooctane sulfonate and
perfluorooctanoate in humans and their species-specific excretion. Environ Res 99:253-261.
Harada K, Nakanishi S, Saito N, et al. 2005b. Airborne perfluorooctanoate may be a substantial source
contamination in Kyoto area, Japan. Bull Environ Contam Toxicol 74:64-69.
Harada K, Nakanishi S, Sasaki K, et al. 2006. Particle size distribution and respiratory deposition
estimates of airborne perfluorooctanoate and perfluorooctanesulfonate in Kyoto area, Japan. Bull
Environ Contam Toxicol 76(2):306-310.
Harada KH, Koizumi A. 2009. Environmental and biological monitoring of persistent fluorinated
compounds in Japan and their toxicities. Environ Health Prev Med 14(1):7-19.
Harada KH, Hashida S, Kaneko T, et al. 2007. Biliary excretion and cerebrospinal fluid partition of
perfluorooctanoate and perfluorooctane sulfonate in humans. Environ Toxicol Pharmacol
24(2):134-139.
Hardell E, Karrman A, van Bavel B, et al. 2014. Case-control study on perfluorinated alkyl acids
(PFAAs) and the risk of prostate cancer. Environ Int 63:35-39.
Hardisty JF, Willson GA, Brown WR, et al. 2010. Pathology Working Group review and evaluation of
proliferative lesions of mammary gland tissues in female rats fed ammonium perfluorooctanoate
(APFO) in the diet for 2 years. Drug Chem Toxicol 33(2):131-137.
Harris LA, Barton HA. 2008. Comparing single and repeated dosimetry data for perfluorooctane
sulfonate in rats. Toxicol Lett 181(3):148-156.
+Harris MW, Birnbaum LS. 1989. Developmental toxicity of perfluorodecanoic acid in C57BL/6N
mice. Fundam Appl Toxicol 12:442-448.
+Harris MW, Uraih LC, Birnbaum LS. 1989. Acute toxicity of perfluorodecanoic acid in C57BL/6 mice
differs from 2,3,7,8-tetrachlorodibenzo-p-dioxin. Fundam Appl Toxicol 13:723-726.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

666
8. REFERENCES

+Haughom B, Spydevold O. 1992. The mechanism underlying the hypolipemic effect of
perfluorooctanoic acid (PFOA), perfluorooctane sulphonic acid (PFOSA) and clofibric acid.
Biochim Biophys Acta 1128(1):65-72.
Heilmann C, Budtz-Jorgensen E, Nielsen F, et al. 2010. Serum concentrations of antibodies against
vaccine toxoids in children exposed perinatally to immunotoxicants. Environ Health Perspect
118(10):1434-1438. 10.1289/ehp.1001975.
Hekster FM, Laane RW, de Voogt P. 2003. Environmental and toxicity effects of perfluoroalkylated
substances. Rev Environ Contam Toxicol 179:99-121.
Henderson WM, Smith MA. 2007. Perfluorooctanoic acid and perfluorononanoic acid in fetal and
neonatal mice following in utero exposure to 8-2 fluorotelomer alcohol. Toxicol Sci 95(2):452-461.
Hinderliter PM, DeLorme MP, Kennedy GL. 2006a. Perfluorooctanoic acid: Relationship between
repeated inhalation exposures and plasma PFOA concentration in the rat. Toxicology 222:80-85.
Hinderliter PM, Han X, Kennedy GL, et al. 2006b. Age effect on perfluorooctanoate (PFOA) plasma
concentration in post-weaning rats following oral gavage with ammonium perfluorooctanoate
(APFO). Toxicology 225:195-203.
Hinderliter PM, Mylchreest E, Gannon SA, et al. 2005. Perfluorooctanoate: Placental and lactational
transport pharmacokinetics in rats. Toxicology 211:139-148.
+Hines EP, White SS, Stanko JP, et al. 2009. Phenotypic dichotomy following developmental exposure
to perfluorooctanoic acid (PFOA) in female CD-1 mice: Low doses induce elevated serum leptin
and insulin, and overweight in mid-life. Mol Cell Endocrinol 304(1-2):97-105.
+Hoberman AM, York RG. 2003. Oral (gavage) combined repeated dose toxicity study of T-7706 with
the reproduction/developmental toxicity screening test. Argus Research.
Hoffman K, Webster TF, Bartell SM, et al. 2011. Private drinking water wells as a source of exposure to
perfluorooctanoic acid (PFOA) in communities surrounding a fluoropolymer production facility.
Environ Health Perspect 119(1):92-97.
Hoffman K, Webster TF, Weisskopf MG, et al. 2010. Exposure to polyfluoroalkyl chemicals and
attention deficit/hyperactivity disorder in U.S. children 12-15 years of age. Environ Health Perspect
118(12):1762-1767.
Hölzer J, Midasch O, Rauchfuss K, et al. 2008. Biomonitoring of perflluorinated compounds in children
and adults exposed to perfluorooctanoate-contaminated drinking water. Environ Health Perspect
116(5):651-657.
Houde M, Bujas TD, Small J, et al. 2006a. Biomagnification of perfluoroalkyl compounds in the
bottlenose dolphin (Tursiops truncatus) food web. Environ Sci Technol 40:4138-4144.
Houde M, Martin JW, Letcher RJ, et al. 2006b. Biological monitoring of polyfluoroalkyl substances: A
Review. Environ Sci Technol 40(11):3463-3473.
Houde M, Wells RS, Fair PA, et al. 2005. Polyfluoroalkyl compounds in free-ranging bottlenose
dolphins (Tursiops truncatus) from the Gulf of Mexico and the Atlantic Ocean. Environ Sci Technol
39:6591-6598.
Høyer BB, Ramlau-Hansen CH, Obel C, et al. 2015a. Pregnancy serum concentrations of perfluorinated
alkyl substances and offspring behaviour and motor development at age 5-9 years--a prospective
study. Environ Health 14:2. 10.1186/1476-069x-14-2.
Høyer BB, Ramlau-Hansen CH, Vrijheid M, et al. 2015b. Anthropometry in 5- to 9-Year-Old
Greenlandic and Ukrainian Children in Relation to Prenatal Exposure to Perfluorinated Alkyl
Substances. Environ Health Perspect 123(8):841-846. 10.1289/ehp.1408881.
Hsu V, de LTVM, Zhao P, et al. 2014. Towards quantitation of the effects of renal impairment and
probenecid inhibition on kidney uptake and efflux transporters, using physiologically based
pharmacokinetic modelling and simulations. Clin Pharmacokinet 53(3):283-293. 10.1007/s40262013-0117-y.
+Hu Q, Strynar MJ, DeWitt JC. 2010. Are developmentally exposed C57BL/6 mice insensitive to
suppression of TDAR by PFOA? J Immunotoxicol 7(4):344-349.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

667
8. REFERENCES

Hu W, Jones PD, Upham BL, et al. 2002. Inhibition of gap junctional intercellular communication by
perfluorinated compounds in rat liver and dolphin kidney epithelial cell lines in vitro and SpragueDawley rats in vivo. Toxicol Sci 68(2):429-426.
Hu XC, Andrews DQ, Lindstrom M, et al. 2016. Detection of poly-and perfluoroalkyl substances
(PFASs) in U.S. drinking water linked to industrial sites, military fire training areas and wastewater
treatment plants. Environ Sci Technol Lett 3:344-350. DOI: 10.1021/acs.estlett.6b00260.
Humblet O, Dias-Ramirez LG, Balmes JR, et al. 2014. Perfluoroalkyl chemicals and asthma among
children 12-19 years of age: NHANES (1999-2008). Environ Health Perspect 122(10):1129-1133.
Hundley SG, Sarrif AM, Kennedy GL. 2006. Absorption, distribution, and excretion of ammonium
perfluorooctanoate (APFO) after oral administration to various species. Drug Chem Toxicol
29(2):137-145.
Hurley MD, Andersen MPS, Wallington TJ, et al. 2004. Atmospheric chemistry of perfluorinated
carboxylic acids: Reaction with OH radicals and atmospheric lifetimes. J Phys Chem 108:615-620.
IARC. 2017. Agents classified by the IARC Monographs, Volumes 1–117. Lyon, France: International
Agency for Research on Cancer.
http://monographs.iarc.fr/ENG/Classification/List_of_Classifications.pdf. February 28, 2017.
ICRP. 1981. Report of the task group on reference man. ICRP Publication 28. International
Commission on Radiological Protection. Oxford: Pergamon Press, 32-40.
+Ikeda T, Aiba K, Fukuda K, et al. 1985. The induction of peroxisome proliferation in rat liver by
perfluorinated fatty acids, metabolically inert derivatives of fatty acids. J Biochem (Tokyo)98:475482.
Innes KE, Ducatman AM, Luster MI, et al. 2011. Association of osteoarthritis with serum levels of the
environmental contaminants perfluorooctanoate and perfluorooctane sulfonate in a large
Appalachian population. Am J Epidemiol 174(4):440-450.
Innes KE, Wimsatt JH, Frisbee S, et al. 2014. Inverse association of colorectal cancer prevalence to
serum levels of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) in a large
Appalachian population. BMC Cancer[electronic resource].
http://www.bilomedcentral.com/1471/14/45. December 10, 2011.
Inoue K, Okada F, Ito R, et al. 2004. Perfluorooctane sulfonate (PFOS) and related perfluorinated
compounds in human maternal and cord blood samples: Assessment of PFOS exposure in a
susceptible population during pregnancy. Environ Health Perspect 112:1204-1207.
IRIS. 2017. IRIS quick list. Integrated Risk Information System. Washington, DC: U.S.
Environmental Protection Agency.
https://cfpub.epa.gov/ncea/iris_drafts/simple_list.cfm?list_type=alpha. February 28, 2017.
Issemann I, Green S. 1990. Activation of a member of a steroid hormone receptor superfamily by
peroxisome proliferators. Nature 347:645-650.
Itoh S, Araki A, Mitsui T, et al. 2016. Association of perfluoroalkyl substances exposure in utero with
reproductive hormone levels in cord blood in the Hokkaido Study on Environment and Children's
Health. Environ Int 94:51-59. 10.1016/j.envint.2016.05.011.
+Iwai H, Hoberman AM. 2014. Oral (gavage) combined developmenal and perinatal/posnatal
reproduction toxicity study of ammonium salt of perfluoroinated hexanoic acid in mice. Int J
Toxicol 33(3):219-237.
+Iwai H, Yamashita K. 2006. A fourteen-day repeated dose oral toxicity study of APFO in rats. Drug
Chem Toxicol 29:323-332.
Jacquet N, Maire MA, Landkocz Y, et al. 2012. Carcinogenic potency of perfluorooctane sulfonate
(PFOS) on Syrian hamster embryo (SHE) cells. Arch Toxicol 86(2):305-314.
Jahnke A, Berger U, Ebinghaus R, et al. 2007a. Latitudinal gradient of airborne polyfluorinated alkyl
substances in the marine atmosphere between Germany and South Africa (53° N-33° S). Environ
Sci Technol 41(9):3055-3061.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

668
8. REFERENCES

Jahnke A, Huber S, Temme C, et al. 2007b. Development and application of a simplified sampling
method for volatile polyfluorinated alkyl substances in indoor and environmental air. J Chromatogr
A 1164:1-9.
Jain RB. 2013. Association between thyroid profile and perfluoroalkyl acids: Data from NHNAES
2007-2008. Environ Res 126:51-59. 10.1016/j.envres.2013.08.006.
Jain RB. 2015. Estimation of the total concentration of perfluoroalkyl acids (PFAA) in human serum:
Data from NHANES 2005-2012. Chemosphere 134:387-394. 10.1016/j.chemosphere.2015.04.104.
Jandacek RJ, Rider T, Keller ER, et al. 2010. The effect of olestra on the absorption, excretion and
storage of 2,2',5,5' tetrachlorobiphenyl; 3,3',4,4' tetrachlorobiphenyl; and perfluorooctanoic acid.
Environ Int 36(8):880-883.
Jensen MS, Norgaard-Pedersen B, Toft G, et al. 2012. Phthalates and perfluorooctanesulfonic acid in
human amniotic fluid: Temporal trends and timing of amniocentesis in pregnancy. Environ Health
Perspect 120(6):897-903.
Jensen TK, Andersen LB, Kyhl HB, et al. 2015. Association between perfluorinated compound exposure
and miscarriage in Danish pregnant women. PLoS ONE 10(4):e0123496.
10.1371/journal.pone.0123496.
Ji K, Kim S, Kho Y, et al. 2012. Serum concentrations of major perfluorinated compounds among the
general population in Korea: Dietary sources and potential impact on thyroid hormones. Environ
Int 45:78-85.
Jiang Q, Lust RM, Dewitt JC. 2013. Perfluorooctanoic acid induced-developmental cardiotoxicity: Are
peroxisome poliferator activated receptor α (PPARα) and bone morphorgenic protein 2 (BMP2)
pathways involved? J Toxicol Environ Health A 76(11):635-650. 10.1080/15287394.2013.789415.
Jiang Q, Lust RM, Strynar MJ, et al. 2012. Perflurooctanoic acid induces developmental cardiotoxicity
in chicken embryos and hatchlings. Toxicology 293(1-3):97-106. 10.1016/j.tox.2012.01.005.
Jiang Q, Ma W, Wu J, et al. 2016. Perfluorooctanoic acid-induced toxicity in primary cultures of
chicken embryo cardiomyocytes. Environ Toxicol 31(11):1580-1590. 10.1002/tox.22162.
http://dx.doi.org/10.1002/tox.22162.
Jin C, Sun Y, Islam A, et al. 2011. Perfluoroalkyl acids including perfluorooctane sulfonate and
perfluorohexane sulfonate in firefighters. J Occup Environ Med 53(3):324-328.
10.1097/JOM.0b013e31820d1314.
Jin YH, Liu W, Sato I, et al. 2009. PFOS and PFOA in environmental and tap water in China.
Chemosphere 77(5):605-611.
Joensen UN, Bossi R, Leffers H, et al. 2009. Do perfluoroalkyl compounds impair human semen
quality? Environ Health Perspect 117(6):923-927.
Joensen UN, Veyrand B, Antignac JP, et al. 2013. PFOS (perfluorooctanesulfonate) in serum is
negatively associated with testosterone levels, but not with semen quality, in healthy men. Hum
Reprod 28(3):599-608.
Jogsten IE, Perello G, Llebaria X, et al. 2009. Exposure to perfluorinated compounds in Catalonia,
Spain, through consumption of various raw and cooked foodstuffs, including packaged food. Food
Chem Toxicol 47:1577-1583.
Johansson JH, Berger U, Vestergren R, et al. 2014. Temporal trends (1999-2010) of perfluoroalkyl acids
in commonly consumed food items. Environ Pollut 188:102-108. 10.1016/j.envpol.2014.01.026.
Johansson N, Eriksson P, Viberg H. 2009. Neonatal exposure to PFOS and PFOA in mice results in
changes in proteins which are important for neuronal growth and synaptogenesis in the developing
brain. Toxicol Sci 108(2):412-418.
+Johansson N, Fredriksson A, Eriksson P. 2008. Neonatal exposure to perfluorooctane sulfonate (PFOS)
and perfluorooctanoic acid (PFOA) causes neurobehavioural defects in adult mice. Neurotoxicology
29:160-169.
Johnson JD. 1995a. Final report. Analytical study, single-dose dermal absorption/toxicity study of
T6049 in rabbits. In vivo reference number: HWI#6329-130. 3M. SCD Division.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

669
8. REFERENCES

Johnson JD. 1995b. Final report. Analytical study, single-dose dermal absorption/toxicity study of
T6053 in rabbits (lithium perfluorooctane sulfonate). In vivo study reference number: HWI#6329137. 3M. SCD Division.
Johnson JD, Ober RE. 1979. Absorption of FC-95-14C in rats after a single oral dose. 3M. Submitted to
the U.S. Environmental Protection Agency's Administrative Record. AR226-0007.
Johnson JD, Ober RE. 1980. Extent and route of excretion and tissue distribution of total carbon-14 in
rats after a single intravenous dose of FC-95-14C. 3M. Submitted to the U.S. Environmental
Protection Agency's Administrative Record. AR226-0006.
Johnson JD, Ober RE. 1999a. Absorption of FC-143-14C in rats after a single oral dose. In: Exploratory
28-day oral toxicity study with telomer alcohol, telomer acrylate, PFHS, and PFOS (POS control) by
daily gavage in the rat, w/CVR LTR DTD, 051500 (Sanitized). 3M. Submitted to the U.S.
Environmental Protection Agency under TSCA Section FYI. OTS05001378S.
Johnson JD, Ober RE. 1999b. Extent and route of excretion and tissue distribution of total carbon-14 in
male and female rats after a single IV dose of FC-143-14C. In: Exploratory 28-day oral toxicity
study with telomer alcohol, telomer acrylate, PFHS, and PFOS (POS control) by daily gavage in the
rat, w/CVR letter dated, 051500 (Sanitized). 3M. Submitted to the U.S. Environmental Protection
Agency under TSCA Section FYI. OTS05001378S.
Johnson JD, Gibson SJ, Ober RE. 1984. Cholestyramine-enhanced fecal elimination of carbon-14 in rats
after administration of ammonium [14C]perfluorooctanoate or potassium [14C]perflorooctanesulfonate. Fundam Appl Toxicol 4:972-976.
Jones CM. 2016. 2010/1015 PFOA Stewardship Program. Docket ID number EPA-HQ-OPPT-20060621. https://www.regulations.gov/document?D=EPA-HQ-OPPT-2006-0621-0115. February 28,
2017.
Jørgensen KT, Specht IO, Lenters V, et al. 2014a. Perfluoroalkyl substances and time to pregnancy in
couples from Greenland, Poland and Ukraine. Environ Health 13:116. 10.1186/1476-069x-13-116.
Jørgensen KT, Specht IO, Lenters V, et al. 2014b. Supplemental files to "Perfluoroalkyl substances and
time to pregnancy in couples from Greenland, Poland and Ukraine" [Environ Health 13:116.
10.1186/1476-069x-13-116]. Environ Health 10.1186/1476-069x-13-116.
Jurado-Sanchez B, Ballesteros E, Gallego M. 2014. Analytical method for biomonitoring of
perfluoroalkyl acids in human urine. Talanta 128:141-146. 10.1016/j.talanta.2014.04.071.
Kaiser MA, Larsen BS, Kao CPC, et al. 2005. Vapor pressures of perfluorooctanoic, nonanoic, -decanoic, -undecanoic, and -dodecanoic acids. J Chem & Eng Data 50:1841-1843.
Kang H, Choi K, Lee HS, et al. 2016. Elevated levels of short carbon-chain PFCAs in breast milk among
Korean women: Current status and potential challenges. Environ Res 148:351-359.
Kannan K, Choi JW, Iseki Net al. 2002a. Concentrations of perfluorinated acids in livers of birds from
Japan and Korea. Chemosphere 49(3):225-231.
Kannan K, Corsolini S, Falandysz J, et al. 2002b. Perfluorooctanesulfonate and related fluorinated
hydrocarbons in marine mammals, fishes, and birds from coasts of the Baltic and the Mediterranean
Seas. Environ Sci Technol 36(15):3210-3216.
Kannan K, Franson JC, Bowerman WW, et al. 2001a. Perfluorooctane sulfonate in fish-eating water
birds including bald eagles and albatrosses. Environ Sci Technol 35(15):3065-3070.
Kannan K, Hansen KJ, Wade TL, et al. 2002c. Perfluorooctane sulfonate in oysters, Crassostrea
virginica, from the Gulf of Mexico and the Chesapeake Bay, USA. Arch Environ Contam Toxicol
42(3):313-318.
Kannan K, Koistinen J, Beckmen K, et al. 2001b. Accumulation of perfluorooctane sulfonate in marine
mammals. Environ Sci Technol 35:1593-1598.
Kannan K, Newsted J, Halbrook RS, et al. 2002d. Perfluorooctanesulfonate and related fluorinated
hydrocarbons in mink and river otters from the United States. Environ Sci Technol 36:2566-2571.
Kannan K, Perrotta E, Thomas NJ. 2006. Association between perfluorinated compounds and
pathological conditions in southern sea otters. Environ Sci Technol 40:4943-4948.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

670
8. REFERENCES

Kannan K, Tao L, Sinclair E, et al. 2005. Perfluorinated compounds in aquatic organisms at various
trophic levels in a Great Lakes food chain. Arch Environ Contam Toxicol 48:559-566.
Karnes C, Winquist A, Steenland K. 2014. Incidence of type II diabetes in a cohort with substantial
exposure to perfluorooctanoic acid. Environ Res 128:78-83. 10.1016/j.envres.2013.11.003.
Kärrman A, Ericson I, van Bavel B, et al. 2007. Exposure of perfluorinated chemicals through lactation:
Levels of matched human milk and serum and a temporal trend, 1996-2004, in Sweden. Environ
Health Perspect 115:226-230.
Kärrman A, van Bavel B, Jarnberg U, et al. 2005. Development of a solid-phase extraction-HPLC/single
quadrupole MS method for quantification of perfluorochemicals in whole blood. Anal Chem
77:864-870.
Katakura M, Kudo N, Tsuda T, et al. 2007. Rat organic anion transporter 3 and organic anion
transporting polypeptide 1 mediate perfluorooctanoic acid transport. J Health Sci 53:77-83.
Kataria A, Trachtman H, Malaga-Dieguez L, et al. 2015. Association between perfluoroalkyl acids and
kidney function in a cross-sectional study of adolescents. Environ Health 14:89. 10.1186/s12940015-0077-9.
Kato K, Calafat AM, Needham LL. 2009a. Polyfluoroalkyl chemicals in house dust. Environ Res
109(12):518-523.
Kato K, Calafat AM, Wong LY, et al. 2009b. Polyfluoroalkyl compounds in pooled sera from children
participating in the National Health and Nutrition Examination Survey 2001-2002. Environ Sci
Technol 43(7):2641-2647.
Kato K, Wong LY, Chen A, et al. 2014. Changes in serum concentrations of maternal poly- and
perfluoroalkyl substances over the course of pregnancy and predictors of exposure in a multiethnic
cohort of Cincinnati, Ohio pregnant women during 2003-2006. Environ Sci Technol 48(16):96009608.
Kauck EA, Diesslin AR. 1951. Some properties of perfluorocarboxylic acids. Ind Eng Chem
43(10):2332-2334.
Kawabata K, Matsuzaki H, Nukui S, et al. 2017. Perfluorododecanoic acid induces cognitive deficit in
adult rats. Toxicol Sci 157(2):421-428. 10.1093/toxsci/kfx058.
Kawamoto K, Oashi T, Oami K, et al. 2010. Perfluorooctanoic acid (PFOA) but not perfluorooctane
sulfonate (PFOS) showed DNA damage in comet assay on Paramecium caudatum. J Toxicol Sci
35(6):835-841.
+Kawamoto K, Sato I, Tsuda S, et al. 2011. Ultrasonic-induced tonic convulsion in rats after subchronic
exposure to perfluorooctane sulfonate (PFOS). J Toxicol Sci 36(1):55-62.
+Kawashima Y, Kobayashi H, Miura H, et al. 1995. Characterization of hepatic responses of rat to
administration of perfluorooctanoic and perfluorodecanoic acids at low levels. Toxicology
99(3):169-178.
+Keil DE, Mehlmann T, Butterworth L, et al. 2008. Gestational exposure to perfluorooctane sulfonate
(PFOS) suppresses immune function in B6C3F1 mice. Toxicol Sci 103(1):77-85.
Keller JM, Kannan K, Taniyasu S, et al. 2005. Perfluorinated compounds in the plasma of loggerhead
and Kemp’s ridley sea turtles from the southeastern coast of the United States. Environ Sci Technol
39:9101-9108.
Kelly J, Solem L. 2009. Identification of a major source of perfluorooctane sulfonate (PFOS) at a
wastewater treatment plant in Brainerd, Minnesota. Reprod Toxicol 27:417-428.
Kelly BC, Ikonomou MG, Blair JD, et al. 2007. Food web-specific biomagnification of persistent
organic pollutants. Science 317:236-238.
Kemper RA. 2003. Perfluorooctanoic acid: Toxicokinetics in the rat. Association of Plastics
Manufactures of Europe. Submitted to the U.S. Environmental Protection Agency's Administrative
Record. AR226-1499.
Kemper RA, Nabb DL. 2005. In vitro studies in microsomes from rat and human liver, kidney, and
intestine suggest that perfluorooctanoic acid is not a substrate for microsomal UDPglucuronosyltransferases. Drug Chem Toxicol 28(3):281-287.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

671
8. REFERENCES

+Kennedy GL. 1985. Dermal toxicity of ammonium perfluorooctanoate. Toxicol Appl Pharmacol
81(2):348-355.
+Kennedy GL. 1987. Increase in mouse liver weight following feeding of ammonium
perfluorooctanoate and related fluorochemicals. Toxicol Lett 39(2-3):295-300.
Kennedy GL, Butenhoff JL, Olsen GW, et al. 2004. The toxicology of perfluorooctanoate. Crit Rev
Toxicol 34(4):351-384.
+Kennedy GL, Hall GT, Brittelli MR, et al. 1986. Inhalation toxicity of ammonium perfluorooctanoate.
Food Chem Toxicol 24(12):1325-1329.
Kerstner-Wood C, Coward L, Gorman G. 2003. Protein binding of perfluorohexane sulfonate,
perfluorooctane sulfonate and perfluorooctanoate to plasma (human, rat, and monkey), and various
human-derived plasma protein fractions. Southern Research Institute. Submitted to the U.S.
Environmental Protection Agency's Administrative Record. AR226-1354.
Khalil N, Chen A, Lee M, et al. 2016. Association of perfluoroalkyl substances, bone mineral density,
and osteoporosis in the U.S. Population in NHANES 2009-2010. Environ Health Perspect
124(1):81-87. 10.1289/ehp.1307909.
Kielsen K, Shamin Z, Ryder LP, et al. 2016. Antibody response to booster vaccination with tetanus and
diphtheria in adults exposed to perfluorinated alkylates. J Immunotoxicol 13(2):270-273.
Kim S, Kannan K. 2007. Perfluorinated acids in air, rain, snow, surface runoff, and lakes: Relative
importance of pathways to contamination of urban lakes. Environ Sci Technol 41:8328-8334.
Kim DH, Kim UJ, Kim HY, et al. 2016a. Perfluoroalkyl substances in serum from South Korean infants
with congenital hypothyroidism and healthy infants - Its relationship with thyroid hormones.
Environ Res 147:399-404. 10.1016/j.envres.2016.02.037.
Kim SJ, Heo SH, Lee DS, et al. 2016b. Gender differences in pharmacokinetics and tissue distribution of
3 perfluoroalkyl and polyfluoroalkyl substances in rats. Food Chem Toxicol 97:243-255.
Kim SK, Lee KT, Kang CS, et al. 2011. Distribution of perfluorochemicals between sera and milk from
the same mothers and implications for prenatal and postnatal exposures. Environ Pollut 159(1):169174.
+Kinney LA, Chromey NC, Kennedy Jr GL. 1989. Acute inhalation toxicity of ammonium
perfluorononanoate. Food Chem Toxicol 21(1):46-68.
Kishi R, Nakajima T, Goudarzi H, et al. 2015. The association of prenatal exposure to perfluorinated
chemicals with maternal essential and long-chain polyunsaturated fatty acids during pregnancy and
the birth weight of their offspring: The Hokkaido Study. Environ Health Perspect 123(10):10381045. 10.1289/ehp.1408834.
Kissa E. 2001. Fluorinated surfactants and repellents. 2nd ed. Revised and expanded. New York, NY:
Marcel Dekker, Inc., 1-101, 198-269, 349-379, 451-487.
Kjeldsen LS, Bonefeld-Jørgensen EC. 2013. Perfluorinated compounds affect the function of sex
hormone receptors. Environ Sci Pollut Res Int 20(11):8031-8044.
Klaunig JE, Babich MA, Baetcke KP, et al. 2003. PPARα agonist-induced rodent tumors: Modes of
action and human relevance. Crit Rev Toxicol 33(6):655-780.
Klaunig JE, Hocevar BA, Kamendulis LM. 2012. Mode of action analysis of perfluorooctanoic acid
(PFOA) tumorigenicity and human relevance. Reprod Toxicol 33(4):410-418.
+Klaunig JE, Shinohara M, Iwai H, et al. 2015. Evaluation of the chronic toxicity and carcinogenicity of
perfluorohexanoic acid (PFHxA) in Sprague-Dawley rats. Toxicol Pathol 43(2):209-220.
10.1177/0192623314530532.
Knox SS, Jackson T, Frisbee SJ, et al. 2011a. Perfluorocarbon exposure, gender and thyroid function in
the C8 Health Project. J Toxicol Sci 36(4):403-410.
Knox SS, Jackson T, Javins B, et al. 2011b. Implications of early menopause in women exposed to
perfluorocarbons. J Clin Endocrinol Metab 96(6):1747-1753. I 10.1210/jc.2010-2401.
Kobayashi Y, Hirokawa N, Ohshiro N, et al. 2002. Differential gene expression of organic anion
transporters in male and female rats. Biochem Biophys Res Commun 290:482-487.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

672
8. REFERENCES

Konwick BJ, Tomy GT, Ismail N, et al. 2008. Concentrations and patterns of perfluoroalkyl acids in
Georgia, USA surface waters near and distant to a major use source. Environ Toxicol Chem
27(10):2011-2018.
+Koskela A, Finnila MA, Korkalainen M, et al. 2016. Effects of developmental exposure to
perfluorooctanoic acid (PFOA) on long bone morphology and bone cell differentiation. Toxicol
Appl Pharmacol 301:14-21. 10.1016/j.taap.2016.04.002.
Kotthoff M, Muller J, Jurling H, et al. 2015. Perfluoroalkyl and polyfluoroalkyl substances in consumer
products. Environ Sci Pollut Res Int 22(19):14546-14559. 10.1007/s11356-015-4202-7.
Koustas E, Lam J, Sutton P, et al. 2014. The Navigation Guide - evidence-based medicine meets
environmental health: Systematic review of nonhuman evidence for PFOA effects on fetal growth.
Environ Health Perspect 122(10):1015-1027.
Krishnan K, Andersen ME, Clewell HJ, et al. 1994. Physiologically based pharmacokinetic modeling of
chemical mixtures. In: Yang RSH, ed. Toxicology of chemical mixtures: Case studies,
mechanisms, and novel approaches. San Diego, CA: Academic Press, 399-437.
Kristensen SL, Ramlau-Hansen CH, Ernst E, et al. 2013. Long-term effects of prenatal exposure to
perfluoroalkyl substances on female reproduction. Hum Reprod 28(12):3337-3348.
10.1093/humrep/det382.
Kroschwitz JI, Howe-Grant M. 1994. Perfluorooctanoic. Kirk-Othmer encyclopedia of chemical
toxicology. 4th ed. Vol. 11. New York, NY: John Wiley & Sons, Inc., 551.
Krusic PJ, Roe DC. 2004. Gas-phase NMR technique for studying the thermolysis of materials:
Thermal decomposition of ammonium perfluorooctanoate. Anal Chem 76(13):3800-3803.
Krusic PJ, Marchione AA, Roe DC. 2005. Gas-phase NMR studies of the thermolysis of
perfluroooctanoic acid. J Fluor Chem 126:1510-1516.
Kubwabo C, Stewart B, Zhu J, et al. 2005. Occurrence of perfluorosulfonates and other
perfluorochemicals in dust from selected homes in the city of Ottawa, Canada. J Environ Monit
7:1074-1078.
Kudo N, Kawashima Y. 1997. Fish oil-feeding prevents perfluorooctanoic acid-induced fatty liver in
mice. Toxicol Appl Pharmacol 145(2):285-293.
Kudo N, Kawashima Y. 2003. Induction of triglyceride accumulation in the liver of rats by
perfluorinated fatty acids with different carbon chain lengths: Comparison with induction of
peroxisomal ß-oxidation. Biol Pharm Bull 26(1):47-51.
Kudo N, Bandai N, Suzuki E, et al. 2000. Induction by perfluorinated fatty acids with different carbon
chain length of peroxisomal β-oxidation in the liver of rats. Chem Biol Interact 124:119-132.
Kudo N, Iwase Y, Okayachi H, et al. 2005. Induction of hepatic peroxisome proliferation by 8-2 telomer
alcohol feeding in mice: Formation of perfluorooctanoic acid in the liver. Toxicol Sci 86(2):231238.
Kudo N, Katakura M, Sato Y, et al. 2002. Sex hormone-regulated renal transport of perfluorooctanoic
acid. Chem Biol Interact 139:301-316.
Kudo N, Mizuguchi H, Yamamoto A, et al. 1999. Alterations by perfluorooctanoic acid of glycerolipid
metabolism in rat liver. Chem Biol Interact 118:69-83.
Kudo N, Sakai A, Mitsumoto A, et al. 2007. Tissue distribution and hepatic subcellular distribution of
perfluorooctanoic acid at low dose are different from those at high dose in rats. Biol Pharm Bull
30(8):1535-1540.
Kudo N, Suzuki E, Katakura M, et al. 2001. Comparison of the elimination between perfluorinated fatty
acids with different carbon chain length in rats. Chem Biol Interact 134:203-216.
Kudo N, Suzuki-Nakajima E, Mitsumoto A, et al. 2006. Responses of the liver to perfluorinated fatty
acids with different carbon chain length in male and female mice: In relation to induction of
hepatomegaly, peroxisomal β-oxidation and microsomal 1-acylglycerophosphocholine
acyltransferase. Biol Pharm Bull 29(9):1952-1957.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

673
8. REFERENCES

Kuklenyik Z, Reich JA, Tully JS, et al. 2004. Automated solid-phase extraction and measurement of
perfluorinated organic acids and amides in human serum and milk. Environ Sci Technol
38(13):3698-3704.
Kunleda H, Shinoda K. 1976. Krafft points, critical micelle concentrations, surface tension, and
solubilizing power of aqueous solutions of fluorinated surfactants. J Phys Chem 80:2468-2470.
Kvist L, Giwercman YL, Jonsson BA, et al. 2012. Serum levels of perfluorinated compounds and sperm
Y:X chromosome ratio in two European populations and in Inuit from Greenland. Reprod Toxicol
34(4):644-650.
Landsteiner A, Huset C, Johnson J, et al. 2014. Biomonitoring for perfluorochemicals in a Minnesota
community with known drinking water contamination. J Environ Health 77(5):14-19.
+Lau C, Thibodeaux JR, Hanson RG, et al. 2003. Exposure to perfluorooctane sulfonate during
pregnancy in rat and mouse. II: Postnatal evaluation. Toxicol Sci 74(2):382-392.
+Lau C, Thibodeaux JR, Hanson RG, et al. 2006. Effects of perfluorooctanoic acid exposure during
pregnancy in the mouse. Toxicol Sci 90(2):510-518.
Lee CK, Kang SG, Lee JT, et al. 2015a. Effects of perfluorooctane sulfuric acid on placental PRLfamily hormone production and fetal growth retardation in mice. Mol Cell Endocrinol 401:165-172.
10.1016/j.mce.2014.10.026.
Lee ES, Han S, Oh JE. 2016. Association between perfluorinated compound concentrations in cord
serum and birth weight using multiple regression models. Reprod Toxicol 59:53-59.
10.1016/j.reprotox.2015.10.020.
Lee YJ, Kim M-K, Bae J, et al. 2013. Concentrations of perfluoroalkyl compounds in maternal and
umbilical cord sera and birth outcomes in Korea. Chemosphere 90(5):1603-1609.
Lee YY, Wong CK, Oger C, et al. 2015b. Prenatal exposure to the contaminant perfluorooctane
sulfonate elevates lipid peroxidation during mouse fetal development but not in the pregnant dam.
Free Radic Res 49(8):1015-1025. 10.3109/10715762.2015.1027199.
+Lefebvre DE, Curran I, Armstrong C, et al. 2008. Immunomodulatory effects of dietary potassium
perfluorooctane sulfonate (PFOS) exposure in adult Sprague-Dawley rats. J Toxicol Environ Health
A 71(23):1516-1525.
Lemieux P, Strynar M, Tabor D, et al. 2007. Emissions of fluorinated compounds from the combustion
of carpeting. In: International conference on incineration and thermalk treatment technologies.
May 14-18, 2007. Phoenix, Arizona. U.S. Environmental Protection Agency.
http://www.epa.gov/nhsrc/pubs/paperFluorCompounds101007.pdf. December 7, 2009.
Lenters V, Portengen L, Rignell-Hydbom A, et al. 2016a. Prenatal phthalate, perfluoroalkyl acid, and
organochlorine exposures and term birth weight in three birth cohorts: Multi-pollutant models based
on elastic net regression. Environ Health Perspect 124(3):365-372. 10.1289/ehp.1408933.
Lenters V, Portengen L, Rignell-Hydbom A, et al. 2016b. Supplemental material to prenatal phthalate,
perfluoroalkyl acid, and organochlorine exposures and term birth weight in three birth cohorts:
Multi-pollutant models based on elastic net regression. [Environ Health Perspect 124(3):365-372].
10.1289/ehp.1408933.
Leonard RC. 2006. Ammonium perfluorooctanoate: Phase II. Retrospective cohort mortality analyses
related to a serum biomarker of exposure in a polymer production plant. Wilmington, DE: E.I. du
pont de Nemours and Company.
Leonard RC, Kreckmann KH, Sakr CJ, et al. 2008. Retrospective cohort mortality study of workers in a
polymer production plant including a reference population of regional workers. Ann Epidemiol
18:15-22.
Lewis RC, Johns LE, Meeker JD. 2015. Serum biomarkers of exposure to perfluoroalkyl substances in
relation to serum testosterone and measures of thyroid function among adults and adolescents from
NHANES 2011-2012. Int J Environ Res Public Health 12(6):6098-6114.
Li L, Zhai Z, Liu J, et al. 2015. Estimating industrial and domestic environmental releases of
perfluorooctanoic acid and its salts in China from 2004 to 2012. Chemosphere 129:100-109.
10.1016/j.chemosphere.2014.11.049.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

674
8. REFERENCES

Li N, Hartley DP, Cherrington NJ, et al. 2002. Tissue expression, ontogeny, and inducibility of rat
organic anion transporting polypeptide. J Pharmacol Exp Ther 301(2):551-560.
+Li X, Ye L, Ge Y, et al. 2016. In utero perfluorooctane sulfonate exposure causes low body weights of
fetal rats: A mechanism study. Placenta 39:125-133. 10.1016/j.placenta.2016.01.010.
Li Y, Fletcher T, Mucs D, et al. 2018. Half-lives of PFOS, PFHxS and PFOA after end of exposure to
contaminated drinking water. Occup Environ Med 75(1):46-51. 10.1136/oemed-2017-104651.
Li Y, Ramdhan DH, Naito H, et al. 2011. Ammonium perfluorooctanoate may cause testosterone
reduction by adversely affecting testis in relation to PPARα. Toxicol Lett 205(3):265-272.
http://dx.doi.org/10.1016/j.toxlet.2011.06.015.
Lide DR. 2005. Pentadecalfluorooctanoic acid, nondecalfluorodecanoic acid, and heptafluroobutanoic
acid. In: CRC handbook of chemistry and physics. 86th ed. Boca Raton, FL: Taylor and Francisk,
3-412, 3-372, 3-398.
+Lieder PH, Chang SC, York RG, et al. 2009a. Toxicological evaluation of potassium
perfluorobutanesulfonate in a 90-day oral gavage study with Sprague-Dawley rats. Toxicology
255:45-52.
+Lieder PH, York RG, Hakes DC, et al. 2009b. A two-generation oral gavage reproduction study with
potassium perfluorobutanesulfonate (K+PFBS) in Sprague-Dawley rats. Toxicology 259:33-45.
Lien GW, Huang CC, Shiu JS, et al. 2016. Perfluoroalkyl substances in cord blood and attention
deficit/hyperactivity disorder symptoms in seven-year-old children. Chemosphere 156:118-127.
10.1016/j.chemosphere.2016.04.102.
Lien GW, Huang CC, Wu KY, et al. 2013. Neonatal-maternal factors and perfluoroalkyl substances in
cord blood. Chemosphere 92(7):843-850.
Liew Z, Ritz B, Bonefeld-Jorgensen EC, et al. 2014. Prenatal exposure to perfluoroalkyl substances and
the risk of congenital cerebral palsy in children. Am J Epidemiol 180(6):574-581.
10.1093/aje/kwu179.
Liew Z, Ritz B, von Ehrenstein OS, et al. 2015. Attention deficit/hyperactivity disorder and childhood
autism in association with prenatal exposure to perfluoroalkyl substances: A nested case-control
study in the Danish National Birth Cohort. Environ Health Perspect 123(4):367-373.
10.1289/ehp.1408412.
Lim TC, Wang B, Huang J, et al. 2011. Emission inventory for PFOS in China: Review of past
methodologies and suggestions. Sci World J 11:1963-1980.
Lin CY, Chen PC, Lin YC, et al. 2009. Association among serum perfluoroalkyl chemicals, glucose
homeostasis, and metabolic syndrome in adolescents and adults. Diabetes Care 32(4):702-707.
Lin CY, Lin LY, Chiang CK, et al. 2010. Investigation of the associations between low-dose serum
perfluorinated chemicals and liver enzymes in US adults. Am J Gastroenterol 105(6):1354-1363.
Lin CY, Lin LY, Wen TW, et al. 2013a. Association between levels of serum perfluorooctane sulfate
and carotid artery intima-media thickness in adolescents and young adults. Int J Cardiol
168(4):3309-3316. 10.1016/j.ijcard.2013.04.042.
Lin CY, Lin LY, Wen TW, et al. 2013b. Supplemental material: Association between levels of serum
perfluoroctane sulfate and carotid artery intima-media thickness in adolescents and young adults. Int
J Cardiol 168:3309-3316.
Lin LY, Wen LL, Su TC, et al. 2014. Negative association between serum perfluorooctane sulfate
concentration and bone mineral density in US premenopausal women: NHANES, 2005-2008. The
Journal of clinical endocrinology and metabolism 99(6):2173-2180. 10.1210/jc.2013-3409.
Lind L, Zethelius B, Salihovic S, et al. 2014. Circulating levels of perfluoroalkyl substances and
prevalent diabetes in the elderly. Diabetologia 57(3):473-479. 10.1007/s00125-013-3126-3.
Lindstrom AB, Strynar MJ, Delinsky AD, et al. 2011. Application of WWTP biosolids and resulting
perfluorinated compound contamination of surface and well water in Decatur, Alabama, USA.
Environ Sci Technol 45:8015-8021.
Liu J, Li J, Liu Y, et al. 2011. Comparison on gestation and lactation exposure of perfluorinated
compounds for newborns. Environ Int 37(7):1206-1212.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

675
8. REFERENCES

Liu J, Lee LS, Nies LF, et al. 2007. Biotransformation of 8:2 fluorotelomer alcohol in soil and by soil
bacteria isolates. Environ Sci Technol 41:8024-8030.
Liu L, Liu W, Song J, et al. 2009. A comparative study on oxidative damage and distributions of
perfluorooctane sulfonate (PFOS) in mice at different postnatal developmental stages. J Toxicol Sci
34(3):245-254.
+Liu RCM, Hurtt ME, Cook JC, et al. 1996. Effect of the peroxisome proliferator, ammonium
perfluorooctanoate (C8), on hepatic aromatase activity in adult male Crl:CD BR (CD) rats. Toxicol
Sci 30(2):220-228.
Liu W, Li X, Xu L, et al. 2010. Influence of gestation, regular bleeding and intermittent exposure on
blood perfluorooctane sulfonate levels in mice: Potential factors inducing sex difference and
affecting exposure evaluation. J Toxicol Sci 35(3):309-316.
Liu X, Guo Z, Krebs KA, et al. 2014. Concentrations and trends of perfluorinated chemicals in potential
indoor sources from 2007 through 2011 in the US. Chemosphere 98:51-57.
10.1016/j.chemosphere.2013.10.001.
Llorca M, Farre M, Pico Y, et al. 2010. Infant exposure of perfluorinated compounds: Levels in breast
milk and commercial baby food. Environ Int 36(6):584-592.
Loccisano AE, Campbell JL, Jr., Andersen ME, et al. 2011. Evaluation and prediction of
pharmacokinetics of PFOA and PFOS in the monkey and human using a PBPK model. Regul
Toxicol Pharmacol 59(1):157-175.
Loccisano AE, Campbell JL, Jr., Butenhoff JL, et al. 2012a. Comparison and evaluation of
pharmacokinetics of PFOA and PFOS in the adult rat using a physiologically based pharmacokinetic
model. Reprod Toxicol 33(4):452-467.
Loccisano AE, Campbell JL, Jr., Butenhoff JL, et al. 2012b. Evaluation of placental and lactational
pharmacokinetics of PFOA and PFOS in the pregnant, lactating, fetal and neonatal rat using a
physiologically based pharmacokinetic model. Reprod Toxicol 33(4):468-490.
Loccisano AE, Longnecker MP, Campbell JL, Jr., et al. 2013. Development of PBPK models for PFOA
and PFOS for human pregnancy and lactation life stages. J Toxicol Environ Health A 76(1):25-57.
Loewen M, Halldorson T, Wang F, et al. 2005. Fluorotelomer carboxylic acids and PFOS in rainwater
from an urban center in Canada. Environ Sci Technol 39:2944-2951.
Loganathan BG, Sajwan KS, Sinclair E, et al. 2007. Perfluoroalkyl sulfonates and perfluorocarboxylates
in two wastewater treatment facilities in Kentucky and Georgia. Water Res 41(20):4611-4620.
doi.org/10.1016/j.watres.2007.06.045.
+Long Y, Wang Y, Ji G, et al. 2013. Neurotoxicity of perfluorooctane sulfonate to hippocampal cells in
adult mice. PLoS ONE 8(1):e54176.
Looker C, Luster MI, Calafat AM, et al. 2014. Influenza vaccine response in adults exposed to
perfluorooctanoate and perfluorooctanesulfonate. Toxicol Sci 138(1):76-88.
Loos R, Locoro G, Huber T, et al. 2008. Analysis of perfluorooctanoate (PFOA) and other
perfluorinated compounds (PFCs) in the River Po watershed in N-Italy. Chemosphere 71:306-313.
López-Doval S, Salgado R, Pereiro N, et al. 2014. Perfluorooctane sulfonate effects on the reproductive
axis in adult male rats. Environ Res 134:158-168.
Lopez-Espinosa MJ, Fletcher T, Armstrong B, et al. 2011. Association of perfluorooctanoic acid
(PFOA) and perfluorooctane sulfonate (PFOS) with age of puberty among children living near a
chemical plant. Environ Sci Technol 45(19):8160-8166.
Lopez-Espinosa MJ, Mondal D, Armstrong B, et al. 2012. Thyroid function and perfluoroalkyl acids in
children living near a chemical plant. Environ Health Perspect 120(7):1036-1041.
10.1289/ehp.1104370.
+Lopez-Espinosa MJ, Mondal D, Armstrong BG, et al. 2016. Perfluoroalkyl substances, sex hormones,
and insulin-like growth factor-1 at 6-9 years of age: A Cross-sectional analysis within the C8 Health
Project. Environ Health Perspect 124:1269-1275. 10.1289/ehp.1509869.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

676
8. REFERENCES

Lorber M, Egeghy PP. 2011. Simple intake and pharmacokinetic modeling to characterize exposure of
Americans to perfluoroctanoic acid, PFOA. Environ Sci Technol 45(19):8006-8014.
10.1021/es103718h.
Lou I, Wambaugh JF, Lau C, et al. 2009. Modeling single and repeated dose pharmacokinetics of PFOA
in mice. Toxicol Sci 107(2):331-341.
Loveless SE, Finlay C, Everds NE, et al. 2006. Comparative responses of rats and mice exposed to
linear/branched, linear, or branched ammonium perfluorooctanoate (APFO). Toxicology 220:203217.
+Loveless SE, Hoban D, Sykes G, et al. 2008. Evaluation of the immune system in rats and mice
administered linear ammonium perfluorooctanoate. Toxicol Sci 105(1):86-96.
Lu R, Kanai N, Bao Y, et al. 1996. Regulation of renal oatp mRNA expression by testosterone. Am J
Physiol 270:F332-F337.
Lubojevic M, Herak-Kramberger CM, Hagos Y, et al. 2004. Rat renal cortical OAT1 and OAT3 exhibit
gender differences determined by both androgen stimulation and estrogen inhibition. Am J Physiol
287:F124-F138.
+Luebker DJ, Case MT, York RG, et al. 2005a. Two-generation reproduction and cross-foster studies of
perfluorooctanesulfonate (PFOS) in rats. Toxicology 215(1-2):126-148.
Luebker DJ, Hansen KJ, Bass NM, et al. 2002. Interactions of fluorochemicals with rat liver fatty acidbinding protein. Toxicology 176:175-185.
+Luebker DJ, York RG, Hansen KJ, et al. 2005b. Neonatal mortality from in utero exposure to
perfluorooctanesulfonate (PFOS) in Sprague-Dawley rats: Dose-response, and biochemical and
pharmacokinetic parameters. Toxicology 215(1-2):149-169.
Lundin JI, Alexander BH, Olsen GW, et al. 2009. Ammonium perfluorooctanoate production and
occupational mortality. Epidemiology 20(6):921-928.
Luo Z, Shi X, Hu Q, et al. 2012. Structural evidence of perfluorooctane sulfonate transport by human
serum albumin. Chem Res Toxicol 25(5):990-992.
Lv Z, Li G, Li Y, et al. 2013. Glucose and lipid homeostasis in adult rat is impaired by early‐life
exposure to perfluorooctane sulfonate. Environ Toxicol 28(9):532-542.
+Lv QY, Wan B, Guo LH, et al. 2015. In vivo immunotoxicity of perfluorooctane sulfonate in BALB/c
mice: Identification of T-cell receptor and calcium-mediated signaling pathway disruption through
gene expression profiling of the spleen. Chem Biol Interact 240:84-93. 10.1016/j.cbi.2015.07.015.
Lyngsø J, Ramlau-Hansen CH, Hoyer BB, et al. 2014. Menstrual cycle characteristics in fertile women
from Greenland, Poland and Ukraine exposed to perfluorinated chemicals: A cross-sectional study.
Human reproduction (Oxford, England) 29(2):359-367. 10.1093/humrep/det390.
MacNeil J, Steenland NK, Shankar A, et al. 2009. A cross-sectional analysis of type II diabetes in a
community with exposure to perfluorooctanoic acid (PFOA). Environ Res 109(8):997-1003.
+Macon MB, Villanueva LR, Tatum-Gibbs K, et al. 2011. Prenatal perfluorooctanoic acid exposure in
CD-1 mice: Low-dose developmental effects and internal dosimetry. Toxicol Sci 122(1):134-145.
MacPherson IR, Bissett D, Petty RD, et al. 2011. A first-in-human phase I clinical trial of CXR1002 in
patients (pts) with advanced cancer. J Clin Oncol 29(Suppl; abstr 3063).
Maestri L, Negri S, Ferrari M, et al. 2006. Determination of perfluorooctanoic acid and
perflluorooctanesulfonate in human tissues by liquid chromatography/single quadrupole mass
spectrometry. Rapid Commun Mass Spectrom 20(18):2728-2734.
Maher JM, Aleksunes LM, Dieter MZ, et al. 2008. Nrf2- and PPARα-mediated regulation of hepatic
Mrp transporters after exposure to perfluorooctanoic acid and perfluorodecanoic acid. Toxicol Sci
106(2):319-328. 10.1093/toxsci/kfn177.
Maisonet M, Terrell ML, McGeehin MA, et al. 2012. Maternal concentrations of polyfluoroalkyl
compounds during pregnancy and fetal and postnatal growth in British girls. Environ Health
Perspect 120(10):1432-1437.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

677
8. REFERENCES

Maisonet M, Calafat AM, Marcus M, et al. 2015a. Prenatal exposure to perfluoroalkyl acids and serum
testosterone concentrations at 15 years of age in female ALSPAC study participants. Environ Health
Perspect 123(12):1325-1330. 10.1289/ehp.1408847.
Maisonet M, Nayha S, Lawlor DA, et al. 2015b. Prenatal exposures to perfluoroalkyl acids and serum
lipids at ages 7 and 15 in females. Environ Int 82:49-60. 10.1016/j.envint.2015.05.001.
Mak YL, Taniyasu S, Yeung LW, et al. 2009. Perfluorinated compounds in tap water from China and
several other countries. Environ Sci Technol 43(13):4824-4829.
Maloney EK, Waxman DJ. 1999. trans-Activation of PPARα and PPARγ by structurally diverse
environmental chemicals. Toxicol Appl Pharmacol 161:209-218.
Mann PC, Frame SR. 2004. FC-143: Two year oral toxicity-oncogenicity study in rats. Peer review of
ovaries. Newark, DE: E.I. du Pont de Nemours and Company. Du Pont Project ID 15261. U.S.
EPA AR226.
Manzano-Salgado CB, Casas M, Lopez-Espinosa MJ, et al. 2015. Transfer of perfluoroalkyl substances
from mother to fetus in a Spanish birth cohort. Environ Res 142:471-478.
10.1016/j.envres.2015.07.020
Martin JW, Ellis DA, Mabury SA. 2006. Atmospheric chemistry of perfluoroalkanesulfonamides:
Kinetic and product studies of the OH radical and CI atom initiated oxidation of N-ethyl
perfluorobutanesulfonamide. Environ Sci Technol 40:864-872.
Martin JW, Smithwick MM, Braune BM, et al. 2004a. Identification of long-chain perfluorinated acids
in biota from the Canadian Arctic. Environ Sci Technol 38(2):373-380.
Martin JW, Whittle DM, Muir DCG, et al. 2004b. Perfluoroalkyl contaminants in a food web from Lake
Ontario. Environ Sci Technol 38:5379-5385.
Martin MT, Brennan RJ, Hu W, et al. 2007. Toxicogenomic study of triazole fungicides and
perfluoroalkyl acids in rat livers predicts toxicity and categorizes chemicals based on mechanisms of
toxicity. Toxicol Sci 97:595-613.
Mattsson K, Rignell-Hydbom A, Holmberg S, et al. 2015. Levels of perfluoroalkyl substances and risk
of coronary heart disease: Findings from a population-based longitudinal study. Environ Res
142:148-154. 10.1016/j.envres.2015.06.033.
MDH. 2007. Groundwater health risk limits. St. Paul, MN: Minnesota Department of Health.
http://www.health.state.mn.us/divs/eh/groundwater/hrltable.html. July 1, 2008.
MDH. 2009. East Metro perfluorochemical biomonitoring pilot project. Minnesota Department of
Health, Minnesota Environment, Exposure and Health.
http://www.health.state.mn.us/divs/hpcd/tracking/biomonitoring/projects/pfcfinalrpt2009.pdf. May
5, 2014.
MDH. 2014. Perfluorochemicals in homes and gardens study. Minnesota Department of Health.
http://www.health.state.mn.us/divs/eh/hazardous/topics/pfcs/index.html.
MDH. 2017a. Human health-based water guidance table. Saint Paul, Minnesota: Minnesota
Department of Health. http://www.health.state.mn.us/divs/eh/risk/guidance/gw/table.html.
December 18, 2017.
MDH. 2017b. Perfluorochemicals (PFCs) and health. Saint Paul, Minnesota: Minnesota Department of
Health. http://www.health.state.mn.us/divs/eh/hazardous/topics/pfcshealth.html. December 18,
2017.
MECDC. 2016. Maine CDC maximum exposure guidelines (MEGs) for drinking water. Augusta, ME:
Maine Center for Disease Control and Prevention.
http://maineenvironmentallaboratory.com/newsite/wp-content/uploads/megtable2016.pdf. December
18, 2017.
Meesters RJ, Schröder HF. 2004. Perfluorooctane sulfonate- a quite mobile anionic anthropogenic
surfactant, ubiquitously found in the environment. Water Sci Technol 50(5):235-242.
Melzer D, Rice N, Depledge MH, et al. 2010. Association between serum perfluorooctanoic acid
(PFOA) and thyroid disease in the U.S. National Health and Nutrition Examination Survey. Environ
Health Perspect 118(5):686-692.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

678
8. REFERENCES

Michigan Department of Environmental Quality. 2016. Rule 57 water quality values. Lansing, MI:
Michigan Department of Environmental Quality. http://www.michigan.gov/documents/deq/wrdswas-rule57_372470_7.pdf. December 18, 2017.
Midasch O, Drexler H, Hart N, et al. 2007. Transplacental exposure of neonates to
perfluorooctanesulfonate and perfluorooctanoate: A pilot study. Int Arch Occup Environ Health
80:643-648.
Min JY, Lee KJ, Park JB, et al. 2012. Perfluorooctanoic acid exposure is associated with elevated
homocysteine and hypertension in US adults. Occup Environ Med 69(9):658-662.
Minata M, Harada KH, Kärrman A, et al. 2010. Role of peroxisome proliferator-activated receptor- α in
hepatobiliary injury induced by ammonium perfluorooctanoate in mouse liver. Ind Health 48:96107.
Mondal D, Lopez-Espinosa MJ, Armstrong B, et al. 2012. Relationships of perfluorooctanoate and
perfluorooctane sulfonate serum concentrations between mother-child pairs in a population with
perfluorooctanoate exposure from drinking water. Environ Health Perspect 120(5):752-757.
Mondal D, Weldon RH, Armstrong BG, et al. 2014. Breastfeeding: A potential excretion route for
mothers and implications for infant exposure to perfluoroalkyl acids. Environ Health Perspect
122(2):187-192.
Monroy R, Morrison K, Teo K, et al. 2008. Serum levels of perfluoroalkyl compounds in human
maternal and umbilical cord blood samples. Environ Res 108:56-62.
Moody CA, Field JA. 1999. Determination of perfluorocarboxylates in groundwater impacted by firefighting activity. Environ Sci Technol 33(16):2800-2806.
Moody CA, Hebert GN, Strauss SH, et al. 2003. Occurrence and persistence of perfluorooctanesulfonate
and other perfluorinated surfactants in groundwater at a fire-training area at Wurtsmith Air Force
Base, Michigan, USA. J Environ Monit 5:341-345.
Mogensen UB, Grandjean P, Nielsen F, et al. 2015b. Breastfeeding as an exposure pathway for
perfluorinated alkylates. Environ Sci Technol 49(17):10466-10473. 10.1021/acs.est.5b02237.
Mogensen UB, Grandjean P, Heilmann C, et al. 2015a. Structural equation modeling of immunotoxicity
associated with exposure to perfluorinated alkylates. Environ Health 14:47 10.1186/s12940-0150032-9.
Morello-Frosch R, Cushing LJ, Jesdale BM, et al. 2016. Environmental chemicals in an urban
population of pregnant women and their newborns from San Francisco. Environ Sci Technol
50:12464-12472.
Moriwaki H, Takata Y, Arakawa R. 2003. Concentrations of perfluorooctane sulfonate (PFOS) and
perfluorooctanoic acid (PFOA) in vacuum cleaner dust collected in Japanese homes. J Environ
Monit 5:753-757.
Mundt DJ, Mundt KA, Luippold RS, et al. 2007. Clinical epidemiological study of employees exposed
to surfactant blend containing perfluorononanoic acid. Occup Environ Med 64:589-594.
Murakami M, Imamura E, Shinohara H, et al. 2008. Occurrence and sources of perfluorinated
surfactants in rivers in Japan. Environ Sci Technol 42(17):6566-6572.
Nabb DL, Szostek B, Himmelstein MW, et al. 2007. In vitro metabolism of 8-2 fluorotelomer alcohol:
Interspecies comparisons and metabolic pathway refinement. Toxicol Sci 100(2):333-344.
Nakagawa H, Hirata T, Terada T, et al. 2008. Roles of organic anion transporters in the renal excretion
of perfluorooctanoic acid. Basic Clin Pharmacol Toxicol 103(1):1-8.
Nakagawa H, Terada T, Harada KH, et al. 2009. Human organic anion transporter hOAT4 is a
transporter of perfluorooctanoic acid. Basic Clin Pharmacol Toxicol 105(2):136-138.
Nakayama S, Strynar MJ, Helfant L, et al. 2007. Perfluorinated compounds in the Cape Fear drainage
basin in North Carolina. Environ Sci Technol 41(15):5271-5276.
Nakagawa T, Ramdhan D, Tanaka N, et al. 2012. Modulation of ammonium perfluorooctanoate-induced
hepatic damage by genetically different PPARα in mice. Arch Toxicol 86(1):63-74.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

679
8. REFERENCES

NAS/NRC. 1989. Report of the oversight committee. In: Biologic markers in reproductive toxicology.
Washington, DC: National Academy of Sciences, National Research Council, National Academy
Press, 15-35.
NCDENR. 2012. Memorandum: NCSAB recommendation for the revision of the IMAC for
perfluorooctanoic acid (PFOA). August 10, 2012. Raleigh, NC: North Carolina Department of
Environment and Natural Resources Division of Air Quality.
https://ncdenr.s3.amazonaws.com/s3fspublic/Air%20Quality/toxics/risk/sab/LAA/20120808%20PFOA%20Final%20Risk%20Assessment
%20signed.pdf. December 18, 2017.
NDEP. 2016. Basic comparison levels. Carson City, NE: Nevada Division of Environmental
Protection. https://ndep.nv.gov/uploads/documents/july-2017-ndep-bcls.pdf. December 18, 2017.
Needham LL, Grandjean P, Heinzow B, et al. 2011. Partition of environmental chemicals between
maternal and fetal blood and tissues. Environ Sci Technol 45(3):1121-1126.
Nelson JW, Hatch EE, Webster TF. 2010. Exposure to polyfluoroalkyl chemicals and cholesterol, body
weight, and insulin resistance in the general U.S. population. Environ Health Perspect 118(2):197202.
+Ngo HT, Hetland RB, Sabaredzovic A, et al. 2014. In utero exposure to perfluorooctanoate (PFOA) or
perfluorooctane sulfonate (PFOS) did not increase body weight or intestinal tumorigenesis in
multiple intestinal neoplasia (Min/+) mice. Environ Res 132:251-263.
10.1016/j.envres.2014.03.033.
Nilsson R, Beije B, Preat V, et al. 1991. On the mechanism of the hepatocarcinogenicity of peroxisome
proliferations. Chem Biol Interact 78:235-250.
NIOSH. 2016. Index of Chemical Abstracts Service Registry Numbers (CAS No.). NIOSH pocket
guide to chemical hazards. Index of Chemical Abstracts Service Registry Numbers (CAS No.).
Atlanta, GA: National Institute for Occupational Safety and Health, Centers for Disease Control and
Prevention. https://www.cdc.gov/niosh/npg/npgdcas.html. February 28, 2017.
Noker PE, Gorman GS. 2003. A pharmacokinetic study of potassium perfluorooctanesulfonate in the
Cynomolgus monkey. St. Paul, MN: 3M Corporation.
Nolan LA, Nolan JM, Shofer FS, et al. 2009. The relationship between birth weight, gestational age and
perfluorooctanoic acid (PFOA)-contaminated public drinking water. Reprod Toxicol 27:231-238.
Nolan LA, Nolan JM, Shofer FS, et al. 2010. Congenital anomalies, labor/delivery complications,
maternal risk factors and their relationship with perfluorooctanoic acid (PFOA)-contaminated public
drinking water. Reprod Toxicol 29(2):147-155.
North Carolina Department of Environmental Quality. 2013. Appendix 1: Interim maximum allowable
concentrations (IMACs). Raleigh, NC: North Carolina Department of Environmental Quality.
https://files.nc.gov/ncdeq/documents/files/IMAC%20table_5-22-13.pdf. December 19, 2017.
NTP. 2016a. CASRN Index in MS Excel. Report on carcinogens, Fourteenth Edition. Research
Triangle Park, NC: U.S. Department of Health and Human Services, Public Health Service,
National Toxicology Program. https://ntp.niehs.nih.gov/pubhealth/roc/index-1.html#P. February
28, 2017.
NTP. 2016b. NTP Monograph on immunotoxicity associated with exposure to perfluorooctanoic acid
(PFOA) and perfluorooctane sulfonate (PFOS). Research Triangle Park, NC: U.S. Department of
Health and Human Services, Public Health Service, National Toxicology Program.
https://ntp.niehs.nih.gov/ntp/ohat/pfoa_pfos/pfoa_pfosmonograph_508.pdf. December 21, 2017.
Oda Y, Nakayama S, Harada KH, et al. 2007. Negative results of umu genotoxicity test of fluorotelomer
alcohols and perfluorinated alkyl acids. Environ Health Prev Med 12:217-219.
Ode A, Kallen K, Gustafsson P, et al. 2014. Fetal exposure to perfluorinated compounds and attention
deficit hyperactivity disorder in childhood. PLoS ONE 9(4):e95891.
10.1371/journal.pone.0095891.
Ode A, Rylander L, Lindh CH, et al. 2013. Determinants of maternal and fetal exposure and temporal
trends of perfluorinated compounds. Environ Sci Pollut Res Int 20(11):7970-7978.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

680
8. REFERENCES

OECD. 2002. Hazard assessment of perfluorooctane sulfonate (PFOS) and its salts. Organisation for
Economic Co-operation and Development. ENV/JM/RD(2002)17/FINAL.
http://www.oecd.org/dataoecd/23/18/2382880.pdf. July 02, 2007.
OECD. 2006a. Results of the 2006 survey on production and use of PFOS, PFAS, PFOA, PFCA, their
related substances and products/mixtures containing these substances. Organisation for Economic
Co-operation and Development.
OECD. 2006b. SIDS initial assessment report after SIAM 22. Ammonium perfluorooctanoate and
perfluorooctanoic acid. Organisation for Economic Co-operation and Development.
http://www.oecd.org/document/63/0,3343,en_2649_34379_1897983_1_1_1_1,00.html. May 18,
2009.
OECD. 2007. Report of an OECD workshop on perfluorocarboxylic acids (PFCAs) and precursors.
Organisation for Economic Co-operation and Development.
http://www.olis.oecd.org/olis/2007doc.nsf/LinkTo/NT00002AB6/$FILE/JT03229256.PDF. April
02, 2008.
Ohmori K, Kudo N, Katayama K, et al. 2003. Comparison of the toxicokinetics between
perfluorocarboxylic acids with different carbon chain length. Toxicology 184:135-140.
Okada E, Sasaki S, Saijo Y, et al. 2012. Prenatal exposure to perfluorinated chemicals and relationship
with allergies and infectious diseases in infants. Environ Res 112:118-125.
Okada E, Sasaki S, Kashino I, et al. 2014. Prenatal exposure to perfluoroalkyl acids and allergic diseases
in early childhood. Environ Int 65:127-134. 10.1016/j.envint.2014.01.007.
Olsen GW. 2015. PFAS biomonitoring in higher exposed populations, Chapter 4. In: Dewitt JC, ed.
Toxicological effects on perfluoroalkyl and polyfouroalkyl substances. London: Humana Press, 77125.
Olsen GW, Zobel LR. 2007. Assessment of lipid, hepatic, and thyroid parameters with serum
perfluorooctanoate (PFOA) concentrations in fluorochemical production workers. Int Arch Occup
Environ Health 81:231-246.
Olsen GW, Burlew MM, Marshall JC, et al. 2004a. Analysis of episodes of care in a
perfluorooctanesulfonyl fluoride production facility. J Occup Environ Hyg 46(8):837-846.
Olsen GW, Burris JM, Burlew MM, et al. 2003a. Epidemiologic assessment of worker serum
perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) concentrations and medical
surveillance examinations. J Occup Environ Med 45(3):260-270.
Olsen GW, Burris JM, Burlew MM, et al. 2000. Plasma cholecystokinin and hepatic enzymes,
cholesterol and lipoproteins in ammonium perfluorooctanoate production workers. Drug Chem
Toxicol 23(4):603-620.
Olsen GW, Burris JM, Ehresman DJ, et al. 2007a. Half-life of serum elimination of
perfluorooctanesulfonate, perfluorohexanesulfonate, and perfluorooctanoate in retired
fluorochemical production workers. Environ Health Perspect 115:1298-1305.
Olsen GW, Burris JM, Mandel JH, et al. 1998a. An epidemiologic investigation of clinical chemistries,
hematology and hormones in relation to serum levels of perfluorooctane sulfonate in male
fluorochemical production employees. St. Paul, MN: 3M Company. AR226-0030.
Olsen GW, Burris JM, Mandel JH, et al. 1999. Serum perfluorooctane sulfonate and hepatic and lipid
clinical chemistry tests in fluorochemical production employees. J Occup Environ Med 41(9):799806.
Olsen GW, Butenhoff JL, Zobel LR. 2009. Perfluoroalkyl chemicals and human fetal development: An
epidemiologic review with clinical and toxicological perspectives. Reprod Toxicol 27(3-4):212-230.
Olsen GW, Church TR, Hansen KJ, et al. 2004b. Quantitative evaluation of perfluorooctanesulfonate
(PFOS) and other fluorochemicals in the serum of children. J Child Health 2(1):53-76.
Olsen GW, Church TR, Larson EB, et al. 2004c. Serum concentrations of perfluorooctanesulfonate and
other fluorochemicals in an elderly population from Seattle, Washington. Chemosphere 54:15991611.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

681
8. REFERENCES

Olsen GW, Church TR, Miller JP, et al. 2003b. Perfluorooctanesulfonate and other fluorochemicals in
the serum of American Red Cross adult blood donors. Environ Health Perspect 111:1892-1901.
Olsen GW, Ehresman DJ, Buehrer BD, et al. 2012. Longitudinal assessment of lipid and hepatic clinical
parameters in workers involved with the demolition of perfluoroalkyl manufacturing facilities. J
Occup Environ Med 54(8):974-983.
Olsen GW, Gilliland FD, Burlew MM, et al. 1998b. An epidemiologic investigation of reproductive
hormones in men with occupational exposure to perfluorooctanoic acid. J Occup Environ Med
40(7):614-622.
Olsen GW, Hansen KJ, Stevenson LA, et al. 2003c. Human donor liver and serum concentrations of
perfluorooctanesulfonate and other perfluorochemicals. Environ Sci Technol 37:888-891.
Olsen GW, Huang HY, Helzlsouer KJ, et al. 2005. Historical comparison of perfluorooctanesulfonate,
perfluorooctanoate, and other fluorochemicals in human blood. Environ Health Perspect
113(5):539-545.
Olsen GW, Mair DC, Church TR, et al. 2008. Decline in perfluorooctanesulfonate and other
polyfluoroalkyl chemicals in American Red Cross adult blood donors, 2000-2006. Environ Sci
Technol 42(13):4989-4995.
Olsen GW, Mair DC, Lange CC, et al. 2017a. Per- and polyfluoroalkyl substances (PFAS) in American
Red Cross adult blood donors, 2000-2015. Environ Res 157:87-95. 10.1016/j.envres.2017.05.013.
Olsen GW, Mair DC, Lange CC, et al. 2017b. Supplemental file. Per-and polyfluoroalkyl substances
(PFAS) in American Red Cross adult blood donors, 2000-2015 [Environ Res 157:87-95]. Environ
Res 10.1016/j.envres.2017.05.013.
Olsen GW, Mair DC, Reagan WK, et al. 2007b. Preliminary evidence of a decline in
perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) concentrations in American Red
Cross blood donors. Chemosphere 68(1):105-111.
+Onishchenko N, Fischer C, Wan Ibrahim WN, et al. 2011. Prenatal exposure to PFOS or PFOA alters
motor function in mice in a sex-related manner. Neurotox Res 19(3):452-461.
Orata F, Quinete N, Werres F, et al. 2009. Determination of perfluorooctanoic acid and perfluorooctane
sulfonate in Lake Victoria Gulf water. Bull Environ Contam Toxicol 82:218-222.
OSHA. 2013. Subpart Z - Toxic and hazardous substances. Air contaminants. Occupational Safety and
Health Standards. Code of Federal Regulations 29 CFR 1910.1000.
http://www.gpo.gov/fdsys/pkg/CFR-2014-title29-vol6/pdf/CFR-2014-title29-vol6-sec19101000.pdf. March 4, 2015.
OSHA. 2014a. Appendix A to Part 1926.55-1970 American Conference of Governmental Industrial
Hygienists’ threshold limit values of airborne contaminants. Occupational Safety and Health
Standards. Code of Federal Regulations 29 CFR 1926.55 http://www.gpo.gov/fdsys/pkg/CFR2014-title29-vol8/pdf/CFR-2014-title29-vol8-sec1926-55.pdf. March 4, 2015.
OSHA. 2014b. Subpart Z - Toxic and hazardous substances. Air contaminants. Table Z - Shipyards.
Occupational Safety and Health Standards. Code of Federal Regulations 29 CFR 1915.1000.
http://www.gpo.gov/fdsys/pkg/CFR-2013-title29-vol7/pdf/CFR-2013-title29-vol7-sec19151000.pdf. March 4, 2015.
Osuna CE, Grandjean P, Weihe P, et al. 2014. Autoantibodies associated with prenatal and childhood
exposure to environmental chemicals in Faroese children. Toxicol Sci 142(1):158-166.
10.1093/toxsci/kfu163.
+Pastoor TP, Lee KP, Perri MA, et al. 1987. Biochemical and morphological studies of ammonium
perfluorooctanoate-induced hepatomegaly and peroxisome proliferation. Exp Mol Pathol 47(1):98109.
Paustenbach DJ, Panko JM, Scott PK, et al. 2007. A methodology for estimating human exposure to
perfluorooctanoic acid (PFOA): A retrospective exposure assessment of a community (1951-2003).
J Toxicol Environ Health A 70(1):28-57.
+Peden-Adams MM, Keller JM, Eudaly JG, et al. 2008. Suppression of humoral immunity in mice
following exposure to perfluorooctane sulfonate. Toxicol Sci 104(1):144-154.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

682
8. REFERENCES

+Pereiro N, Moyano R, Blanco A, et al. 2014. Regulation of corticosterone secretion is modified by
PFOS exposure at different levels of the hypothalamic-pituitary-adrenal axis in adult male rats.
Toxicol Lett 230(2):252-262. 10.1016/j.toxlet.2014.01.003.
Perez F, Nadal M, Navarro-Ortega A, et al. 2013. Accumulation of perfluoroalkyl substances in human
tissues. Environ Int 59:354-362. 10.1016/j.envint.2013.06.004.
+Perkins RG, Butenhoff JL, Kennedy GL, et al. 2004. 13-Week dietary toxicity study of ammonium
perfluorooctanoate (APFO) in male rats. Drug Chem Toxicol 27(4):361-378.
+Permadi H, Lundgren B, Andersson K, et al. 1992. Effects of perfluoro fatty acids on xenobioticmetabolizing enzymes, enzymes which detoxify reactive forms of oxygen and lipid peroxidation in
mouse liver. Biochem Pharmacol 44(6):1183-1191.
+Permadi H, Lundgren B, Andersson K, et al. 1993. Effects of perfluoro fatty acids on peroxisome
proliferation and mitochondrial size in mouse liver: Dose and time factors and effect of chain
length. Xenobiotica 23(7):761-770.
Peters JM, Gonzalez FJ. 2011. Why toxic equivalency factors are not suitable for perfluoroalkyl
chemicals. American Chemical Society. Chem Res Toxicol 24(10):1601-1609.
10.1021/tx200316x.
Pinney SM, Biro FM, Windham GC, et al. 2014. Serum biomarkers of polyfluoroalkyl compound
exposure in young girls in Greater Cincinnati and the San Francisco Bay Area, USA. Environ Pollut
184:327-334. 10.1016/j.envpol.2013.09.008.
Plastics Europe. 2012. Guide for the safe handling of fluoropolymer resins, November 2012. Plastics
Europe, Association of Plastics Manufacturers.
Porpora MG, Lucchini R, Abballe A, et al. 2013. Placental transfer of persistent organic pollutants: A
preliminary study on mother-newborn pairs. Int J Environ Res Public Health 10(2):699-711.
Post GB, Louis JB, Cooper KR, et al. 2009. Occurrence and potential significance of perfluorooctanoic
acid (PFOA) detected in New Jersey public drinking water systems. Environ Sci Technol
43(12):4547-4554.
Post GB, Louis JB, Lippincott RL, et al. 2013. Occurrence of perfluorinated compounds in raw water
from New Jersey public drinking water systems. Environ Sci Technol 47(23):13266-13275.
10.1021/es402884x.
Power MC, Webster TF, Baccarelli AA, et al. 2013. Cross-sectional association between polyfluoroalkyl
chemicals and cognitive limitation in the National Health and Nutrition Examination Survey.
Neuroepidemiology 40(2):125-132.
Powley CR, George SW, Russell MH, et al. 2008. Polyfluorinated chemicals in a spatially and
temporally integrated food web in the Western Arctic. Chemosphere 70:664-672.
Powley CR, Michalczyk MJ, Kaiser MA, et al. 2005. Determination of perfluorooctanoic acid (PFOA)
extractable from the surface of commercial cookware under simulated cooking conditions by
LC/MS/MS. Analyst 130(9):1299-1302.
Predieri B, Iughetti L, Guerranti C, et al. 2015. High levels of perfluorooctane sulfonate in children at
the onset of diabetes. Int J Endocrinol 2015:234358. 10.1155/2015/234358.
Prevedouros K, Cousins I T, Buck RC, et al. 2006. Sources, fate and transport of perfluorocarboxylates.
Environ Sci Technol 40(1):32-44.
Qazi MR, Abedi MR, Nelson BD, et al. 2010a. Dietary exposure to perfluorooctanoate or
perfluorooctane sulfonate induces hypertrophy in centrilobular hepatocytes and alters the hepatic
immune status in mice. Int Immunopharmacol 10(11):1420-1427.
+Qazi MR, Nelson BD, Depierre JW, et al. 2010b. 28-Day dietary exposure of mice to a low total dose
(7 mg/kg) of perfluorooctanesulfonate (PFOS) alters neither the cellular compositions of the thymus
and spleen nor humoral immune responses: Does the route of administration play a pivotal role in
PFOS-induced immunotoxicity? Toxicology 267(1-3):132-139. 10.1016/j.tox.2009.10.035.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

683
8. REFERENCES

Qazi MR, Bogdanska J, Butenhoff JL, et al. 2009a. High-dose, short-term exposure of mice to
perfluorooctanesulfonate (PFOS) or perfluorooctanoate (PFOA) affects the number of circulating
neutrophils differently, but enhances the inflammatory responses of macrophages to
lipopolysaccharide (LPS) in a similar fashion. Toxicology 262(3):207-214.
Qazi MR, Nelson BD, DePierre JW, et al. 2012. High-dose dietary exposure of mice to
perfluorooctanoate or perfluorooctane sulfonate exerts toxic effects on myeloid and B-lymphoid
cells in the bone marrow and these effects are partially dependent on reduced food consumption.
Food Chem Toxicol 50(9):2955-2963.
Qazi MR, Xia Z, Bogdanska J, et al. 2009b. The atrophy and changes in the cellular compositions of the
thymus and spleen observed in mice subjected to short-term exposure to perfluorooctanesulfonate
are high-dose phenomena mediated in part by peroxisome proliferator-activated receptor-alpha
(PPARα). Toxicology 260(1-3):68-76.
Qin XD, Qian Z, Vaughn MG, et al. 2016. Positive associations of serum perfluoroalkyl substances with
uric acid and hyperuricemia in children from Taiwan. Environ Pollut 212:519-524.
10.1016/j.envpol.2016.02.050.
Quaak I, de Cock M, de Boer M, et al. 2016. Prenatal exposure to perfluoroalkyl substances and
behavioral development in children. Int J Environ Res Public Health 13(5).
10.3390/ijerph13050511.
Quinete N, Wu Q, Zhang T, et al. 2009. Specific profiles of perfluorinated compounds in surface and
drinking waters and accumulation in mussels, fish, and dolphins from southeastern Brazil.
Chemosphere 77(6):863-869. 10.1016/j.chemosphere.2009.07.079.
Quinones O, Snyder SA. 2009. Occurrence of perfluoroalkyl carboxylates and sulfonates in drinking
water utilities and related waters from the United States. Environ Sci Technol 43(24):9089-9095.
Quist EM, Filgo AJ, Cummings CA, et al. 2015a. Hepatic mitochondrial alteration in CD-1 mice
associated with prenatal exposures to low doses of perfluorooctanoic acid (PFOA). Toxicol Pathol
43(4):546-557. 10.1177/0192623314551841.
Quist EM, Filgo AJ, Cummings CA, et al. 2015b. Supplemental data: Hepatic mitochondrial alteration
in CD-1 mice associated with prenatal exposures to low doses of perfluorooctanoic acid (PFOA).
(Toxicol Pathol 43(4):546-557). Toxicol Pathol 43:546-557.
Raleigh KK, Alexander BH, Olsen GW, et al. 2014. Mortality and cancer incidence in ammonium
perfluorooctanoate production workers. Occup Environ Med 71(7):500-506. 10.1136/oemed-2014102109.
Rankin K, Lee H, Tseng PJ, et al. 2014. Investigating the biodegradability of a fluorotelomer-based
acrylate polymer in a soil-plant microcosm by indirect and direct analysis. Environ Sci Technol
48(21):12783-12790.
Raymer JH, Michael LC, Studabaker WB, et al. 2012. Concentrations of perfluorooctane sulfonate
(PFOS) and perfluorooctanoate (PFOA) and their associations with human semen quality
measurements. Reprod Toxicol 33(4):419-427.
+Rebholz SL, Jones T, Herrick RL, et al. 2016. Hypercholesterolemia with consumption of PFOA-laced
Western diets is dependent on strain and sex of mice. Toxicology reports 3:46-54.
10.1016/j.toxrep.2015.11.004.
Reiner JL, Nakayama SF, Delinsky AD, et al. 2009. Analysis of PFOA in dosed CD1 mice. Part 1.
Methods development for the analysis of tissues and fluids from pregnant and lactating mice and
their pups. Reprod Toxicol 27(3-4):360-364. 10.1016/j.reprotox.2008.10.006.
Ren H, Vallanat B, Nelson DM, et al. 2009. Evidence for the involvement of xenobiotic-responsive
nuclear receptors in transcriptional effects upon perfluoroalkyl acid exposure in diverse species.
Reprod Toxicol 27(3-4):266-277. 10.1016/j.reprotox.2008.12.011.
RePORTER. 2017. Perflouroalkyls. National Institutes of Health, Research Portfolio Online Reporting
Tools. http://projectreporter.nih.gov/reporter.cfm. February 28, 2017.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

684
8. REFERENCES

Ribes D, Fuentes S, Torrente M, et al. 2010. Combined effects of perfluorooctane sulfonate (PFOS) and
maternal restraint stress on hypothalamus adrenal axis (HPA) function in the offspring of mice.
Toxicol Appl Pharmacol 243(1):13-18.
Robledo CA, Yeung E, Mendola P, et al. 2015a. Preconception maternal and paternal exposure to
persistent organic pollutants and birth size: The LIFE study. Environ Health Perspect 123(1):88-94.
10.1289/ehp.1308016.
Robledo CA, Yeung E, Mendola P, et al. 2015b. Supplemental material to preconception maternal and
paternal exposure to persistent organic pollutants and birth size: The LIFE study. [Environ Health
Perspect 123(1):88-94]. Environ Health Perspect 10.1289/ehp.1308016.
Rockwell CE, Turley AE, Cheng X, et al. 2013. Acute immunotoxic effects of perfluorononanoic acid
(PFNA) in C57BL/6 Mice. Clin Exp Pharmacol Suppl 4. 10.4172/2161-1459.S4-002.
http://www.ncbi.nlm.nih.gov/pubmed/25568816.
Rockwell CE, Turley AE, Cheng X, et al. 2017. Persistent alterations in immune cell populations and
function from a single dose of perfluorononanoic acid (PFNA) in C57B1/6 mice. Food Chem
Toxicol 100:24-33.
Rodriguez CE, Setzer RW, Barton HA. 2009. Pharmacokinetic modeling of perfluorooctanoic acid
during gestation and lactation in the mouse. Reprod Toxicol 27(3-4):373-386.
+Rogers JM, Ellis-Hutchings RG, Grey BE, et al. 2014. Elevated blood pressure in offspring of rats
exposed to diverse chemicals during pregnancy. Toxicol Sci 137(2):436-446.
10.1093/toxsci/kft248.
Romano ME, Xu Y, Calafat AM, et al. 2016. Maternal serum perfluoroalkyl substances during
pregnancy and duration of breastfeeding. Environ Res 10.1016/j.envres.2016.04.034.
Rosen MB, Abbott BD, Wolf DC, et al. 2008a. Gene profiling in the livers of wild-type and PPARα-null
mice exposed to perfluorooctanoic acid. Toxicol Pathol 36(4):592-607.
Rosen MB, Das KP, Rooney J, et al. 2017. PPARα-independent transcriptional targets of perfluoroalkyl
acids revealed by transcript profiling. Toxicology [In press].
Rosen MB, Lee JS, Ren H, et al. 2008b. Toxicogenomic dissection of the perfluorooctanoic acid
transcript profile in mouse liver: Evidence for the involvement of nuclear receptors PPARα and
CAR. Toxicol Sci 103(1):46-56.
Rosen MB, Schmid JR, Corton JC, et al. 2010. Gene expression profiling in wild-type and pparalphanull mice exposed to perfluorooctane sulfonate reveals PPARα-independent effects. PPAR Res
10.1155/2010/794739.
+Rosen MB, Schmid JE, Das KP, et al. 2009. Gene expression profiling in the liver and lung of
perfluorooctane sulfonate-exposed mouse fetuses: Comparison to changes induced by exposure to
perfluorooctanoic acid. Reprod Toxicol 27:278-288.
Rosen MB, Thibodeaux JR, Wood CR, et al. 2007. Gene expression profiling in the lung and liver of
PFOA-exposed mouse fetuses. Toxicology 239:15-33.
Rotander A, Toms LM, Aylward L, et al. 2015. Elevated levels of PFOS and PFHxS in firefighters
exposed to aqueous film forming foam (AFFF). Environ Int 82:28-34.
10.1016/j.envint.2015.05.005.
RTECS. 2008. Perfluoroalkyls. Hamilton, Ontario: Registry of Toxic Effects on Chemical Substances.
Canadian Centre for Occupational Health and Safety. Symyx Software, Inc. May 29, 2008.
Rumsby PC, McLaughlin CL, Hall T. 2009. Perfluorooctane sulphonate and perfluorooctanoic acid in
drinking and environmental waters. Philos Trans A Math Phys Eng Sci 367(1904):4119-4136.
10.1098/rsta.2009.0109.
Ryu MH, Jha A, Ojo OO, et al. 2014. Chronic exposure to perfluorinated compounds: Impact on airway
hyperresponsiveness and inflammation. Am J Physiol Lung Cell Mol Physiol 307(10):L765-L774.
Sakr CJ, Kreckmann KH, Green JW, et al. 2007b. Cross-sectional study of lipids and liver enzymes
related to a serum biomarker of exposure (ammonium perfluorooctanoate or APFO) as part of a
general health survey in a cohort of occupationally exposed workers. J Occup Environ Med
49:1086-1096.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

685
8. REFERENCES

Sakr CJ, Leonard RC, Kreckmann KH, et al. 2007a. Longitudinal study of serum lipids and liver
enzymes in workers with occupational exposure to ammonium perfluorooctanoate. J Occup Environ
Med 49:872-879.
Sakr CJ, Symons JM, Kreckmann KH, et al. 2009. Ischaemic heart disease mortality study among
workers with occupational exposure to ammonium perfluorooctanoate. Occup Environ Med 66:699703.
+Salgado R, Pereiro N, Lopez-Doval S, et al. 2015. Initial study on the possible mechanisms involved in
the effects of high doses of perfluorooctane sulfonate (PFOS) on prolactin secretion. Food Chem
Toxicol 83:10-16. 10.1016/j.fct.2015.05.013.
Salvalaglio M, Muscionico I, Cavallotti C. 2010. Determination of energies and sites of binding of
PFOA and PFOS to human serum albumin. J Phys Chem B 114(46):14860-14874.
Sato I, Kawamoto K, Nishikawa Y, et al. 2009. Neurotoxicity of perfluorooctane sulfonate (PFOS) in
rats and mice after single oral exposure. J Toxicol Sci 34(5):569-574.
Savitz DA, Stein CR, Bartell SM, et al. 2012a. Perfluorooctanoic acid exposure and pregnancy outcome
in a highly exposed community. Epidemiology 23(3):386-392.
Savitz DA, Stein CR, Elston B, et al. 2012b. Relationship of perfluorooctanoic acid exposure to
pregnancy outcome based on birth records in the Mid-Ohio Valley. Environ Health Perspect
120(8):1201-1207.
Savu PM. 1994a. Fluorinated higher carboxylic acids. In: Kirk-Othmer encyclopedia of chemical
technology. John Wiley & Sons, Inc., 1-7.
http://mrw.interscience.wiley.com/emrw/9780471238966/kirk/article/fluosavu.a01/current/pdf.
April 01, 2008.
Savu P. 1994b. Fluorine-containing polymers, perfluoroalkanesulfonic acids. In: Kirk-Othmer
encyclopedia of chemical technology. John Wiley & Sons, Inc., 1-7.
http://mrw.interscience.wiley.com/emrw/9780471238966/kirk/article/perfsavu.a01.current/pdf.
April 01, 2008.
Schaider LA, Balan SA, Blum A, et al. 2017. Fluorinated compounds in U.S. fast food packaging.
Environ Sci Tech Lett 4(3):105-111. 10.1021/acs.estlett.6b00435.
Schecter A, Colacino J, Haffner D, et al. 2010. Perfluorinated compounds, polychlorinated biphenyls,
and organochlorine pesticide contamination in composite food samples from Dallas, Texas, USA.
Environ Health Perspect 118(6):796-802.
Schultz MM, Barofsky DF, Field JA. 2003. Fluorinated alkyl surfactants. Environ Eng Sci 20(5):487501.
Schultz MM, Barofsky DF, Field JA. 2006a. Quantitative determination of fluorinated alkyl substances
by large-volume-injection liquid chromatography tandem mass spectrometry. Characterization of
municipal wastewaters. Environ Sci Technol 40:289-295.
Schultz MM, Higgins CP, Huset CA, et al. 2006b. Fluorochemical mass flows in a municipal
wastewater treatment facility. Environ Sci Technol 40(23):7350-7367.
+Seacat AM, Luebker DJ. 2000. Toxicokinetic study of perfluorooctane sulfonamide (PFOS; T-7132.2)
in rats. 3M Strategic Toxicology Laboratory. Submitted to the U.S. Environmental Protection
Agency's Administrative Record. AR226-1030A011.
+Seacat AM, Thomford PJ, Hansen KJ, et al. 2002. Subchronic toxicity studies on
perfluorooctanesulfonate potassium salt in Cynomolgus monkeys. Toxicol Sci 68(1):249-264.
+Seacat AM, Thomford PJ, Hansen KJ, et al. 2003. Sub-chronic dietary toxicity of potassium
perfluorooctanesulfonate in rats. (Erratum in: Toxicology 2003 192(2-3):263-264). Toxicology
183(1-3):117-133.
Seals R, Bartell SM, Steenland K. 2011. Accumulation and clearance of perfluorooctanoic acid (PFOA)
in current and former residents of an exposed community. Environ Health Perspect 119(1):119-124.
Sepulvado JG, Blaine AC, Hundal LS, et al. 2011. Occurrence and fate of perfluorochemicals in soil
following the land application of municipal biosolids. Environ Sci Technol 45(19):8106-8112.
10.1021/es103903d.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

686
8. REFERENCES

Shankar A, Xiao J, Ducatman A. 2011a. Perfluoroalkyl chemicals and chronic kidney disease in US
adults. Am J Epidemiol 174(8):893-900.
Shankar A, Xiao J, Ducatman A. 2011b. Perfluoroalkyl chemicals and elevated serum uric acid in US
adults. Clin Epidemiol 3:251-258.
Shankar A, Xiao J, Ducatman A. 2012. Perfluorooctanoic acid and cardiovascular disease in US adults.
Arch Intern Med 172(18):1397-1403.
Shapiro GD, Dodds L, Arbuckle TE, et al. 2016. Exposure to organophosphorus and organochlorine
pesticides, perfluoroalkyl substances, and polychlorinated biphenyls in pregnancy and the
association with impaired glucose tolerance and gestational diabetes mellitus: The MIREC Study.
Environ Res 147:71-81. 10.1016/j.envres.2016.01.040.
+Shi Z, Zhang H, Ding L, et al. 2009. The effect of perfluorododecanoic acid on endocrine status, sex
hormones and expression of steroidogenic genes in pubertal female rats. Reprod Toxicol 27(34):352-359.
+Shi Z, Zhang H, Liu Y, et al. 2007. Alterations in gene expression and testosterone synthesis in the
testes of male rats exposed to perfluorododecanoic acid. Toxicol Sci 98(1):206-215.
Shin HM, Vieira VM, Ryan PB, et al. 2011. Environmental fate and transport modeling for
perfluorooctanoic acid emitted from the Washington Works Facility in West Virginia. Environ Sci
Technol 45(4):1435-1442.
Shipley JM, Hurst CH, Tanaka SS, et al. 2004. trans-Activation of PPAR(beta) and induction of
PPAR(beta) target genes by perfluorooctane-based chemicals. Toxicol Sci 80(1):151-160.
Shoeib M, Harner T, Ikonomou M, et al. 2004. Indoor and outdoor air concentrations and phase
partitioning of perfluoroalkyl sulfonamides and polybrominated diphenyl ethers. Environ Sci
Technol 38:1313-1320.
Shoeib M, Harner T, Vlahos P. 2006. Perfluorinated chemicals in the arctic atmosphere. Environ Sci
Technol 40:7577-7583.
Shrestha S, Bloom MS, Yucel R, et al. 2015. Perfluoroalkyl substances and thyroid function in older
adults. Environ Int 75:206-214. 10.1016/j.envint.2014.11.018.
Siegemund G, Schwertfeger W, Feiring A, et al. 2005. Fluorine compounds, organic. Ullmann's
Encyclopedia of Industrial Chemistry. Wiley-VCH Verlag GmbH & Co.
http://mrw.interscience.wiley.com/emrw/9783527306732/ueic/article/a11_349/current/abstract.
April 01, 2008.
Simcik MF. 2005. Global transport and fate of perfluorochemicals. J Environ Monit 7:759-763.
Simcik MF, Dorweiler KJ. 2005. Ratio of perfluorochemical concentrations as a tracer of atmospheric
deposition to surface waters. Environ Sci Technol 39:8678-8683.
Simpson C, Winquist A, Lally C, et al. 2013. Relation between perfluorooctanoic acid exposure and
strokes in a large cohort living near a chemical plant. Environ Res 127:22-28.
10.1016/j.envres.2013.10.002.
Sinclair E, Kannan K. 2006. Mass loading and fate of perfluoroalkyl surfactants in wastewater treatment
plants. Environ Sci Technol 40(5):1408-1414.
Sinclair E, Kim SK, Akinleye HB, et al. 2007. Quantitation of gas-phase perfluoroalkyl surfactants and
fluorotelomer alcohols released from nonstick cookware and microwave popcorn bags. Environ Sci
Technol 41:1180-1185.
Sinclair E, Mayack DT, Roblee K, et al. 2006. Occurrence of perfluoroalkyl surfactants in water, fish,
and birds from New York State. Arch Environ Contam Toxicol 50:398-401.
Sinclair E, Taniyasu S, Yamashita N, et al. 2004. Perfluorooctanoic acid and perfluorooctane sulfonate
in Michigan and New York waters. Organohalogen Compounds 66:4069-4073.
Skuladottir M, Ramel A, Rytter D, et al. 2015. Examining confounding by diet in the association
between perfluoroalkyl acids and serum cholesterol in pregnancy. Environ Res 143(Pt A):33-38.
10.1016/j.envres.2015.09.001.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

687
8. REFERENCES

Small MJ. 2009. Final report of the peer consultation panel conducting the review for the scientific peer
consultation process for a site environmental assessment program as part of the Dupont-EPA
memorandum of understanding and Phase II Workplan. Pittsburgh, PA: Carnegie Mellon
University, Civil and Environmental Engineering and Engineering and Public Policy.
Smit LA, Lenters V, Hoyer BB, et al. 2015. Prenatal exposure to environmental chemical contaminants
and asthma and eczema in school-age children. Allergy 70(6):653-660. 10.1111/all.12605.
Smithwick M, Mabury SA, Solomon KR, et al. 2005a. Circumpolar study of perfluoroalkyl
contaminants in polar bears (Ursus maritimus). Environ Sci Technol 39(15):5517-5523.
Smithwick M, Muir DC, Mabury SA, et al. 2005b. Perfluoroalkyl contaminants in liver tissue from East
Greenland polar bears (Ursus maritimus). Environ Toxicol Chem 24(4):981-986.
Smithwick M, Norstrom RJ, Mabury SA, et al. 2006. Temporal trends of perfluoroalkyl contaminants in
polar bears (Ursus maritimus) from two locations in the North American arctic, 1972-2002. Environ
Sci Technol 40:1139-1143.
So MK, Taniyasu S, Lam PKS, et al. 2006a. Alkaline digestion and solid phase extraction method for
perfluorinated compounds in mussels and oysters from South China and Japan. Arch Environ
Contam Toxicol 50:240-248.
So MK, Yamashita N, Taniyasu S, et al. 2006b. Health risks in infants associated with exposure to
perfluorinated compounds in human breast milk from Zhoushan, China. Environ Sci Technol
40:2924-2929.
+Sobolewski M, Conrad K, Allen JL, et al. 2014. Sex-specific enhanced behavioral toxicity induced by
maternal exposure to a mixture of low dose endocrine-disrupting chemicals. Neurotoxicology
45:121-130. 10.1016/j.neuro.2014.09.008.
+Son H, Kim S, Shin HI, et al. 2008. Perfluorooctanoic acid-induced hepatic toxicity following 21-day
oral exposure in mice. Arch Toxicol 82:239-246.
+Son HY, Lee S, Tak EN, et al. 2009. Perfluorooctanoic acid alters T lymphocyte phenotypes and
cytokine expression in mice. Environ Toxicol 24(6):580-588.
SPARC. 2008. Macroscopic pka. Sparc Performs Automated Reasoning in Chemistry.
http://sparc.chem.ugs.edu/sparc/display/ShowMacro.cfm. May 08, 2008.
Specht IO, Hougaard KS, Spano D, et al. 2012. Sperm DNA integrity in relation to exposure to
environmental perfluoralkyl substances. A study of spouses of pregnant women in three
geographical regions. Reprod Toxicol 33:577-583.
Spliethoff HM, Tao L, Shaver SM, et al. 2008. Use of newborn screening program blood spots for
exposure assessment: Declining levels of perfluorinated compounds in New York State infants.
Environ Sci Technol 42(14):5361-5367.
Staels B, Dallongeville J, Auwerx J, et al. 1998. Mechanism of action of fibrates on lipid and lipoprotein
metabolism. Circulation 98:2088-2093.
+Staples RE, Burgess BA, Kerns WD. 1984. The embryo-fetal toxicity and teratogenic potential of
ammonium perfluorooctanoate (APFO) in the rat. Fundam Appl Toxicol 4:429-440.
Starkov AA, Wallace KB. 2002. Structural determinants of fluorochemical-induced mitochondrial
dysfunction. Toxicol Sci 66(2):244-252.
Starling AP, Engel SM, Richardson DB, et al. 2014b. Perfluoroalkyl substances during pregnancy and
validated preeclampsia among nulliparous women in the Norwegian Mother and Child Cohort
Study. Am J Epidemiol 179(7):824-833. 10.1093/aje/kwt432.
Starling AP, Engel SM, Whitworth KW, et al. 2014a. Perfluoroalkyl substances and lipid concentrations
in plasma during pregnancy among women in the Norwegian Mother and Child Cohort Study.
Environ Int 62:104-112. 10.1016/j.envint.2013.10.004.
Stasinakis AS, Petalas AV, Mamais D, et al. 2008. Application of the OECD 301F respirometric test for
the biodegradability assessment of various potential endocrine disrupting chemicals. Bioresour
Technol 99:3458-3467.
Steenland K, Woskie S. 2012. Cohort mortality study of workers exposed to perfluorooctanoic acid. Am
J Epidemiol 176(10):909-917.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

688
8. REFERENCES

Steenland K, Fletcher T, Savitz DA. 2010a. Epidemiologic evidence on the health effects of
perfluorooctanoic acid (PFOA). Environ Health Perspect 118(8):1100-1108.
Steenland K, Jin C, MacNeil J, et al. 2009a. Predictors of PFOA levels in a community surrounding a
chemical plant. Environ Health Perspect 117(7):1083-1088.
Steenland K, Tinker S, Frisbee S, et al. 2009b. Association of perfluorooctanoic acid and
perfluorooctane sulfonate with serum lipids among adults living near a chemical plant. Am J
Epidemiol 170(10):1268-1278.
Steenland K, Tinker S, Shankar A, et al. 2010b. Association of perfluorooctanoic acid (PFOA) and
perfluorooctane sulfonate (PFOS) with uric acid among adults with elevated community exposure to
PFOA. Environ Health Perspect 118(2):229-233.
Steenland K, Zhao L, Winquist A, et al. 2013. Ulcerative colitis and perfluorooctanoic acid (PFOA) in a
highly exposed population of community residents and workers in the Mid-Ohio Valley. Environ
Health Perspect 121(8):900-905.
Steenland K, Zhao L, Winquist A. 2015. A cohort incidence study of workers exposed to
perfluorooctanoic acid (PFOA). Occup Environ Med 72(5):373-380. 10.1136/oemed-2014-102364.
Stein CR, Savitz DA. 2011. Serum perfluorinated compound concentration and attention
deficit/hyperactivity disorder in children 5-18 years of age. Environ Health Perspect 119(10):14661471.
Stein CR, McGovern KJ, Pajak AM, et al. 2016a. Perfluoroalkyl and polyfluoroalkyl substances and
indicators of immune function in children aged 12-19 y: National Health and Nutrition Examination
Survey. Pediatr Res 79(2):348-357. 10.1038/pr.2015.213.
Stein CR, Ge Y, Wolff MS, et al. 2016b. Perfluoroalkyl substance serum concentrations and immune
response to FluMist vaccination among healthy adults. Environ Res 149:171-178.
10.1016/j.envres.2016.05.020.
Stein CR, Savitz DA, Bellinger DC. 2013. Perfluorooctanoate and neuropsychological outcomes in
children. Epidemiology 24(4):590-599.
Stein CR, Savitz DA, Bellinger DC. 2014a. Perfluorooctanoate exposure in a highly exposed community
and parent and teacher reports of behaviour in 6-12-year-old children. Paediatr Perinat Epidemiol
28(2):146-156. 10.1111/ppe.12097.
Stein CR, Savitz DA, Bellinger DC. 2014b. Supporting information to perfluorooctanoate exposure in a
highly exposed community and parent and teacher reports of behaviour in 6-12-year-old children.
Paediatr Perinat Epidemiol 10.1111/ppe.12097.
Stein CR, Savitz DA, Elston B, et al. 2014c. Perfluorooctanoate exposure and major birth defects.
Reprod Toxicol 47:15-20. 10.1016/j.reprotox.2014.04.006.
Stein CR, Savitz DA, Dougan M. 2009. Serum levels of perfluorooctanoic acid and perfluorooctane
sulfonate and pregnancy outcome. Am J Epidemiol 170(7):837-846.
Stock NL, Lau FK, Ellis DA, et al. 2004. Polyfluorinated telomer alcohols and sulfonamides in the
North American troposphere. Environ Sci Technol 38:991-996.
Strøm M, Hansen S, Olsen SF, et al. 2014. Persistent organic pollutants measured in maternal serum and
offspring neurodevelopmental outcomes--a prospective study with long-term follow-up. Environ Int
68:41-48. 10.1016/j.envint.2014.03.002.
Strynar MJ, Lindstrom AB. 2008. Perfluorinated compounds in house dust from Ohio and North
Carolina, USA. Environ Sci Technol 42:3751-3756.
Strynar MJ, Lindstrom AB, Nakayama SF, et al. 2012. Pilot scale application of a method for the
analysis of perfluorinated compounds in surface soils. Chemosphere 86:252-257.
Su TC, Kuo CC, Hwang JJ, et al. 2016. Serum perfluorinated chemicals, glucose homeostasis and the
risk of diabetes in working-aged Taiwanese adults. Environ Int 88:15-22.
10.1016/j.envint.2015.11.016.
Suh CH, Cho NK, Lee CK, et al. 2011. Perfluorooctanoic acid-induced inhibition of placental prolactinfamily hormone and fetal growth retardation in mice. Mol Cell Endocrinol 337(1-2):7-15.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

689
8. REFERENCES

Sundström M, Chang SC, Noker PE, et al. 2012. Comparative pharmacokinetics of
perfluorohexanesulfonate (PFHxS) in rats, mice, and monkeys. Reprod Toxicol 33(4):441-451.
Sweeney LM, Tyler TR, Kirman CR, et al. 2001. Proposed occupational exposure limits for select
ethylene glycol ethers using PBPK models and Monte Carlo simulations. Toxicol Sci 62(1):124139.
Takacs ML, Abbott BD. 2007. Activation of mouse and human peroxisome proliferator-activated
receptors (a, b/d, c) by perfluorooctanoic acid and perfluorooctane sulfonate. Toxicol Sci 95(1):108117.
Takagi A, Sai K, Umemura T, et al. 1991. Short-term exposure to the peroxisome proliferators,
perfluorooctanoic acid and perfluorodecanoic acid, causes significant increase of 8hydroxydeoxyguanosine in liver DNA of rats. Cancer Lett 57(1):55-60.
+Takahashi M, Ishida S, Hirata-Koizumi M, et al. 2014. Repeated dose and reproductive/developmental
toxicity of perfluoroundecanoic acid in rats. J Toxicol Sci 39(1):97-108.
+Tan X, Xie G, Sun X, et al. 2013. High fat diet feeding exaggerates perfluorooctanoic acid-induced
liver injury in mice via modulating multiple metabolic pathways. PLoS ONE 8(4):e61409.
Tan Y, Clewell HJ, Andersen ME. 2008. Time dependencies in perfluorooctylacids disposition in rat
and monkeys: A kinetic analysis. Toxicol Lett 177:38-47.
Tatum-Gibbs K, Wambaugh JF, Das KP, et al. 2011. Comparative pharmacokinetics of
perfluorononanoic acid in rat and mouse. Toxicology 281(1-3):48-55. 10.1016/j.tox.2011.01.003.
Tao L, Kannan K, Aldous KM, et al. 2008a. Biomonitoring of perfluorochemicals in plasma of New
York State personnel responding to the World Trade Center disaster. Environ Sci Technol
42(9):3472-3478.
Tao L, Kannan K, Wong CM, et al. 2008b. Perfluorinated compounds in human milk from
Massachusetts, U.S.A. Environ Sci Technol 42(8):3096-3101.
Tarazona JV, Rodriguez C, Alonso E, et al. 2016. Toxicokinetics of perfluorooctane sulfonate in rabbits
under environmentally realistic exposure conditions and comparative assessment between mammals
and birds. Toxicol Lett 241:200-206. 10.1016/j.toxlet.2015.11.002.
Tardiff RG, Carson ML, Sweeney LM, et al. 2009. Derivation of a drinking water equivalent level
(DWEL) related to the maximum contaminant level goal for perfluorooctanoic acid (PFOA), a
persistent water soluble compound. Food Chem Toxicol 47(10):2557-2589.
Taylor KW, Hoffman K, Thayer KA, et al. 2014. Polyfluoroalkyl chemicals and menopause among
women 20-65 years of age (NHANES). Environ Health Perspect 122(2):145-150.
10.1289/ehp.1306707.
+Thibodeaux JR, Hanson RG, Rogers JM, et al. 2003. Exposure to perfluorooctane sulfonate during
pregnancy in rat and mouse. I: Maternal and prenatal evaluations. Toxicol Sci 74(2):369-381.
+Thomford PJ. 2001. 4-Week capsule toxicity study with ammonium perfluorooctanoate (APFO) in
Cynomolgus monkeys. APME Ad-Hoc APFO toxicology working group.
+Thomford PJ. 2002a. 4-week capsule toxicity study with perfluorooctane sulfonic acid potassium salt
(PFOS; T-6295) in Cynomolgus monkeys. St. Paul, MN: 3M.
+Thomford PJ. 2002b. 104-Week dietary chronic toxicity and carcinogenicity study with
perfluorooctane sulfonic acid potassium salt (PFOS; T-6295) in rats. St. Paul, MN: 3M.
Thompson J, Lorber M, Toms LM, et al. 2010. Use of simple pharmacokinetic modeling to characterize
exposure of Australians to perfluorooctanoic acid and perfluorooctane sulfonic acid. Environ Int
36(4):390-397. 10.1016/j.envint.2010.02.008.
Thomsen C, Haug LS, Stigum H, et al. 2010. Changes in concentrations of perfluorinated compounds,
polybrominated diphenyl ethers, and polychlorinated biphenyls in Norwegian breast-milk during
twelve months of lactation. Environ Sci Technol 44(24):9550-9556.
Timmermann CA, Rossing LI, Grontved A, et al. 2014. Adiposity and glycemic control in children
exposed to perfluorinated compounds. J Clin Endocrinol Metab 99(4):E608-614. 10.1210/jc.20133460.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

690
8. REFERENCES

Tittlemier S, Ryan JJ, Van Oostdam J. 2004. Presence of anionic perfluorinated organic compounds in
serum collected from northern Canadian populations. Organohalogen Compounds 66:3959-3964.
Tittlemier SA, Pepper K, Seymour C, et al. 2007. Dietary exposure of Canadians to perfluorinated
carboxylates and perfluorooctane sulfonate via consumption of meat, fish, fast foods, and food items
prepared in their packaging. J Agric Food Chem 55:3203-3210.
Toft G, Jonsson BA, Bonde JP, et al. 2016. Perfluorooctane sulfonate concentrations in amniotic fluid,
biomarkers of fetal Leydig Cell Function, and cryptorchidism and hypospadias in Danish boys
(1980-1996). Environ Health Perspect 124(1):151-156. 10.1289/ehp.1409288.
Toft G, Jonsson BA, Lindh CH, et al. 2012. Exposure to perfluorinated compounds and human semen
quality in Arctic and European populations. Hum Reprod 27(8):2532-2540.
Toms LL, Calafat AM, Kato K, et al. 2009. Polyfluoroalkyl chemicals in pooled blood serum from
infants, children, and adults in Australia. Environ Sci Technol 43(11):4194-4199.
Tomy GT, Budakowski W, Halldorson T, et al. 2004. Fluorinated organic compounds in an eastern
arctic marine food web. Environ Sci Technol 38:6475-6481.
Trudel D, Horowitz L, Wormuth M, et al. 2008. Estimating consumer exposure to PFOS and PFOA.
Risk Anal 28(2):251-269.
Tsai MS, Lin CY, Lin CC, et al. 2015. Association between perfluoroalkyl substances and reproductive
hormones in adolescents and young adults. Int J Hyg Environ Health 218(5):437-443.
10.1016/j.ijheh.2015.03.008.
+Tucker DK, Macon MB, Strynar MJ, et al. 2015. The mammary gland is a sensitive pubertal target in
CD-1 and C57Bl/6 mice following perinatal perfluorooctanoic acid (PFOA) exposure. Reprod
Toxicol 54:26-36. 10.1016/j.reprotox.2014.12.002.
Uhl SA, James-Todd T, Bell ML. 2013. Association of osteoarthritis with perfluorooctanoate and
perfluorooctane sulfonate in NHANES 2003-2008. Environ Health Perspect 121(4):447-452.
Upham BL, Deocampo ND, Wurl B, et al. 1998. Inhibition of gap junctional intercellular
communication by perfluorinated fatty acids is dependent on the chain length of the fluorinated tail.
Int J Cancer 78:491-495.
Upham BL, Park JS, Babica P, et al. 2009. Structure-activity-dependent regulation of cell
communication by perfluorinated fatty acids using in vivo and in vitro model systems. Environ
Health Perspect 117(4):545-551. 10.1289/ehp.11728.
URS. 2012. MOU Phase III - future work plan data assessment DuPont Washington Works,
Washington, West Virginia, PFOA site-related environmental assessment program (OPPT-20040113 PFOA). In: Docket EPA-HQ-2004-0113-0517. Supporting documents: Letter to Dr. Maria
Doa, Director, Chemical Control Division, USEPA, from Andrew S. Hartten, Principal Remediation
Project Manager, DuPont Corporate Remediation Group. URS Corporation.
https://www.regulations.gov/document?D=EPA-HQ-OPPT-2004-0113-0516 February 28, 2017.
Vagi SJ, Azziz-Baumgartner E, Sjodin A, et al. 2014. Exploring the potential association between
brominated diphenyl ethers, polychlorinated biphenyls, organochlorine pesticides, perfluorinated
compounds, phthalates, and bisphenol A in polycystic ovary syndrome: A case-control study. BMC
Endoc Disord 14:86. 10.1186/1472-6823-14-86.
Vanden Heuvel JP, Davis JW, Sommers R, et al. 1992a. Renal excretion of perfluorooctanoic acid in
male rats: Inhibitory effect of testosterone. J Biochem Toxicol 7(1):31-36.
Vanden Heuvel JP, Kuslikis BI, Peterson RE. 1992b. Covalent binding of perkuorinated fatty acids to
proteins in the plasma, liver and testes of rats. Chem Biol Interact 82:318-328.
Vanden Heuvel JP, Kuslikis BI, Shrago E, et al. 1991a. Inhibition of long-chain acyl-CoA synthetase by
the peroxisome proliferator perfluorodecanoic acid in rat hepatocytes. Biochem Pharmacol
42(2):295-302.
Vanden Heuvel JP, Kuslikis BI, Van Rafelghem MJ, et al. 1991b. Disposition of perfluorodecanoic acid
in male and female rats. Toxicol Appl Pharmacol 107:450-459.
Vanden Heuvel JP, Kuslikis BI, Van Rafelghem MJ, et al. 1991c. Tissue distribution, metabolism, and
elimination of perfluorooctanoic acid in male and female rats. J Biochem Toxicol 6(2):83-92.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

691
8. REFERENCES

Vanden Heuvel JP, Thompson JT, Frame SR, et al. 2006. Differential activation of nuclear receptors by
perfluorinated fatty acid analogs and natural fatty acids: A comparison of human, mouse, and rat
peroxisome proliferator-activated receptor-α, -ß, and -γ, liver x receptor-ß, and retinoid x receptor-α.
Toxicol Sci 92(2):476-489.
+van Otterdijk FM. 2007a. Repeated dose 28-day oral toxicity study with MTDID-8391 by daily gavage
in the rat, followed by a 21-day recovery period. 3M.
van Otterdijk FM. 2007b. Repeated dose 90-day oral toxicity study with MTDID 8391 by daily gavage
in the rat followed by a 3-week recovery period. 3M.
Vélez MP, Arbuckle TE, Fraser WD. 2015. Maternal exposure to perfluorinated chemicals and reduced
fecundity: The MIREC study. Hum Reprod 30(3):701-709. 10.1093/humrep/deu350.
Vermont Department of Health. 2017. Drinking water guidance. Burlington, VT: Vermont Department
of Health.
http://www.healthvermont.gov/sites/default/files/documents/pdf/ENV_DW_Guidance.pdf.
December 19, 2017.
Verner MA, Loccisano AE, Morken NH, et al. 2015. Associations of perfluoroalkyl substances (PFAS)
with lower birth weight: An evaluation of potential confounding by glomerular filtration rate using a
physiologically based pharmacokinetic model (PBPK). Environ Health Perspect 123(12):13171324.
Verner MA, Longnecker MP. 2015. Comment on "Enhanced elimination of perfluorooctanesulfonic acid
by menstruating women: Evidence from population-based pharmacokinetic modeling." Environ Sci
Technol 49(9):5836-5837. 10.1021/acs.est.5b00187.
Verner MA, Ngueta G, Jensen ET, et al. 2016. A simple pharmacokinetic model of prenatal and
postnatal exposure to perfluoroalkyl substances (PFASs). Environ Sci Technol 50(2):978-986.
Verreault J, Berger U, Gabrielsen GW. 2007. Trends of perfluorinated alkyl substances in herring gull
eggs from two coastal colonies in northern Norway: 1983-2003. Environ Sci Technol 41:66716677.
Verreault J, Houde M, Gabrielsen GW, et al. 2005. Perfluorinated alkyl substances in plasma, liver,
brain, and eggs of glaucous gulls (Larus hyperboreus) from the Norwegian Arctic. Environ Sci
Technol 39:7439-7445.
Vested A, Ramlau-Hansen CH, Olsen SF, et al. 2013. Associations of in utero exposure to
perfluorinated alkyl acids with human semen quality and reproductive hormones in adult men.
Environ Health Perspect 121(4):453-458.
Vestergaard S, Nielsen F, Andersson AM, et al. 2012. Association between perfluorinated compounds
and time to pregnancy in a prospective cohort of Danish couples attempting to conceive. Hum
Reprod 27(3):873-880.
Vestergren R, Cousins IT. 2009. Tracking the pathways of human exposure to perfluorocarboxylates.
Environ Sci Technol 43(15):5565-5575.
Vestergren R, Cousins IT, Trudel D, et al. 2008. Estimating the contribution of precursor compounds in
consumer exposure to PFOS and PFOA. Chemosphere 73(10):1617-1624.
Vesterholm Jensen D, Christensen J, Virtanen HE, et al. 2014. No association between exposure to
perfluorinated compounds and congenital cryptorchidism: A nested case-control study among 215
boys from Denmark and Finland. Reproduction 147(4):411-417.
+Vetvicka V, Vetvickova J. 2013. Reversal of perfluorooctanesulfonate-induced immunotoxicity by a
glucan-resverarol-vitamin C combination. Oriental Pharmacy and Experimental Medicine 13(1):7784.
+Viberg H, Lee I, Eriksson P. 2013. Adult dose-dependent behavioral and cognitive disturbances after a
single neonatal PFHxS dose. Toxicology 304:185-191.
Vieira VM, Hoffman K, Shin M, et al. 2013. Perfluorooctanoic acid exposure and cancer outcomes in a
contaminated community: A geographic analysis. Environ Health Perspect 121(3):318-323.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

692
8. REFERENCES

Völkel W, Genzel-Boroviczeny O, Demmelmair H, et al. 2008. Perfluorooctane sulphonate (PFOS) and
perfluorooctanoic acid (PFOA) in human breast milk: Results of a pilot study. Int J Hyg Environ
Health 211:440-446.
von Ehrenstein OS, Fenton SE, Kato K, et al. 2009. Polyfluoroalkyl chemicals in the serum and milk of
breastfeeding women. Reprod Toxicol 27(3-4):239-245.
Vuong AM, Yolton K, Webster GM, et al. 2016. Prenatal polybrominated diphenyl ether and
perfluoroalkyl substance exposures and executive function in school-age children. Environ Res
147:556-564. 10.1016/j.envres.2016.01.008.
Wallace K, Kissling G, Melnick R, et al. 2013. Structure-activity relationships for perfluoroalkaneinduced in vitro interference with rat liver mitochondrial respiration. Toxicol Lett 222(3):257-264.
Wallington TJ, Hurley MD, Xia J, et al. 2006. Formation of C7F15COOH (PFOA) and other
perfluorocarboxylic acids during the atmospheric oxidation of 8:2 fluorotelomer alcohol. Environ
Sci Technol 40:924-930.
Walters M, Bjork J, Wallace K. 2009. Perfluorooctanoic acid stimulated mitochondrial biogenesis and
gene transcription in rats. Toxicology 264(1):10-15.
Wambaugh JF, Barton HA, Setzer RW. 2008. Comparing models for perfluorooctanoic acid
pharmacokinetics using Bayesian analysis. J Pharmacokinet Pharmacodyn 35(6):683-712.
10.1007/s10928-008-9108-2.
Wambaugh JF, Setzer RW, Pitruzzello AM, et al. 2013. Dosimetric anchoring of in vivo and in vitro
studies for perfluorooctanoate and perfluorooctanesulfonate. Toxicol Sci 136(2):308-327.
10.1093/toxsci/kft204.
Wan HT, Mruk DD, Wong CK, et al. 2014a. Perfluorooctanesulfonate (PFOS) perturbs male rat Sertoli
cell blood-testis barrier function by affecting F-actin organization via p-FAK-Tyr(407): An in vitro
study. Endocrinology 155(1):249-262. 10.1210/en.2013-1657.
+Wan HT, Zhao YG, Wong MH, et al. 2011. Testicular signaling is the potential target of
perfluorooctanesulfonate-mediated subfertility in male mice. Biol Reprod 84(5):1016-1023.
+Wan HT, Zhao YG, Leung PY, et al. 2014b. Perinatal exposure to perfluorooctane sulfonate affects
glucose metabolism in adult offspring. PLoS ONE 9(1):e87137. 10.1371/journal.pone.0087137.
Wang IJ, Hsieh W-S, Chen C-Y, et al. 2011. The effect of prenatal perfluorinated chemicals exposures
on pediatric atopy. Environ Res 111(6):785-791.
Wang J, Zhang Y, Zhang W, et al. 2012. Association of perfluorooctanoic acid with HDL cholesterol
and circulating miR-26b and miR-199-3p in workers of a fluorochemical plant and nearby residents.
Environ Sci Technol 46(17):9274-9281.
Wang J, Yan S, Zhang W, et al. 2015a. Integrated proteomic and miRNA transcriptional analysis reveals
the hepatotoxicity mechanism of PFNA exposure in mice. J Proteome Res 14(1):330-341.
10.1021/pr500641b.
Wang N, Szostek B, Buck RC, et al. 2005a. Fluorotelomer alcohol biodegradation–direct evidence that
perfluorinated carbon chains breakdown. Environ Sci Technol 39:7516-7528.
Wang N, Szostek B, Folsom PW, et al. 2005b. Aerobic biotransformation of 14c-labeled 8-2 telomer B
alcohol by activated sludge from a domestic sewage treatment plant. Environ Sci Technol 39:531538.
Wang Y, Adgent M, Su PH, et al. 2016. Prenatal exposure to perfluorocarboxylic acids (PFCAs) and
fetal and postnatal growth in the Taiwan maternal and infant cohort study. Environ Health Perspect
10.1289/ehp.1509998.
Wang Y, Liu W, Zhang Q, et al. 2015c. Effects of developmental perfluorooctane sulfonate exposure on
spatial learning and memory ability of rats and mechanism associated with synaptic plasticity. Food
Chem Toxicol 76:70-76. 10.1016/j.fct.2014.12.008.
Wang Y, Rogan WJ, Chen PC, et al. 2014. Association between maternal serum perfluoroalkyl
substances during pregnancy and maternal and cord thyroid hormones: Taiwan maternal and infant
cohort study. Environ Health Perspect 122(5):529-534. 10.1289/ehp.1306925.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

693
8. REFERENCES

Wang Y, Rogan WJ, Chen HY, et al. 2015b. Prenatal exposure to perfluroalkyl substances and children's
IQ: The Taiwan maternal and infant cohort study. Int J Hyg Environ Health 218(7):639-644.
10.1016/j.ijheh.2015.07.002.
Wang Y, Starling AP, Haug LS, et al. 2013a. Association between perfluoroalkyl substances and thyroid
stimulating hormone among pregnant women: A cross-sectional study. Environ Health 12(1):76.
10.1186/1476-069x-12-76.
Wang Y, Yeung LWY, Taniyasu S, et al. 2008. Perfluorooctane sulfonate and other fluorochemicals in
waterbird eggs from south China. Environ Sci Technol 42(21):8146-8151.
Wang Z, Cousins IT, Scheringer M, et al. 2013b. Fluorinated alternatives to long-chain perfluoroalkyl
carboxylic acids (PFCAs), perfluoroalkane, sulfonic acids (PFSAs) and their potential precursors.
Environ Int 60:242-248.
Wania F. 2007. A global mass balance analysis of the source of perfluorocarboxylic acids in the Arctic
Ocean. Environ Sci Technol 41:4529-4535.
Washburn ST, Bingman TS, Braitwaite SK, et al. 2005. Exposure assessment and risk characterization
for perfluorooctanoate in selected consumer articles. Environ Sci Technol 39:3904-3910.
Washington JW, Ellington J, Jenkins TM, et al. 2009. Degradability of an acrylate-linked, fluorotelomer
polymer in soil. Environ Sci Technol 43(17):6617-6623.
Washington JW, Jenkins TM. 2015. Abiotic hydrolysis of fluorotelomer-based polymers as a source of
perfluorocarboxylates at the global scale. Environ Sci Technol 49(24):14129-14135.
10.1021/acs.est.5b03686.
Washington JW, Jenkins TM, Rankin K, et al. 2015. Decades-scale degradation of commercial, sidechain, fluorotelomer-based polymers in soils and water. Environ Sci Technol 49(2):915-823.
Washino N, Saijo Y, Sasaki S, et al. 2009. Correlations between prenatal exposure to perfluorinated
chemicals and reduced fetal growth. Environ Health Perspect 117:660-667.
Watkins DJ, Josson J, Elston B, et al. 2013. Exposure to perfluoroalkyl acids and markers of kidney
function among children and adolescents living near a chemical plant. Environ Health Perspect
121(5):625-630.
Weaver YM, Ehresman DJ, Butenhoff JL, et al. 2010. Roles of rat renal organic anion transporters in
transporting perflluorinated carboxylates with different chain lengths. Toxicol Sci 113(2):305-314.
Webster GM, Rauch SA, Marie NS, et al. 2016. Cross-sectional associations of serum perfluoroalkyl
acids and thyroid hormones in U.S. adults: Variation according to TPOAb and iodine status
(NHANES 2007-2008). Environ Health Perspect 124(7):935-942. 10.1289/ehp.1409589.
Wei S, Chen LQ, Taniyasu S, et al. 2007a. Distribution of perfluorinated compounds in surface
seawaters between Asia and Antarctica. Mar Pollut Bull 54:1813-1838.
Wei Y, Dai J, Liu M, et al. 2007b. Estrogen-like properties of perfluorooctanoic acid as revealed by
expressing hepatic estrogen-responsive genes in rare minnows (Gobiocyris rarus). Environ Toxicol
Chem 26(11):2440-2447.
Wen LL, Lin LY, Su TC, et al. 2013. Association between serum perfluorinated chemicals and thyroid
function in U.S. adults: The national health and nutrition examination survey 2007-2010. J Clin
Endocrinol Metab 98(9):E1456-1464. 10.1210/jc.2013-1282.
+White SS, Calafat AM, Kuklenyik Z, et al. 2007. Gestational PFOA exposure of mice is associated
with altered mammary gland development in dams and female offspring. Toxicol Sci 96(1):133144.
+White SS, Kato K, Jia LT, et al. 2009. Effects of perfluorooctanoic acid on mouse mammary gland
development and differentiation resulting from cross-foster and restricted gestational exposures.
Reprod Toxicol 27(3-4):289-298.
+White SS, Stanko JP, Kato K, et al. 2011. Gestational and chronic low-dose PFOA exposures and
mammary gland growth and differentiation in three generations of CD-1 mice. Environ Health
Perspect 119(8):1070-1076.
Whitworth KW, Haug LS, Baird DD, et al. 2012a. Perfluorinated compounds in relation to birth weight
in the Norwegian Mother and Child Cohort Study. Am J Epidemiol 175(12):1209-1216.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

694
8. REFERENCES

Whitworth KW, Haug LS, Baird DD, et al. 2012b. Perfluorinated compounds and subfecundity in
pregnant women. Epidemiology 23(2):257-263.
WHO. 2010. Guidelines for indoor air quality: Selected pollutants. Geneva, Switzerland: World Health
Organization. http://www.euro.who.int/__data/assets/pdf_file/0009/128169/e94535.pdf. January
08, 2014.
WHO. 2017. Guidelines for drinking-water quality. Geneva, Switzerland: World Health Organization.
http://apps.who.int/iris/bitstream/10665/254637/1/9789241549950-eng.pdf?ua=1. February 28,
2017.
Wilhelm M, Kraft M, Rauchfuss K, et al. 2008. Assessment and management of the first German case of
a contamination with perfluorinated compounds (PFC) in the Region Sauerland, North RhineWestphalia. J Toxicol Environ Health A 71:725-733.
Wilhelm M, Holzer J, Dobler L, et al. 2009. Preliminary observations on perfluorinated compounds in
plasma samples (1977-2004) of young German adults from an area with perfluorooctanoatecontaminated drinking water. Int J Hyg Environ Health 212(2):142-145.
Winquist A, Steenland K. 2014a. Modeled PFOA exposure and coronary artery disease, hypertension,
and high cholesterol in community and worker cohorts. Environ Health Perspect 122(12):12991305. 10.1289/ehp.1307943.
Winquist A, Steenland K. 2014b. Perfluorooctanoic acid exposure and thyroid disease in community and
worker cohorts. Epidemiology 25(2):255-264. 10.1097/ede.0000000000000040.
+Wolf CJ, Fenton SE, Schmid JE, et al. 2007. Developmental toxicity of perfluorooctanoic acid in the
CD-1 mouse after cross-foster and restricted gestational exposures. Toxicol Sci 95(2):462-473.
Wolf CJ, Rider CV, Lau C, et al. 2014. Evaluating the additivity of perfluoroalkyl acids in binary
combinations on peroxisome proliferator-activated receptor-alpha activation. Toxicology 316:43-54.
10.1016/j.tox.2013.12.002.
Wolf CJ, Schmid JE, Lau C, et al. 2012. Activation of mouse and human peroxisome proliferatoractivated receptor-alpha (PPARa) by perfluoroalkyl acids (PFAAs): Further investigation of C4C12 compounds. Reprod Toxicol 33:546-551.
Wolf CJ, Takacs ML, Schmid JE, et al. 2008a. Activation of mouse and human peroxisome proliferatoractivated receptor alpha by perfluoroalkyl acids of different functional groups and chain lengths.
Toxicol Sci 106(1):162-171.
+Wolf CJ, Zehr RD, Schmid JE, et al. 2010. Developmental effects of perfluorononanoic Acid in the
mouse are dependent on peroxisome proliferator-activated receptor-alpha. PPAR Res 2010
10.1155/2010/282896.
Wolf DC, Moore T, Abbott BD, et al. 2008b. Comparative hepatic effects of perfluorooctanoic acid and
WY 14,643 in PPAR-α knockout and wild-type mice. Toxicol Pathol 36(4):632-639.
10.1177/0192623308318216.
Wong F, MacLeod M, Mueller JF, et al. 2014. Enhanced elimination of perfluorooctane sulfonic acid by
menstruating women: Evidence from population-based pharmacokinetic modeling. Environ Sci
Technol 48(15):8807-8814.
Wong F, MacLeod M, Mueller JF, et al. 2015. Response to comment on "Enhanced elimination of
perfluorooctane sulfonic acid by menstruating women: Evidence from population-based
pharmacokinetic modeling." Environ Sci Technol 49(9):5838-5839.
Worley RR, Fisher J. 2015a. Application of physiologically-based pharmacokinetic modeling to explore
the role of kidney transporters in renal reabsorption of perfluorooctanoic acid in the rat. Toxicol
Appl Pharmacol 289(3):428-441. 10.1016/j.taap.2015.10.017.
Worley RR, Fisher J. 2015b. Supplementary material: Application of physiologically-based
pharmacokinetic modeling to explore the role of kidney transporters in renal reabsorption of
perfluorooctanoic acid in the rat. Toxicol Appl Pharmacol 289(3):428-441.
Worley RR, Moore SM, Tierney BC, et al. 2017a. Per- and polyfluoroalkyl substances in human serum
and urine samples from a residentially exposed community. Environ Int 106:135-143.
10.1016/j.envint.2017.06.007.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

695
8. REFERENCES

Worley RR, Yang X, Fisher J. 2017b. Physiologically based pharmacokinetic modeling of human
exposure to perfluorooctanoic acid suggests historical non drinking-water exposures are important
for predicting current serum concentrations. Toxicol Appl Pharmacol 330:9-21.
10.1016/j.taap.2017.07.001.
Woskie SR, Gore R, Steenland K. 2012. Retrospective exposure assessment of perfluorooctanoic acid
serum concentrations at a fluoropolymer manufacturing plant. Ann Occup Hyg 56(9):1025-1037.
10.1093/annhyg/mes023.
Wu LL, Gao HW, Gao NY, et al. 2009. Interaction of perfluorooctanoic acid with human serum
albumin. BMC Struct Biol 9:31.
Wu XM, Bennett DH, Calafat AM, et al. 2015. Serum concentrations of perfluorinated compounds
(PFC) among selected populations of children and adults in California. Environ Res 136:264-273.
10.1016/j.envres.2014.09.026.
+Xia W, Wan Y, Li YY, et al. 2011. PFOS prenatal exposure induce mitochondrial injury and gene
expression change in hearts of weaned SD rats. Toxicology 282(1-2):23-29.
Xiao F, Simcik MF, Halbach TR, et al. 2015. Perfluorooctane sulfonate (PFOS) and perfluorooctanoate
(PFOA) in soils and groundwater of a U.S. metropolitan area: Migration and implications for human
exposure. Water Res 72:64-74.
+Xie Y, Yang Q, Nelson BD, et al. 2003. The relationship between liver peroxisome proliferation and
adipose tissue atrophy induced by peroxisome proliferator exposure and withdrawal in mice.
Biochem Pharmacol 66(5):749-756.
+Xing J, Wang G, Zhao J, et al. 2016. Toxicity assessment of perfluorooctane sulfonate using acute and
subchronic male C57BL/6J mouse models. Environ Pollut 210:388-396.
10.1016/j.envpol.2015.12.008.
Xu J, Shimpi P, Armstrong L, et al. 2016. PFOS induces adipogenesis and glucose uptake in association
with activation of Nrf2 signaling pathway. Toxicol Appl Pharmacol 290:21-30.
+Yahia D, El-Nasser MA, Abedel-Latif M, et al. 2010. Effects of perfluorooctanoic acid (PFOA)
exposure to pregnant mice on reproduction. J Toxicol Sci 35(4):527-533.
+Yahia D, Tsukuba C, Yoshida M, et al. 2008. Neonatal death of mice treated with perfluorooctane
sulfonate. J Toxicol Sci 33(2):219-226.
Yamada A, Maeda K, Kamiyama E, et al. 2007. Multiple human isoforms of drug transporters contribute
to the hepatic and renal transport of olmesartan, a selective antagonist of the angiotensin II AT1receptor. Drug Metab Dispos 35(12):2166-2176. 10.1124/dmd.107.017459.
Yamada T, Taylor PH, Buck RC, et al. 2005. Thermal degradation of fluorotelomer treated articles and
related materials. Chemosphere 61:974-984.
Yamaguchi M, Arisawa K, Uemura H, et al. 2013. Consumption of seafood, serum liver enzymes, and
blood levels of PFOS and PFOA in the Japanese population. J Occup Health 55(3):184-194.
Yamashita N, Kannan K, Taniyasu S, et al. 2005. A global survey of perfluorinated acids in oceans.
Mar Pollut Bull 51:658-668.
Yamashita N, Taniyasu S, Petrick G, et al. 2008. Perfluorinated acids as novel chemical tracers of global
circulation of ocean waters. Chemosphere 70:1247-1255.
Yang CH, Glover KP, Han X. 2009. Organic anion transporting polypeptide (Oatp) 1a1-mediated
perfluorooctanoate transport and evidence for a renal reabsorption mechanism of Oatp1a1 in renal
elimination of perfluorocarboxylates in rats. Toxicol Lett 190(2):163-171.
Yang CH, Glover KP, Han X. 2010. Characterization of cellular uptake of perfluorooctanoate via
organic anion-transporting polypeptide 1A2, organic anion transporter 4, and urate transporter 1 for
their potential roles in mediating human renal reabsorption of perfluorocarboxylates. Toxicol Sci
117(2):294-302.
Yang L, Li J, Lai J, et al. 2016a. Placental transfer of perfluoroalkyl substances and associations with
thyroid hormones: Beijing prenatal exposure study. Sci Rep 6:21699. 10.1038/srep21699.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

696
8. REFERENCES

Yang L, Wang Z, Shi Y, et al. 2016b. Human placental transfer of perfluoroalkyl acid precursors:
Levels and profiles in paired maternal and cord serum. Chemosphere 144:1631-1638.
10.1016/j.chemosphere.2015.10.063.
+Yang Q, Abedi-Valugerdi M, Xie Y, et al. 2002a. Potent suppression of the adaptive immune response
in mice upon dietary exposure to the potent peroxisome proliferator, perfluorooctanoic acid. Int
Immunopharmacol 2(2-3):389-397.
+Yang Q, Xie Y, Alexson SE, et al. 2002b. Involvement of the peroxisome proliferator-activated
receptor alpha in the immunomodulation caused by peroxisome proliferators in mice. Biochem
Pharmacol 63(10):1893-1900.
+Yang Q, Xie Y, Depierre JW. 2000. Effects of peroxisome proliferators on the thymus and spleen of
mice. Clin Exp Immunol 122(2):219-226.
+Yang Q, Xie Y, Eriksson AM, et al. 2001. Further evidence for the involvement of inhibition of cell
proliferation and development in thymic and splenic atrophy induced by the peroxisome proliferator
perfluorooctanoic acid in mice. Biochem Pharmacol 62(8):1133-1140.
Yao PL, Ehresman DJ, Rae JM, et al. 2014. Comparative in vivo and in vitro analysis of possible
estrogenic effects of perfluorooctanoic acid. Toxicology 326:62-73. 10.1016/j.tox.2014.10.008.
Yao X, Zhong L. 2005. Genotoxic risk and oxidative DNA damage in HepG2 cells exposed to
perfluorooctanoic acid. Mutat Res 587:38-44.
Yarwood G, Kemball-Cook S, Keinath M, et al. 2007. High-resolution atmospheric modeling of
fluorotelomer alcohols and perfluorocarboxylic acids in the North American troposphere. Environ
Sci Technol 41:5756-5762.
Ye X, Kato K, Wong LY, et al. 2018a. Per- and polyfluoroalkyl substances in sera from children 3 to 11
years of age participating in the National Health and Nutrition Examination Survey 2013-2014. Int J
Hyg Environ Health 221(1):9-16. 10.1016/j.ijheh.2017.09.011.
Ye X, Kato K, Wong LY, et al. 2018b. Supporting information to "Per- and polyfluoroalkyl substances
in sera from children 3 to 11 years of age participating in the National Health and Nutrition
Examination Survey 2013-2014" [Int J Hyg Environ Health 221(1):9-16.
10.1016/j.ijheh.2017.09.011]. Int J Hyg Environ Health
https://www.sciencedirect.com/science/article/pii/S1438463917305886#sec0045. December 19,
2017.
Ye X, Strynar MJ, Nakayama SF, et al. 2008. Perfluorinated compounds in whole fish homogenates
from the Ohio, Missouri, and Upper Mississippi Rivers, USA. Environ Pollut 156(3):1227-1232.
10.1016/j.envpol.2008.03.014.
Yeung LW, Robinson SJ, Koschorreck J, et al. 2013. Part II. A temporal study of PFOS and its
precursors in human plasma from two German cities in 1982-2009. Environ Sci Technol
47(8):3875-3882. 10.1021/es4004153.
Ylinen M, Auriola S. 1990. Tissue distribution and elimination of perfluorodecanoic acid in the rat after
single intraperitoneal administration. Pharmacol Toxicol 66:45-48.
Ylinen M, Kojo A, Hanhijarvi H, et al. 1990. Disposition of perfluorooctanoic acid in the rat after single
and subchronic administration. Bull Environ Contam Toxicol 44:46-53.
Yoo H, Washington JW, Jenkins TM, et al. 2011. Quantitative determination of perfluorochemicals and
flurotelomer alcohols in plants from biosolid-amended fields using LC/MS/MS and GC/MS.
Environ Sci Technol 45.
Young CJ, Furdui VI, Franklin J, et al. 2007. Perfluorinated acids in Arctic snow: New evidence for
atmospheric formation. Environ Sci Technol 41:3455-3461.
Yu J, Hu J, Tanaka S, et al. 2009c. Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid
(PFOA) in sewage treatment plants. Water Res 43(9):2399-2408.
+Yu WG, Liu W, Jin YH. 2009a. Effects of perfluorooctane sulfonate on rat thyroid hormone
biosynthesis and metabolism. Environ Toxicol Chem 28(5):990-996.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

697
8. REFERENCES

Yu WG, Liu W, Jin YH, et al. 2009b. Prenatal and postnatal impact of perfluorooctane sulfonate (PFOS)
on rat development: A cross-foster study on chemical burden and thyroid hormone system. Environ
Sci Technol 43(21):8416-8422.
Yuan G, Peng H, Huang C, et al. 2016. Ubiquitous occurrence of fluorotelomer alcohols in eco-friendly
paper-made food-contact materials and their implication for human exposure. Environ Sci Technol
50:942-950.
Zeng H-c, Zhang L, Li Y-y, et al. 2011. Inflammation-like glial response in rat brain induced by prenatal
PFOS exposure. Neurotoxicology 32(1):130-139.
Zeng XW, Qian Z, Emo B. 2015. The association between perfluoroalkyl chemicals and serum lipid
levels in children. Sci Total Environ 512-513:364-370
Zhang C, Sundaram R, Maisog J, et al. 2015a. A prospective study of prepregnancy serum
concentrations of perfluorochemicals and the risk of gestational diabetes. Fertil Steril 103(1):184189.
Zhang H, Shi Z, Liu Y, et al. 2008. Lipid homeostasis and oxidative stress in the liver of male rats
exposed to perfluorododecanoic acid. Toxicol Appl Pharmacol 227:16-25.
Zhang L, Ren X-M, Wan B, et al. 2014. Structure-dependent binding and activation of perfluorinated
compounds on human peroxisome proliferator-activated receptor γ. Toxicol Appl Pharmacol
279(3):275-283.
Zhang T, Sun H, Lin Y, et al. 2011. Perfluorinated compounds in human blood, water, edible freshwater
fish, and seafood in China: Daily intake and regional differences in human exposures. J Agric Food
Chem 59:11168-11176.
Zhang T, Sun H, Qin X, et al. 2015b. PFOS and PFOA in paired urine and blood from general adults and
pregnant women: Assessment of urinary elimination. Environ Sci Pollut Res Int 22(7):5572-5579.
10.1007/s11356-014-3725-7.
Zhang X, Chen L, Fei XC, et al. 2009. Binding of PFOS to serum albumin and DNA: Insight into the
molecular toxicity of perfluorochemicals. BMC Mol Biol 10:16.
Zhang X, Lohmann R, Dassuncao C, et al. 2016. Source attribution of poly-and perfluoroalkyl
substances (PFASs) in surface waters from Rhode Island and New York metropolitan area. Environ
Sci Technol Lett 3:316-321. 10.1021/acs.estlett.6b00255.
Zhang Y, Beesoon S, Zhu L, et al. 2013. Biomonitoring of perfluoroalkyl acids in human urine and
estimates of biological half-life. Environ Sci Technol 47(18):10619-10627. 10.1021/es401905e.
Zhao G, Wang J, Wang X, et al. 2011. Mutagenicity of PFOA in mammalian cells: Role of
mitochondria-dependent reactive oxygen species. Environ Sci Technol 45(4):1638-1644.
Zhao H, Qu B, Guan Y, et al. 2016. Influence of salinity and temperature on uptake of perfluorinated
carboxylic acids (PFCAs) by hydroponically grown wheat (Triticum aestivum L.). Springer Plus
5:541.
+Zheng L, Dong GH, Jin YH, et al. 2009. Immunotoxic changes associated with a 7-day oral exposure
to perfluorooctanesulfonate (PFOS) in adult male C57BL/6 mice. Arch Toxicol 83(7):679-689.
Zhu Y, Qin XD, Zeng XW, et al. 2016. Associations of serum perfluoroalkyl acid levels with T-helper
cell-specific cytokines in children: By gender and asthma status. Sci Total Environ 559:166-173.
10.1016/j.scitotenv.2016.03.187.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-1

APPENDIX A. ATSDR MINIMAL RISK LEVEL WORKSHEETS
MRLs are derived when reliable and sufficient data exist to identify the target organ(s) of effect or the
most sensitive health effect(s) for a specific duration for a given route of exposure. An MRL is an
estimate of the daily human exposure to a hazardous substance that is likely to be without appreciable risk
of adverse noncancer health effects over a specified route and duration of exposure. MRLs are based on
noncancer health effects only; cancer effects are not considered. These substance-specific estimates,
which are intended to serve as screening levels, are used by ATSDR health assessors to identify
contaminants and potential health effects that may be of concern at hazardous waste sites. It is important
to note that MRLs are not intended to define clean-up or action levels.

MRLs are derived for hazardous substances using the NOAEL/uncertainty factor approach. They are
below levels that might cause adverse health effects in the people most sensitive to such chemicalinduced effects. MRLs are derived for acute (1–14 days), intermediate (15–364 days), and chronic
(≥365 days) durations and for the oral and inhalation routes of exposure. Currently, MRLs for the dermal
route of exposure are not derived because ATSDR has not yet identified a method suitable for this route
of exposure. MRLs are generally based on the most sensitive substance-induced endpoint considered to
be of relevance to humans. Serious health effects (such as irreparable damage to the liver or kidneys, or
birth defects) are not used as a basis for establishing MRLs. Exposure to a level above the MRL does not
mean that adverse health effects will occur.

MRLs are intended only to serve as a screening tool to help public health professionals decide where to
look more closely. They may also be viewed as a mechanism to identify those hazardous waste sites that
are not expected to cause adverse health effects. Most MRLs contain a degree of uncertainty because of
the lack of precise toxicological information on the people who might be most sensitive (e.g., infants,
elderly, nutritionally or immunologically compromised) to the effects of hazardous substances. ATSDR
uses a conservative (i.e., protective) approach to address this uncertainty consistent with the public health
principle of prevention. Although human data are preferred, MRLs often must be based on animal studies
because relevant human studies are lacking. In the absence of evidence to the contrary, ATSDR assumes
that humans are more sensitive to the effects of hazardous substance than animals and that certain persons
may be particularly sensitive. Thus, the resulting MRL may be as much as 100-fold below levels that
have been shown to be nontoxic in laboratory animals.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-2
APPENDIX A

Proposed MRLs undergo a rigorous review process: Health Effects/MRL Workgroup reviews within the
Division of Toxicology and Human Health Sciences, expert panel peer reviews, and agency-wide MRL
Workgroup reviews, with participation from other federal agencies and comments from the public. They
are subject to change as new information becomes available concomitant with updating the toxicological
profiles. Thus, MRLs in the most recent toxicological profiles supersede previously published MRLs.
For additional information regarding MRLs, please contact the Division of Toxicology and Human
Health Sciences, Agency for Toxic Substances and Disease Registry, 1600 Clifton Road NE, Mailstop
F-57, Atlanta, Georgia 30329-4027.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-3
APPENDIX A

INTRODUCTION
Overview of Epidemiological Studies
A large number of epidemiology studies have evaluated a wide range of potential health outcomes
resulting from exposure to perfluoroalkyls, particularly PFOA and PFOS. The epidemiology studies fall
into three broad categories: occupational exposure primarily to airborne PFOA and PFOS, exposure to
PFOA-contaminated drinking water by residents living near a PFOA production facility, and general
population exposure to background levels of perfluoroalkyls. Most of the occupational exposure studies
were conducted in workers at four facilities in Minnesota, Alabama, West Virginia, and the Netherlands.
Studies of the highly-exposed residents primarily come from several large-scale studies (C8 Health
Project, C8 Health Study) of Mid-Ohio Valley residents living near the Washington Works facility in
West Virginia who were exposed to high levels of PFOA in the drinking water. General population
studies primarily utilized data collected in NHANES in the United States and several large-scale health
studies conducted in Europe.
Most of the epidemiology studies lack environmental monitoring data and there is a potential for multiple
sources of exposure (inhalation and oral). However, the majority of the epidemiology studies used serum
perfluoroalkyl levels as a biomarker of exposure. One limitation of the C8 Health Studies is that they
used blood samples collected in 2005–2006. However, the facility started using PFOA in the 1950s and
peak usage was in the 1990s and by 2003, there was an 87% decline in PFOA emissions, as compared to
1999 levels (Emmett et al. 2006a). Therefore, serum PFOA levels measured in 2005–2006 likely do not
represent earlier higher exposures, which may have contributed to observed health outcomes. As an
alternative to using older serum PFOA levels, several C8 Health Studies estimated serum levels based on
data on the release of PFOA from the facility and pharmacokinetic modeling. Of the three categories of
subjects examined in the epidemiology studies, workers have the highest potential exposure to
perfluoroalkyls, followed by the highly-exposed residents in the Mid-Ohio Valley (referred to as
community exposure), and then the general population. In one study of workers at the Washington Works
facility in West Virginia, the average serum PFOA level in 2001–2004 was 1,000 ng/mL (Sakr et al.
2007a); the mean PFOA level in community residents (without occupational exposure) near this facility
was 423 ng/mL in 2004–2005 (Emmett et al. 2006a). By comparison, the geometric mean concentration
of PFOA in the U.S. population was 3.92 ng/mL in 2005–2006 (CDC 2013).
Identification of Adverse Health Effects Based on Epidemiological Studies—Weight-ofEvidence Approach. Although a large number of epidemiology studies have examined the potential
of perfluoroalkyl compounds to induce adverse health effects, most of the studies were cross-sectional in
design and do not establish causality. Epidemiology studies have found statistically significant
associations between serum perfluoroalkyl levels and several health effects, although the results were not
consistent across studies. Many of the studies reported dose-related trends, but these trends were not as
apparent when comparing across studies; some effects were observed in populations with background
PFOA levels but not in populations with high serum PFOA levels. Given the inconsistencies, a weightof-evidence approach was used to evaluate whether the available data supported a link between
perfluoroalkyl exposure and a particular health effect, taking into consideration the consistency of the
findings across studies, the quality of the studies, dose-response, and plausibility. It should be noted that
although the data may provide strong evidence for an association, it does not imply that the observed
effect is biologically relevant because the magnitude of the change is within the normal limits or not
indicative of an adverse health outcome. Plausibility depends primarily on experimental toxicology
studies that establish a biological mechanism for the observed effects.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-4
APPENDIX A

Using this weight-of-evidence approach, the available epidemiology data identify several potential health
hazards of PFOA, PFOS, PFHxS, PFNA, and PFDeA in humans as listed below.
PFOA
• Pregnancy-induced hypertension/pre-eclampsia
• Liver damage, as evidenced by increases in serum enzymes and decreases in serum bilirubin
levels
• Increases in serum lipids, particularly total cholesterol and LDL cholesterol
• Increased risk of thyroid disease
• Decreased antibody response to vaccines
• Increased risk of asthma diagnosis
• Increased risk of decreased fertility
• Small (<20 g or 0.7 ounces per 1 ng/mL increase in blood perfluoroalkyl level) decreases in birth
weight
PFOS
• Pregnancy-induced hypertension/pre-eclampsia
• Liver damage, as evidenced by increases in serum enzymes and decreases in serum bilirubin
levels
• Increases in serum lipids, particularly total cholesterol and LDL cholesterol
• Increased risk of thyroid disease
• Decreased antibody response to vaccines
• Increased risk of decreased fertility
• Small (<20 g or 0.7 ounces per 1 ng/mL increase in blood perfluoroalkyl level) decreases in birth
weight
PFHxS
• Liver damage, as evidenced by increases in serum enzymes and decreases in serum bilirubin
levels
• Decreased antibody response to vaccines
PFNA
• Increases in serum lipids, particularly total cholesterol and LDL cholesterol
• Decreased antibody response to vaccines
PFDeA
• Increases in serum lipids, particularly total cholesterol and LDL cholesterol
• Decreased antibody response to vaccines
Limitations of Epidemiological Data. There are sufficient epidemiology data to identify possible
sensitive targets for many of the perfluoroalkyls; however, there are two major limitations to establishing
dose-response relationships for these effects and using the epidemiology studies to derive MRLs:
accurate identification of environmental exposure levels producing increased risk for adverse effects
(exposure estimates and routes of exposure) and likely co-exposure to mixtures of perfluoroalkyls.
Uncertainty in Exposure Estimates. In general, the epidemiology studies provide a one-time serum
perfluoroalkyl concentration, but lack information on actual environmental exposure concentration or
doses, routes of exposure, and exposure duration. Although serum perfluoroalkyl levels provide reliable
information on recent exposure (weeks to years, depending on the elimination t1/2 for the perfluoroalkyl),

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-5
APPENDIX A

they likely do not reflect historical exposure levels or exposure levels at the onset of the effect. This is
especially true for occupational exposure cohorts where past exposure levels were higher before industrial
hygiene improved and in the C8 community studies since peak PFOA levels in drinking water occurred at
least 10 years prior to the onset of the studies. Additionally, data from NHANES suggest that most
perfluoroalkyl levels are declining in the general population; for example, the geometric mean serum
levels of PFOA and PFOS declined from 5.21 and 30.4 ng/mL, respectively, in 1999–2000 to 2.08 and
6.31 ng/mL in 2011–2012. In contrast, levels of PFNA have increased during that time frame; the
geometric mean went from 0.551 ng/mL in 1999–2000 to 1.26 ng/mL in 2009–2010 and then decreased
to 0.881 ng/mL in 2011–2012. Most studies do not provide adequate information to determine whether
perfluoroalkyl levels reflect a steady state and relatively constant exposure, since most designs only
include a single measurement. An added uncertainty occurs in studies that used maternal serum levels as
the biomarker of exposure for effects in children or for effects on fertility.
It is assumed that workers were primarily exposed via inhalation; however, oral exposure may have also
contributed to the total perfluoroalkyl body burden, particularly since workers frequently lived in
communities with elevated levels of PFOA in the drinking water. It has been determined that drinking
water was the primary source of perfluoroalkyls in residents living near a PFOA facility (Emmett et al.
2006a); however, it is likely that airborne PFOA contributed to overall body burden. Drinking water is
the likely primary route of exposure for the general population.
Uncertainty due to Co-Exposure to Other Perfluoroalkyl Compounds. Based on NHANES data, the
U.S. general population is exposed to a variety of perfluoroalkyl compounds. A number of studies
reported a high degree of correlation between perfluoroalkyl compounds; however, most studies did not
control for exposure to other perfluoroalkyls. Given that many of the perfluoroalkyl compounds have
similar targets of toxicity, it is likely that several perfluoroalkyls contributed to the observed effects. The
potential interactions between different perfluoroalkyl compounds have not been fully elucidated.
In summary, the epidemiology databases for several perfluoroalkyls provide valuable information on
hazard identification; however, uncertainties regarding doses associated with adverse effects and possible
interactions between compounds preclude use of these data to derive MRLs.
Overview of Laboratory Animal Studies
Laboratory animal studies are available for 11 perfluoroalkyl compounds; however, more than 80% of the
studies examined PFOA and/or PFOS. The laboratory animal studies primarily involved oral exposure
and examined a wide range of potential health outcomes. The primary health effects observed in
laboratory animals were liver toxicity, developmental toxicity, and immune toxicity. Other effects
typically observed at higher doses included weight loss, histological alterations in reproductive tissues,
and histological alterations in the thyroid gland. The sensitive targets of toxicity identified in the
laboratory animals are similar to those observed in epidemiology studies.
Limitations of Laboratory Animal Studies for Derivation of MRLs. Use of controlled animal
studies eliminates the uncertainties regarding effective doses and co-exposure to other perfluoroalkyl
compounds. However, there are uncertainties associated with derivation of MRLs based on animal
studies, in part, because of large interspecies differences in the toxicokinetics of perfluoroalkyls for which
mechanisms are not completely understood. Available information on the toxicokinetics of
perfluoroalkyls in humans, nonhuman primates, and various rodent species indicate that elimination rates
(and very likely elimination mechanisms and hormonal regulation of these mechanisms) vary
substantially across chemical species (i.e., carbon chain length) and animal species (i.e., slower in humans
compared to nonhuman primates and rodents), and show pronounced sex differences within certain
species (e.g., faster elimination in female rats). As a result, there is some uncertainty associated with

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-6
APPENDIX A

extrapolation of external dose-response relationships from animals to humans. Although progress has
been made in modeling toxicokinetics of PFOA and PFOS in rats and nonhuman primates, no models for
humans have been developed to simulate the substantial differences in toxicokinetics of these compounds
between humans and nonhuman primates or between humans and rats (Andersen et al. 2006; Tan et al.
2008). An additional uncertainty in the animal data is the relevance of effects associated with activation
of PPARα. Many of the effects observed in rodents, particularly liver and developmental effects, involve
the activation of PPARα; humans and nonhuman primates are less responsive to PPARα agonists than rats
and mice. However, studies in PPARα-null mice suggest that PPARα-independent mechanisms also play
a role in the liver, immunological, and developmental toxicity.
MRL Approach
The following approach was used for derivation of MRLs:
• Identify sensitive endpoints from epidemiology studies
• Identify laboratory animal studies that have evaluated dose-response relationships for toxicity
targets identified in epidemiology studies
• Estimate a point of departure (POD) using animal serum perfluoroalkyl levels for sensitive
endpoints
• Calculate human equivalent doses (HEDs) using the assumption that a serum concentration
resulting in an effect in a laboratory animal would also result in an effect in humans. An
empirical pharmacokinetic model was used to estimate a human dose associated with this serum
concentration for PFOA and PFOS. Measured serum concentrations in laboratory animal studies
were used to calculate the HEDs for PFHxS and PFNA.
• Apply appropriate uncertainty factors informed by comparison of the POD to serum
perfluoroalkyl levels reported in epidemiology studies
Predicting Mean Serum PFOA and PFOS Concentrations in Laboratory Animals. Timeweighted average (TWA) serum concentrations corresponding to external doses (mg/kg/day) and
exposure durations (days) were predicted with a pharmacokinetic (PK) model for the animal species,
strain, and sex used in the studies (Wambaugh et al. 2013). The TWA serum concentration was
calculated as follows (Equation A-1):

CTWA =

C AUC
ED

Eq. (A-1)

where CTWA is the predicted TWA serum concentration (mg/L), CAUC is the predicted area under the curve
(AUC) of the serum concentration-time profile for the exposure (mg hour/L), and ED is the exposure
duration (hour). Gavage studies were simulated as a single dose (e.g., gavage) given once every 24 hours.
Daily drinking water exposures of were simulated as 12 hourly doses of 1/12 of the total daily dose,
followed by 12 hours with no dosing. This assumes that the animals consumed water during a 12-hour
active period.
The Wambaugh et al. (2013) model was originally implemented in R (v2.10.0) and was migrated to
MATLAB (vR2016) for calculations of MRLs. Wambaugh et al. (2013) reported mean and confidence
limits for parameter values estimated from a Bayesian Markov Chain Monte Carlo (MCMC) analysis.
The posterior means were used as point estimates for parameters in the MATLAB version. Function of
the point estimate implementation in MATLAB was verified by comparing predictions of CAUC obtained
from the MATLAB version with predictions from the MCMC analysis reported in EPA (2016e, 2016f).
This comparison for PFOA included a total of 18 predictions of CAUC for female CD-1 mice (Lau et al.
2006; Wolf et al. 2007), female C57Bl6 mice (DeWitt et al. 2008), and male Sprague-Dawley rats

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-7
APPENDIX A

(Butenhoff et al. 2004b). The r2 for MATLAB vs R predictions of CAUC was 0.99 and the average relative
percent difference (MATLAB - R) was 2.8% (range: -6.6–13.5). The comparison for PFOS included a
total of 28 predictions of CAUC for female CD-1 mice (Lau et al. 2003), female Sprague-Dawley rats
(Butenhoff et al. 2009b; Lau et al. 2003; Luebker et al. 2005a, 2005b), and male and female Cynomolgus
monkeys (Seacat et al. 2002). The r2 for MATLAB vs R predictions of CAUC was 1.00 and the average
relative percent difference (MATLAB - R) was 4.6% (range: -11–20).
The Wambaugh et al. (2013) model was selected over other available pharmacokinetic models because it
provided sex- and strain-specific parameters for rats, mice, and monkeys, which allowed for comparisons
of HEDs across strains and species.
Estimating TWA Serum PFHxS and PFNA Concentrations in Laboratory Animals. Because
a PK model for predicting the TWA serum concentrations was not identified for PFHxS and PFNA, a
TWA serum concentration was estimated from measured serum concentrations. ATSDR estimated the
TWA values from the areas under the curve calculated using the trapezoid rule. Since most studies did
not report pre-exposure levels, serum concentrations in the control group were used as the baseline
concentration.
Estimating HEDs for Perfluoroalkyls. The serum concentration PODs identified from the
laboratory animal data were converted to an equivalent dose in humans, which is defined as the
continuous ingestion dose (mg/kg/day) that would result in steady-state serum concentrations of
perfluoroalkyl equal to the serum concentration (µg/mL) selected as the POD.
The relationship between perfluoroalkyl external dosage (mg/kg/day) and steady-state serum
concentration (Css, mg/L) in humans was estimated assuming a single-compartment first-order model in
which elimination kinetics are adequately represented by observed serum elimination t1/2 values for the
specific perfluoroalkyl compound, an assumed apparent volume of distribution (Vd, L/kg) and
gastrointestinal absorption fraction. In the first-order single-compartment model, continuous exposure
will result in a steady-state body burden (BBSS, mg/kg) for PFOA or PFOS, which will be distributed in a
single volume of distribution to yield a steady-state serum concentration (Equation A-2):

C SS =

BBSS
Vd

Eq. (A-2)

At steady state, the rate of first-order elimination rate (a constant fraction of the body burden, ke per day)
will equal the absorbed dosage (Dss, mg/kg/day) adjusted for gastrointestinal absorption (AF)
(Equation A-3):

DSS ⋅ AF = BBSS ⋅ ke

Eq. (A-3)

Rearrangement of Equation A-3 allows calculation of the steady-state body burden corresponding to a
given external dosage (Equation A-4):

BBSS =

DSS ⋅ AF
ke

Eq. (A-4)

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-8
APPENDIX A

The relationship between the elimination rate constant (ke, day-1) and the elimination half-life (t1/2, day), is
given in Equation A-5:

ke =

ln(2)
t1 / 2

Eq. (A-5)

Combining Equations A-2 and A-3 yields an expression relating the external steady-state dosage and
steady-state serum concentration (Equation A-6):

DSS =

CSS ⋅ ke ⋅ Vd
AF

Eq. (A-6)

The above estimates of CSS/DSS are sensitive to the input parameters, t1/2, AF, and Vd.
PFOA and PFOS. Several studies have estimated PFOA and PFOS half-lives (t1/2) in workers (Costa et
al. 2009; Olsen et al. 2007a) or highly exposed residents (Bartell et al. 2010). Estimates of the half-lives
based on Olsen et al. (2007a) were derived from longitudinal measurements of serum concentrations of
PFOA and PFOS in a group of fluorochemical production workers (24 males, 2 females); the estimated
half-lives were 3.8 years (1,387 days) and 5.4 years (1,971 days), respectively. Costa et al. (2009)
reported a half-life for PFOA of 5.1 years (1,862 days) for a group of workers (n=16) following their
cessation of PFOA production work. A longitudinal study by Bartell et al. (2010) followed serum PFOA
concentrations in 200 subjects recruited from the Lubeck Public Service District and Little Hocking
Water Association and followed for a period of 6–12 months after mitigation of exposures from drinking
water. The estimated half-life for PFOA was 2.3 years (840 days). A fourth study estimated half-lives in
a cross-sectional study of residents served by the Lubeck Public Service District and Little Hocking
Water Association (Seals et al. 2011). The estimated half-lives ranged from 2.9 to 10.1 years (1,059–
3,687 days) for PFOA. Results from the longitudinal studies are shown in Table A-1. For the MRL
calculations, the PFOA half-life estimated by Olsen et al. (2007a) was selected over the half-life
estimated by Bartell et al. (2010) because the Olsen et al. (2007a) study had a longer follow-up time
(>5 years compared to 6–12 months) and estimates of the terminal half-life appear to increase with longer
follow-ups because slower kinetics make a larger contribution to the terminal half-life (Seals et al. 2011).
Estimates of the half-life for PFOA and PFOS are most applicable to serum concentrations within the
above ranges and would be less certain if applied to serum concentrations substantially below or above
these range. Serum concentrations during the 5-year observation period in the Olsen et al. (2007a) study
are provided in Table A-2.

Table A-1. Half-Life PFOA and PFOS Levels in Humans
PFOA t1/2 (days)
1,387

PFOS t1/2 (days)
1971

Exposure type
Occupational

Number
26

Source
Olsen et al. (2007a)

1,862
840

NA
NA

Occupational
Environmental

16
200

Costa et al. (2009)
Bartell et al. (2010)

PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-9
APPENDIX A

Table A-2. Serum PFOA and PFOS Concentrations Measured in Fluorochemical
Production Workers

Initial

PFOA (ppb)
408 (72, 5,100)

PFOS (ppb)
626 (145, 3,490)

Final

148 (17, 2,435)

295 (37, 1,740)

PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid
Source: Olsen et al. 2007a

Estimates of volume of distribution (Vd) are based on non-compartmental modeling of serum
concentration kinetics in monkeys and are assumed to be applicable to humans at the above serum
concentrations (Table A-3).

Table A-3. Apparent Volume of Distribution for PFOA and PFOS
PFOA Vd (L/kg)
0.18 (male)
0.20 (female)
NA
NA
0.3

PFOS Vd (L/kg)
NA
NA
0.20 (male)
0.27 (female)
0.3

Source
Butenhoff et al. (2004c)
Chang et al. (2012)
Harada et al. (2005a)

NA = not applicable; PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid

Numerous studies conducted in various animal models provide evidence for approximately complete
absorption of oral doses of PFOA and PFOS (i.e., AF≈1, see Section 3.1.1).
PFHxS. For PFHxS, the estimate of the elimination t½ was derived from longitudinal measurements of
serum concentrations of PFHxS in a group of retired fluorochemical production workers (24 males and
2 females) observed for a 5-year period; the estimated half-life was 8.5 years (3,102 days) (Olsen et al.
2007a). The range of initial serum concentrations was 16–1,295 ng/mL (mean of 290 ng/mL), and the
final concentrations ranged from 10 to 1,740 ng/mL (mean of 182 ng/mL). Estimates of the t½ for
PFHxS are most applicable to serum concentrations within the above ranges and would be less certain if
applied to serum concentrations substantially below or above these range.
Estimates of volume of distribution (𝑉𝑉𝑑𝑑 ) are based on non-compartmental modeling of serum
concentration kinetics in monkeys and are assumed to be applicable to humans at the above serum
concentrations. Sundström et al. (2012) estimate the apparent Vd for PFHxS at 0.287 L/kg for male
Cynomolgus monkeys and at 0.213 L/kg for female Cynomolgus monkeys.
Few studies have been conducted in animals that provide estimates for a gastrointestinal absorption factor
of oral doses of PFHxS. Sundström et al. (2012), based on comparison of the AUC for oral and
intravenous administration, estimate an oral absorption fraction for PFHxS (administered as a single
10 mg/kg dose) of 50% in female rats. However, as the authors point out that this estimate may not be
reliable due to the short (24 hours) observation period (Sundström et al. 2012) and that “female Cmax
values did not differ significantly between the oral and IV doses, and Tmax after oral dosing was estimated
to be at approximately 30 min.” These latter observations suggest approximately complete
bioavailability. The AUC for male rats following oral exposure was not available, and the AUC after

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-10
APPENDIX A

intravenous administration was done in only one male rat. (Sundström et al. 2012). A study conducted by
Kim et al. (2016) in rats estimated an approximately 100% oral bioavailability based on Cmax value and
AUC comparison between oral and intravenous doses. Therefore, an absorption fraction (AF) of 1 was
used for PFHxS.
PFNA. For PFNA, the elimination half-life estimates were derived by paired blood and urine samples
(n=86) from Chinese adults in a study that measured the concentrations of a number of perfluoroalkyl
compounds, including PFNA (Zhang et al. 2013). The participants were first divided into four groups;
young females (age ≤50 years, n=20), older females (>50 years, n=19), young males (≤50 years, n=32),
and older males (>50 years, n=15). The group of young females had significantly lower levels of
perfluoroalkyls than the other groups; therefore, the three other groups were combined. The lower
perfluoroalkyl levels were likely due to the elimination via menstrual bleeding, pregnancy, and lactation.
The estimated arithmetic mean elimination half-lives for the young female group and the combined male
and older female group for PFNA were 2.5 and 4.3 years (913 and 1,570 days), respectively.
Toxicokinetics parameters for perfluorocarboxylic acids, among them PFOA and PFNA analogs, were
investigated in rats by Ohmori et al. (2003). The authors estimated that the 𝑉𝑉𝑑𝑑 values in steady state were
not much different between the perfluorocarboxylic acids and between the sexes. Based on this, the
estimate volume distribution for PFNA in humans will be assumed to be the same for PFOA, 0.2 L/kg.
There are no studies on absorption of PFNA in humans. In rodents, oral absorption occurs rapidly as
indicated by its presence in the serum of rodents soon after oral administration (Tatum-Gibbs et al. 2011).
Therefore, based on animal studies of PFNA and other perfluorocarboxylic acid analogs, as well as
sufficient findings of PFNA and other perfluorocarboxylic acids in human blood, it can be assumed that
PFNA is well absorbed after oral exposure; therefore, an AF of 1 was used.
Model Input Parameters. The first-order one-compartment model input parameters (t1/2, Vd, and AF) are
provided in Table A-4.

Table A-4. First Order One-Compartment Model Parameters
Parameter

PFOA

PFOS

PFHxS

PFNA

1,400a

2,000a

3,100a

Serum elimination rate constantc, ke (day-1)
Gastrointestinal absorption fractiond, AF

4.95x10-4

3.47x10-4

2.23x10-4

900b
7.59x10-4

1

1

1

1

Apparent volume of distribution, Vd (L/kg)

0.2e

0.2e

0.287f

0.2e

Serum elimination

half-lifea;

t1/2 (day)

aEstimates

from Olsen et al. (2007a).
from Zhang et al. (2013) for young females.
cCalculated using Equation 5.
dBased on studies in rodents and nonhuman primates.
eEstimates based on studies in nonhuman primates (Butenhoff et al. 2004c; Chang et al. 2012; Harada et al. 2005a).
fEstimates based on studies in nonhuman male primates (Sundström et al. 2012).
bEstimates

PFOA = perfluorooctanoic acid; PFOS = perfluorooctane sulfonic acid; PFHxS = perfluorohexane sulfonic acid;
PFNA = perfluorononanoic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-11
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanoic acid (PFOA)
335-67-1
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFOA.
Rationale for Not Deriving an MRL: Derivation of an inhalation MRL was precluded because
inhalation-specific PBPK/pharmacokinetic model parameters are not available for PFOA and none of the
studies reported serum PFOA concentrations.
Four studies have examined the acute toxicity of airborne PFOA in laboratory animals (Griffith and Long
1980; Kennedy et al. 1986; Staples et al. 1984). The observed effects included excessive salivation and
eye and nose irritation in rats exposed to 18,600 mg/m3 for 1 hour (Griffith and Long 1980), weight loss
and pulmonary edema in rats exposed to 380 mg/m3 for 4 hours (Kennedy et al. 1986), weight loss in rats
exposed nose-only to 84 mg/m3 6 hours/day, 5 days/week for 2 weeks (Kennedy et al. 1986), and
decreases in maternal weight gain at 10 mg/m3 and maternal deaths and decreases in neonatal body
weight at 25 mg/m3 in rats exposed 6 hours/day on GDs 6–15 (Staples et al. 1984). The 2-week study
also reported increases in liver weight and hepatocellular hypertrophy in rats exposed to 7.6 mg/m3.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-12
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanoic acid (PFOA)
335-67-1
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFOA.
Rationale for Not Deriving an MRL: No intermediate-duration inhalation studies in laboratory animals
were identified for PFOA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-13
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanoic acid (PFOA)
335-67-1
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFOA.
Rationale for Not Deriving an MRL: No chronic-duration inhalation studies in laboratory animals were
identified for PFOA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-14
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanoic acid (PFOA)
335-67-1
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFOA.
Rationale for Not Deriving an MRL: An acute-duration oral MRL cannot be derived for PFOA because
the modeling approach used for estimating HEDs cannot be used to estimate acute human exposure where
the exposure duration of 14 days is 1% of the elimination half-life in humans.
Acute-duration oral studies are available in rats and mice and provide information on body weight,
hepatic, immunological, reproductive, and developmental effects. The liver effects consisted of increases
in liver weight, hepatocellular hypertrophy, and/or decreases in serum cholesterol and triglyceride in rats
and mice exposed to ≥1 mg/kg/day (Cook et al. 1992; Elcombe et al. 2010; Haughom and Spydevold
1992; Ikeda et al. 1985; Kawashima et al. 1995; Kennedy 1987; Liu et al. 1996; Pastoor et al. 1987; Iwai
and Yamashita 2006; Permadi et al. 1992, 1993; Vetvicka and Vetvickova 2013; White et al. 2009; Wolf
et al. 2007; Xie et al. 2003; Yang et al. 2000, 2001, 2002b). Consistent with the Hall et al. (2012) criteria
(see Section 2.9 for a discussion of the criteria), the liver weight increases and hypertrophy observed in
rats and mice were not considered relevant to human risk assessment. The immunological effects
consisted of impaired responses to T-dependent antigens, such as sRBCs, altered antibody response, and
decreases in spleen and thymus weights at 11.5 mg/kg/day and higher (DeWitt et al. 2009; Vetvicka and
Vetvickova 2013; Yang et al. 2001, 2002a). Information on the potential reproductive toxicity of PFOA
is limited to three studies that reported increases in serum estradiol levels in rats exposed to ≥2 mg/kg/day
for 14 days (Biegel et al. 1995; Cook et al. 1992; Liu et al. 1996). A number of studies have evaluated the
developmental toxicity of PFOA. In the only acute-duration developmental toxicity study in rats, no
alterations in fetal body weight or malformations were observed at 100 mg/kg/day (Staples et al. 1984).
Mice appear to be more sensitive to PFOA’s developmental toxicity; observed effects include decreases
in litter weight (Hu et al. 2010), decreases in pup body weight (White et al. 2007, 2009; Wolf et al. 2007),
alterations in spontaneous activity (Johansson et al. 2008), and delays in mammary gland development
(White et al. 2007, 2009; Wolf et al. 2007). The lowest LOAEL for developmental effects in mice was
0.5 mg/kg/day for decreased litter weight. A list of the NOAEL and LOAEL values for the
immunological, reproductive, and developmental effects is presented in Table A-5.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-15
APPENDIX A

Table A-5. Summary of the Adverse Effects Observed in Laboratory Animals
Following Acute-Duration Oral Exposure
Species and
exposure
duration
Immunological
Mouse
10 days
Mouse
10 days
Mouse
7 days
Mouse
7 days
Reproductive
Rat
14 days
Rat
14 days
Rat
14 days
Developmental
Mouse
GDs 6–17
Mouse
PND 10

NOAEL
LOAEL
(mg/kg/day) (mg/kg/day) Effect
11.5
7.5

15
20
24

0.2
1

2
10
25

Reference

Decreased spleen and thymus
weights
Altered response to sRBC

Yang et al. 2001

Altered response to sRBC,
decreased antibody formation
Decreased response to horse red
blood cells

Vetvicka and
Vetvickova 2013
Yang et al. 2002a

DeWitt et al. 2009

2-Fold increase in serum estradiol Liu et al. 1996
levels
63% increase in serum estradiol
Cook et al. 1992
levels
184% increase in serum estradiol Biegel et al. 1995
levels

0.5

Decreased litter weight on PND 2

0.58

Decreased spontaneous behavior Johansson et al.
and altered response to cholinergic 2008
stimulant
Altered mammary gland
White et al. 2007
development, decreased pup body
weight on PND 20

Mouse
GDs 8–17 or 12–
17

5

Mouse
GDs 8–17

5

Mouse
Various GDs

5

Hu et al. 2010

Delayed mammary gland
White et al. 2007
development, decreased pup body
weight on PND 20
Delayed mammary gland
White et al. 2009
development; decreased pup body Wolf et al. 2007
weight at weaning

GD = gestation day; LOAEL = lowest-observed-effect level; NOAEL = no-observed-adverse-effect level;
PND = postnatal day; sRBC = sheep red blood cell

The lowest LOAEL values were identified for developmental effects; Hu et al. (2010) identified a
LOAEL of 0.5 mg/kg/day for decreases in litter weight on PND 2 and Johansson et al. (2008) identified a
LOAEL of 0.58 mg/kg for decreases in spontaneous behavior (locomotion and total activity) and
decreased response to a cholinergic stimulant in adult mice exposed to PFOA on PND 10. Neither study
identified NOAEL values.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-16
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:
Provisional MRL:
Critical Effect:
Reference:
Point of Departure:
Uncertainty Factor:
LSE Graph Key:
Species:

Perfluorooctanoic acid (PFOA)
335-67-1
June 2018
Final, Pre-Public Comment
Oral
Intermediate
3x10-6 mg/kg/day
Altered activity and skeletal alterations in offspring
Onishchenko et al. 2011; Koskela et al. 2016
0.000821 mg/kg/day
300
61, 67
Mouse

MRL Summary: A provisional intermediate-duration oral MRL of 3x10-6 mg/kg/day was derived for
PFOA based on altered activity at 5–8weeks of age and skeletal alterations at 13 and 17 months of age in
the offspring of mice fed a diet containing PFOA on GD 1 through GD 21 (Koskela et al. 2016;
Onishchenko et al. 2011). The MRL is based on a HED LOAEL of 0.000821 mg/kg/day and a total
uncertainty factor of 300 (10 for use of a LOAEL, 3 for extrapolation from animals to humans with
dosimetric adjustments, and 10 for human variability).
Selection of the Critical Effect: Intermediate-duration oral studies of PFOA in animals indicate that the
liver, immune system, reproductive system, and the developing organism are the primary targets of
toxicity because adverse outcomes were observed at lower doses than other effects and have been
consistently observed across studies. A summary of the lower LOAEL values (and associated NOAEL
values) for these tissues/systems is presented in Table A-6; given the large number of studies, this table is
limited to studies that identified LOAEL values of ≤4 mg/kg/day. Although these studies identified the
lowest LOAEL values, not all were considered suitable as the basis of an intermediate-duration oral
MRL.
Exposure to low levels of PFOA results in increases in liver weight, hepatocellular hypertrophy, and
decreases in serum lipids in rats, mice, and monkeys exposed to PFOA for intermediate durations. The
increases in liver weight, hepatocellular hypertrophy, and alterations in serum lipid levels observed in the
rodents are likely due to peroxisome proliferation and are not considered relevant for human risk
assessment (Hall et al. 2012). Consistent with the Hall et al. (2012) criteria, the increases in liver weight
and hepatocellular hypertrophy, in the absence of other degenerative alterations, were not considered
adverse. A small number of animal studies have reported degenerative lesions, lesions to specialty cells,
bile duct lesions, or inflammation; these endpoints were considered relevant for human risk assessment
(Butenhoff et al. 2004b; Cui et al. 2009; Loveless et al. 2008). The lowest LOAEL for adverse liver
effects was 0.96 mg/kg/day for increased liver weight, hepatocellular hypertrophy, and focal necrosis in
mice exposed for 28 days (Loveless et al. 2008). In utero exposure has also resulted in liver effects in
offspring (Filgo et al. 2015a; Quist et al. 2015a); the lowest maternal LOAEL identified in these studies
was 0.01 mg/kg/day (Quist et al. 2015a). Because the Quist et al. (2015a) study did not provide incidence
data for the reported inflammation, this study was not considered suitable for derivation of an MRL.
Hepatic effects consisting of increases in absolute liver weight at ≥3 mg/kg/day and increases in serum
triglyceride levels at 30/20 mg/kg/day have also been observed in monkeys administered capsules
containing PFOA (Butenhoff et al. 2002); no histological alterations were observed in surviving animals,

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-17
APPENDIX A

but hepatocellular degeneration and necrosis was noted in a monkey sacrificed early due to morbidity.
The small number of animals examined and early deaths at several dose levels precludes using this study
as the basis of an MRL. Two studies examining the immunotoxicity of PFOA following intermediateduration oral exposure found decreases in antigen-specific antibody responses in mice exposed for
15 days (DeWitt et al. 2008, 2016); the lowest LOAEL was 1.88 mg/kg/day (DeWitt et al. 2016).
Reproductive and developmental toxicity studies have identified very low LOAELs of
≥0.0024 mg/kg/day for delays in mammary gland development in dams and offspring (Macon et al. 2011;
Tucker et al. 2015; White et al. 2011). However, the mammary gland effect did not result in an adverse
effect on lactational support at maternal doses as high as 1 mg/kg/day, based on normal growth and
survival in F2 pups (White et al. 2011). Given that milk production was adequate to support growth, the
biological significance of the delayed development of the mammary gland is uncertain and was not
considered a suitable basis for the MRL. Other developmental effects include increases in locomotor
activity (Cheng et al. 2013; Onishchenko et al. 2011; Sobolewski et al. 2014) at ≥0.1 mg/kg/day, reduced
ossification of proximal phalanges and advanced preputial separation at ≥1 mg/kg/day (Lau et al. 2006),
altered long bone morphology and decreased bone mineral density in 13- and 17-month-old mice
following in utero exposure to 0.3 mg/kg/day (Koskela et al. 2016), decreases in pup survival at
≥0.6 mg/kg/day (Abbott et al. 2007; Albrecht et al. 2013), decreases in the number of successful births at
≥3 mg/kg/day (Ngo et al. 2014), and reduced neonatal weight gain and delayed eye opening at ≥3
mg/kg/day (Wolf et al. 2007).

Table A-6. Summary of the Adverse Effects Observed in Laboratory Animals
Following Intermediate-Duration Oral Exposure
Species and
exposure
duration
Hepatic
Mouse
GDs 1–17
Mouse
28 days
Mouse
GDs 1–17
(examined at
18 months of
age)
Rat
70–90 days
Monkey
26 weeks
Immunological
Mouse
15 days
Mouse
15 days

NOAEL
(mg/kg/day)

LOAEL
(mg/kg/day)

Effect

Reference

Hepatocellular hypertrophy and
periportal inflammation in offspring
Moderate to severe hepatocellular
hypertrophy and focal necrosis
Increased severity of chronic
inflammation in liver

Quist et al.
2015a, 2015b
Loveless et al.
2008
Filgo et al.
2015a, 2015b

Butenhoff et al.
2004b

3

Increased liver weight,
hepatocellular hypertrophy and
necrosis
Increased absolute liver weight

0.94

1.88

Reduced antibody response

1.88

3.75

Reduced sRBC response

DeWitt et al.
2016
DeWitt et al.
2008

0.01
0.29

0.96

0.3

1

1

3

***DRAFT FOR PUBLIC COMMENT***

Butenhoff et al.
2002

 PERFLUOROALKYLS

A-18
APPENDIX A

Table A-6. Summary of the Adverse Effects Observed in Laboratory Animals
Following Intermediate-Duration Oral Exposure
Species and
exposure
duration

NOAEL
(mg/kg/day)

Reproductive
Mouse
GDs 1–17

LOAEL
(mg/kg/day)

Effect

Reference

0.0024

Delayed mammary gland
White et al.
development in dams (3-generation 2011
study)

1

Delayed mammary gland
development in dams (singlegeneration study)

Mouse
GD 7–PND 22

0.0024

Impaired development of mammary White et al.
glands
2011

Mouse
GDs 10–17
Mouse
GDs 1–17
Mouse
GD 7–PND 21
Mouse
GDs 1–21
Mouse
GDs 1–21
Mouse
GDs 1–17
Mouse
GDs 1–17
Mouse
GDs 1–17

0.01

Impaired development of mammary
glands
Impaired development of mammary
glands
Neurodevelopmental

Mouse
GDs 1–17

White et al.
2011

Developmental

Rat
GD 1–PND 21
Mouse
GDs 1–17
Mouse
GDs 1–17
Mouse
GDs 1–17

0.01
0.1
0.3
0.3
0.3
0.3

0.6
1

1.6
0.1

3
3
3

Macon et al.
2011
Tucker et al.
2015
Sobolewski et
al. 2014
Neurodevelopmental
Onishchenko et
al. 2011
Skeletal alterations
Koskela et al.
2016
Impaired development of mammary Macon et al.
glands
2011
Decreased pup survival
Abbott et al.
2007
Reduced ossification of proximal
Lau et al. 2006
phalanges and advanced preputial
separation
Neurodevelopmental
Cheng et al.
2013
Decreased number of successful
Ngo et al. 2014
births
Reduced pups per litter on PND 20 Albrecht et al.
2013
Reduced weight gain, delayed eye Wolf et al. 2007
opening

GD = gestation day; LOAEL = lowest-observed-effect level; NOAEL = no-observed-adverse-effect level;
PND = postnatal day; sRBC = sheep red blood cell

Selection of the Principal Study: As outlined in the MRL approach section, serum PFOA levels were
predicted for the administered doses for most of the studies listed in Table A-6. Mean serum PFOA
levels could not be predicted for four studies because pharmacokinetic model parameters were not
available for Wistar rats (Cheng et al. 2013), male CD-1 mice (Loveless et al. 2008), or 129S1/SvlmJ

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-19
APPENDIX A

wild-type mice (Abbott et al. 2007; Albrecht et al. 2013). A summary of the predicted serum PFOA
levels is presented in Table A-7.

Table A-7. Summary of the Predicted TWA Serum PFOA levels in Laboratory
Animals Following Intermediate-Duration Oral Exposure
Species and
exposure
Dose
duration
(mg/kg/day)
Hepatic
CD
Mouse
28 days
129/Sv
Mouse
GDs 1–17

SpragueDawley
Rat
70–90 days
Cynomolgus
Monkey
6 months

0.29
0.96
9.6
0.01
0.1
0.3
1
5
1
3
10
30
3
10
20/30

Predicted TWA
serum PFOA
(µg/mL)

Not calculated

Effect

Reference

Moderate to severe
Loveless et al.
hepatocellular hypertrophy and 2008
focal necrosis at 0.96 mg/kg/day

0.423
4.21
12.5
39.2
102
60.4
136
222
242
68.5
93.8
113

Increased severity of chronic
Filgo et al.
inflammation in liver of offspring 2015a, 2015b
aged 18 months at 1 mg/kg/day

Increased liver weight,
hepatocellular hypertrophy and
necrosis at 10 mg/kg/day

Butenhoff et al.
2004b

Increased liver weight at
3 mg/kg/day

Butenhoff et al.
2002

Immunological
C58BL/6N
Mouse
15 days

0.94
1.88
3.75
7.5

21.4
42.5
58.4
83.5

Reduced antibody response at
1.88 mg/kg/day

DeWitt et al.
2016

C57BL/6N
Mouse
15 days

0.94
1.88
3.75
7.5
15
30

21.4
42.5
58.4
83.5
109
149

Reduced sRBC response at
3.75 mg/kg/day

DeWitt et al.
2008

2.23

Neurodevelopmental effects
(increased horizontal and vertical
activity and decreased resting
activity) at 0.1 mg/kg/day
Skeletal alterations at
0.3 mg/kg/day

Sobolewski et al.
2014

Developmental
C57BL/6
0.1
Mouse
GD 7–
PND 21
C57BL/6
0.3
Mouse
GDs 1–21
C57BL/6
0.3
Mouse

8.29

8.29

Koskela et al.
2016

Neurodevelopmental (decreased Onishchenko et
number of inactive periods,
al. 2011

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-20
APPENDIX A

Table A-7. Summary of the Predicted TWA Serum PFOA levels in Laboratory
Animals Following Intermediate-Duration Oral Exposure
Species and
exposure
Dose
duration
(mg/kg/day)

Predicted TWA
serum PFOA
(µg/mL)

GDs 1–21
129S1/SvlmJ
Mouse
GDs 1–17

CD-1
Mouse
GDs 1–17

Effect

Reference

altered novelty induced activity)
at 0.3 mg/kg/day
0.1
0.3
0.6
1
3
5
10
20
1
3
5
10
20
40

Decreased pup survival at
0.6 mg/kg/day

Abbott et al. 2007

Not calculated

39.2
83.6
102
125
155
205

Reduced ossification of proximal Lau et al. 2006
phalanges and advanced
preputial separation at
1 mg/kg/day

Neurodevelopmental effects
Cheng et al. 2013
(increased locomotor activity in
males and decreased activity in
females) at 1.6 mg/kg/day
Decreased number of successful Ngo et al. 2014
births at 3 mg/kg/day

Wistar
Rat
GD 1–PND 21

1.6

Not calculated

C57BL/6J
Mouse
GDs 1–17
SV/129
Mouse
GDs 1–17
CD-1
Mouse
GDs 1–17

0.1
3

2.43
62.0

3

Not calculated

Reduced pups per litter on
PND 20 at 3 mg/kg/day

Albrecht et al.
2013

3
5

84.8
102

Reduced weight gain, delayed
eye opening at 3 mg/kg/day

Wolf et al. 2007

GD = gestation day; PFOA = perfluorooctanoic acid; PND = postnatal day; sRBC = sheep red blood cell;
TWA = time-weighted average

Selection of the Point of Departure for the MRL: The NOAEL/LOAEL and the benchmark dose
(BMD) approaches were utilized to identify potential PODs for derivation of the intermediate-duration
oral MRL for PFOA. The only datasets amenable to BMD modeling were from the DeWitt et al. (2008)
and DeWitt et al. (2016) immunotoxicity studies. The results of the BMD modeling are presented at the
end of this worksheet.
HEDs were calculated for each potential PODs (NOAEL, LOAEL, or BMDL value) identified in
laboratory animal studies using the first order single-compartment model previously discussed and the
assumption that humans would have similar effects as the laboratory animal at a given serum

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-21
APPENDIX A

concentration. The HEDs for each POD are presented in Table A-8. The potential PODHED values were
divided by an uncertainty factor to calculate candidate MRLs; these values are also presented in
Table A-8. The candidate MRLs range from 7.4x10-7 mg/kg/day for neurodevelopmental effects in mice
(Sobolewski et al. 2014) to 4.5x10-4 mg/kg/day for liver effects in male rats (Butenhoff et al. 2004b). The
lowest LOAEL (expressed as predicted serum concentration) was identified in the Sobolewski et al.
(2014) study, which found neurodevelopmental effects in mouse offspring at predicted serum PFOA
concentration of 2.23 µg/mL. However, this study was not considered suitable as the basis of the MRL
because the subroute and vehicle used for the controls (peanut oil with anisole administered via gavage)
were different from the PFOA group (PFOA dissolved in water and added to diet). Rather, the
Onishchenko et al. (2011) and Koskela et al. (2016) studies, which identified the second lowest LOAEL
(serum PFOA concentration) of 8.29 µg/mL, were selected as the basis for the MRL; it is noted that these
two studies used the offspring from the same animals.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-22
APPENDIX A

Table A-8. Summary of Potential Points of Departures (PODs) and Human
Equivalent Doses (HEDs) for Intermediate-Duration Oral MRL for PFOA

Endpoint (reference)

Predicted serum
concentrations
(µg/mL)
NOAEL or
BMDL
LOAEL

PODHEDa
(mg/kg/day)

Candidate
MRLs
Total UF (mg/kg/day)

Neurodevelopmental effects
(increased horizontal and vertical
activity and decreased resting
activity) in mice (Sobolewski et al.
2014)
Neurodevelopmental effects
(decreased number of inactive
periods, altered novelty induced
activity) in mice (Onishchenko et al.
2011)
Skeletal alterations in mice (Koskela
et al. 2016)
Decreased number of successful
2.43
births in mice (Ngo et al. 2014)
Reduced ossification of proximal
phalanges and advanced preputial
separation in mice (Lau et al. 2006)

2.23

0.000221

300b

7.4x10-7

8.29

0.000821

300b

2.7x10-6

8.29

0.000821

300b

2.7x10-6

62.0

0.000241

30c

8.0x10-6

39.2

0.00388

300b

1.3x10-5

Reduced weight gain and delayed
eye opening (Wolf et al. 2007)

84.8

0.00840

300b

2.8x10-5

0.00121

30c

4.0x10-5

39.2

0.00124

30c

4.1x10-5

33.49
(BMDL1SD)
136
222

0.00332

30c

1.1x10-4

0.0135

30c

4.5x10-4

Reduced response to dinitrophenylficoll (DNP) antigen in female mice
(DeWitt et al. 2016)
Increased severity of chronic
inflammation in liver of offspring
aged 18 months (Filgo et al. 2015a,
2015b)
Reduced response to sRBC in
female mice (DeWitt et al. 2008)
Increased liver weight,
hepatocellular hypertrophy, and
necrosis in male rats (Butenhoff et
al. 2004b)

12.23
(BMDL1SD)
12.5

aBased

on the NOAEL or BMDL; based on LOAEL if there was no NOAEL or BMDL.
of 10 for extrapolation from a LOAEL, 3 for extrapolation from animals to humans with dosimetric adjustments,
and UF of 10 for human variability.
cUF of 3 for extrapolation from animals to humans with dosimetric adjustments, and 10 for human variability.
bUF

BMDL = lower confidence limit on the BMD; HED = human equivalent dose; LOAEL = lowest-observed-adverseeffect level; MRL = Minimal Risk Level; NOAEL = no-observed-adverse-effect level; POD = point of departure;
sRBC = sheep red blood cell; UF = uncertainty factor

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-23
APPENDIX A

Summary of the Principal Study:
Onishchenko et al. 2011. Prenatal exposure to PFOS or PFOA alters motor function in mice in a sexrelated manner. Neurotox Res 19:452-461.
Koskela et al. 2016. Effects of developmental exposure to perfluorooctanoic acid (PFOA) on long bone
morphology and bone cell differentiation. Toxicol Appl Pharmacol 301:14-21.
Neurobehavioral effects were assessed in offspring of C57BL/6/Bk1 mice that were exposed to PFOA
(96% pure) in food at dose levels of 0 mg/kg/day (n=10) or 0.3 mg/kg/day (n=6) from GD 1 throughout
pregnancy (presumed GD 21). PFOA was dissolved in ethanol and applied to palatable food in volumes
adjusted according to individual body weights to provide 0.3 mg/kg/day, followed by evaporation of
ethanol; controls received food with ethanol applied and then evaporated.
Onishchenko et al. (2011) study: Endpoints examined in the included maternal body weight during
pregnancy; litter size and sex ratio; body, brain and liver weights in pups at birth; and PFOA
concentrations in whole brain and liver tissue samples from pups at birth. Pups were subcutaneously
injected with microtransponders on PND 21, and tested (6–10/sex/group) for novelty-induced locomotor
activity and circadian activity over 48 hours at age 5–8 weeks, elevated maze and forced swimming
performance at age 5–8 weeks, and muscle strength and coordination at age 3–4 months. The locomotor
test measured distance traveled in the first 30 minutes in a new cage. Circadian activity was measured as
number of antenna crossings in the new cage over 48 hours, with activity in the first 3 hours analyzed
separately as adaptation to a novel environment. The elevated plus maze test recorded number of entries
and time spent in open or closed areas as indicators of anxiety-like behavior. The forced swimming test
measured immobility duration (total of times floating passively for ≥2 seconds) as an indicator of
depression-like behavior. Muscle strength was evaluated by latency to fall in a hanging wire test. Motor
coordination was evaluated by latency to fall in an accelerating rotarod test.
Prenatal PFOA exposure was associated with increases in global activity and exploratory activity in adult
offspring as indicated by statistically significant (p<0.05 compared with controls) increased circadian
activity in males (44–48% higher number of crossings over 48 hours in a new cage, occurring in both
12-hour dark phases and in the intervening 12-hour light phase, but not in the initial 3-hour novelty
phase), and decreased total number of inactive periods in males and females (approximately 40 and 25%
lower, respectively). No alterations were observed in the forced swimming test or hanging wire test.
Mean PFOA concentrations in brain and liver tissue samples at birth were 0.7 and 16.3 µg/g in exposed
pups and below detection limits in control pups.
Koskela et al. (2016) study: Groups of five offspring were sacrificed at either 13 or 17 months of age.
The following parameters were used to assess toxicity: body weight and morphometric/biochemical
properties in bone (femurs and tibias) of offspring. Offspring body weight was significantly higher in
comparison with controls at 13 and 17 months of age (9.9 and 7.8%, respectively). In 17-month-old
offspring, there was a 6.8% increase in periosteal area of the femoral cortical bone and increases in the
peri- and endosteal perimeters (3.2 and 5.2%, respectively) and the marrow area (10.0%); an increase in
medullary area was also observed. There were no differences in femoral cortical bone area or femoral
mineral density. In the tibia, the total area inside the periosteal envelope and the periosteal perimeter
were increased (4.9 and 3.5%, respectively). Although the investigators noted in the text that tibial
medullary areas were “essentially the same between groups,” data in Figure 2 of the paper show a
significant increase at 17 months. Significant decreases in tibial mineral density were observed at 13 and
17 months. There were no significant differences in the tibial medullary area or the endosteal perimeter.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-24
APPENDIX A

There was a trend for increasing maximum force (Fmax); however, the effect was not statistically
significant. There were no significant effects on any other measured biochemical parameter in the femur
or tibia (stiffness, maximum energy, absorption). Concentrations of PFOA in the femurs and tibias of
treated animals were significantly higher (4–5 times) than controls at 13 and 17 months.
Calculation of Internal Dosimetric: TWA serum PFOA concentrations corresponding to external doses
and exposure durations were predicted from a pharmacokinetic model (Wambaugh et al. 2013) using
animal species-, strain-, and sex-specific parameters (see MRL approach section for details).
Human Equivalent Dose: HEDs were calculated based on the assumption that humans would have
similar effects as the laboratory animal at a given serum concentration. HEDs that would result in steadystate serum concentrations of PFOA equal to the serum concentration selected as the POD were
calculated using the first-order single-compartment model (see MRL approach section for details).
Uncertainty Factor: The LOAELHED is divided by a total uncertainty factor of 300:
• 10 for the use of a LOAEL
• 3 for extrapolation from animals to humans with dosimetric adjustment
• 10 for human variability
Provisional MRL = LOAELHED ÷ UFs
0.000821 mg/kg/day ÷ (10 x 3 x 10) = 3x10-6 mg/kg/day
Other Additional Studies or Pertinent Information that Lend Support to this MRL: A discussion of the
findings from epidemiology studies is presented in the MRL introduction section.
Benchmark Dose Modeling: BMD modeling was conducted for the DeWitt et al. (2008) and DeWitt et
al. (2016) immunotoxicity studies. Using predicted TWA serum PFOA levels as the internal dosimetric,
the IgM response data (summarized in Tables A-9 and A-10) were fit to all available continuous models
in EPA’s Benchmark Dose Software (BMDS, version 2.6.0). The following procedure for fitting
continuous data was used: the simplest model (linear) was first applied to the data while assuming
constant variance; if the data were consistent with the assumption of constant variance (p≥0.1), then the
fit of the linear model to the means was evaluated and the polynomial, power, and Hill models were fit to
the data while assuming constant variance. Adequate model fit was judged by three criteria: goodnessof-fit p-value (p>0.1), visual inspection of the dose-response curve, and scaled residual at the data point
(except the control) closest to the predefined benchmark dose response (BMR). Among all of the models
providing adequate fit to the data, the lowest BMDL (the lower limit of a one-sided 95% CI on the BMD)
was selected as a reasonably conservative POD when differences between the BMDLs estimated from
these models are >2–3-fold; otherwise, the BMDL from the model with the lowest Akaike's information
criterion (AIC) was chosen. If the test for constant variance was negative, the linear model was run again
while applying the power model integrated into the BMDS to account for nonhomogenous variance. If
the nonhomogenous variance model provided an adequate fit (p≥0.1) to the variance data, then the fit of
the linear model to the means was evaluated and the polynomial, power, and Hill models were fit to the
data and evaluated while the variance model was applied. Model fit and POD selection proceeded as
described earlier. For both datasets, a BMR of 1 SD change from the control was used.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-25
APPENDIX A

Table A-9. T-Cell Independent IgM Antibody Response in C57BL/6N Female Mice
Immunized with Sheep Red Blood Cells
Predicted TWA serum IgM antibody titersa
PFOA concentration [mean serum IgM titer
(µg/mL)
(log2) to reach 0.5 OD] SEa

Administered
Number of
dose
animals per group (mg/kg/day)
8
8
8
8

0
0.94
1.88
3.75

0
21.4
42.5
58.4

7.28
7.39
7.08
6.75b

0.13
0.07
0.10
0.09

8

7.5

83.5

6.61b

0.12

aData

taken from Figure 3-C using GrabIt.
different from controls (p<0.05).

bStatistically

PFOA = perfluorooctanoic acid; SE = standard error; TWA = time-weighted-average
Source: DeWitt et al. 2008

Table A-10. T-Cell Independent IgM Antibody Response In C57BL/6 Female Mice
Immunized with Dinitrophenyl-Ficoll
Number of Administered
animals per dose
group
(mg/kg/day)

Predicted TWA serum T-cell independent IgM antibody
PFOA concentration responsea [mean serum titer
(µg/mL)
(log2) to reach 0.5 OD]
SDa

8

0

0

11.38

0.56

8
8
8
8

0.94
1.88
3.75
7.5

21.4
42.5
58.4
83.5

11.01
9.67b
9.81b
9.62b

1.11
1.34
1.46
0.97

aData

taken from Figure 3b using GrabIt.
different from controls (p<0.05).

bStatistically

PFOA = perfluorooctanoic acid; SE = standard error; TWA = time-weighted-average
Source: DeWitt et al. 2016

The results of the BMD analysis of the DeWitt et al. (2008) and DeWitt et al. (2016) datasets are
presented in Tables A-11 and A-12. For the DeWitt et al. (2008) data, the Hill model with constant
variance provided the best fit to the IgM response data, as judged by the model with the lowest AIC since
the range of BMDL values were sufficiently close; the fit of this model is presented in Figure A-1. For
the DeWitt et al. (2016) IgM response data, constant variance models provided adequate fit; since the
estimated BMDL values were not sufficiently close, the model with the lowest BMDL (Exponential 4)
was selected; the fit of this model is presented in Figure A-2.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-26
APPENDIX A

Table A-11. T-Cell Independent IgM Antibody Response in C57BL/6N Female
Mice Immunized With Sheep Red Blood Cells Using Predicted TWA Serum
PFOA as the Dose Metric (DeWitt et al. 2008)

Model

Test for
significant
difference Variance Means
p-valuea p-valueb p-valueb

Scaled residualsc
Dose Dose
below above Overall
BMC BMC largest AIC

BMD1SD BMDL1SD
(ng/mL) (ng/mL)

Constant variance
Exponential
(model 2)d
Exponential
(model 3)d
Exponential
(model 4)d
Exponential
(model 5)d
Hilld,e
Linearf
Polynomial
(2-degree)f
Polynomial
(3-degree)f
Polynomial
(4-degree)f
Powerd

<0.0001

0.46

0.08

1.73

<0.0001

0.46

0.10

<0.0001

0.46

<0.0001

0.72 1.73

-50.41 ND

ND

0.16 -1.30 1.33

-50.61 42.55

26.37

0.08

1.73

0.72 1.73

-50.41 ND

ND

0.46

0.36

-0.14

0.09 0.67

-52.34 44.11

33.33

<0.0001
<0.0001
<0.0001

0.46
0.46
0.46

0.41
0.09
0.08

-0.05 0.06 0.60
1.69 0.67 1.69
0.10 -1.41 1.43

-52.47 43.57
-50.66 ND
-50.00 ND

33.49
ND
ND

<0.0001

0.46

0.08

0.10 -1.41 1.43

-50.00 ND

ND

<0.0001

0.46

0.08

0.10 -1.41 1.43

-50.00 ND

ND

<0.0001

0.46

0.10

0.16 -1.32 1.35

-50.49 42.62

26.23

aValues

>0.05 fail to meet conventional goodness-of-fit criteria.
<0.10 fail to meet conventional goodness-of-fit criteria.
cScaled residuals at doses immediately below and above the benchmark dose; also the largest residual at any dose.
dPower restricted to ≥1.
eSelected model. Constant variance model provided adequate fit to variance data. With constant variance model
applied, the only models that provided adequate fit to the means were the Exponential 3 and 5, Hill, and Power
models. BMDLs for models providing adequate fit were considered to be sufficiently close (differed by <2–3-fold);
therefore, the model with the lowest AIC was selected (Hill).
fCoefficients restricted to be negative.
bValues

AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the exposure concentration associated
with the selected benchmark response; BMDL = 95% lower confidence limit on the BMD (subscripts denote
benchmark response: i.e., 10 = exposure concentration associated with 10% extra risk); ND = not determined, model
did not provide adequate fit; PFOA = perfluorooctanoic acid; SD = standard deviation; TWA = time-weighted average

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-27
APPENDIX A

Figure A-1. Predicted (Hill Model with Constant Variance, 1 Standard Deviation
Benchmark Response) and Observed IgM Response Using Predicted TWA
Serum PFOA as the Dose Metric
Hill Model, with BMR of 1 Std. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
Hill
7.6

7.4

Mean Response

7.2

7

6.8

6.6

6.4

BMDL

BMD

6.2
0

10

20

30

40

50

dose
09:21 02/16 2017

***DRAFT FOR PUBLIC COMMENT***

60

70

80

 PERFLUOROALKYLS

A-28
APPENDIX A

Table A-12. T-Cell Independent IgM Antibody Response In C57BL/6 Female Mice
Immunized With Dinitrophenyl-Ficoll Using Predicted TWA Serum PFOA as the
Dose Metric (DeWitt et al. 2016)

Model

Scaled residualsc
Test for
significant
Dose Dose
difference Variance Means below above Overall
p-valuea p-valueb p-valueb BMC BMC largest AIC

BMD1SD BMDL1SD
(ng/mL) (ng/mL)

0.0029

0.12

0.34

-1.47

-0.17 -1.47

54.00

45.00

29.56

0.0029

0.12

0.34

-1.47

-0.17 -1.47

54.00

45.00

29.56

0.0029

0.12

0.31

1.01

-1.06 -1.06

54.97

29.22

12.23

0.0029

0.12

0.71

0.00

-0.08

0.29

54.76

26.62

18.75

0.0029
0.0029
0.0029

0.12
0.12
0.12

0.93
0.30
0.30

0.00
-1.52
-1.52

-0.08 0.29
-0.23 -1.52
-0.23 -1.52

52.76
54.27
54.27

23.66
47.96
47.96

19.11
32.63
32.63

0.0029

0.12

0.30

-1.52

-0.23 -1.52

54.27

47.96

32.63

0.0029

0.12

0.30

-1.52

-0.23 -1.52

54.27

47.96

32.63

0.0029

0.12

0.30

-1.52

-0.23 -1.52

54.27

47.96

32.63

Constant variance
Exponential
(model 2)d
Exponential
(model 3)d
Exponential
(model 4)d,e
Exponential
(model 5)d
Hilld
Linearf
Polynomial
(2-degree)f
Polynomial
(3-degree)f
Polynomial
(4-degree)f
Powerd
aValues

>0.05 fail to meet conventional goodness-of-fit criteria.
<0.10 fail to meet conventional goodness-of-fit criteria.
cScaled residuals at doses immediately below and above the benchmark dose; also the largest residual at any dose.
dPower restricted to ≥1.
eSelected model. Constant variance model provided adequate fit to variance data. With constant variance model
applied, all models provided adequate fit to the means. BMDLs for models providing adequate fit were not
considered to be sufficiently close (differed by >2-fold, but <3-fold). In order to remain conservative, the model with
the lowest BMDL was selected (Exponential 4).
fCoefficients restricted to be negative.
bValues

AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the exposure concentration associated
with the selected benchmark response; BMDL = 95% lower confidence limit on the BMD (subscripts denote
benchmark response: i.e., 10 = exposure concentration associated with 10% extra risk); NA = not applicable (BMDL
computation failed); SD = standard deviation; TWA = time-weighted average

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-29
APPENDIX A

Figure A-2. Predicted (Exponential 4 Model with Constant Variance, 1 Standard
Deviation Benchmark Response) and Observed IgM Response Using Predicted
TWA Serum PFOA as the Dose Metric
Exponential 4 Model, with BMR of 1 Std. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
Exponential 4
12

11.5

Mean Response

11

10.5

10

9.5

9

8.5
BMDL
0

10

BMD
20

30

40

50

dose
08:25 02/16 2017

Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

60

70

80

 PERFLUOROALKYLS

A-30
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanoic acid (PFOA)
335-67-1
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFOA.
Rationale for Not Deriving an MRL: Intermediate-duration studies have identified the immune system
as a sensitive target of PFOA toxicity. However, the potential immunotoxicity of PFOA has not been
investigated in chronic-duration studies. Given the uncertainty that the chronic studies identified the most
sensitive endpoint, a chronic-duration oral MRL for PFOA is not recommended at this time.
The chronic oral animal database for PFOA is limited to dietary exposure studies in male rats (3M 1983)
or male and female rats (Biegel et al. 2001). The lowest LOAEL identified in the 3M (1983) study was
1.5 mg/kg/day for tubular hyperplasia in the ovaries and inflammation of salivary gland in rats exposed to
PFOA in the diet for 2 years. At 15 mg/kg/day, hepatocellular necrosis was observed after 1 year of
exposure and vascular mineralization was observed in the testes. In the Biegel et al. (2001) study,
exposure to 13.6 mg/kg/day PFOA in the diet for 2 years resulted in decreases in body weight gain,
increases in Leydig cell hyperplasia, and pancreatic acinar cell hyperplasia in male rats.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-31
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctane sulfonic acid (PFOS)
1763-23-1
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFOS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFOS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-32
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctane sulfonic acid (PFOS)
1763-23-1
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFOS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFOS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-33
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctane sulfonic acid (PFOS)
1763-23-1
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFOS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFOS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-34
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctane sulfonic acid (PFOS)
1763-23-1
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFOS.
Rationale for Not Deriving an MRL: An acute-duration oral MRL for PFOS cannot be derived because
the modeling approach used for estimating HEDs cannot be used to estimate acute human exposure where
the exposure duration of 14 days is <1% of the PFOS elimination half-life in humans.
A number of studies have examined the toxicity of PFOS in laboratory animals following acute-duration
exposure. The available data suggest that the liver, developing organism, and immune system are
sensitive targets. The liver effects consisted of decreases in serum lipids, increases in liver weight, and
hepatocellular hypertrophy (Elcombe et al. 2012a, 2012b; Era et al. 2009; Fuentes et al. 2006; Haughom
and Spydevold 1992; Vetvicka and Vetvickova 2013; Wan et al. 2011); using the Hall et al. (2012)
criteria (see Section 2.9 for a discussion of the criteria) for assessing the adversity of liver alterations for
peroxisome proliferators, these effects were not considered relevant for human risk assessment.
Immunological effects included altered responses to sRBC and decreased IgM antibody formation in
response to antigen exposure in mice (Vetvicka and Vetvickova 2013; Zheng et al. 2009). Developmental
effects consisted of decreases in neonatal survival (Abbott et al. 2009; Grasty et al. 2003), increases in
post-implantation losses (Lee et al. 2015a), decreases in fetal body weight (Case et al. 2001; Era et al.
2009; Fuentes et al. 2007b; Lee et al. 2015a), increases in malformations (Era et al. 2009), and alterations
in motor activity (Hallgren et al. 2015; Johansson et al. 2008). The lowest LOAEL identified was
0.5 mg/kg/day for increased post-implantation losses in mice (Lee et al. 2015a). A summary of the
adverse effect levels for the immunological and developmental effects are presented in Table A-13.

Table A-13. Summary of the Adverse Effects Observed in Laboratory Animals
Following Acute-Duration Oral Exposure to PFOS
Species and
exposure
duration
Immunological
Mouse
7 days
Mouse
7 days

NOAEL
LOAEL
(mg/kg/day) (mg/kg/day) Effect
5

20

Reference

Impaired response to T-cell
Zheng et al. 2009
mitogens; suppressed response to
sRBC
Inhibition of T lymphocyte
Vetvicka and
proliferation in response to sRBC; Vetvickova 2013
decreased phagocytosis by
peripheral blood cells and NK cell
activity; decreased IgM antibody
formation in response to OVA

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-35
APPENDIX A

Table A-13. Summary of the Adverse Effects Observed in Laboratory Animals
Following Acute-Duration Oral Exposure to PFOS
Species and
exposure
duration
Developmental
Mouse
GDs 11–16
Mouse
Once
Rabbit
GDs 6–20
Mouse
GDs 15–18
Mouse
GDs 12–18
Mouse
Once

NOAEL
LOAEL
(mg/kg/day) (mg/kg/day) Effect

1

Reference

0.5

Increased post-implantation losses Lee et al. 2015a

0.75

Decreased motor activity

2.5

Decreased fetal body weight

4.5

Decreased number of live pups per
litter on PND 15
Reduced pup body weight on
PNDs 4 and 8
Altered spontaneous behavior in
pups

Abbott et al. 2009

6
11.3

Johansson et al.
2008
Case et al. 2001

Fuentes et al.
2007b
Hallgren et al.
2015

Rat
GDs 19–20

25

Decreased neonatal survival

Grasty et al. 2003

Rat
GDs 2–5, 6–9,
10–13, 14–17, or
17–20
Mouse
GDs 11–15

25

Decreased neonatal survival

Grasty et al. 2003

50

Cleft palate and reduced fetal
body weight

Era et al. 2009

GD = gestation day; LOAEL = lowest-observed-effect level; NK = natural killer; NOAEL = no-observed-adverseeffect level; perfluorooctane sulfonic acid; PND = postnatal day; sRBC = sheep red blood cell

Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-36
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:
Provisional MRL:
Critical Effect:
Reference:
Point of Departure:
Uncertainty Factor:
Modifying Factor:
LSE Graph Key:
Species:

Perfluorooctane sulfonic acid (PFOS)
1763-23-1
June 2018
Final, Pre-Public Comment
Oral
Intermediate
2x10-6 mg/kg/day
Delayed eye opening and decreased pup body weight
Luebker et al. 2005a
0.000515 mg/kg/day
30
10
34
Rat

MRL Summary: A provisional intermediate-duration oral MRL of 2x10-6 mg/kg/day was derived for
PFOS based on delayed eye opening and transient decrease in F2 body weight during lactation in the
offspring of rats administered PFOS via gavage in a 2-generation study (Luebker et al. 2005a). The MRL
is based on a HED NOAEL of 0.000515 mg/kg/day and a total uncertainty factor of 30 (3 for
extrapolation from animals to humans with dosimetric adjustments and 10 for human variability) and a
modifying factor of 10 for concern that immunotoxicity may be a more sensitive endpoint than
developmental toxicity).
Selection of the Critical Effect: Intermediate-duration studies in monkeys, rats, and mice have identified
several sensitive targets of PFOS toxicity including the liver, nervous system, immune system, and the
developing organism; adverse outcomes occurred in these tissues at lower doses than other effects. The
lowest LOAEL and NOAEL values for these outcomes are presented in Table A-14; given the large
number of intermediate-duration studies, this table was limited to studies which identified LOAEL values
of ≤3 mg/kg/day. The liver effects observed in monkeys, rats, and mice included increases in liver
weight, decreases in serum lipids, hepatocellular degeneration, and focal necrosis (Cui et al. 2009; Curran
et al. 2008; Elcombe et al. 2012a; Lefebvre et al. 2008; Seacat et al. 2002, 2003; Thibodeaux et al. 2003;
Wan et al. 2011; Yahia et al. 2008). In the absence of degenerative changes such as necrosis, the liver
hypertrophy observed in rodent studies was not considered relevant to human risk assessment (Hall et al.
2012). Several studies have examined potential neurological endpoints and found overt signs of
neurotoxicity (cachexia, lethargy, and tonic convulsions in response to stimuli) in rats exposed to 5 or
8.5 mg/kg/day (Cui et al. 2009; Kawamoto et al. 2011) and impaired spatial learning and memory in mice
exposed to 2.15 mg/kg/day (Long et al. 2013). Four studies have evaluated the immune response of
PFOS exposed mice following exposure to an antigen (sRBC) or a virus (Dong et al. 2009, 2011; Guruge
et al. 2009; Peden-Adams et al. 2008). Although the studies have consistently reported adverse effects,
there is considerable overlap in LOAEL values. Peden-Adams et al. (2008) identified the lowest LOAEL
of 0.00166 mg/kg/day with a NOAEL of 0.000166 mg/kg/day for a suppressed response to sRBC in mice
administered PFOS for 28 days. Longer duration studies (Dong et al. 2009, 2011) have identified
NOAEL values (0.0083 and 0.0167 mg/kg/day) in mice exposed to PFOS for 60 days that are higher than
the LOAELs identified in the Peden-Adams et al. (2008) study. It is noted that the studies used different
mouse strains (B6C3F1 in the Peden-Adams study and C57BL/6N in the Dong studies), which may
account for this difference. A variety of developmental effects have been observed in rats and mice; these
include increases in postnatal mortality (Chen et al. 2012b; Lau et al. 2003; Luebker et al. 2005a, 2005b;
Xia et al. 2011; Yahia et al. 2008), neurodevelopmental alterations (locomotor activity and impaired

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-37
APPENDIX A

learning) (Butenhoff et al. 2009b; Onishchenko et al. 2011; Wang et al. 2015c), developmental delays
(Lau et al. 2003; Luebker et al. 2005a), and malformations and anomalies (sternal defects and cleft palate)
(Era et al. 2009; Thibodeaux et al. 2003; Yahia et al. 2008). The Wang et al. (2015c) study showed that
decreases in spatial learning were observed in rats exposed in utero and in rat pups exposed postnatally
(PND 7). Other effects that occur at similar doses include decreases in body weight (Lefebvre et al. 2008;
Luebker et al. 2005a, 2005b; Seacat et al. 2002) and alterations in thyroid hormone levels (decreases in
T3 and T4 levels and increases in TSH levels) (Curran et al. 2008; Luebker et al. 2005b; Thibodeaux et al.
2003).
The most sensitive targets of PFOS toxicity in laboratory animals are similar to those identified in longer
term epidemiology studies. These effects include liver damage and increases in serum lipids, decreased
antibody response to vaccines, and small decreases in birth weight; epidemiology studies have not
consistently found neurological effects to be associated with serum PFOS levels.

Table A-14. Summary of the Adverse Effects Observed in Laboratory Animals
Following Intermediate-Duration Oral Exposure to PFOS
Species and
exposure
duration
Hepatic
Monkey
26 weeks

NOAEL
LOAEL
(mg/kg/day) (mg/kg/day) Effect

Reference

0.03

0.75

Increased liver weight, decreased
serum cholesterol, hepatocellular
hypertrophy, lipid vacuolation

Seacat et al.
2002

0.43

2.15

Impaired spatial learning and
memory

Long et al. 2013

Mouse
28 days

0.00016

0.00166 Suppressed response to sRBC

Mouse
21 days

0.005

0.025

Decreased resistance to influenza Guruge et al.
virus
2009

Mouse
60 days

0.0083

0.083

Impaired response to sRBC

Dong et al. 2009

Mouse
60 days

0.0167

0.083

Impaired response to sRBC

Dong et al. 2011

0.3

Onishchenko et
al. 2011

0.4

Decreased locomotion, muscle
strength, motor coordination in
adult offspring
Delayed eye opening

0.4

Decreased pup weight

0.8

Decreases spatial learning

1

Increased locomotor activity and
concurrent failure to habituate to

Neurological
Mouse
3 months
Immunological

Developmental
Mouse
GDs 1–21
Rat
84 days
Rat
67 days
Rat
GD 1–PND 1
Rat
GD 0–PND 20

0.1

0.3

***DRAFT FOR PUBLIC COMMENT***

Peden-Adams et
al. 2008

Luebker et al.
2005a
Luebker et al.
2005b
Wang et al.
2015c
Butenhoff et al.
2009b

 PERFLUOROALKYLS

A-38
APPENDIX A

Table A-14. Summary of the Adverse Effects Observed in Laboratory Animals
Following Intermediate-Duration Oral Exposure to PFOS
Species and
exposure
duration

NOAEL
LOAEL
(mg/kg/day) (mg/kg/day) Effect

Reference

test environment in male pups on
PND 17
Mouse
GDs 1–17

1

Delayed eye opening

Lau et al. 2003

Mouse
GDs 0–17

1

Increased sternal defects

Yahia et al. 2008

Rat
GDs 1–21

0.1

2

Increased postnatal mortality and
severe lung histopathology

Chen et al. 2012b

Rat
GDs 2–21

0.6

2

Increased neonatal mortality

Xia et al. 2011

GD = gestation day; LOAEL = lowest-observed-effect level; NK = natural killer; NOAEL = no-observed-adverseeffect level; PFOS = perfluorooctane sulfonic acid; PND = postnatal day; sRBC = sheep red blood cell

Selection of the Principal Study: Using the Wambaugh et al. (2013) pharmacokinetic model, TWA
serum concentrations corresponding to external doses (mg/kg/day) and exposure durations (days) were
predicted for the studies listed in Table A-14. Pharmacokinetic model parameters were not available for
C57BL/6N mice, B6C3F1 mice, or Wistar rats, which precluded predicting TWA serum concentrations
for the Long et al. (2013), Dong et al. (2009, 2011), Guruge et al. (2009), Peden-Adams et al. (2008),
Wang et al. (2015c), Onishchenko et al. (2011), and Yahia et al. (2008) studies. The predicted serum
PFOS levels for each administered dose is presented in Table A-15.

Table A-15. Summary of the Predicted TWA Serum PFOS levels in Laboratory
Animals Following Intermediate-Duration Oral Exposure
Species and
exposure
Dose
duration
(mg/kg/day)
Hepatic
Cynomolgus 0.03 (males)
Monkey
0.15 (males)
26 weeks
0.75 (males)
0.03 (females)
0.15 (females)
0.75 (females)
Neurological
C57BL/6
0.43
Mouse
2.15
3 months
10.75

Predicted TWA
serum PFOS
(µg/mL)
7.81
37.8
150
7.72
37.6
146

Not calculated

Effect

Reference

Increased liver weight,
Seacat et al.
decreased serum cholesterol,
2002
hepatocellular hypertrophy, mild
bile stasis, lipid vacuolation at
0.75 mg/kg/day

Impaired spatial learning and
memory at 2.15 mg/kg/day

***DRAFT FOR PUBLIC COMMENT***

Long et al. 2013

 PERFLUOROALKYLS

A-39
APPENDIX A

Table A-15. Summary of the Predicted TWA Serum PFOS levels in Laboratory
Animals Following Intermediate-Duration Oral Exposure
Species and
exposure
Dose
duration
(mg/kg/day)
Immunological
B6C3F1
Mouse
28 days

B6C3F1
Mouse
21 days
C57BL/6N
Mouse
60 days

C57BL/6N
Mouse
60 days

Developmental
C57BL/6
Mouse
GDs 1–21
SpragueDawley
Rat
84 days
SpragueDawley
Rat
67 days

Wistar
Rat
GD 1–PND 1

0.00016
0.00166
0.00331
0.0166
0.0331
0.005
0.025
0.0083
0.083
0.41667
0.8333
2.0833
0.0083
0.0167
0.083
0.41667
0.8333

Predicted TWA
serum PFOS
(µg/mL)

Effect

Reference

Suppressed response to sRBC
at 0.00166 mg/kg/day

Peden-Adams et
al. 2008

Decreased resistance to
influenza virus at
0.025 mg/kg/day
Impaired response to sRBC at
0.083 mg/kg/day

Guruge et al.
2009

Impaired response to sRBC at
0.083 mg/kg/day

Dong et al. 2011

Decreased locomotion, muscle
strength, motor coordination in
adult offspring at 0.3 mg/kg/day
Delayed eye opening in F1 pups
and transient decrease in F2 pup
body weight during lactation at
0.4 mg/kg/day

Onishchenko et
al. 2011

Decreased pup weight per litter
at birth and on LD 5 at 0.4
mg/kg/day

Luebker et al.
2005b

Decreases spatial learning at
0.8 mg/kg/day

Wang et al.
2015c

Not calculated

Not calculated

Dong et al. 2009

Not calculated

Not calculated

0.3

Not calculated

0.1
0.4
1.6
3.2
0.4
0.8
1
1.2
1.6
3.2
0.8

7.43
29.7
119
238
24.1
48.1
60.1
72.2
96.2
120
Not calculated

***DRAFT FOR PUBLIC COMMENT***

Luebker et al.
2005a

 PERFLUOROALKYLS

A-40
APPENDIX A

Table A-15. Summary of the Predicted TWA Serum PFOS levels in Laboratory
Animals Following Intermediate-Duration Oral Exposure
Species and
exposure
Dose
duration
(mg/kg/day)
SpragueDawley
Rat
GD 0–
PND 20
CD-1
Mouse
GDs 1–17

ICR
Mouse
GDs 0–17
SpragueDawley
Rat
GDs 1–21
SpragueDawley
Rat
GDs 2–21

0.1
0.3
1

Predicted TWA
serum PFOS
(µg/mL)

Effect

Reference

3.75
11.3
37.5

Increased locomotor activity and Butenhoff et al.
concurrent failure to habituate to 2009b
test environment in male pups
on PND 17 at 1 mg/kg/day

31.9
146
216
244
260
Not calculated

Delayed eye opening at
1 mg/kg/day

Lau et al. 2003

Increased sternal defects at
1 mg/kg/day

Yahia et al. 2008

0.1
2

2.01
40.1

Increased postnatal mortality
and severe lung histopathology
at 2 mg/kg/day

Chen et al. 2012b

0.1
0.6
2

1.92
11.5
38.3

Increased neonatal mortality at
2 mg/kg/day

Xia et al. 2011

1
5
10
15
20
1

GD = gestation day; LD = lactation day; perfluorooctane sulfonic acid; PND = postnatal day; sRBC = sheep red
blood cell; TWA – time-weighted average

Selection of the Point of Departure for the MRL: None of the studies with predicted serum PFOS levels
had datasets that were amenable for BMD modeling; thus, the NOAEL/LOAEL approach was used to
identify PODs for derivation of the intermediate-duration MRL for PFOS. A summary of the PODs is
presented in Table A-16. HEDs were calculated for each potential POD (NOAEL or LOAEL) identified
in laboratory animal studies using the first-order single-compartment model previously discussed and the
assumption that humans would have similar effects as the laboratory animal at a given serum
concentration. The HEDs for each POD are presented in Table A-16. The potential PODHED values were
divided by a total uncertainty factor to calculate candidate MRLs; these values are also presented in
Table A-16. The lowest administered doses associated with adverse effects were found in the
immunotoxicity studies conducted by Dong et al. (2009, 2011), Guruge et al. (2009), and Peden-Adams et
al. (2008). These data could not be considered as PODs because TWA serum PFOS values could not be
predicted due to the lack of pharmacokinetic model parameters for the two mouse strains tested.
Although there is considerable overlap between the LOAEL for IgM response to sRBC
(0.00166 mg/kg/day) identified in the Peden-Adams et al. (2008) 28-day study and the NOAELs for IgM
response to sRBC (0.0083 and 0.0167 mg/kg/day) identified in the Dong et al. (2009, 2011) 60-day
studies, the data do suggest that immunotoxicity could occur at <0.3 mg/kg/day (the lowest LOAEL
identified in developmental toxicity studies).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-41
APPENDIX A

Table A-16. Summary of Potential Points of Departures Human Equivalent Doses
(PODHED) for Intermediate-Duration Oral MRL for PFOS

Endpoint (reference)

Predicted serum
concentrations
(µg/mL)
NOAEL LOAEL

Increased rat pup mortality and
2.01
lung histopathology (Chen et al.
2012b)
Decreased rat pup weight at birth
and on PND 4 (Luebker et al.
2005b)
Delayed eye opening in mouse
pups (Lau et al. 2003)
Neurodevelopmental effects in
11.3
male rat pups (Butenhoff et al.
2009b)
Delayed eye opening and
7.43
decreased F2 rat pup body weight
(Luebker et al. 2005a)
Increased neonatal mortality in rat 11.5
pups (Xia et al. 2011)
Hepatic effects in monkeys
37.8
(Seacat et al. 2002)

PODHEDa
(mg/kg/day)

Candidate
MRLs
Total UF (mg/kg/day)
300b

4.6x10-7

0.00167

3,000c

5.6x10-7

31.9

0.00221

3,000c

7.4x10-7

37.5

0.000780

300b

2.6x10-6

29.7

0.000515

300b

1.7x10-6

38.3

0.000797

300b

2.7x10-6

0.00262

300b

8.7x10-6

40.1

0.000139

24.1

150

aBased

on the NOAEL or the LOAEL if NOAEL was not identified.
factors (UF) of 3 for extrapolation from animals to humans with dosimetric adjustments and 10 for
human variability and modifying factor (MF) of 10 for concern that immunotoxicity may be a more sensitive endpoint
than developmental toxicity.
cUF of 10 for extrapolation from a LOAEL, 3 for extrapolation from animals to humans with dosimetric adjustments,
and UF of 10 for human variability and MF of 10 for concern that immunotoxicity may be a more sensitive endpoint
than developmental toxicity.
bUncertainty

LOAEL = lowest-observed-effect level; MRL = Minimal Risk Level; NOAEL = no-observed-adverse-effect level;
perfluorooctane sulfonic acid; PND = postnatal day; UF = uncertainty factor

The serum PFOS concentrations predicted to occur at the lowest LOAEL values were 24.1, 29.7, and
31.9 µg/mL identified in the Luebker et al. (2005b), Luebker et al. (2005a), and Lau et al. (2003) studies;
decreases in pup body weight and delays in eye opening were observed at these levels. Luebker et al.
(2005a) was the only study that identified a NOAEL for these effects. The predicted serum concentration
for this NOAEL dose was selected as the basis for the MRL.
Summary of the Principal Study:
Luebker DJ, Case MT, York RG, et al. 2005a. Two-generation reproduction and cross-foster studies of
perfluorooctanesulfonate (PFOS) in rats. Toxicol 215: 126-148.
Groups of Sprague-Dawley rats (P generation) (35/sex/dose level) were administered PFOS (86.9% pure)
by gavage in deionized water with 2% Tween-80 at doses of 0, 0.1, 0.4, 1.6, or 3.2 mg/kg/day for 6 weeks
before mating and until sacrifice (after mating for males, GD 10 for some females, and PND 21 for the

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-42
APPENDIX A

remaining females). Body weight and feed consumption were evaluated during the dosing period. Prior
to mating, 15 females per dose group were evaluated for estrous cycling. Ten females/dose group were
sacrificed on GD 10 and the remaining females were allowed to give birth (F1 generation). Parental rats
sacrificed on GD 10 were examined for number of corpora lutea, implantations, and viable and non-viable
fetuses. Body weight of F1 was evaluated during lactation; also, F1 rats were assessed for developmental
landmarks during lactation. On PND 22, the F1 rats were started on the same diet as the parental rats. At
approximately PND 90, F1 were mated to produce the F2 generation. F1 males and females were killed
as the P generation. F1 females and males were evaluated for vaginal patency and preputial separation,
respectively. At age 24 days, F1 rats were administered three neurobehavioral tests (learning, memory
retention, and avoidance memory). At the age of 70 days, F1 were administered three different
neurobehavioral tests (neuromuscular coordination, swimming ability, learning, and memory). PFOS was
analyzed in liver and blood from parental females and in liver from F1on PND 21; and in liver and serum
from parental males after mating and after 42–56 days of dosing.
There were no deaths in parental males or females and no clinical signs in parental males. High-dose
parental males had significantly reduced terminal body weight (11% reduction). Absolute and relative
food consumption was reduced during treatment in males by less than 10%. Parental females at
0.4 mg/kg/day and higher had localized areas of partial alopecia. Body weight of high-dose parental
females was significantly lower during cohabitation and gestation (11% reduced). Absolute and relative
food consumption were significantly reduced in high-dose parental females during premating and
gestation (>15%) and in 1.6 mg/kg/day parental females during lactation. Administration of PFOS did
not affect any mating or fertility parameter. Estrous cycling was not affected. Examination of parental
females sacrificed on GD 10 showed no significant effect on numbers of corpora lutea or implantations or
viable and non-viable fetuses. Significant delivery observations for high-dose parental females included
reduced number of implantations per delivered litter, decreased gestational length, increased number of
dams with all pups dying on PND 1–4 (also at 1.6 mg/kg/day). Observation of F1 pups during PNDs 1–
21 showed significantly reduced weight and decreased viability (≥1.6 mg/kg/day). Examination of dead
F1 pups did not reveal a cause of death; no labored breathing was noted in pups at birth. Developmental
delays were noted at 1.6 mg/kg/day (pinna unfolding, surface righting, and air righting), and
0.4 mg/kg/day (eye opening). Follow-up observations of 0.1 and 0.4 mg/kg/day offspring showed no
alterations in body weight or food consumption, including F1 females during gestation and lactation.
Sexual maturation was not affected in F1 males or females; no effects were noted in the neurobehavioral
tests. Reproductive performance of F1 were not affected. Viability of F2 pups during PNDs 1–21 was
not affected. F2 pup weight was significantly reduced at 0.4 mg/kg/day on PND 7 (13%). Serum and
liver PFOS increased with dose.
Calculations of Internal Dosimetric: TWA serum PFOS concentrations corresponding to external doses
and exposure durations were predicted from a pharmacokinetic model (Wambaugh et al. 2013) using
animal species-, strain-, and sex-specific parameters (see MRL approach section for details).
Human Equivalent Dose: HEDs were calculated based on the assumption that humans would have
similar effects as the laboratory animal at a given serum concentration. HEDs that would result in steadystate serum concentrations of PFOS equal to the serum concentration selected as the POD were calculated
using the first order single-compartment model (see MRL approach section for details).
Uncertainty Factor and Modifying Factor: The NOAELHED is divided by a total uncertainty factor (UF)
of 30 and modifying factor (MF) of 10:
• 3 UF for extrapolation from animals to humans with dosimetric adjustment
• 10 UF for human variability
• 10 MF for concern that immunotoxicity may be a more sensitive endpoint of PFOS toxicity than
developmental toxicity

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-43
APPENDIX A

Provisional MRL = NOAELHED ÷ (UFs x MF)
0.000515 mg/kg/day ÷ ((3 x 10) x 10) = 2x10-6 mg/kg/day
Although pharmacokinetic model parameters were not available for the strain/sex of the animals tested in
the immunotoxicity studies, most of the studies did provide measured serum PFOS levels. The serum
PFOS levels at the NOAEL and LOAEL doses are presented in Table A-17. The measured serum PFOS
levels associated with altered immune responses are approximately 1–10 times lower than the serum
concentration predicted to occur at the NOAEL dose. These data suggest that immunotoxicity may be a
more sensitive effect than developmental toxicity.

Table A-17. Measured Serum PFOS Levels at the NOAEL and LOAEL Doses for
Immunological Effects

Effect, species and
exposure duration

Dose
(mg/kg/day)

Measured mean
serum PFOS
(µg/mL)
Reference

Impaired response to sRBC in NOAEL
mice exposed for 60 days
LOAEL
Impaired response to sRBC in NOAEL
mice exposed for 60 days
LOAEL

0.0083
0.083
0.0167
0.083

0.674
7.132
2.36
10.75

Dong et al. 2009

Decreased resistance to
influenza virus in mice
exposed for 21 days
Suppressed response to sRBC
in mice exposed for 28 days

NOAEL
LOAEL

0.005
0.025

0.189
0.670

Guruge et al.
2009

NOAEL
LOAEL

0.00016
0.00166

0.0178
0.0915

Peden-Adams et
al. 2008

Dong et al. 2011

PFOS = perfluorooctane sulfonic acid; sRBC = sheep red blood cell

A candidate MRL was calculated using the NOAEL of 0.0167 mg/kg/day identified in the Dong et al.
(2011). This study was selected over the other immunotoxicity studies because it identified the highest
NOAEL for immunotoxicity and it had the longest exposure duration; the Peden-Adams et al. (2008) was
not selected because the LOAEL of 0.00166 mg/kg/day is not supported by the other three studies. A
TWA concentration was estimated using a similar approach described for PFHxS and PFNA in the MRL
approach section. The estimated TWA concentration was 1.2 µg/mL for the 0.0167 mg/kg/day; this
estimated TWA concentration was used to calculated a HED of 0.000083 mg/kg/day. A candidate MRL
of 3x10-6 was calculated using an uncertainty factor of 30 (3 for extrapolation from animals to humans
using dosimetric adjustments and 10 for human variability). This MRL is similar to the MRL calculated
from the Luebker et al. (2005a) study and lends to support to using the additional uncertainty factor of
10 to account for the lack of pharmacokinetic modeling parameters for the mouse strains tested for
immunotoxicity.
Other Additional Studies or Pertinent Information that Lend Support to this MRL: A discussion of the
findings from epidemiology studies is presented in the MRL introduction section.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-44
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctane sulfonic acid (PFOS)
1763-23-1
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFOS.
Rationale for Not Deriving an MRL: Immune function was not examined following chronic-duration
oral exposure in laboratory animal studies; in intermediate-duration animal studies, the lowest LOAEL
doses were for immunological effects. Given the concern that immunotoxicity may occur at lower doses
than liver toxicity, a chronic-duration oral MRL for PFOS is not recommended at this time.
One study has evaluated the chronic toxicity of PFOS in laboratory animals. Histological alterations in
the liver were the primary effects observed in rats exposed to PFOS in the diet for 2 years (Butenhoff et
al. 2012b; Thomford 2002b). Centrilobular hepatocellular hypertrophy was observed in rats exposed to
≥0.1 mg/kg/day. At 1.04 mg/kg/day, increases in the incidence of single cell necrosis and cystic
degeneration were observed in the liver. Decreases in body weight were observed at 1.04 mg/kg/day in
female rats. Thus, the 1.04 mg/kg/day dose was identified as the lowest LOAEL for this study.
Epidemiology data (Dalsager et al. 2016; Dong et al. 2013; Fei et al. 2010; Grandjean et al. 2012, 2016;
Granum et al. 2013; Kielsen et al. 2016; Mogensen et al. 2015a; Stein et al. 2016a; Zhu et al. 2016)
suggest that the immune system is a sensitive target of PFOS toxicity following long-term exposures,
which is supported by intermediate-duration PFOS laboratory animal studies (Dong et al. 2009, 2011;
Guruge et al. 2009; Peden-Adams et al. 2008).
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-45
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexane sulfonic acid (PFHxS)
355-46-4
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFHxS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFHxS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-46
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexane sulfonic acid (PFHxS)
355-46-4
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFHxS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFHxS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-47
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexane sulfonic acid (PFHxS)
355-46-4
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFHxS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFHxS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-48
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexane sulfonic acid (PFHxS)
355-46-4
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFHxS.
Rationale for Not Deriving an MRL: The acute oral database for PFHxS was not considered adequate
for derivation of an MRL due to the short duration of the only available study and the lack of
pharmacokinetic model parameters for calculating an HED.
In the only available study of PFHxS in laboratory animals, Viberg et al. (2013) reported altered
spontaneous behavior and habituation in adult mice administered a single gavage dose of 9.2 mg/kg/day
PFHxS on PND 10; no alterations were observed at 6.1 mg/kg/day. This single exposure study was not
considered adequate as the basis of an acute-duration MRL for PFHxS due to the uncertainty of whether
an MRL based on this study would be protective for repeated exposures or for other potential sensitive
endpoints, such as immunotoxicity.
For perfluoroalkyls, ATSDR has used the approach of predicting TWA serum perfluoroalkyl levels in
laboratory animals and calculating HEDs for these serum concentrations. For PFOA and PFOS, the
Wambaugh et al. (2013) pharmacokinetic model was utilized for predicting the TWA serum
perfluoroalkyl concentrations. However, strain-, sex-, and compound-specific model parameters are not
available for other perfluoroalkyls, thus precluding deriving MRLs for other perfluoroalkyls. Other
approaches such as “read across” (i.e., using data for a particular endpoint from one chemical to predict
the same endpoint for another chemical that has similar chemical structure or mechanisms of action) or
equivalency factors were considered for the other perfluoroalkyl compounds; however, there are limited
data available that would allow for comparison of the toxicity and toxicokinetic properties of different
perfluoroalkyl compounds. Peters and Gonzalez (2011) noted that the toxic equivalency factor approach
would not be suitable for perfluoroalkyls because the current data suggest that the toxicity of these
compounds appear to be mediated by multiple receptors, including PPARα, CAR, and PXR, and that
there may be species differences in the response mediated by different receptors. Additionally, available
data suggest that there are qualitative differences in the toxicities of various perfluoroalkyls.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-49
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:
Provisional MRL:
Critical Effect:
Reference:
Point of Departure:
Uncertainty Factor:
Modifying Factor:
LSE Graph Key:
Species:

Perfluorohexane sulfonic acid (PFHxS)
355-46-4
June 2018
Final, Pre-Public Comment
Oral
Intermediate
2x10-5 mg/kg/day
Thyroid follicular cell damage
Butenhoff et al. 2009a; Hoberman and York 2003
0.0047 mg/kg/day
30
10
32
Rat

MRL Summary: A provisional intermediate-duration oral MRL of 2x10-5 mg/kg/day was derived for
PFHxS based on thyroid follicular cell damage in adult male rats administered via gavage PFHxS for a
minimum of 42 days (Butenhoff et al. 2009a; Hoberman and York 2003). The MRL is based on a HED
NOAEL of 0.0047 mg/kg/day and a total uncertainty factor of 30 (3 for extrapolation from animals to
humans with dosimetric adjustments and 10 for human variability) and a modifying factor of 10 for
database limitations.
Selection of the Critical Effect: Two intermediate-duration studies in laboratory animals have been
identified for PFHxS. In a developmental toxicity study, increased incidences of thyroid follicular cells
hypertrophy, and hyperplasia were observed in F0 male rats administered ≥3 mg/kg/day (Butenhoff et al.
2009a; Hoberman and York 2003). Increased liver weight and centrilobular hepatocellular hypertrophy
were also observed in the males at ≥3 mg/kg/day. No reproductive or developmental effects were
reported. Liver effects (decreases in serum lipids, increases in hepatic triglyceride levels, and increases in
liver weight) were also observed in mice exposed to 6 mg/kg/day PFHxS in the diet for 4–6 weeks
(Bijland et al. 2011). Using the Hall et al. (2012) criteria (see Section 2.9 for a discussion of the criteria),
the liver effects were not considered relevant for human risk assessment. Thus, the lowest LOAEL
identified in intermediate-duration studies was 3 mg/kg/day for thyroid effects.
Selection of the Principal Study: Since the liver effects were not considered relevant to humans, the
Butenhoff et al. (2009a) study was selected as the principal study. Thyroid effects in male rats was noted
as the most sensitive endpoint in this study, with a LOAEL of 3 mg/kg/day and a NOAEL of
1 mg/kg/day.
Summary of the Principal Study:
Butenhoff JL, Chang SC, Ehresman DJ, et al. 2009a. Evaluation of potential reproductive and
developmental toxicity of potassium perfluorohexanesulfonate in Sprague Dawley rats. Reprod Toxicol
27:331-341.
Hoberman AM, York RG. 2003. Oral (gavage) combined repeated dose toxicity study of T-7706 with
the reproduction/developmental toxicity screening test. Argus Research.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-50
APPENDIX A

The reproductive/developmental effects of PFHxS were studied in Sprague-Dawley rats (15/sex/group).
Doses of 0, 0.3, 1, 3, or 10 mg/kg/day PFHxS were administered by gavage in an aqueous vehicle. Male
rats were dosed beginning 14 days before cohabitation and continued until 1 day before sacrifice (a
minimum of 42 days). Females were dosed beginning 14 days before cohabitation and continued until
1 day before sacrifice on PND 21 or GD 25 (rats that did not deliver a litter). Endpoints evaluated
included: body weight, food consumption, estrous cycling, functional observational battery (FOB; tests of
autonomic function, reactivity and sensitivity, excitability, gait and sensorimotor coordination, grip
strength, and clinical signs), hematology and clinical chemistry, gross necropsy, organ weights,
histopathology, and sperm evaluations. At parturition, litters were evaluated for size and viability; weight
of the pups was also recorded. Pups were sacrificed on PND 22.
The following are findings for male F0 rats. Treatment with PFHxS did not affect survival and did not
induce clinical signs that could be attributed to the chemical. Terminal body weight in the 10 mg/kg/day
groups was approximately 6% lower than controls. Food consumption was not affected. Necropsy did
not reveal any treatment-related changes. Histopathological effects were restricted to the liver and
thyroid of males treated with 3 and 10 mg/kg/day. Liver effects consisted of minimal to moderate
hypertrophy of centrilobular hepatocytes. The affected hepatocytes were enlarged with an increased
amount of dense eosinophilic granular cytoplasm. In the thyroid, the changes consisted of hypertrophy
and hyperplasia of follicular cells. These effects could have been associated with the liver effects.
Significant organ weight changes consisted of increased absolute and relative liver weight at 3 and
10 mg/kg/day and decreased heart/brain weight at 10 mg/kg/day. Significant hematology changes
consisted of decreased hemoglobin at 1 mg/kg/day, decreased red cell count and hematocrit at
3 mg/kg/day, and increased prothrombin time at 0.3 mg/kg/day. Increases in albumin, BUN, alkaline
phosphatase, calcium, and albumin/globulin ratio were seen at 10 mg/kg/day. There were no significant
effects on the FOB or on motor activity and no significant effects on sperm parameters. There were no
significant effects in any parameter monitored in F0 females or in pups. Treatment with PFHxS had no
significant effect on the gross or microscopic morphology of the spleen, thymus, or lymph nodes. There
were no significant effects on sex organ weights or gross or microscopic lesions in the reproductive
organs of males and females. Fertility was not affected by treatment with PFHxS and there were no
significant effects on sperm parameters. Estrous cycling was not affected by dosing with PFHxS.
Treatment with PFHxS did not significantly affect any of the developmental parameters evaluated
including gestation length, number of dams delivering litters, averages for implantation sites per delivered
litter, number of dams with stillborn pups, number of dams with no live pups, dams with all pups dying,
number of pups surviving per litter, sex ratios, litter size, or pup weight. Also, necropsy of the pups
showed no treatment-related effects, and pup liver weight was not affected. Treatment with PFHxS had
no significant effect on the FOB or motor activity. The battery tested autonomic functions, reactivity and
sensitivity to stimuli, excitability, gait and sensorimotor coordination, limb grip strength, and abnormal
clinical signs.
Selection of the Point of Departure for the MRL: The HED of the NOAEL of 1 mg/kg/day identified in
the Butenhoff et al. (2009a) developmental toxicity study was selected as the POD for the MRL. A TWA
serum PFHxS concentration of 73.22 µg/mL was estimated for the adult males exposed to 1 mg/kg/day
(Butenhoff et al. 2009a).
Human Equivalent Dose: HEDs were calculated based on the assumption that humans would have
similar effects as the laboratory animal at a given serum concentration. HEDs that would result in steadystate serum concentrations of PFHxS equal to the estimated TWA serum concentration selected as the
POD were calculated using the first-order single-compartment model (see MRL approach section for
details). The PFHxS model parameters for humans are presented in Table A-4. The NOAELHED is
0.0047 mg/kg/day

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-51
APPENDIX A

Uncertainty Factor and Modifying Factor: The NOAELHED is divided by a total uncertainty factor (UF)
of 30 and a modifying factor (MF) of 10:
• 3 UF for extrapolation from animals to humans with dosimetric adjustment
• 10 UF for human variability
• 10 MF for database limitations to account for small number of studies examining the toxicity of
PFHxS following intermediate-duration exposure and the limited scope of these studies in
particular studies examining immunotoxicity, a sensitive endpoint for other perfluoroalkyl
compounds, and general toxicity.
Provisional MRL = NOAELHED ÷ (UFs x MF)
0.0047 mg/kg/day ÷ ((10 x 3) x 10) = 2x10-5 mg/kg/day
Other Additional Studies or Pertinent Information that Lend Support to this MRL: A discussion of the
findings from epidemiology studies is presented in the MRL introduction section.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-52
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexane sulfonic acid (PFHxS)
355-46-4
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFHxS.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFHxS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-53
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorononanoic acid (PFNA)
375-95-1
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFNA.
Rationale for Not Deriving an MRL: The only available inhalation exposure study for PFNA (Kinney et
al. 1989) was not considered suitable for derivation of an inhalation MRL due to its lack of
histopathological examination and short exposure duration.
In the only available inhalation exposure study for PFNA, Kinney et al. (1989) noted labored breathing in
rats during and after a 4-hour nose-only exposure to 590 mg/m3 exposure; the study also reported an
increase in relative liver weight 5 days after exposure to ≥67 mg/m3.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-54
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorononanoic acid (PFNA)
375-95-1
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFNA.
Rationale for Not Deriving an MRL: No intermediate-duration inhalation studies in laboratory animals
were identified for PFNA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-55
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorononanoic acid (PFNA)
375-95-1
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFNA.
Rationale for Not Deriving an MRL: No chronic-duration inhalation studies in laboratory animals were
identified for PFNA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-56
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorononanoic acid (PFNA)
375-95-1
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFNA.
Rationale for Not Deriving an MRL: An acute-duration oral MRL cannot be derived for PFNA because
the study identifying the lowest dose for a non-hepatic effect (Fang et al. 2009) did not measure serum
PFNA levels, which are needed for estimating an HED.
A number of studies of examined the toxicity of PFNA in rats and mice exposed for acute durations.
These studies reported immune, liver, and body weight effects. Immune effects included increases in
thymus weight in rats at 1 mg/kg/day (Fang et al. 2009), decreases in thymus and spleen weights in rats at
3 mg/kg/day (Fang et al. 2009, 2010), and an alteration in splenic lymphocyte phenotypes in mice at 1
mg/kg/day (Fang et al. 2008). In the only study examining immune function, no alterations in splenic
lymphocyte response to ConA were observed at doses as high as 5 mg/kg/day in mice (Fang et al. 2008).
Liver effects included increases in hepatic lipid levels at ≥0.2 mg/kg/day (Wang et al. 2015a), increases in
liver weights at ≥0.2 mg/kg/day (Wang et al. 2015a; Kennedy 1987), serum lipid levels at ≥1 mg/kg/day
(Fang et al. 2012a, 2012b), hepatocellular vacuolation at 5 mg/kg/day (Fang et al. 2012b), and increases
in serum aminotransferases at 5 mg/kg/day (Wang et al. 2015a). Decreases in body weight were observed
in rats and mice administered 5 mg/kg/day (Hadrup et al. 2016; Wang et al. 2015a).
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-57
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:
Provisional MRL:
Critical Effect:
Reference:
Point of Departure:
Uncertainty Factor:
Modifying Factor:
LSE Graph Key:
Species:

Perfluorononanoic acid (PFNA)
375-95-1
June 2018
Final, Pre-Public Comment
Oral
Intermediate
3x10-6 mg/kg/day
Decreased body weight and developmental delays
Das et al. 2015
0.001 mg/kg/day
30
10
35
Mouse

MRL Summary: A provisional intermediate-duration oral MRL of 3x10-6 mg/kg/day was derived for
PFNA based on decreased body weight gain and developmental delays in the offspring of mice
administered via gavage PFNA on GDs 1–17 (Das et al. 2015). The MRL is based on a HED NOAEL of
0.001 mg/kg/day and a total uncertainty factor of 30 (3 for extrapolation from animals to humans with
dosimetric adjustments and 10 for human variability), and a modifying factor of 10 for database
limitations.
Selection of the Critical Effect: The intermediate-duration database consists of three developmental
toxicity studies in rats and mice. The lowest LOAEL for developmental toxicity was 1.1 mg/kg/day in
mice administered PFNA on GDs 1–18; at this dose, decreases in litter size and pup survival were
observed (Wolf et al. 2010). At higher doses (2–5 mg/kg/day), decreases in pup body weight, delays in
postnatal development (Das et al. 2015; Rogers et al. 2014; Wolf et al. 2010), increases in pup systolic
blood pressure (Rogers et al. 2014), and reduced nephron endowment (Rogers et al. 2014) were observed.
A study of PPARα knockout mice did not find alterations pup body weight or postnatal development at
2 mg/kg/day (Wolf et al. 2010). A summary of developmental effects is presented in Table A-18.

Table A-18. Summary of the Adverse Effects Observed in Laboratory Animals
Following Intermediate-Duration Oral Exposure to PFNA
Species and exposure
duration

Dose
(mg/kg/day)

129S1/svlm mouse
GDs 1–18
(offspring followed until
PND 21)

0.83
1.1

CD-1 mouse
GDs 1–17
(offspring followed until
PND 287

1
3

1.5
2.0

Effect

Reference

No effects reported
Wolf et al. 2010
Decreased litter size and pup
survival
No effects reported
Decreased number of live pups per
litter and decreased pup body
weight gain
No effects reported
Das et al. 2015
Decreased body weight gain and
delayed eye opening, preputial
separation, and vaginal opening

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-58
APPENDIX A

Table A-18. Summary of the Adverse Effects Observed in Laboratory Animals
Following Intermediate-Duration Oral Exposure to PFNA
Species and exposure
duration

Dose
(mg/kg/day)
5

Reference

Decreased postnatal survival, 80%
mortality between PND 2 and 10
Full litter resorption
Decreased birth weight, increased Rogers et al. 2014
blood pressure at 10 weeks of age;
reduced nephron endowment

10
5

Sprague-Dawley rat
GDs 1–20
offspring followed through
PND 434)

Effect

GD = gestation day; PFNA perfluorononanoic acid; PND = postnatal day

Selection of the Principal Study: Developmental toxicity, including decreases in pup survival,
developmental delays, and decreases in birth weight have been observed in three studies. A comparison
of the estimated TWA serum PFNA levels (Table A-19) for the Wolf et al. (2010) and Das et al. (2015)
studies (measured serum levels were not available from the Rogers et al. 2014 study) showed that the
lowest LOAEL for developmental effects was 10.9 µg/mL (Das et al. 2015); this study reported a
NOAEL of 6.8 µg/mL. Thus, the Das et al. (2015) study was selected as the principal study for the MRL.

Table A-19. Summary of Estimated TWA Serum PFNA levels in Laboratory
Animals Following Intermediate-Duration Oral Exposure
Estimated
TWA serum
Species and exposure Dose
PFNA
duration
(mg/kg/day) (µg/mL)
Effect
129S1/svlm mouse
GDs 1–18
(offspring followed until
PND 21)

0.83
1.1

4.47
11.6

1.5
2.0

10.5
17.6

CD-1 mouse
GDs 1–17
(offspring followed until
PND 287

1
3

6.8
10.9

5

39.7

Sprague-Dawley rat
GDs 1–20
(offspring followed
through PND 434)

10
5

NA
Not
calculated

No effects reported
Decreased litter size and pup
survival
No effects reported
Decreased number of live pups
per litter and decreased pup body
weight gain
No effects reported
Decreased body weight gain and
delayed eye opening, preputial
separation, and vaginal opening
Decreased postnatal survival, 80%
mortality between PND 2 and 10
Full litter resorption
Decreased birth weight, increased
blood pressure at 10 weeks of
age; reduced nephron endowment

GD = gestation day; PFNA perfluorononanoic acid; PND = postnatal day

***DRAFT FOR PUBLIC COMMENT***

Reference
Wolf et al.
2010

Das et al.
2015

Rogers et
al. 2014

 PERFLUOROALKYLS

A-59
APPENDIX A

Summary of the Principal Study:
Das KP, Grey BE, Rosen MB, Wood CR, Tatum-Gibbs KR, Zehr RD, Strynar MJ, Lindstrom AB, Lau C.
2015. Developmental toxicity of perfluorononanoic acid in mice. Reprod Toxicol. 51:133-44.
Groups of 8–10 timed-pregnant female CD-1 mice were administered via gavage 1, 3, 5, or 10 mg/kg/day
PNFA at a dosing volume of 10 ml/kg body weight in deionized water on GDs 1–17. On GD 17, selected
mice from each group were sacrificed for maternal and fetal examination, while the remaining mice were
allowed to give birth. Pups were observed for postnatal survival up to PND 24 as well as growth and
development up to PND 287. The following parameters were used to assess toxicity: clinical
observations, maternal body weight, pup body weight (pre- and postnatal), organ weights (liver, gravid
uterus weight), number of implantation sites, percent of live fetuses, percent of prenatal loss per litter, and
morphological changes (eye opening, vaginal opening, preputial separation).
Maternal weight loss beginning on GD 8 was observed at 10 mg/kg/day; on GD 13, the 10 mg/kg/day
group weighed approximately 30% less than controls. The 10 mg/kg/day group was terminated on
GD 13. Significant increases in full litter resorptions occurred at 10 mg/kg/day (7/7 compared to 2/8 in
controls). There were no adverse effects on pregnancy outcome following in utero exposure to 5 mg/kg.
Statistically significant dose-related increases in absolute and relative liver weights were observed in
dams in the 1, 3, and 5 mg/kg/day groups examined on GD 17, as well as in dams examined on postweaning day 28. There were no effects on the number of implants, number of live fetuses, or fetal body
weight. Relative and absolute fetal liver weight were significantly increased; however, the increase did
not appear to be dose-related. Visceral and skeletal examination of fetuses revealed no treatment-related
effects. Increases in postnatal deaths were observed in the 5 mg/kg/day offspring between PND 2 and 10;
postnatal survival was approximately 20% on PND 10. Weight gain was significantly reduced in male
pups from the 3 and 5 mg/kg dose groups and in females from the 5 mg/kg dose group. The changes in
males were dose-related and persisted from PND 25 to 287. Weight reduction in females was less
substantial in comparison with males and returned to control levels by 7 weeks of age. Relative pup liver
weights were significantly increased at all doses up to PND 24 and at 3 and 5 mg/kg/day on PND 42. No
significant effects on liver weight were detectable by PND 70. Postnatal development (eye opening,
preputial separation, and vaginal opening) was significantly delayed (by 2–7 days) at 3 and 5 mg/kg/day.
Selection of the Point of Departure for the MRL: The HED of the NOAEL of 1 mg/kg/day identified in
the Das et al. (2015) developmental toxicity study was selected as the POD for the MRL. A TWA serum
PFNA concentration was estimated for dams using the serum concentration in the control group
(0.015 µg/mL) as the baseline concentrations and the terminal concentration for the 1 mg/kg/day group
(13.67 µg/mL) resulting in an estimated TWA serum concentration of 6.8 µg/mL (serum concentrations
were provided to ATSDR).
Human Equivalent Dose: HEDs were calculated based on the assumption that humans would have
similar effects as the laboratory animal at a given serum concentration. HEDs that would result in steadystate serum concentrations of PFNA equal to the estimated TWA serum concentration selected as the
POD were calculated using the first-order single-compartment model (see MRL approach section for
details). The PFNA model parameters for humans are presented in Table A-4. The NOAELHED is
0.001 mg/kg/day
Uncertainty Factor and Modifying Factor: The NOAELHED is divided by a total uncertainty factor (UF)
of 30 and modifying factor (MF) of 10:
• 3 UF for extrapolation from animals to humans with dosimetric adjustment
• 10 UF for human variability

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-60
APPENDIX A

•

10 MF for database limitations to account for small number of studies examining the toxicity of
PFNA following intermediate-duration exposure and the limited scope of these studies. In
particular, no studies were identified that examined immunotoxicity, a sensitive endpoint for
other perfluoroalkyl compounds, or general toxicity.
Provisional MRL = NOAELHED ÷ (UFs x MF)
0.001 mg/kg/day ÷ ((10 x 3) x 10) = 3x10-6 mg/kg/day

Other Additional Studies or Pertinent Information that Lend Support to this MRL: A discussion of the
findings from epidemiology studies is presented in the MRL introduction section.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-61
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorononanoic acid (PFNA)
375-95-1
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFNA.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFNA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-62
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorodecanoic acid (PFDeA)
335-76-2
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFDeA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFDeA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-63
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorodecanoic acid (PFDeA)
335-76-2
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFDeA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFDeA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-64
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorodecanoic acid (PFDeA)
335-76-2
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFDeA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFDeA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-65
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorodecanoic acid (PFDeA)
335-76-2
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFDeA.
Rationale for Not Deriving an MRL: The available acute oral database for PFDeA was not considered
adequate for derivation of an MRL because the study identifying the lowest adverse effect level did not
measure serum PFDeA levels, which are needed to estimate HEDs.
Several laboratory animal studies have examined the acute oral toxicity of PFDeA; most were limited in
scope. The lowest LOAEL was 1 mg/kg/day for decreases in fetal weight in mice administered PFDeA
on GDs 6–15 (Harris and Birnbaum 1989). At 12.8 mg/kg/day, decreases in the number of live fetuses
per litter were observed; maternal weight loss was also observed at this dose level (Harris and Birnbaum
1989). Another developmental toxicity study did not report alterations in performance on
neurobehavioral tests in 2–4-month-old mice administered 10.8 mg/kg/day PFDeA on PND 10
(Johansson et al. 2008). Other effects observed in acute exposure studies include decreases in maternal
weight gain at 6.4 mg/kg/day (Harris and Birnbaum 1989), weight loss at ≥9.5 mg/kg/day in rats
(Kawashima et al. 1995) and mice (Harris and Birnbaum 1989; Permadi et al. 1992, 1993), increases in
T3 and T4 levels in mice at 80 mg/kg/day (Harris et al. 1989), decreases in spleen weight in mice at
80 mg/kg/day (Harris et al. 1989), and atrophy and lymphoid depletion in thymus and spleen in mice at
160 mg/kg/day (Harris et al. 1989). Liver effects included increases in liver weight at ≥2.4 mg/kg/day
(Brewster and Birnbaum 1989; Harris et al. 1989; Kawashima et al. 1995; Permadi et al. 1992, 1993),
increases in hepatic lipid levels at ≥9.5 mg/kg/day (Brewster and Birnbaum 1989; Kawashima et al.
1995), and hepatocellular hypertrophy at ≥20 mg/kg/day (Harris et al. 1989).
To derive MRLs for perfluoroalkyls, ATSDR used the approach of predicting TWA serum perfluoroalkyl
levels in laboratory animals or measured serum perfluoroalkyl levels and calculating HEDs for these
serum concentrations. For PFOA and PFOS, the Wambaugh et al. (2013) pharmacokinetic model was
utilized for predicting the TWA serum perfluoroalkyl concentrations. However, strain-, sex-, and
compound-specific model parameters are not available for other perfluoroalkyls. The Harris and
Birnbaum (1989) study, which identified the lowest adverse effect level, did not measure maternal serum
PFDeA levels. Thus, HEDs could not be calculated using animal serum PFDeA levels. Other approaches
such as “read across” or equivalency factors were considered; however, there are limited data available
that would allow for comparison of the toxicity and toxicokinetic properties of different perfluoroalkyl
compounds. Peters and Gonzalez (2011) noted that the toxic equivalency factor approach would not be
suitable for perfluoroalkyls because the current data suggest that the toxicity of these compounds appear
to be mediated by multiple receptors, including PPARα, CAR, and PXR, and that there may be species
differences in the response mediated by different receptors. Additionally, available data suggest that there
are qualitative differences in the toxicities of various perfluoroalkyls.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-66
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorodecanoic acid (PFDeA)
335-76-2
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFDeA.
Rationale for Not Deriving an MRL: No intermediate-duration oral studies in laboratory animals were
identified for PFDeA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-67
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorodecanoic acid (PFDeA)
335-76-2
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFDeA.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFDeA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-68
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroundecanoic acid (PFUA)
2058-94-8
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFUA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFUA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-69
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroundecanoic acid (PFUA)
2058-94-8
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFUA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFUA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-70
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroundecanoic acid (PFUA)
2058-94-8
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFUA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFUA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-71
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroundecanoic acid (PFUA)
2058-94-8
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFUA.
Rationale for Not Deriving an MRL: No acute-duration oral studies in laboratory animals were
identified for PFUA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-72
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroundecanoic acid (PFUA)
2058-94-8
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFUA.
Rationale for Not Deriving an MRL: Intermediate-duration oral database was considered inadequate for
derivation of an MRL for PFUA because the only available study did not measure serum PFUA levels,
which are needed to calculated HEDs (see MRL approach in Appendix A introduction).
One study was identified that examined the oral toxicity of PFUA in laboratory animals. In this study,
decreases in body weight, hematological alterations, increases in liver weight, and centrilobular
hypertrophy were observed in rat dams administered 1.0 mg/kg/day for 41–46 days (Takahashi et al.
2014). The study also found decreases in pup body weight on PNDs 0 and 4 at 1.0 mg/kg/day.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-73
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroundecanoic acid (PFUA)
2058-94-8
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFUA.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFUA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-74
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroheptanoic acid (PFHpA)
375-85-9
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFHpA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFHpA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-75
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroheptanoic acid (PFHpA)
375-85-9
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFHpA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFHpA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-76
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroheptanoic acid (PFHpA)
375-85-9
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFHpA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFHpA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-77
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroheptanoic acid (PFHpA)
375-85-9
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFHpA.
Rationale for Not Deriving an MRL: No oral studies in laboratory animals were identified for PFHpA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-78
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroheptanoic acid (PFHpA)
375-85-9
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFHpA.
Rationale for Not Deriving an MRL: No oral studies in laboratory animals were identified for PFHpA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-79
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluoroheptanoic acid (PFHpA)
375-85-9
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFHpA.
Rationale for Not Deriving an MRL: No oral studies in laboratory animals were identified for PFHpA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-80
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutane sulfonic acid (PFBuS)
375-73-5
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFBuS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFBuS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-81
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutane sulfonic acid (PFBuS)
375-73-5
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFBuS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFBuS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-82
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutane sulfonic acid (PFBuS)
375-73-5
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFBuS.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFBuS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-83
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutane sulfonic acid (PFBuS)
375-73-5
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFBuS.
Rationale for Not Deriving an MRL: No acute-duration oral studies in laboratory animals were
identified for PFBuS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-84
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutane sulfonic acid (PFBuS)
375-73-5
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFBuS.
Rationale for Not Deriving an MRL: Laboratory animal studies for other perfluoroalkyl compounds
have identified immunotoxicity and developmental toxicity as sensitive endpoints; these potential targets
have not been investigated for PFBuS. Thus, the database was considered inadequate for identifying a
critical endpoint and evaluating dose-response relationships.
Limited data available on the toxicity of PFBuS in laboratory animals have identified the liver, kidneys,
stomach, and hematological systems as targets of toxicity. Decreases in hemoglobin and hematocrit
levels were observed in male rats administered 200 mg/kg/day PFBuS for 90 days (Lieder et al. 2009a);
decreases in erythrocyte levels were observed at 600 mg/kg/day. Administration of 600 mg/kg/day for
90 days also resulted in tubular and ductal papillary epithelial hyperplasia in the kidneys and necrosis and
hyperplasia/hyperkeratosis in the forestomach (Lieder et al. 2009a). Effects in the liver consisted of
decreases in plasma triglyceride levels in mice exposed to 30 mg/kg/day for 4–6 weeks (Bijland et al.
2011), increases in absolute and relative liver weight in male rats administered 300 mg/kg/day for at least
70 days (Lieder et al. 2009b) or 900 mg/kg/day for 28 days (3M 2001), and hepatocellular hypertrophy in
rats administered 1,000 mg/kg/day in a 2-generation study (Lieder et al. 2009b). In general, no
biologically relevant alterations in performance on FOB tests or motor activity tests were observed in rats
administered 900 mg/kg/day PFBuS for 28 days (3M 2001) or 600 mg/kg/day for 90 days (Lieder et al.
2009a).
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-85
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutane sulfonic acid (PFBuS)
375-73-5
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFBuS.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFBuS.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-86
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutyric acid (PFBA)
375-22-4
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFBA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFBA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-87
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutyric acid (PFBA)
375-22-4
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFBA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFBA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-88
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutyric acid (PFBA)
375-22-4
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFBA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFBA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-89
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutyric acid (PFBA)
375-22-4
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFBA.
Rationale for Not Deriving an MRL: Laboratory animal studies for other perfluoroalkyl compounds
have identified immunotoxicity and developmental toxicity as sensitive endpoints following acuteduration oral exposure; these potential targets have not been investigated for PFBA. Thus, the database
was considered inadequate for identifying a critical endpoint and evaluating dose-response relationships.
Three studies have examined the acute toxicity of PFBA in laboratory animals for a limited number of
potential endpoints. Ikeda et al. (1985) reported that administration of approximately 20 mg/kg/day
PFBA in the diet to male rats for 2 weeks did not significantly affect relative liver weight, but increased
catalase activity in liver homogenates by 42% and induced peroxisome proliferation, as assessed by
electron microscopy. In a similar study, dietary administration of approximately 78 mg/kg/day PFBA to
male mice for 10 days induced a 63% increase in absolute liver weight (Permadi et al. 1992). The
increase in liver weight was accompanied by changes in enzymes involved in drug metabolism and/or in
deactivation of reactive oxygen species; however, PFBA did not have a significant effect on parameters of
peroxisomal fatty acid β-oxidation (Permadi et al. 1993). In a more comprehensive study, no significant
effect on a wide range of endpoints including body and organ weights, hematology and clinical chemistry,
and histopathology were observed in rats administered 184 mg/kg/day for 5 days (3M 2007a).
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-90
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutyric acid (PFBA)
375-22-4
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFBA.
Rationale for Not Deriving an MRL: The available intermediate-duration database was not considered
adequate for derivation of an MRL. Although the available studies have examined potentially sensitive
endpoints and developmental toxicity and both studies measured serum PFBA levels, the database is
missing a reliable estimate of elimination half-life in humans. Chang et al. (2008b) reported serum halflives in small groups of subjects (<10 subjects); only 2 of the subjects were females. Because
developmental toxicity is one on the more sensitive endpoints, data from females is needed in order to
estimate the HED.
The intermediate-duration oral database for PFBA consists of a developmental study in mice (Das et al.
2008) and 28- and 90-day gavage studies in rats (Butenhoff et al. 2012a; van Otterdijk 2007a, 2007b). In
the developmental study, PFBA administered to pregnant mice on GDs 1–17 did not affect newborn
weight gain or viability (Das et al. 2008). The most sensitive response was a delay in eye opening in the
pups at maternal doses of PFBA of 35 mg/kg/day. In the 28- and 90-day studies, hyperplasia/hypertrophy
of the follicular epithelium of the thyroid and hepatocellular hypertrophy were observed at ≥30 mg/kg/day
(Butenhoff et al. 2012a; van Otterdijk 2007a, 2007b). In addition, the 90-day study reported
hematological alterations in male rats dosed with 30 mg/kg/day PFBA. The NOAEL for these effects was
6 mg/kg/day.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-91
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorobutyric acid (PFBA)
375-22-4
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFBA.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFBA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-92
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorododecanoic acid (PFDoA)
307-55-1
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFDoA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFDoA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-93
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorododecanoic acid (PFDoA)
307-55-1
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFDoA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFDoA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-94
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorododecanoic acid (PFDoA)
307-55-1
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFDoA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFDoA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-95
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorododecanoic acid (PFDoA)
307-55-1
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFDoA.
Rationale for Not Deriving an MRL: The database was considered inadequate for derivation of an MRL.
Two studies have examined the acute-oral toxicity of PFDoA. Shi et al. (2007) reported decreases in
body weight and decreases in serum testosterone and estradiol levels in rats following a 14-day gavage
administration of 5 mg/kg/day (Shi et al. 2007). The study also reported an increase in serum cholesterol
levels at 10 mg/kg/day. In the second study, Zhang et al. (2008) found increases in liver weight and
hepatic triglyceride and cholesterol levels in rats administered via gavage ≥5 mg/kg/day for 14 days; these
liver effects were not considered relevant to humans. Given the limited number of endpoints examined in
Shi et al. (2007) this study, including the lack of histopathological examination, this study was not
considered suitable for the derivation of an MRL.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-96
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorododecanoic acid (PFDoA)
307-55-1
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFDoA.
Rationale for Not Deriving an MRL: One intermediate-duration study examined the oral toxicity of
PFDoA. In this study, decreases in serum estradiol and increases in serum cholesterol were observed in
pubertal females exposed to 3 mg/kg/day PFDoA on PNDs 24–72 (Shi et al. 2009).
Given the limited number of endpoints examined in this study, including the lack of histopathological
examination, the database was not considered adequate for identifying the critical target of toxicity and
thus for derivation of an MRL.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-97
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorododecanoic acid (PFDoA)
307-55-1
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFDoA.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFDoA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-98
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanesulfonamide (PFOSA)
754-91-6
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFOSA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFOSA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-99
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanesulfonamide (PFOSA)
754-91-6
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFOSA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFOSA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-100
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanesulfonamide (PFOSA)
754-91-6
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFOSA.
Rationale for Not Deriving an MRL: No inhalation studies in laboratory animals were identified for
PFOSA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-101
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanesulfonamide (PFOSA)
754-91-6
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFOSA.
Rationale for Not Deriving an MRL: The acute-duration database for PFOSA was not considered
adequate for identifying critical targets of toxicity because the Seacat and Luebker (2000) study only
examined a limited number of potential endpoints and the potential developmental and immunological
effects (sensitive targets for other perfluoroalkyl compounds) were not examined.
One laboratory animal study evaluated the acute oral toxicity of PFOSA. Seacat and Luebker (2000) did
not find alterations in body weight or liver weight in rats administered a single dose of 5 mg/kg/day
PFOSA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-102
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanesulfonamide (PFOSA)
754-91-6
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration oral MRL for
PFOSA.
Rationale for Not Deriving an MRL: No intermediate-duration oral studies in laboratory animals were
identified for PFOSA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-103
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorooctanesulfonamide (PFOSA)
754-91-6
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFOSA.
Rationale for Not Deriving an MRL: No chronic-duration oral studies in laboratory animals were
identified for PFOSA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-104
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexanoic acid (PFHxA)
307-24-4
June 2018
Final, Pre-Public Comment
Inhalation
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration inhalation MRL for
PFHxA.
Rationale for Not Deriving an MRL: No acute-duration inhalation studies in laboratory animals were
identified for PFHxA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-105
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexanoic acid (PFHxA)
307-24-4
June 2018
Final, Pre-Public Comment
Inhalation
Intermediate

MRL Summary: There are insufficient data for derivation of an intermediate-duration inhalation MRL
for PFHxA.
Rationale for Not Deriving an MRL: No intermediate-duration inhalation studies in laboratory animals
were identified for PFHxA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-106
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexanoic acid (PFHxA)
307-24-4
June 2018
Final, Pre-Public Comment
Inhalation
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration inhalation MRL for
PFHxA.
Rationale for Not Deriving an MRL: No chronic-duration inhalation studies in laboratory animals were
identified for PFHxA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-107
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexanoic acid (PFHxA)
307-24-4
June 2018
Final, Pre-Public Comment
Oral
Acute

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFHxA.
Rationale for Not Deriving an MRL: The acute database for PFHxA was not considered adequate for
derivation of an MRL because serum PFHxA levels at the lowest LOAEL were below the detection limit
and an elimination half-life has not been estimated for humans; both of these toxicokinetic parameters are
needed to estimate HEDs.
Two developmental toxicity studies conducted by Iwai and Hoberman (2014) examined the acute toxicity
of PFHxA following gavage administration. Increases in stillborn pups and decreases in pup body weight
were observed at 175 mg/kg/day; no effects were observed at 35 mg/kg/day. In the second study,
decreases in birth weight and delayed eye opening was observed at 350 mg/kg/day; the NOAEL was
100 mg/kg/day. These studies were not considered adequate for derivation of an MRL because the
measured serum PFHxA levels at 35 and 175 mg/kg/day groups were below the limit of detection.
Additionally, the elimination half-life has not been estimated in humans.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-108
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexanoic acid (PFHxA)
307-24-4
June 2018
Final, Pre-Public Comment
Oral
Intermediate

MRL Summary: There are insufficient data for derivation of an acute-duration oral MRL for PFHxA.
Rationale for Not Deriving an MRL: No intermediate-duration oral studies in laboratory animals were
identified for PFHxA.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

A-109
APPENDIX A

MINIMAL RISK LEVEL (MRL) WORKSHEET
Chemical Name:
CAS Numbers:
Date:
Profile Status:
Route:
Duration:

Perfluorohexanoic acid (PFHxA)
307-24-4
June 2018
Final, Pre-Public Comment
Oral
Chronic

MRL Summary: There are insufficient data for derivation of a chronic-duration oral MRL for PFHxA.
Rationale for Not Deriving an MRL: The chronic duration oral database for PFHxA is not considered
adequate for derivation of a chronic MRL because the only study available did not measure serum PFHxA
levels and elimination half-life data are not available for humans. These toxicokinetic data are needed to
derive HEDs.
One study has evaluated the chronic oral toxicity of PFHxA in laboratory animals (Klaunig et al. 2015).
Exposure to female rats to 200 mg/kg/day resulted in hematological alterations (decreases in red blood
cells and hemoglobin levels and increases in reticulocyte counts), renal effects (tubular degeneration,
necrosis, increased urine volume and reduced specific gravity), and liver effects (necrosis); no adverse
alterations were observed at 30 mg/kg/day or at 100 mg/kg/day in males. This study was not considered
suitable for derivation of an MRL because serum PFHxA levels were not measured. Additionally, an
elimination half-life has not been estimated in humans.
Agency Contacts (Chemical Managers): Selene Chou

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-1

APPENDIX B. LITERATURE SEARCH FRAMEWORK FOR
PERFLUOROALKYLS
The objective of the toxicological profile is to evaluate the potential for human exposure and the potential
health hazards associated with inhalation, oral, or dermal/ocular exposure to perfluoroalkyls.
B.1 LITERATURE SEARCH AND SCREEN
A literature search and screen was conducted to identify studies examining health effects, toxicokinetics,
mechanisms of action, susceptible populations, biomarkers, chemical interactions, physical and chemical
properties, production, use, environmental fate, environmental releases, and environmental and biological
monitoring data for perfluoroalkyls. ATSDR primarily focused on peer-reviewed articles without
publication date or language restrictions. Non-peer-reviewed studies that were considered relevant to the
assessment of the health effects of perfluoroalkyls have undergone peer review by at least three ATSDRselected experts who have been screened for conflict of interest. The inclusion criteria used to identify
relevant studies examining the health effects of perfluoroalkyls are presented in Table B-1.

Table B-1. Inclusion Criteria for the Literature Search and Screen
Health Effects
Species
Human
Laboratory mammals
Route of exposure
Inhalation
Oral
Dermal (or ocular)
Parenteral (these studies will be considered supporting data)
Health outcome
Death
Systemic effects
Body weight effects
Respiratory effects
Cardiovascular effects
Gastrointestinal effects
Hematological effects
Musculoskeletal effects
Hepatic effects
Renal effects
Dermal effects
Ocular effects
Endocrine effects
Immunological effects
Neurological effects
Reproductive effects
Developmental effects

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-2
APPENDIX B

Table B-1. Inclusion Criteria for the Literature Search and Screen
Other noncancer effects
Cancer
Toxicokinetics
Absorption
Distribution
Metabolism
Excretion
PBPK models
Biomarkers
Biomarkers of exposure
Biomarkers of effect
Interactions with other chemicals
Potential for human exposure
Releases to the environment
Air
Water
Soil
Environmental fate
Transport and partitioning
Transformation and degradation
Environmental monitoring
Air
Water
Sediment and soil
Other media
Biomonitoring
General populations
Occupation populations

B.1.1 Literature Search
The current literature search was intended to update the draft toxicological profile for perfluoroalkyls
released for public comment in 2015. The following main databases were searched in March 2008,
September/October 2013, and May 2016:
•
•
•

PubMed
National Library of Medicine’s TOXLINE
Scientific and Technical Information Network’s TOXCENTER

The search strategy used the chemical names, Chemical Abstracts Service (CAS) numbers,
synonyms, and Medical Subject Headings (MeSH) terms for perfluoroalkyls. The query strings used
for the literature search are presented in Table B-2.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-3
APPENDIX B

The search was augmented by searching the Toxic Substances Control Act Test Submissions (TSCATS),
NTP website, and National Institute of Health Research Portfolio Online Reporting Tools Expenditures
and Results (NIH RePORTER) databases using the queries presented in Table B-3. Additional databases
were searched in the creation of various tables and figures, such as the TRI Explorer, the Substance
priority list (SPL) resource page, and other items as needed. Regulations applicable to perfluoroalkyls
were identified by searching international and U.S. agency websites and documents.
Review articles were identified and used for the purpose of providing background information and
identifying additional references. ATSDR also identified reports from the grey literature, which included
unpublished research reports, technical reports from government agencies, conference proceedings and
abstracts, and theses and dissertations.

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
PubMed
05/24/2016

((((("1,1,2,2,3,3,4,4,4-Nonafluoro-1-butanesulfonic acid"[tw] OR "1,1,2,2,3,3,4,4,4-Nonafluorobutane1-sulphonic acid"[tw] OR "1,1,2,2,3,3,4,4,5,5,6,6,6-Tridecafluorohexane-1-sulfonic acid"[tw] OR "1Perfluorobutanesulfonic acid"[tw] OR "2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,9-heptadecafluoro-Nonanoic
acid"[tw] OR "2-(N-Ethyl-perfluorooctane sulfonamido) acetic acid"[tw] OR "2-(N-Methylperfluorooctane sulfonamido) acetic acid"[tw] OR "C11-PFA"[tw] OR "et-pfosa-acoh"[tw] OR
"Glycine, N-ethyl-N-((1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-heptadecafluorooctyl)sulfonyl)-"[tw] OR
"Henicosafluoroundecanoic acid"[tw] OR "heptadecafluoro-1-octane sulfonic acid"[tw] OR
"Heptadecafluoro-1-octanesulfonic acid"[tw] OR "heptadecafluorooctane sulfonic acid"[tw] OR
"Heptadecafluorooctane-1-sulphonic acid"[tw] OR "Heptadecafluorooctanesulphonamide"[tw] OR
"Heptafluoro-1-butanoic acid"[tw] OR "Heptafluorobutanoic acid"[tw] OR "Heptafluorobutyric acid"[tw]
OR "me-pfosa-acoh"[tw] OR "N-Ethyl-N-((heptadecafluorooctyl)sulphonyl)glycine"[tw] OR "Ndfda"[tw]
OR "Nonadecafluoro-n-decanoic acid"[tw] OR "Nonadecafluorodecanoic acid"[tw] OR "Nonafluoro-1butanesulfonic acid"[tw] OR "Nonafluorobutanesulfonic acid"[tw] OR "Pentadecafluoro-1-octanoic
acid"[tw] OR "Pentadecafluoro-n-octanoic acid"[tw] OR "Pentadecafluorooctanoic acid"[tw] OR
"Pentyl perfluorobutanoate"[tw] OR "Perfluoro-n-decanoic acid"[tw] OR "Perfluoro-n-heptanoic
acid"[tw] OR "Perfluoro-n-nonanoic acid"[tw] OR "Perfluoro-n-undecanoic acid"[tw] OR
"Perfluorobutane sulfonic acid"[tw] OR "Perfluorobutanesulfonic acid"[tw] OR "Perfluorobutanoic
acid"[tw] OR "Perfluorobutyric acid"[tw] OR "Perfluorocaprylic acid"[tw] OR "Perfluoroctanoic acid"[tw]
OR "Perfluoroctylsulfonamide"[tw] OR "Perfluorodecanoic acid"[tw] OR "Perfluorododecanoic acid
"[tw] OR "Perfluorododecanoic acid"[tw] OR "Perfluoroheptanecarboxylic acid"[tw] OR
"Perfluoroheptanoic acid"[tw] OR "Perfluorohexane sulfonic acid"[tw] OR "Perfluorohexane-1sulphonic acid"[tw] OR "perfluorohexanesulfonate"[tw] OR "perfluorohexanesulfonic acid"[tw] OR
"Perfluorolauric acid"[tw] OR "Perfluorononan-1-oic acid"[tw] OR "Perfluorononanoic acid"[tw] OR
"Perfluorooctane sulfonamide"[tw] OR "Perfluorooctane sulfonate"[tw] OR "Perfluorooctane sulfonic
acid "[tw] OR "perfluorooctane sulphonic acid"[tw] OR "Perfluorooctanesulfonamide"[tw] OR
"Perfluorooctanesulfonate"[tw] OR "Perfluorooctanesulfonic acid amide"[tw] OR
"perfluorooctanesulfonic acid"[tw] OR "Perfluorooctanoic acid"[tw] OR "Perfluorooctylsulfonic
acid"[tw] OR "Perfluoropropanecarboxylic acid"[tw] OR "Perfluoroundecanoic acid"[tw] OR "pfbus"[tw]
OR "PFDA"[tw] OR "pfdea"[tw] OR "pfdoa"[tw] OR "Pfhpa"[tw] OR "PFHS cpd"[tw] OR "pfhxs"[tw] OR
"pfna"[tw] OR "PFOA"[tw] OR "PFOS"[tw] OR "pfsoa"[tw] OR "Pfua"[tw] OR
"Tricosafluorododecanoic acid"[tw] OR "Tridecafluoro-1-heptanoic acid"[tw] OR
"Tridecafluoroheptanoic acid"[tw]) AND (2013/09/01:3000[crdat] OR 2013/09/01:3000[edat])) NOT
medline[sb])))) OR (((("1,1,2,2,3,3,4,4,4-Nonafluoro-1-butanesulfonic acid"[tw] OR "1,1,2,2,3,3,4,4,4Nonafluorobutane-1-sulphonic acid"[tw] OR "1,1,2,2,3,3,4,4,5,5,6,6,6-Tridecafluorohexane-1-sulfonic
acid"[tw] OR "1-Perfluorobutanesulfonic acid"[tw] OR "2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,9heptadecafluoro-Nonanoic acid"[tw] OR "2-(N-Ethyl-perfluorooctane sulfonamido) acetic acid"[tw] OR
"2-(N-Methyl-perfluorooctane sulfonamido) acetic acid"[tw] OR "C11-PFA"[tw] OR "et-pfosa-acoh"[tw]
OR "Glycine, N-ethyl-N-((1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-heptadecafluorooctyl)sulfonyl)-"[tw] OR
"Henicosafluoroundecanoic acid"[tw] OR "heptadecafluoro-1-octane sulfonic acid"[tw] OR
"Heptadecafluoro-1-octanesulfonic acid"[tw] OR "heptadecafluorooctane sulfonic acid"[tw] OR

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-4
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
"Heptadecafluorooctane-1-sulphonic acid"[tw] OR "Heptadecafluorooctanesulphonamide"[tw] OR
"Heptafluoro-1-butanoic acid"[tw] OR "Heptafluorobutanoic acid"[tw] OR "Heptafluorobutyric acid"[tw]
OR "me-pfosa-acoh"[tw] OR "N-Ethyl-N-((heptadecafluorooctyl)sulphonyl)glycine"[tw] OR "Ndfda"[tw]
OR "Nonadecafluoro-n-decanoic acid"[tw] OR "Nonadecafluorodecanoic acid"[tw] OR "Nonafluoro-1butanesulfonic acid"[tw] OR "Nonafluorobutanesulfonic acid"[tw] OR "Pentadecafluoro-1-octanoic
acid"[tw] OR "Pentadecafluoro-n-octanoic acid"[tw] OR "Pentadecafluorooctanoic acid"[tw] OR
"Pentyl perfluorobutanoate"[tw] OR "Perfluoro-n-decanoic acid"[tw] OR "Perfluoro-n-heptanoic
acid"[tw] OR "Perfluoro-n-nonanoic acid"[tw] OR "Perfluoro-n-undecanoic acid"[tw] OR
"Perfluorobutane sulfonic acid"[tw] OR "Perfluorobutanesulfonic acid"[tw] OR "Perfluorobutanoic
acid"[tw] OR "Perfluorobutyric acid"[tw] OR "Perfluorocaprylic acid"[tw] OR "Perfluoroctanoic acid"[tw]
OR "Perfluoroctylsulfonamide"[tw] OR "Perfluorodecanoic acid"[tw] OR "Perfluorododecanoic acid
"[tw] OR "Perfluorododecanoic acid"[tw] OR "Perfluoroheptanecarboxylic acid"[tw] OR
"Perfluoroheptanoic acid"[tw] OR "Perfluorohexane sulfonic acid"[tw] OR "Perfluorohexane-1sulphonic acid"[tw] OR "perfluorohexanesulfonate"[tw] OR "perfluorohexanesulfonic acid"[tw] OR
"Perfluorolauric acid"[tw] OR "Perfluorononan-1-oic acid"[tw] OR "Perfluorononanoic acid"[tw] OR
"Perfluorooctane sulfonamide"[tw] OR "Perfluorooctane sulfonate"[tw] OR "Perfluorooctane sulfonic
acid "[tw] OR "perfluorooctane sulphonic acid"[tw] OR "Perfluorooctanesulfonamide"[tw] OR
"Perfluorooctanesulfonate"[tw] OR "Perfluorooctanesulfonic acid amide"[tw] OR
"perfluorooctanesulfonic acid"[tw] OR "Perfluorooctanoic acid"[tw] OR "Perfluorooctylsulfonic
acid"[tw] OR "Perfluoropropanecarboxylic acid"[tw] OR "Perfluoroundecanoic acid"[tw] OR "pfbus"[tw]
OR "PFDA"[tw] OR "pfdea"[tw] OR "pfdoa"[tw] OR "Pfhpa"[tw] OR "PFHS cpd"[tw] OR "pfhxs"[tw] OR
"pfna"[tw] OR "PFOA"[tw] OR "PFOS"[tw] OR "pfsoa"[tw] OR "Pfua"[tw] OR
"Tricosafluorododecanoic acid"[tw] OR "Tridecafluoro-1-heptanoic acid"[tw] OR
"Tridecafluoroheptanoic acid"[tw] OR 1763-23-1[rn] OR 2058-94-8[rn] OR 2355-31-9[rn] OR 299150-6[rn] OR 307-55-1[rn] OR 335-67-1[rn] OR 335-76-2[rn] OR 355-46-4[rn] OR 375-22-4[rn] OR
375-73-5[rn] OR 375-85-9[rn] OR 375-95-1[rn] OR 754-91-6[rn]) AND (2013/09/01:3000[mhda] OR
2013/09/01:3000[crdat] OR 2013/09/01:3000[edat])) AND (to[sh] OR po[sh] OR ae[sh] OR pk[sh] OR
ai[sh] OR ci[sh] OR bl[sh] OR cf[sh] OR ur[sh] OR "pharmacology"[sh:noexp] OR "environmental
exposure"[mh] OR "endocrine system"[mh] OR "hormones, hormone substitutes, and hormone
antagonists"[mh] OR "endocrine disruptors"[mh] OR "Computational biology"[mh] OR "Medical
Informatics"[mh] OR Genomics[mh] OR Genome[mh] OR Proteomics[mh] OR Proteome[mh] OR
Metabolomics[mh] OR Metabolome[mh] OR Genes[mh] OR "Gene expression"[mh] OR
Phenotype[mh] OR genetics[mh] OR genotype[mh] OR Transcriptome[mh] OR ("Systems
Biology"[mh] AND ("Environmental Exposure"[mh] OR "Epidemiological Monitoring"[mh] OR
analysis[sh])) OR "Transcription, Genetic "[mh] OR "Reverse transcription"[mh] OR "Transcriptional
activation"[mh] OR "Transcription factors"[mh] OR ("biosynthesis"[sh] AND (RNA[mh] OR DNA[mh]))
OR "RNA, Messenger"[mh] OR "RNA, Transfer"[mh] OR "peptide biosynthesis"[mh] OR "protein
biosynthesis"[mh] OR "Reverse Transcriptase Polymerase Chain Reaction"[mh] OR "Base
Sequence"[mh] OR "Trans-activators"[mh] OR "Gene Expression Profiling"[mh] OR cancer[sb] OR
(me[sh] AND ("humans"[mh] OR "animals"[mh])))
10/03/2013

("Computational biology"[mh] OR "Medical Informatics"[mh] OR Genomics[mh] OR Genome[mh] OR
Proteomics[mh] OR Proteome[mh] OR Metabolomics[mh] OR Metabolome[mh] OR Genes[mh] OR
"Gene expression"[mh] OR Phenotype[mh] OR genetics[mh] OR genotype[mh] OR
Transcriptome[mh] OR ("Systems Biology"[mh] AND ("Environmental Exposure"[mh] OR
"Epidemiological Monitoring"[mh] OR analysis[sh])) OR "Transcription, Genetic "[mh] OR "Reverse
transcription"[mh] OR "Transcriptional activation"[mh] OR "Transcription factors"[mh] OR
("biosynthesis"[sh] AND (RNA[mh] OR DNA[mh])) OR " RNA, Messenger "[mh] OR " RNA,
Transfer"[mh] OR "peptide biosynthesis"[mh] OR "protein biosynthesis"[mh] OR "Reverse
Transcriptase Polymerase Chain Reaction"[mh] OR "Base Sequence"[mh] OR "Trans-activators"[mh]
OR "Gene Expression Profiling"[mh]) AND ((335-67-1[rn] OR 1763-23-1[rn] OR 355-46-4[rn] OR
2991-50-6[rn] OR 2355-31-9[rn] OR 335-76-2[rn] OR 375-73-5[rn] OR 375-85-9[rn] OR 375-95-1[rn]
OR 754-91-6[rn] OR 2058-94-8[rn] OR 307-55-1[rn] OR 375-22-4[rn] OR 80AM718FML[rn]) AND
2007/05/01:2013/10/03[dp])

09/19/2013

(((335-67-1[rn] OR 1763-23-1[rn] OR 355-46-4[rn] OR 2991-50-6[rn] OR 2355-31-9[rn] OR 335-762[rn] OR 375-73-5[rn] OR 375-85-9[rn] OR 375-95-1[rn] OR 754-91-6[rn] OR 2058-94-8[rn] OR 307-

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-5
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
55-1[rn] OR 375-22-4[rn] OR 80AM718FML[rn]) AND 2007/05/01:2013/09/19[dp]) AND
(((Caprylates/metabolism[MeSH Terms] OR Fluorocarbons/metabolism[MeSH Terms] OR
"Alkanesulfonic Acids/metabolism"[MeSH Terms] OR "Sulfonic Acids/metabolism"[MeSH Terms] OR
"Decanoic Acids/metabolism"[MeSH Terms] OR "Heptanoic Acids/metabolism"[MeSH Terms] OR
"Hydrocarbons, Fluorinated/metabolism"[MeSH Terms] OR "Fatty Acids/metabolism"[MeSH Terms]
OR Sulfonamides/metabolism[MeSH Terms]) AND ("humans"[MeSH Terms] OR "animals"[MeSH
Terms])) OR ((Caprylates[MeSH Terms] OR Fluorocarbons[MeSH Terms] OR "Alkanesulfonic
Acids"[MeSH Terms] OR "Sulfonic Acids"[MeSH Terms] OR "Decanoic Acids"[MeSH Terms] OR
"Heptanoic Acids"[MeSH Terms] OR "Hydrocarbons, Fluorinated"[MeSH Terms] OR "Fatty
Acids"[MeSH Terms] OR Sulfonamides[MeSH Terms]) AND (Endocrine System[mh] OR
Hormones[mh] OR Endocrine disruptors[mh])) OR ((Caprylates[MeSH Terms] OR
Fluorocarbons[MeSH Terms] OR "Alkanesulfonic Acids"[MeSH Terms] OR "Sulfonic Acids"[MeSH
Terms] OR "Decanoic Acids"[MeSH Terms] OR "Heptanoic Acids"[MeSH Terms] OR "Hydrocarbons,
Fluorinated"[MeSH Terms] OR "Fatty Acids"[MeSH Terms] OR Sulfonamides[MeSH Terms]) AND
"environmental exposure"[MeSH Terms]) OR ((Caprylates[MeSH Terms] OR Fluorocarbons[MeSH
Terms] OR "Alkanesulfonic Acids"[MeSH Terms] OR "Sulfonic Acids"[MeSH Terms] OR "Decanoic
Acids"[MeSH Terms] OR "Heptanoic Acids"[MeSH Terms] OR "Hydrocarbons, Fluorinated"[MeSH
Terms] OR "Fatty Acids"[MeSH Terms] OR Sulfonamides[MeSH Terms]) AND "chemically
induced"[MeSH Subheading]) OR ((((((((((("caprylates/adverse effects"[MeSH Terms] OR
"caprylates/antagonists and inhibitors"[MeSH Terms] OR "caprylates/blood"[MeSH Terms] OR
"caprylates/cerebrospinal fluid"[MeSH Terms] OR "caprylates/pharmacokinetics"[MeSH Terms] OR
"caprylates/poisoning"[MeSH Terms] OR "caprylates/toxicity"[MeSH Terms] OR
"caprylates/urine"[MeSH Terms]))) OR (("fluorocarbons/adverse effects"[MeSH Terms] OR
"fluorocarbons/antagonists and inhibitors"[MeSH Terms] OR "fluorocarbons/blood"[MeSH Terms] OR
"fluorocarbons/pharmacokinetics"[MeSH Terms] OR "fluorocarbons/poisoning"[MeSH Terms] OR
"fluorocarbons/toxicity"[MeSH Terms] OR "fluorocarbons/urine"[MeSH Terms]))) OR
(("alkanesulfonic acids/adverse effects"[MeSH Terms] OR "alkanesulfonic acids/antagonists and
inhibitors"[MeSH Terms] OR "alkanesulfonic acids/blood"[MeSH Terms] OR "alkanesulfonic
acids/cerebrospinal fluid"[MeSH Terms] OR "alkanesulfonic acids/pharmacokinetics"[MeSH Terms]
OR "alkanesulfonic acids/poisoning"[MeSH Terms] OR "alkanesulfonic acids/toxicity"[MeSH Terms]
OR "alkanesulfonic acids/urine"[MeSH Terms]))) OR (("sulfonic acids/adverse effects"[MeSH Terms]
OR "sulfonic acids/antagonists and inhibitors"[MeSH Terms] OR "sulfonic acids/blood"[MeSH Terms]
OR "sulfonic acids/cerebrospinal fluid"[MeSH Terms] OR "sulfonic acids/pharmacokinetics"[MeSH
Terms] OR "sulfonic acids/poisoning"[MeSH Terms] OR "sulfonic acids/toxicity"[MeSH Terms] OR
"sulfonic acids/urine"[MeSH Terms]))) OR (("decanoic acids/adverse effects"[MeSH Terms] OR
"decanoic acids/antagonists and inhibitors"[MeSH Terms] OR "decanoic acids/blood"[MeSH Terms]
OR "decanoic acids/pharmacokinetics"[MeSH Terms] OR "decanoic acids/poisoning"[MeSH Terms]
OR "decanoic acids/toxicity"[MeSH Terms] OR "decanoic acids/urine"[MeSH Terms]))) OR
(("heptanoic acids/adverse effects"[MeSH Terms] OR "heptanoic acids/antagonists and
inhibitors"[MeSH Terms] OR "heptanoic acids/blood"[MeSH Terms] OR "heptanoic
acids/cerebrospinal fluid"[MeSH Terms] OR "heptanoic acids/pharmacokinetics"[MeSH Terms] OR
"heptanoic acids/poisoning"[MeSH Terms] OR "heptanoic acids/toxicity"[MeSH Terms] OR
"heptanoic acids/urine"[MeSH Terms]))) OR (("hydrocarbons, fluorinated/adverse effects"[MeSH
Terms] OR "hydrocarbons, fluorinated/antagonists and inhibitors"[MeSH Terms] OR "hydrocarbons,
fluorinated/blood"[MeSH Terms] OR "hydrocarbons, fluorinated/cerebrospinal fluid"[MeSH Terms]
OR "hydrocarbons, fluorinated/pharmacokinetics"[MeSH Terms] OR "hydrocarbons,
fluorinated/toxicity"[MeSH Terms] OR "hydrocarbons, fluorinated/urine"[MeSH Terms]))) OR (("fatty
acids/adverse effects"[MeSH Terms] OR "fatty acids/antagonists and inhibitors"[MeSH Terms] OR
"fatty acids/blood"[MeSH Terms] OR "fatty acids/cerebrospinal fluid"[MeSH Terms] OR "fatty
acids/pharmacokinetics"[MeSH Terms] OR "fatty acids/poisoning"[MeSH Terms] OR "fatty
acids/toxicity"[MeSH Terms] OR "fatty acids/urine"[MeSH Terms]))) OR (("sulfonamides/adverse
effects"[MeSH Terms] OR "sulfonamides/antagonists and inhibitors"[MeSH Terms] OR
"sulfonamides/blood"[MeSH Terms] OR "sulfonamides/cerebrospinal fluid"[MeSH Terms] OR
"sulfonamides/pharmacokinetics"[MeSH Terms] OR "sulfonamides/poisoning"[MeSH Terms] OR
"sulfonamides/toxicity"[MeSH Terms] OR "sulfonamides/urine"[MeSH Terms]))))) OR
(("Perfluorooctanoic acid"[tw] OR "Pentadecafluoro-1-octanoic acid"[tw] OR "Pentadecafluoro-n-

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-6
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
octanoic acid"[tw] OR "Pentadecafluorooctanoic acid"[tw] OR "Perfluorocaprylic acid"[tw] OR
"Perfluoroctanoic acid"[tw] OR "Perfluoroheptanecarboxylic acid"[tw] OR "Perfluorooctanoic acid"[tw]
OR "Pentadecafluorooctanoic acid"[tw] OR "Perfluorooctanoic acid"[tw] OR "Perfluorooctane sulfonic
acid "[tw] OR "Heptadecafluoro-1-octanesulfonic acid"[tw] OR "heptadecafluoro-1-octane sulfonic
acid"[tw] OR "Heptadecafluorooctane-1-sulphonic acid"[tw] OR "heptadecafluorooctane sulfonic
acid"[tw] OR "Perfluorooctane sulfonate"[tw] OR "Perfluorooctylsulfonic acid"[tw] OR
"perfluorooctane sulphonic acid"[tw] OR "perfluorooctanesulfonic acid"[tw] OR
"Perfluorooctanesulfonate"[tw] OR "Heptadecafluorooctane-1-sulphonic acid1Perfluorooctanesulfonic acid"[tw] OR "Perfluorohexane sulfonic acid"[tw] OR "pfhxs"[tw] OR
"perfluorohexanesulfonic acid"[tw] OR "perfluorohexanesulfonate"[tw] OR "1,1,2,2,3,3,4,4,5,5,6,6,6Tridecafluorohexane-1-sulfonic acid"[tw] OR "Perfluorohexane-1-sulphonic acid"[tw] OR "PFHS
cpd"[tw] OR "2-(N-Ethyl-perfluorooctane sulfonamido) acetic acid"[tw] OR "et-pfosa-acoh"[tw] OR "NEthyl-N-((heptadecafluorooctyl)sulphonyl)glycine"[tw] OR "Glycine, N-ethyl-N((1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-heptadecafluorooctyl)sulfonyl)-"[tw] OR "2-(N-Methylperfluorooctane sulfonamido) acetic acid"[tw] OR "me-pfosa-acoh"[tw] OR "Perfluorodecanoic
acid"[tw] OR "Nonadecafluoro-n-decanoic acid"[tw] OR "Nonadecafluorodecanoic acid"[tw] OR
"Perfluoro-N-decanoic acid"[tw] OR "Perfluoro-n-decanoic acid"[tw] OR "Perfluorodecanoic acid"[tw]
OR "Nonadecafluorodecanoic acid"[tw] OR "Perfluoro-N-decanoic acid"[tw] OR "Perfluorobutane
sulfonic acid"[tw] OR "Perfluorobutanesulfonic acid"[tw] OR "1,1,2,2,3,3,4,4,4-Nonafluoro-1butanesulfonic acid"[tw] OR "1-Perfluorobutanesulfonic acid"[tw] OR "Nonafluoro-1-butanesulfonic
acid"[tw] OR "Nonafluorobutanesulfonic acid"[tw] OR "Pentyl perfluorobutanoate"[tw] OR
"1,1,2,2,3,3,4,4,4-Nonafluorobutane-1-sulphonic acid"[tw] OR "Nonafluoro-1-butanesulfonic acid"[tw]
OR "Perfluoroheptanoic acid"[tw] OR "Tridecafluoro-1-heptanoic acid"[tw] OR "Perfluoro-n-heptanoic
acid"[tw] OR "Perfluoroheptanoic acid"[tw] OR "Tridecafluoroheptanoic acid"[tw] OR
"Perfluorononanoic acid"[tw] OR "Perfluoro-n-nonanoic acid"[tw] OR
"2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,9-heptadecafluoro-Nonanoic acid"[tw] OR "Perfluorononan-1-oic
acid"[tw] OR "Perfluorooctane sulfonamide"[tw] OR "Perfluorooctanesulfonamide"[tw] OR
"Perfluoroctylsulfonamide"[tw] OR "Perfluorooctanesulfonic acid amide"[tw] OR
"Heptadecafluorooctanesulphonamide"[tw] OR "Perfluoroundecanoic acid"[tw] OR "Perfluoro-nundecanoic acid"[tw] OR "Henicosafluoroundecanoic acid"[tw] OR "Perfluorododecanoic acid "[tw]
OR "Perfluorododecanoic acid"[tw] OR "Tricosafluorododecanoic acid"[tw] OR "Perfluorolauric
acid"[tw] OR "Perfluorobutyric acid"[tw] OR "Heptafluorobutyric acid"[tw] OR "Heptafluoro-1-butanoic
acid"[tw] OR "Heptafluorobutanoic acid"[tw] OR "Heptafluorobutyric acid"[tw] OR "Perfluorobutanoic
acid"[tw] OR "Perfluoropropanecarboxylic acid"[tw]) NOT medline[sb]) OR (("PFOA"[tw] OR
"PFOS"[tw] OR "Pfua"[tw] OR "pfdoa"[tw] OR "C11-PFA"[tw] OR "pfsoa"[tw] OR "pfna"[tw] OR
"Pfhpa"[tw] OR "pfbus"[tw] OR "PFDA"[tw] OR "pfdea"[tw] OR "Ndfda"[tw]) NOT medline[sb])
Toxcenter
05/25/2016

FILE 'TOXCENTER' ENTERED AT 08:37:55 ON 25 MAY 2016
5994 SEA 335-67-1 OR 1763-23-1 OR 355-46-4 OR 2991-50-6 OR 2355-31-9
OR 335-76-2 OR 375-73-5 OR 375-85-9 OR 375-95-1 OR 754-91-6
OR 2058-94-8 OR 307-55-1 OR 375-22-4
L2
5967 SEA L1 NOT TSCATS/FS
L3
5731 SEA L2 NOT PATENT/DT
L4
1847 SEA L3 AND ED>=20130701
ACT TOXQUERY/Q
L5
QUE (CHRONIC OR IMMUNOTOX? OR NEUROTOX? OR TOXICOKIN? OR
BIOMARKER? OR NEUROLOG?)
L6
QUE (PHARMACOKIN? OR SUBCHRONIC OR PBPK OR EPIDEMIOLOGY/ST,CT,
IT)
L7
QUE (ACUTE OR SUBACUTE OR LD50# OR LD(W)50 OR LC50# OR
LC(W)50)
L8
QUE (TOXICITY OR ADVERSE OR POISONING)/ST,CT,IT
L9
QUE (INHAL? OR PULMON? OR NASAL? OR LUNG? OR RESPIR?)
L10
QUE ((OCCUPATION? OR WORKPLACE? OR WORKER?) AND EXPOS?)
L11
QUE (ORAL OR ORALLY OR INGEST? OR GAVAGE? OR DIET OR DIETS OR
L1

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-7
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
DIETARY OR DRINKING(W)WATER?)
QUE (MAXIMUM AND CONCENTRATION? AND (ALLOWABLE OR PERMISSIBLE))
QUE (ABORT? OR ABNORMALIT? OR EMBRYO? OR CLEFT? OR FETUS?)
QUE (FOETUS? OR FETAL? OR FOETAL? OR FERTIL? OR MALFORM? OR
OVUM?)
L15
QUE (OVA OR OVARY OR PLACENTA? OR PREGNAN? OR PRENATAL?)
L16
QUE (PERINATAL? OR POSTNATAL? OR REPRODUC? OR STERIL? OR
TERATOGEN?)
L17
QUE (SPERM OR SPERMAC? OR SPERMAG? OR SPERMATI? OR SPERMAS? OR
SPERMATOB? OR SPERMATOC? OR SPERMATOG?)
L18
QUE (SPERMATOI? OR SPERMATOL? OR SPERMATOR? OR SPERMATOX? OR
SPERMATOZ? OR SPERMATU? OR SPERMI? OR SPERMO?)
L19
QUE (NEONAT? OR NEWBORN? OR DEVELOPMENT OR DEVELOPMENTAL?)
L20
QUE (ENDOCRIN? AND DISRUPT?)
L21
QUE (ZYGOTE? OR CHILD OR CHILDREN OR ADOLESCEN? OR INFANT?)
L22
QUE (WEAN? OR OFFSPRING OR AGE(W)FACTOR?)
L23
QUE (DERMAL? OR DERMIS OR SKIN OR EPIDERM? OR CUTANEOUS?)
L24
QUE (CARCINOG? OR COCARCINOG? OR CANCER? OR PRECANCER? OR
NEOPLAS?)
L25
QUE (TUMOR? OR TUMOUR? OR ONCOGEN? OR LYMPHOMA? OR CARCINOM?)
L26
QUE (GENETOX? OR GENOTOX? OR MUTAGEN? OR GENETIC(W)TOXIC?)
L27
QUE (NEPHROTOX? OR HEPATOTOX?)
L28
QUE (ENDOCRIN? OR ESTROGEN? OR ANDROGEN? OR HORMON?)
L29
QUE (OCCUPATION? OR WORKER? OR WORKPLACE? OR EPIDEM?)
L30
QUE L5 OR L6 OR L7 OR L8 OR L9 OR L10 OR L11 OR L12 OR L13 OR
L14 OR L15 OR L16 OR L17 OR L18 OR L19 OR L20 OR L21 OR L22 OR
L23 OR L24 OR L25 OR L26 OR L27 OR L28 OR L29
L31
QUE (RAT OR RATS OR MOUSE OR MICE OR GUINEA(W)PIG? OR MURIDAE
OR DOG OR DOGS OR RABBIT? OR HAMSTER? OR PIG OR PIGS OR SWINE
OR PORCINE OR MONKEY? OR MACAQUE?)
L32
QUE (MARMOSET? OR FERRET? OR GERBIL? OR RODENT? OR LAGOMORPHA
OR BABOON? OR CANINE OR CAT OR CATS OR FELINE OR MURINE)
L33
QUE L30 OR L31 OR L32
L34
QUE (NONHUMAN MAMMALS)/ORGN
L35
QUE L33 OR L34
L36
QUE (HUMAN OR HUMANS OR HOMINIDAE OR MAMMALS OR MAMMAL? OR
PRIMATES OR PRIMATE?)
L37
QUE L35 OR L36
L38
1318 SEA L4 AND L37
L39
1148 SEA L4 AND L30
L40
356 SEA L38 AND MEDLINE/FS
L41
297 SEA L38 AND BIOSIS/FS
L42
664 SEA L38 AND CAPLUS/FS
L43
1 SEA L38 NOT (MEDLINE/FS OR BIOSIS/FS OR CAPLUS/FS)
L44
931 DUP REM L40 L41 L43 L42 (387 DUPLICATES REMOVED)
L*** DEL 356 S L38 AND MEDLINE/FS
L*** DEL 356 S L38 AND MEDLINE/FS
L45
356 SEA L44
L*** DEL 297 S L38 AND BIOSIS/FS
L*** DEL 297 S L38 AND BIOSIS/FS
L46
190 SEA L44
L*** DEL 664 S L38 AND CAPLUS/FS
L*** DEL 664 S L38 AND CAPLUS/FS
L47
385 SEA L44
L48
575 SEA (L45 OR L46 OR L47) NOT MEDLINE/FS
D SCAN L48
L12
L13
L14

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-8
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
09/19/2013

FILE 'TOXCENTER' ENTERED AT 09:10:51 ON 19 SEP 2013
L1
3993 SEA 335-67-1 OR 1763-23-1 OR 355-46-4 OR 2991-50-6 OR 2355-31-9
OR 335-76-2 OR 375-73-5 OR 375-85-9 OR 375-95-1 OR 754-91-6
OR 2058-94-8 OR 307-55-1 OR 375-22-4
L2
3966 SEA L1 NOT TSCATS/FS
L3
3782 SEA L2 NOT PATENT/DT
L4
2796 SEA L3 AND PY>2006
ACTIVATE TOXBROAD/Q
L5
QUE (CHRONIC OR IMMUNOTOX? OR NEUROTOX? OR TOXICOKIN? OR
BIOMARKER? OR NEUROLOG?)
L6
QUE (PHARMACOKIN? OR SUBCHRONIC OR PBPK OR EPIDEMIOLOGY/ST,CT)
L7
QUE (ACUTE OR SUBACUTE OR LD50 OR LC50)
L8
QUE (TOXICITY OR ADVERSE OR POISONING)/ST,CT
L9
QUE (INHAL? OR PULMON? OR NASAL? OR LUNG? OR RESPIR?)
L10
QUE (VAPOR? OR VAPOUR? OR AEROSOL?)
L11
QUE ((OCCUPATION? OR WORKPLACE? OR WORKER?) AND EXPOS?)
L12
QUE (ORAL OR ORALLY OR INGEST? OR GAVAGE? OR DIET? OR DRINKING(
W)WATER?)
L13
QUE (MAXIMUM AND CONCENTRATION? AND (ALLOWABLE OR PERMISSIBLE))
L14
QUE (ABORT? OR ABNORMALIT? OR EMBRYO? OR CLEFT? OR FETUS?)
L15
QUE (FOETUS? OR FETAL? OR FOETAL? OR FERTIL? OR MALFORM? OR
OVUM?)
L16
QUE (OVA OR OVARY OR PLACENTA? OR PREGNAN? OR PRENATAL?)
L17
QUE (PERINATAL? OR POSTNATAL? OR REPRODUC? OR STERIL? OR
TERATOGEN?)
L18
QUE (SPERM? OR NEONAT? OR NEWBORN? OR DEVELOPMENT OR
DEVELOPMEN
TAL?)
L19
QUE (ENDOCRIN? AND DISRUPT?)
L20
QUE (ZYGOTE? OR CHILD OR CHILDREN OR ADOLESCEN? OR INFANT?)
L21
QUE (WEAN? OR OFFSPRING OR AGE(W)FACTOR?)
L22
QUE (DERMAL? OR DERMIS OR SKIN OR EPIDERM? OR CUTANEOUS?)
L23
QUE (CARCINOG? OR COCARCINOG? OR CANCER? OR PRECANCER? OR
NEOPLAS?)
L24
QUE (TUMOR? OR TUMOUR? OR ONCOGEN? OR LYMPHOMA? OR CARCINOM?)
L25
QUE (GENETOX? OR GENOTOX? OR MUTAGEN?)
L26
QUE GENETIC(W)TOXIC?
L27
QUE L5 OR L6 OR L7 OR L9 OR L10 OR L11 OR L12 OR L13 OR L14 OR
L15 OR L16 OR L17 OR L18 OR L19 OR L20 OR L21
L28
QUE L27 OR L22 OR L23 OR L24 OR L25 OR L26
L29
QUE L28 OR L8
L30
QUE NEPHROTOX? OR HEPATOTOX? OR ENDOCRIN? OR ESTROGEN? OR
ANDROGEN? OR HORMON?
L31
QUE L29 OR L30
L32
QUE RAT OR RATS OR MOUSE OR MICE OR GUINEA PIG OR MURIDAE OR
DOG OR DOGS OR RABBIT? OR HAMSTER? OR PIG OR PIGS OR SWINE OR
PORCINE OR GOAT OR GOATS OR SHEEP OR MONKEY? OR MACAQUE?
L33
QUE MARMOSET? OR FERRET? OR GERBIL? OR HAMSTER? OR RODENT? OR
LAGOMORPHA OR BABOON? OR BOVINE OR CANINE OR CAT OR CATS OR
FELINE OR PIGEON?
L34
QUE OCCUPATION? OR WORKER? OR WORKPLACE? OR EPIDEM?
L35
QUE L31 OR L32 OR L33 OR L34
L36
QUE NONHUMAN MAMMALS/ORGN
L37
QUE L35 OR L36
L38
QUE HUMAN? OR HOMINIDAE OR MAMMAL? OR PRIMATE?
L39
QUE L37 OR L38

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-9
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
L40
2012 SEA L4 AND L39
L41
619 SEA L40 AND MEDLINE/FS
L42
417 SEA L40 AND BIOSIS/FS
L43
975 SEA L40 AND CAPLUS/FS
L44
1 SEA L40 NOT (MEDLINE/FS OR BIOSIS/FS OR CAPLUS/FS)
L45
1308 DUP REM L41 L42 L44 L43 (704 DUPLICATES REMOVED)
L*** DEL 619 S L40 AND MEDLINE/FS
L*** DEL 619 S L40 AND MEDLINE/FS
L46
619 SEA L45
L*** DEL 417 S L40 AND BIOSIS/FS
L*** DEL 417 S L40 AND BIOSIS/FS
L47
217 SEA L45
L*** DEL 975 S L40 AND CAPLUS/FS
L*** DEL 975 S L40 AND CAPLUS/FS
L48
471 SEA L45
L*** DEL
1 S L40 NOT (MEDLINE/FS OR BIOSIS/FS OR CAPLUS/FS)
L*** DEL
1 S L40 NOT (MEDLINE/FS OR BIOSIS/FS OR CAPLUS/FS)
L49
1 SEA L45
L50
689 SEA (L46 OR L47 OR L48 OR L49) NOT MEDLINE/FS
SAVE TEMP L50 PERFLUOROALKYLS/A PFOA/A
L51
217 SEA L50 AND BIOSIS/FS
L52
471 SEA L50 AND CAPLUS/FS
L53
220 SEA L52 AND 4-?/CC
L54
1 SEA L50 NOT (MEDLINE/FS OR BIOSIS/FS OR CAPLUS/FS)
L55
438 SEA L51 OR L53 OR L54
L56
689 SEA L55 OR L52
D SCAN L55
ToxLine
05/24/2016

((("C11-PFA" OR "et-pfosa-acoh" OR "Henicosafluoroundecanoic acid" OR "heptadecafluoro-1octane sulfonic acid" OR "Heptadecafluoro-1-octanesulfonic acid" OR "heptadecafluorooctane
sulfonic acid" OR "Heptadecafluorooctane-1-sulphonic acid" OR "Heptadecafluorooctane-1-sulphonic
acid1-Perfluorooctanesulfonic acid" OR "Heptadecafluorooctanesulphonamide" OR "Heptafluoro-1butanoic acid" OR "Heptafluorobutanoic acid" OR "Heptafluorobutyric acid" OR "me-pfosa-acoh" OR
"N-Ethyl-N-((heptadecafluorooctyl)sulphonyl)glycine" OR "Ndfda" OR "Nonadecafluoro-n-decanoic
acid" OR "Nonadecafluorodecanoic acid" OR "Nonafluoro-1-butanesulfonic acid" OR
"Nonafluorobutanesulfonic acid" OR "Pentadecafluoro-1-octanoic acid" OR "Pentadecafluoro-noctanoic acid" OR "Pentadecafluorooctanoic acid" OR "Pentyl perfluorobutanoate" OR "Perfluoro-ndecanoic acid" OR "Perfluoro-n-heptanoic acid" OR "Perfluoro-n-nonanoic acid" OR "Perfluoro-nundecanoic acid" OR "Perfluorobutane sulfonic acid" OR "Perfluorobutanesulfonic acid" OR
"Perfluorobutanoic acid" OR "Perfluorobutyric acid" OR "Perfluorocaprylic acid" OR "Perfluoroctanoic
acid" OR "Perfluoroctylsulfonamide" OR "Perfluorodecanoic acid" OR "Perfluorododecanoic acid "
OR "Perfluorododecanoic acid" OR "Perfluoroheptanecarboxylic acid" OR "Perfluoroheptanoic acid"
OR "Perfluorohexane sulfonic acid" OR "Perfluorohexane-1-sulphonic acid" OR
"perfluorohexanesulfonate" OR "perfluorohexanesulfonic acid" OR "Perfluorolauric acid" OR
"Perfluorononan-1-oic acid" OR "Perfluorononanoic acid" OR "Perfluorooctane sulfonamide" OR
"Perfluorooctane sulfonate" OR "Perfluorooctane sulfonic acid " OR "perfluorooctane sulphonic acid"
OR "Perfluorooctanesulfonamide" OR "Perfluorooctanesulfonate" OR "Perfluorooctanesulfonic acid
amide" OR "perfluorooctanesulfonic acid" OR "Perfluorooctanoic acid" OR "Perfluorooctylsulfonic
acid" OR "Perfluoropropanecarboxylic acid" OR "Perfluoroundecanoic acid" OR "pfbus" OR "PFDA"
OR "pfdea" OR "pfdoa" OR "Pfhpa" OR "PFHS cpd" OR "pfhxs" OR "pfna" OR "PFOA" OR "PFOS"
OR "pfsoa" OR "Pfua" OR "Tricosafluorododecanoic acid" OR "Tridecafluoro-1-heptanoic acid" OR
"Tridecafluoroheptanoic acid") AND ( ANEUPL [org] OR BIOSIS [org] OR CIS [org] OR DART [org]
OR EMIC [org] OR EPIDEM [org] OR HEEP [org] OR HMTC [org] OR IPA [org] OR RISKLINE [org]
OR MTGABS [org] OR NIOSH [org] OR NTIS [org] OR PESTAB [org] OR PPBIB [org] ) OR ("176323-1" OR "2058-94-8" OR "2355-31-9" OR "2991-50-6" OR "307-55-1" OR "335-67-1" OR "335-76-2"

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-10
APPENDIX B

Table B-2. Database Query Strings Post Public Comment Searches
Database
search date Query string
OR "355-46-4" OR "375-22-4" OR "375-73-5" OR "375-85-9" OR "375-95-1" OR "754-91-6") AND (
ANEUPL [org] OR BIOSIS [org] OR CIS [org] OR DART [org] OR EMIC [org] OR EPIDEM [org] OR
HEEP [org] OR HMTC [org] OR IPA [org] OR RISKLINE [org] OR MTGABS [org] OR NIOSH [org]
OR NTIS [org] OR PESTAB [org] OR PPBIB [org] ))) AND 2013:2016 [yr]
("1,1,2,2,3,3,4,4,4-Nonafluoro-1-butanesulfonic acid" OR "1,1,2,2,3,3,4,4,4-Nonafluorobutane-1sulphonic acid" OR "1,1,2,2,3,3,4,4,5,5,6,6,6-Tridecafluorohexane-1-sulfonic acid" OR "1Perfluorobutanesulfonic acid" OR "2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,9-heptadecafluoro-Nonanoic acid"
OR "2-(N-Ethyl-perfluorooctane sulfonamido) acetic acid" OR "2-(N-Methyl-perfluorooctane
sulfonamido) acetic acid" OR "Glycine, N-ethyl-N-((1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8heptadecafluorooctyl)sulfonyl)-") AND 2013:2016 [yr] AND ( ANEUPL [org] OR BIOSIS [org] OR CIS
[org] OR DART [org] OR EMIC [org] OR EPIDEM [org] OR HEEP [org] OR HMTC [org] OR IPA [org]
OR RISKLINE [org] OR MTGABS [org] OR NIOSH [org] OR NTIS [org] OR PESTAB [org] OR PPBIB
[org] )
09/18/2013

( "perfluorooctanoic acid" OR "pentadecafluoro 1 octanoic acid" OR "pentadecafluoro n octanoic
acid" OR "pentadecafluorooctanoic acid" OR "perfluorocaprylic acid" OR "perfluoroctanoic acid" OR
"perfluoroheptanecarboxylic acid" OR "perfluorooctanoic acid" OR "pentadecafluorooctanoic acid"
OR "perfluorooctanoic acid" OR "perfluorooctane sulfonic acid " OR "heptadecafluoro 1
octanesulfonic acid" OR "heptadecafluoro 1 octane sulfonic acid" OR "heptadecafluorooctane 1
sulphonic acid" OR "heptadecafluorooctane sulfonic acid" OR "perfluorooctane sulfonate" OR
"perfluorooctylsulfonic acid" OR "perfluorooctane sulphonic acid" OR "perfluorooctanesulfonic acid"
OR "perfluorooctanesulfonate" OR "heptadecafluorooctane 1 sulphonic acid1 perfluorooctanesulfonic
acid" OR "perfluorohexane sulfonic acid pfhxs " OR "perfluorohexanesulfonic acid" OR
"perfluorohexanesulfonate" OR "perfluorohexane 1 sulphonic acid" OR "pfhs cpd" OR "2 ( n ethyl
perfluorooctane sulfonamido ) acetic acid" OR "et pfosa acoh" OR "n ethyl n ( ( heptadecafluorooctyl
) sulphonyl ) glycine" OR "2 ( n methyl perfluorooctane sulfonamido ) acetic acid" OR "me pfosa
acoh" OR "perfluorodecanoic acid" OR "nonadecafluoro n decanoic acid" OR
"nonadecafluorodecanoic acid" OR "perfluoro n decanoic acid" OR "perfluoro n decanoic acid" OR
"perfluorodecanoic acid" OR "nonadecafluorodecanoic acid" OR "perfluoro n decanoic acid" OR
"perfluorobutane sulfonic acid" OR "perfluorobutanesulfonic acid" OR OR "pentyl perfluorobutanoate"
OR "nonafluoro 1 butanesulfonic acid" OR "perfluoroheptanoic acid" OR "tridecafluoro 1 heptanoic
acid" OR "perfluoro n heptanoic acid" OR "perfluoroheptanoic acid" OR "tridecafluoroheptanoic acid"
OR "perfluorononanoic acid" OR "perfluoro n nonanoic acid" OR "perfluorononan 1 oic acid" OR
"perfluorooctane sulfonamide" OR "perfluorooctanesulfonamide" OR "perfluoroctylsulfonamide" OR
"perfluorooctanesulfonic acid amide" OR "heptadecafluorooctanesulphonamide" OR
"perfluoroundecanoic acid" OR "perfluoro n undecanoic acid" OR "hennone ) AND 2007:2013 [yr]
AND ( ANEUPL [org] OR BIOSIS [org] OR CIS [org] OR NIH RePORTER [org] OR DART [org] OR
EMIC [org] OR EPIDEM [org] OR FEDRIP [org] OR HEEP [org] OR HMTC [org] OR IPA [org] OR
RISKLINE [org] OR MTGABS [org] OR NIOSH [org] OR NTIS [org] OR PESTAB [org] OR PPBIB
[org] ) NOT PubMed [org] NOT pubdart [org]

Table B-3. Strategies to Augment the Literature Search
Source

Query and number screened when available

TSCATSa

Compounds searched: 335-67-1; 1763-23-1; 355-46-4; 2991-50-6; 2355-31-9; 335-76-2; 375-735; 375-85-9; 375-95-1; 754-91-6; 2058-94-8; 307-55-1; 375-22-4

5/23/2016
9/18/2013
NTP
5/23/2016

"335-67-1" OR "1763-23-1" OR "355-46-4" OR "2991-50-6" OR "2355-31-9" OR "335-76-2" OR
"375-73-5" OR "375-85-9" OR "375-95-1" OR "754-91-6" OR "2058-94-8" OR "307-55-1" OR
"375-22-4" OR "Perfluorooctanoic acid" OR "PFOA" OR "Pentadecafluoro-1-octanoic acid" OR
"Pentadecafluorooctanoic acid" OR "Perfluoroctanoic acid" OR "Perfluorooctane sulfonic acid"
OR "Heptadecafluorooctane-1-sulphonic acid" OR "PFOS" OR "Perfluorooctane sulfonate" OR

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-11
APPENDIX B

Table B-3. Strategies to Augment the Literature Search
Source

Query and number screened when available
"Perfluorooctylsulfonic acid" OR "perfluorooctane sulphonic acid" OR "perfluorooctanesulfonic
acid" OR "Perfluorooctanesulfonate" OR "Heptadecafluorooctane-1-sulphonic acid" OR
"1,1,2,2,3,3,4,4,5,5,6,6,6-Tridecafluorohexane-1-sulfonic acid" OR "Perfluorohexane sulfonic
acid" OR "Perfluorohexane-1-sulphonic acid" OR "perfluorohexanesulfonate" OR
"perfluorohexanesulfonic acid" OR "pfhxs" OR "Ndfda" OR "Nonadecafluorodecanoic acid" OR
"Nonadecafluoro-n-decanoic acid" OR "Perfluorodecanoic acid" OR "Perfluoro-n-decanoic acid"
OR "PFDA" OR "pfdea" OR "1-Perfluorobutanesulfonic acid" OR "Nonafluoro-1-butanesulfonic
acid" OR "Perfluorobutane sulfonic acid" OR "Perfluorobutanesulfonic acid" OR "pfbus" OR
"Perfluorononanoic acid" OR "pfna" OR "Perfluorooctanesulfonamide" OR "Perfluorododecanoic
acid" OR "pfdoa" OR "Heptafluorobutyric acid" OR "Perfluorobutanoic acid" OR "Perfluorobutyric
acid"
Screened: 146 hits
"1,1,2,2,3,3,4,4,4-Nonafluoro-1-butanesulfonic acid" OR "1,1,2,2,3,3,4,4,4-Nonafluorobutane-1sulphonic acid" OR "Nonafluorobutanesulfonic acid" OR "Pentyl perfluorobutanoate" OR
"Pentadecafluoro-n-octanoic acid" OR "Perfluorocaprylic acid" OR "Perfluoroheptanecarboxylic
acid" OR "Heptadecafluoro-1-octanesulfonic acid" OR "heptadecafluoro-1-octane sulfonic acid"
OR "heptadecafluorooctane sulfonic acid" OR "1-Perfluorooctanesulfonic acid" OR "PFHS cpd"
OR "2-(N-Ethyl-perfluorooctane sulfonamido) acetic acid" OR "et-pfosa-acoh" OR "N-Ethyl-N((heptadecafluorooctyl)sulphonyl)glycine" OR "Glycine, N-ethyl-N((1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-heptadecafluorooctyl)sulfonyl)-" OR "2-(N-Methylperfluorooctane sulfonamido) acetic acid" OR "me-pfosa-acoh" OR "Perfluoroheptanoic acid" OR
"Perfluoro-n-heptanoic acid" OR "Pfhpa" OR "Tridecafluoro-1-heptanoic acid" OR
"Tridecafluoroheptanoic acid" OR "2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,9-heptadecafluoro-Nonanoic
acid" OR "Perfluoro-n-nonanoic acid" OR "Perfluorononan-1-oic acid" OR
"Heptadecafluorooctanesulphonamide" OR "Perfluoroctylsulfonamide" OR "Perfluorooctane
sulfonamide" OR "Perfluorooctanesulfonic acid amide" OR "Pfsoa" OR "C11-PFA" OR
"Henicosafluoroundecanoic acid" OR "Perfluoro-n-undecanoic acid" OR "Perfluoroundecanoic
acid" OR "Pfua" OR "Perfluorolauric acid" OR "Tricosafluorododecanoic acid" OR "Heptafluoro1-butanoic acid" OR "Heptafluorobutanoic acid" OR "Perfluoropropanecarboxylic acid"
Screened: 0 hits

NIH RePORTER
2/28/2017

Text Search: "1,1,2,2,3,3,4,4,4-Nonafluoro-1-butanesulfonic acid" OR "1,1,2,2,3,3,4,4,4Nonafluorobutane-1-sulphonic acid" OR "1,1,2,2,3,3,4,4,5,5,6,6,6-Tridecafluorohexane-1sulfonic acid" OR "1-Perfluorobutanesulfonic acid" OR "2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,9heptadecafluoro-Nonanoic acid" OR "2-(N-Ethyl-perfluorooctane sulfonamido) acetic acid" OR
"2-(N-Methyl-perfluorooctane sulfonamido) acetic acid" OR "C11-PFA" OR "et-pfosa-acoh" OR
"Glycine, N-ethyl-N-((1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-heptadecafluorooctyl)sulfonyl)-" OR
"Henicosafluoroundecanoic acid" OR "heptadecafluoro-1-octane sulfonic acid" OR
"Heptadecafluoro-1-octanesulfonic acid" OR "heptadecafluorooctane sulfonic acid" OR
"Heptadecafluorooctane-1-sulphonic acid" OR "Heptadecafluorooctanesulphonamide" OR
"Heptafluoro-1-butanoic acid" OR "Heptafluorobutanoic acid" OR "Heptafluorobutyric acid" OR
"me-pfosa-acoh" OR "N-Ethyl-N-((heptadecafluorooctyl)sulphonyl)glycine" OR "Ndfda" OR
"Nonadecafluoro-n-decanoic acid" OR "Nonadecafluorodecanoic acid" OR "Nonafluoro-1butanesulfonic acid" OR "Nonafluorobutanesulfonic acid" OR "Pentadecafluoro-1-octanoic acid"
OR "Pentadecafluoro-n-octanoic acid" OR "Pentadecafluorooctanoic acid" OR "Pentyl
perfluorobutanoate" OR "Perfluoro-n-decanoic acid" OR "Perfluoro-n-heptanoic acid" OR
"Perfluoro-n-nonanoic acid" OR "Perfluoro-n-undecanoic acid" OR "Perfluorobutane sulfonic
acid" OR "Perfluorobutanesulfonic acid" OR "Perfluorobutanoic acid" OR "Perfluorobutyric acid"
OR "Perfluorocaprylic acid" OR "Perfluoroctanoic acid" OR "Perfluoroctylsulfonamide" OR
"Perfluorodecanoic acid" OR "Perfluorododecanoic acid" OR "Perfluorododecanoic acid" OR
"Perfluoroheptanecarboxylic acid" OR "Perfluoroheptanoic acid" OR "Perfluorohexane sulfonic
acid" OR "Perfluorohexane-1-sulphonic acid" OR "perfluorohexanesulfonate" OR
"perfluorohexanesulfonic acid" OR "Perfluorolauric acid" OR "Perfluorononan-1-oic acid" OR
"Perfluorononanoic acid" OR "Perfluorooctane sulfonamide" OR "Perfluorooctane sulfonate" OR
"Perfluorooctane sulfonic acid" OR "perfluorooctane sulphonic acid" OR
"Perfluorooctanesulfonamide" OR "Perfluorooctanesulfonate" OR "Perfluorooctanesulfonic acid

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-12
APPENDIX B

Table B-3. Strategies to Augment the Literature Search
Source

4/7/2014

Other

Query and number screened when available
amide" OR "perfluorooctanesulfonic acid" OR "Perfluorooctanoic acid" OR
"Perfluorooctylsulfonic acid" OR "Perfluoropropanecarboxylic acid" OR "Perfluoroundecanoic
acid" OR "pfbus" OR "PFDA" OR "pfdea" OR "pfdoa" OR "Pfhpa" OR "PFHS cpd" OR "pfhxs"
OR "pfna" OR "PFOA" OR "PFOS" OR "pfsoa" OR "Pfua" OR "Tricosafluorododecanoic acid"
OR "Tridecafluoro-1-heptanoic acid" OR "Tridecafluoroheptanoic acid" (Advanced), Search in:
Projects Admin IC: All, Fiscal Year: Active Projects, 2017, 2016, 2015, 2014, 2013, 2012
Screened: 80
Compounds searched: 335-67-1; 1763-23-1; 355-46-4; 2991-50-6; 2355-31-9; 335-76-2; 375-735; 375-85-9; 375-95-1; 754-91-6; 2058-94-8; 307-55-1; 375-22-4
Screened: 82 hits
Identified throughout the assessment process

aSeveral

versions of the TSCATS database were searched, as needed, by CASRN including TSCATS1 via Toxline
(no date limit), TSCATS2 via https://yosemite.epa.gov/oppts/epatscat8.nsf/ReportSearch?OpenForm (date restricted
by EPA receipt date), and TSCATS via CDAT (date restricted by ‘Mail Received Date Range’), as well as google for
recent TSCA submissions.

The May 2016 results were:
• Number of records identified from PubMed, TOXLINE, and TOXCENTER (after duplicate
removal): 1,153
• Number of records identified from other strategies: 209
• Total number of records to undergo literature screening: 1,363
B.1.2 Literature Screening
A two-step process was used to screen the literature search to identify relevant studies on perfluoroalkyls:
•
•

Title and abstract screen
Full text screen

Title and Abstract Screen. Within the reference library, titles and abstracts were screened manually for
relevance. Studies that were considered relevant (see Table B-1 for inclusion criteria) were moved to the
second step of the literature screening process. Studies were excluded when the title and abstract clearly
indicated that the study was not relevant to the toxicological profile.
•
•

Number of titles and abstracts screened: 1,363
Number of studies considered relevant and moved to the next step: 457

Full Text Screen. The second step in the literature screening process was a full text review of individual
studies considered relevant in the title and abstract screen step. Each study was reviewed to determine
whether it was relevant for inclusion in the toxicological profile.
•
•
•

Number of studies undergoing full text review: 457
Number of studies cited in the previous draft of the toxicological profile: 639
Total number of studies cited in the profile: 968

A summary of the results of the literature search and screening is presented in Figure B-1.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

B-13
APPENDIX B

Figure B-1. May 2016 Literature Search Results and Screen for Perfluoroalkyls

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

C-1

APPENDIX C. USER'S GUIDE
Chapter 1. Relevance to Public Health
This chapter provides an overview of U.S. exposures, a summary of health effects based on evaluations of
existing toxicologic, epidemiologic, and toxicokinetic information, and an overview of the minimal risk
levels. This is designed to present interpretive, weight-of-evidence discussions for human health
endpoints by addressing the following questions:
1. What effects are known to occur in humans?
2. What effects observed in animals are likely to be of concern to humans?
3. What exposure conditions are likely to be of concern to humans, especially around hazardous
waste sites?
Minimal Risk Levels (MRLs)
Where sufficient toxicologic information is available, ATSDR derives MRLs for inhalation and oral
routes of entry at each duration of exposure (acute, intermediate, and chronic). These MRLs are not
meant to support regulatory action, but to acquaint health professionals with exposure levels at which
adverse health effects are not expected to occur in humans.
MRLs should help physicians and public health officials determine the safety of a community living near
a hazardous substance emission, given the concentration of a contaminant in air or the estimated daily
dose in water. MRLs are based largely on toxicological studies in animals and on reports of human
occupational exposure.
MRL users should be familiar with the toxicologic information on which the number is based.
Section 1.2, Summary of Health Effects, contains basic information known about the substance. Other
sections, such as Section 3.2 Children and Other Populations that are Unusually Susceptible and
Section 3.4 Interactions with Other Substances, provide important supplemental information.
MRL users should also understand the MRL derivation methodology. MRLs are derived using a
modified version of the risk assessment methodology that the Environmental Protection Agency (EPA)
provides (Barnes and Dourson 1988) to determine reference doses (RfDs) for lifetime exposure.
To derive an MRL, ATSDR generally selects the most sensitive endpoint which, in its best judgement,
represents the most sensitive human health effect for a given exposure route and duration. ATSDR
cannot make this judgement or derive an MRL unless information (quantitative or qualitative) is available
for all potential systemic, neurological, and developmental effects. If this information and reliable
quantitative data on the chosen endpoint are available, ATSDR derives an MRL using the most sensitive
species (when information from multiple species is available) with the highest no-observed-adverse-effect
level (NOAEL) that does not exceed any adverse effect levels. When a NOAEL is not available, a
lowest-observed-adverse-effect level (LOAEL) can be used to derive an MRL, and an uncertainty factor
of 10 must be employed. Additional uncertainty factors of 10 must be used both for human variability to
protect sensitive subpopulations (people who are most susceptible to the health effects caused by the
substance) and for interspecies variability (extrapolation from animals to humans). In deriving an MRL,
these individual uncertainty factors are multiplied together. The product is then divided into the
inhalation concentration or oral dosage selected from the study. Uncertainty factors used in developing a

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

C-2
APPENDIX C

substance-specific MRL are provided in the footnotes of the levels of significant exposure (LSE) tables
that are provided in Chapter 2. Detailed discussions of the MRLs are presented in Appendix A.
Chapter 2. Health Effects
Tables and Figures for Levels of Significant Exposure (LSE)
Tables and figures are used to summarize health effects and illustrate graphically levels of exposure
associated with those effects. These levels cover health effects observed at increasing dose
concentrations and durations, differences in response by species and MRLs to humans for noncancer
endpoints. The LSE tables and figures can be used for a quick review of the health effects and to locate
data for a specific exposure scenario. The LSE tables and figures should always be used in conjunction
with the text. All entries in these tables and figures represent studies that provide reliable, quantitative
estimates of NOAELs, LOAELs, or Cancer Effect Levels (CELs).
The legends presented below demonstrate the application of these tables and figures. Representative
examples of LSE tables and figures follow. The numbers in the left column of the legends correspond to
the numbers in the example table and figure.
TABLE LEGEND
See Sample LSE Table (page C-5)
(1)

Route of exposure. One of the first considerations when reviewing the toxicity of a substance
using these tables and figures should be the relevant and appropriate route of exposure.
Typically, when sufficient data exist, three LSE tables and two LSE figures are presented in the
document. The three LSE tables present data on the three principal routes of exposure
(i.e., inhalation, oral, and dermal). LSE figures are limited to the inhalation and oral routes. Not
all substances will have data on each route of exposure and will not, therefore, have all five of the
tables and figures. Profiles with more than one chemical may have more LSE tables and figures.

(2)

Exposure period. Three exposure periods—acute (<15 days), intermediate (15–364 days), and
chronic (≥365 days)—are presented within each relevant route of exposure. In this example, two
oral studies of chronic-duration exposure are reported. For quick reference to health effects
occurring from a known length of exposure, locate the applicable exposure period within the LSE
table and figure.

(3)

Figure key. Each key number in the LSE table links study information to one or more data points
using the same key number in the corresponding LSE figure. In this example, the study
represented by key number 51 identified NOAELs and less serious LOAELs (also see the three
"51R" data points in sample LSE Figure 2-X).

(4)

Species (strain) No./group. The test species (and strain), whether animal or human, are identified
in this column. The column also contains information on the number of subjects and sex per
group. Chapter 1, Relevance to Public Health, covers the relevance of animal data to human
toxicity and Section 3.1, Toxicokinetics, contains any available information on comparative
toxicokinetics. Although NOAELs and LOAELs are species specific, the levels are extrapolated
to equivalent human doses to derive an MRL.

(5)

Exposure parameters/doses. The duration of the study and exposure regimens are provided in
these columns. This permits comparison of NOAELs and LOAELs from different studies. In
this case (key number 51), rats were orally exposed to “Chemical X” via feed for 2 years. For a

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

C-3
APPENDIX C

more complete review of the dosing regimen, refer to the appropriate sections of the text or the
original reference paper (i.e., Aida et al. 1992).
(6)

Parameters monitored. This column lists the parameters used to assess health effects. Parameters
monitored could include serum (blood) chemistry (BC), behavioral (BH), biochemical changes
(BI), body weight (BW), clinical signs (CS), developmental toxicity (DX), enzyme activity (EA),
food intake (FI), fetal toxicity (FX), gross necropsy (GN), hematology (HE), histopathology
(HP), lethality (LE), maternal toxicity (MX), organ function (OF), ophthalmology (OP), organ
weight (OW), teratogenicity (TG), urinalysis (UR), and water intake (WI).

(7)

Endpoint. This column lists the endpoint examined. The major categories of health endpoints
included in LSE tables and figures are death, body weight, respiratory, cardiovascular,
gastrointestinal, hematological, musculoskeletal, hepatic, renal, dermal, ocular, endocrine,
immunological, neurological, reproductive, developmental, other noncancer, and cancer. "Other
noncancer" refers to any effect (e.g., alterations in blood glucose levels) not covered in these
systems. In the example of key number 51, three endpoints (body weight, hematological, and
hepatic) were investigated.

(8)

NOAEL. A NOAEL is the highest exposure level at which no adverse effects were seen in the
organ system studied. The body weight effect reported in key number 51 is a NOAEL at
25.5 mg/kg/day. NOAELs are not reported for cancer and death; with the exception of these two
endpoints, this field is left blank if no NOAEL was identified in the study.

(9)

LOAEL. A LOAEL is the lowest dose used in the study that caused an adverse health effect.
LOAELs have been classified into "Less Serious" and "Serious" effects. These distinctions help
readers identify the levels of exposure at which adverse health effects first appear and the
gradation of effects with increasing dose. A brief description of the specific endpoint used to
quantify the adverse effect accompanies the LOAEL. Key number 51 reports a less serious
LOAEL of 6.1 mg/kg/day for the hepatic system, which was used to derive a chronic exposure,
oral MRL of 0.008 mg/kg/day (see footnote "c"). MRLs are not derived from serious LOAELs.
A cancer effect level (CEL) is the lowest exposure level associated with the onset of
carcinogenesis in experimental or epidemiologic studies. CELs are always considered serious
effects. The LSE tables and figures do not contain NOAELs for cancer, but the text may report
doses not causing measurable cancer increases. If no LOAEL/CEL values were identified in the
study, this field is left blank.

(10)

Reference. The complete reference citation is provided in Chapter 8 of the profile.

(11)

Footnotes. Explanations of abbreviations or reference notes for data in the LSE tables are found
in the footnotes. For example, footnote "c" indicates that the LOAEL of 6.1 mg/kg/day in key
number 51 was used to derive an oral MRL of 0.008 mg/kg/day.

FIGURE LEGEND
See Sample LSE Figure (page C-6)
LSE figures graphically illustrate the data presented in the corresponding LSE tables. Figures help the
reader quickly compare health effects according to exposure concentrations for particular exposure
periods.
(13)

Exposure period. The same exposure periods appear as in the LSE table. In this example, health
effects observed within the chronic exposure period are illustrated.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

C-4
APPENDIX C

(14)

Endpoint. These are the categories of health effects for which reliable quantitative data exist.
The same health effect endpoints appear in the LSE table.

(15)

Levels of exposure. Concentrations or doses for each health effect in the LSE tables are
graphically displayed in the LSE figures. Exposure concentration or dose is measured on the log
scale "y" axis. Inhalation exposure is reported in mg/m3 or ppm and oral exposure is reported in
mg/kg/day.

(16)

LOAEL. In this example, the half-shaded circle that is designated 51R identifies a LOAEL
critical endpoint in the rat upon which a chronic oral exposure MRL is based. The key number
51 corresponds to the entry in the LSE table. The dashed descending arrow indicates the
extrapolation from the exposure level of 6.1 mg/kg/day (see entry 51 in the sample LSE table) to
the MRL of 0.008 mg/kg/day (see footnote "c" in the sample LSE table).

(17)

CEL. Key number 59R is one of studies for which CELs were derived. The diamond symbol
refers to a CEL for the test species (rat). The number 59 corresponds to the entry in the LSE
table.

(18)

Key to LSE figure. The key provides the abbreviations and symbols used in the figure.

***DRAFT FOR PUBLIC COMMENT***

  C-5
APPENDIX 

Table 2-K. Levels of Significant Exposure to [Chemical - Grate-

 

 

 

 


a a Ht
Sp cies serious erious
Figure I(strain) Exposure Doses Parameters NOAEL LOAEL LOAEL

my: No.igroup parameters [mgikgiday] monitored Endpoint (mgikgiday) imgikgiday) Effect

I-cunomc exposune

 

 

51 Rat 2 years M: 0, 8.1, CS, WI, ?g wt 25.5 138.0 Decreased body weight gain in
1 (Wistar) (F) 25.5, 138.0 BW, OW, males (23?25%) and females {31?
40 M, F: 0, 8.0, HE, BC, HP 38%)
40 311 158.4 1331]
Hepatic 6.1? Increases in absolute and relative

weights at 2808.0 mgikgiday after
12 months of exposure; fatty
generation at 28.1 mgl'kgiday in
males and at 231 mgikgiday in
females, and granulomas in
females at 31.? and
188.4 mgikgiday after 12, 18, or
24 months of exposure and in
males at 28.1 mgikgi'day only after
24 months of exposure

Aida et al. 1992

 

 

52 Rat 104 weeks 0, 3.9, 20.8, CS, BW, Fl, Hepatic 38.3
(F344) 35-3 I301 OW, Renal 20.8 38.3 Increased incidence of renal tubular
T8 HP cell hyperplasia

E00991: 36.3

George et al. 2002

58 Flat Lifetime M: 0, 90 BW, Cancer 190 Inc reased incidence of hepatic
(Wistar) (W) 0, 100 neoplastic nodules in females only;
58M, 58F no additional description of the

tumors was provided

et al. 1985

{[00 number corresponds to entries in Figure 2~x.

50mm derive an acute-duration oral minimal risk level of 0.1 mgikgiday based on the of 10 mgixgiday and an uncertainty factor of 100 {10 for
extrapolation from animals to humans and 10 for human variability).
Maggie derive a chronic?duration oral MRL of 0.008 mgixgiday based on the of 0.?8 mgiicgiday and an uncertainty factor of 100 {10 for extrapolation
from animals to humans and 10 for human variability).

FOR PUBLIC 



1am -
0.01 2
11cm ..

APPENDIX 

0-6

Figure 2-K. Levels of Significant Exposure to [Chemical - Oral

-?>Chmnic (33365 clays}

 

 

 

FOR PUBLIC 

Death Body Weight Respiratory Cardin Gastre Hemate

em 56M

559:0 559 55::
56M 59R 0 
5 gnaw: (in: 55M 51R
53M 52HHepatic Cancer

 

59R
Eg?wo?em

55M 5m

511::

 

 

M?Mouse
R?Rat
H-Rabb'rt

eAnimaI NDAEL

oAnimaJ LOAEL, Lees Serious

uAnimaI LOAEL. More Belieus

.Animal Cancer Eilect Level

- Minimal Risk Level for e?ecte ether than cancer

 

 

PERFLUOROALKYLS

D-1

APPENDIX D. QUICK REFERENCE FOR HEALTH CARE PROVIDERS
Toxicological Profiles are a unique compilation of toxicological information on a given hazardous
substance. Each profile reflects a comprehensive and extensive evaluation, summary, and interpretation
of available toxicologic and epidemiologic information on a substance. Health care providers treating
patients potentially exposed to hazardous substances may find the following information helpful for fast
answers to often-asked questions.

Primary Chapters/Sections of Interest
Chapter 1: Relevance to Public Health: The Relevance to Public Health Section provides an overview
of exposure and health effects and evaluates, interprets, and assesses the significance of toxicity
data to human health. A table listing minimal risk levels (MRLs) is also included in this chapter.
Chapter 2: Health Effects: Specific health effects identified in both human and animal studies are
reported by type of health effect (e.g., death, hepatic, renal, immune, reproductive), route of
exposure (e.g., inhalation, oral, dermal), and length of exposure (e.g., acute, intermediate, and
chronic).
NOTE: Not all health effects reported in this section are necessarily observed in the clinical
setting.
Pediatrics:
Section 3.2
Section 3.3

Children and Other Populations that are Unusually Susceptible
Biomarkers of Exposure and Effect

ATSDR Information Center
Phone: 1-800-CDC-INFO (800-232-4636) or 1-888-232-6348 (TTY)
Internet: http://www.atsdr.cdc.gov
The following additional materials are available online:
Case Studies in Environmental Medicine are self-instructional publications designed to increase primary
health care providers’ knowledge of a hazardous substance in the environment and to aid in the
evaluation of potentially exposed patients (see https://www.atsdr.cdc.gov/csem/csem.html).
Managing Hazardous Materials Incidents is a three-volume set of recommendations for on-scene
(prehospital) and hospital medical management of patients exposed during a hazardous materials
incident (see https://www.atsdr.cdc.gov/MHMI/index.asp). Volumes I and II are planning guides
to assist first responders and hospital emergency department personnel in planning for incidents
that involve hazardous materials. Volume III—Medical Management Guidelines for Acute
Chemical Exposures—is a guide for health care professionals treating patients exposed to
hazardous materials.
Fact Sheets (ToxFAQs™) provide answers to frequently asked questions about toxic substances (see
https://www.atsdr.cdc.gov/toxfaqs/Index.asp).

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

D-2
APPENDIX D

Other Agencies and Organizations
The National Center for Environmental Health (NCEH) focuses on preventing or controlling disease,
injury, and disability related to the interactions between people and their environment outside the
workplace. Contact: NCEH, Mailstop F-29, 4770 Buford Highway, NE, Atlanta, GA
30341-3724 • Phone: 770-488-7000 • FAX: 770-488-7015 • Web Page:
https://www.cdc.gov/nceh/.
The National Institute for Occupational Safety and Health (NIOSH) conducts research on occupational
diseases and injuries, responds to requests for assistance by investigating problems of health and
safety in the workplace, recommends standards to the Occupational Safety and Health
Administration (OSHA) and the Mine Safety and Health Administration (MSHA), and trains
professionals in occupational safety and health. Contact: NIOSH, 395 E Street, S.W., Suite 9200,
Patriots Plaza Building, Washington, DC 20201 • Phone: 202-245-0625 or 1-800-CDC-INFO
(800-232-4636) • Web Page: https://www.cdc.gov/niosh/.
The National Institute of Environmental Health Sciences (NIEHS) is the principal federal agency for
biomedical research on the effects of chemical, physical, and biologic environmental agents on
human health and well-being. Contact: NIEHS, PO Box 12233, 104 T.W. Alexander Drive,
Research Triangle Park, NC 27709 • Phone: 919-541-3212 • Web Page:
https://www.niehs.nih.gov/.

Clinical Resources (Publicly Available Information)
The Association of Occupational and Environmental Clinics (AOEC) has developed a network of clinics
in the United States to provide expertise in occupational and environmental issues. Contact:
AOEC, 1010 Vermont Avenue, NW, #513, Washington, DC 20005 • Phone: 202-347-4976
• FAX: 202-347-4950 • e-mail: AOEC@AOEC.ORG • Web Page: http://www.aoec.org/.
The American College of Occupational and Environmental Medicine (ACOEM) is an association of
physicians and other health care providers specializing in the field of occupational and
environmental medicine. Contact: ACOEM, 25 Northwest Point Boulevard, Suite 700, Elk
Grove Village, IL 60007-1030 • Phone: 847-818-1800 • FAX: 847-818-9266 • Web Page:
http://www.acoem.org/.
The American College of Medical Toxicology (ACMT) is a nonprofit association of physicians with
recognized expertise in medical toxicology. Contact: ACMT, 10645 North Tatum Boulevard,
Suite 200-111, Phoenix AZ 85028 • Phone: 844-226-8333 • FAX: 844-226-8333 • Web Page:
http://www.acmt.net.
The Pediatric Environmental Health Specialty Units (PEHSUs) is an interconnected system of specialists
who respond to questions from public health professionals, clinicians, policy makers, and the
public about the impact of environmental factors on the health of children and reproductive-aged
adults. Contact information for regional centers can be found at http://pehsu.net/findhelp.html.
The American Association of Poison Control Centers (AAPCC) provide support on the prevention and
treatment of poison exposures. Contact: AAPCC, 515 King Street, Suite 510, Alexandria VA
22314 • Phone: 701-894-1858 • Poison Help Line: 1-800-222-1222 • Web Page:
http://www.aapcc.org/.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

E-1

APPENDIX E. GLOSSARY
Absorption—The process by which a substance crosses biological membranes and enters systemic
circulation. Absorption can also refer to the taking up of liquids by solids, or of gases by solids or liquids.
Acute Exposure—Exposure to a chemical for a duration of ≤14 days, as specified in the Toxicological
Profiles.
Adsorption—The adhesion in an extremely thin layer of molecules (as of gases, solutes, or liquids) to the
surfaces of solid bodies or liquids with which they are in contact.
Adsorption Coefficient (Koc)—The ratio of the amount of a chemical adsorbed per unit weight of
organic carbon in the soil or sediment to the concentration of the chemical in solution at equilibrium.
Adsorption Ratio (Kd)—The amount of a chemical adsorbed by sediment or soil (i.e., the solid phase)
divided by the amount of chemical in the solution phase, which is in equilibrium with the solid phase, at a
fixed solid/solution ratio. It is generally expressed in micrograms of chemical sorbed per gram of soil or
sediment.
Benchmark Dose (BMD) or Benchmark Concentration (BMC)—is the dose/concentration
corresponding to a specific response level estimate using a statistical dose-response model applied to
either experimental toxicology or epidemiology data. For example, a BMD10 would be the dose
corresponding to a 10% benchmark response (BMR). The BMD is determined by modeling the doseresponse curve in the region of the dose-response relationship where biologically observable data are
feasible. The BMDL or BMCL is the 95% lower confidence limit on the BMD or BMC.
Bioconcentration Factor (BCF)—The quotient of the concentration of a chemical in aquatic organisms
at a specific time or during a discrete time period of exposure divided by the concentration in the
surrounding water at the same time or during the same period.
Biomarkers—Indicators signaling events in biologic systems or samples, typically classified as markers
of exposure, effect, and susceptibility.
Cancer Effect Level (CEL)—The lowest dose of a chemical in a study, or group of studies, that
produces significant increases in the incidence of cancer (or tumors) between the exposed population and
its appropriate control.
Carcinogen—A chemical capable of inducing cancer.
Case-Control Study—A type of epidemiological study that examines the relationship between a
particular outcome (disease or condition) and a variety of potential causative agents (such as toxic
chemicals). In a case-control study, a group of people with a specified and well-defined outcome is
identified and compared to a similar group of people without the outcome.
Case Report—A report that describes a single individual with a particular disease or exposure. These
reports may suggest some potential topics for scientific research, but are not actual research studies.
Case Series—Reports that describe the experience of a small number of individuals with the same
disease or exposure. These reports may suggest potential topics for scientific research, but are not actual
research studies.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

E-2
APPENDIX E

Ceiling Value—A concentration that must not be exceeded.
Chronic Exposure—Exposure to a chemical for ≥365 days, as specified in the Toxicological Profiles.
Clastogen—A substance that causes breaks in chromosomes resulting in addition, deletion, or
rearrangement of parts of the chromosome.
Cohort Study—A type of epidemiological study of a specific group or groups of people who have had a
common insult (e.g., exposure to an agent suspected of causing disease or a common disease) and are
followed forward from exposure to outcome, and who are disease-free at start of follow-up. Often, at
least one exposed group is compared to one unexposed group, while in other cohorts, exposure is a
continuous variable and analyses are directed towards analyzing an exposure-response coefficient.
Cross-sectional Study—A type of epidemiological study of a group or groups of people that examines
the relationship between exposure and outcome to a chemical or to chemicals at a specific point in time.
Data Needs—Substance-specific informational needs that, if met, would reduce the uncertainties of
human health risk assessment.
Developmental Toxicity—The occurrence of adverse effects on the developing organism that may result
from exposure to a chemical prior to conception (either parent), during prenatal development, or
postnatally to the time of sexual maturation. Adverse developmental effects may be detected at any point
in the life span of the organism.
Dose-Response Relationship—The quantitative relationship between the amount of exposure to a
toxicant and the incidence of the response or amount of the response.
Embryotoxicity and Fetotoxicity—Any toxic effect on the conceptus as a result of prenatal exposure to
a chemical; the distinguishing feature between the two terms is the stage of development during which the
effect occurs. Effects include malformations and variations, altered growth, and in utero death.
Epidemiology—The investigation of factors that determine the frequency and distribution of disease or
other health-related conditions within a defined human population during a specified period.
Excretion—The process by which metabolic waste products are removed from the body.
Genotoxicity—A specific adverse effect on the genome of living cells that, upon the duplication of
affected cells, can be expressed as a mutagenic, clastogenic, or carcinogenic event because of specific
alteration of the molecular structure of the genome.
Half-life—A measure of rate for the time required to eliminate one-half of a quantity of a chemical from
the body or environmental media.
Health Advisory—An estimate of acceptable drinking water levels for a chemical substance derived by
EPA and based on health effects information. A health advisory is not a legally enforceable federal
standard, but serves as technical guidance to assist federal, state, and local officials.
Immediately Dangerous to Life or Health (IDLH)—A condition that poses a threat of life or health, or
conditions that pose an immediate threat of severe exposure to contaminants that are likely to have
adverse cumulative or delayed effects on health.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

E-3
APPENDIX E

Immunotoxicity—Adverse effect on the functioning of the immune system that may result from
exposure to chemical substances.
Incidence—The ratio of new cases of individuals in a population who develop a specified condition to
the total number of individuals in that population who could have developed that condition in a specified
time period.
Intermediate Exposure—Exposure to a chemical for a duration of 15–364 days, as specified in the
Toxicological Profiles.
In Vitro—Isolated from the living organism and artificially maintained, as in a test tube.
In Vivo—Occurring within the living organism.
Lethal Concentration(LO) (LCLO)—The lowest concentration of a chemical in air that has been reported
to have caused death in humans or animals.
Lethal Concentration(50) (LC50)—A calculated concentration of a chemical in air to which exposure for
a specific length of time is expected to cause death in 50% of a defined experimental animal population.
Lethal Dose(LO) (LDLo)—The lowest dose of a chemical introduced by a route other than inhalation that
has been reported to have caused death in humans or animals.
Lethal Dose(50) (LD50)—The dose of a chemical that has been calculated to cause death in 50% of a
defined experimental animal population.
Lethal Time(50) (LT50)—A calculated period of time within which a specific concentration of a chemical
is expected to cause death in 50% of a defined experimental animal population.
Lowest-Observed-Adverse-Effect Level (LOAEL)—The lowest exposure level of chemical in a study,
or group of studies, that produces statistically or biologically significant increases in frequency or severity
of adverse effects between the exposed population and its appropriate control.
Lymphoreticular Effects—Represent morphological effects involving lymphatic tissues such as the
lymph nodes, spleen, and thymus.
Malformations—Permanent structural changes that may adversely affect survival, development, or
function.
Metabolism—Process in which chemical substances are biotransformed in the body that could result in
less toxic and/or readily excreted compounds or produce a biologically active intermediate.
Minimal Risk Level (MRL)—An estimate of daily human exposure to a hazardous substance that is
likely to be without an appreciable risk of adverse noncancer health effects over a specified route and
duration of exposure.
Modifying Factor (MF)—A value (greater than zero) that is applied to the derivation of a Minimal Risk
Level (MRL) to reflect additional concerns about the database that are not covered by the uncertainty
factors. The default value for a MF is 1.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

E-4
APPENDIX E

Morbidity—The state of being diseased; the morbidity rate is the incidence or prevalence of a disease in
a specific population.
Mortality—Death; the mortality rate is a measure of the number of deaths in a population during a
specified interval of time.
Mutagen—A substance that causes mutations, which are changes in the DNA sequence of a cell’s DNA.
Mutations can lead to birth defects, miscarriages, or cancer.
Necropsy—The gross examination of the organs and tissues of a dead body to determine the cause of
death or pathological conditions.
Neurotoxicity—The occurrence of adverse effects on the nervous system following exposure to a
hazardous substance.
No-Observed-Adverse-Effect Level (NOAEL)—The dose of a chemical at which there were no
statistically or biologically significant increases in frequency or severity of adverse effects seen between
the exposed population and its appropriate control. Although effects may be produced at this dose, they
are not considered to be adverse.
Octanol-Water Partition Coefficient (Kow)—The equilibrium ratio of the concentrations of a chemical
in n-octanol and water, in dilute solution.
Odds Ratio (OR)—A means of measuring the association between an exposure (such as toxic substances
and a disease or condition) that represents the best estimate of relative risk (risk as a ratio of the incidence
among subjects exposed to a particular risk factor divided by the incidence among subjects who were not
exposed to the risk factor). An odds ratio that is greater than 1 is considered to indicate greater risk of
disease in the exposed group compared to the unexposed group.
Permissible Exposure Limit (PEL)—An Occupational Safety and Health Administration (OSHA)
regulatory limit on the amount or concentration of a substance not to be exceeded in workplace air
averaged over any 8-hour work shift of a 40-hour workweek.
Pesticide—General classification of chemicals specifically developed and produced for use in the control
of agricultural and public health pests (insects or other organisms harmful to cultivated plants or animals).
Pharmacokinetics—The dynamic behavior of a material in the body, used to predict the fate
(disposition) of an exogenous substance in an organism. Utilizing computational techniques, it provides
the means of studying the absorption, distribution, metabolism, and excretion of chemicals by the body.
Pharmacokinetic Model—A set of equations that can be used to describe the time course of a parent
chemical or metabolite in an animal system. There are two types of pharmacokinetic models: data-based
and physiologically-based. A data-based model divides the animal system into a series of compartments,
which, in general, do not represent real, identifiable anatomic regions of the body, whereas the
physiologically-based model compartments represent real anatomic regions of the body.
Physiologically Based Pharmacodynamic (PBPD) Model—A type of physiologically based doseresponse model that quantitatively describes the relationship between target tissue dose and toxic
endpoints. These models advance the importance of physiologically based models in that they clearly
describe the biological effect (response) produced by the system following exposure to an exogenous
substance.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

E-5
APPENDIX E

Physiologically Based Pharmacokinetic (PBPK) Model—A type of physiologically based doseresponse model that is comprised of a series of compartments representing organs or tissue groups with
realistic weights and blood flows. These models require a variety of physiological information, including
tissue volumes, blood flow rates to tissues, cardiac output, alveolar ventilation rates, and possibly
membrane permeabilities. The models also utilize biochemical information, such as blood:air partition
coefficients, and metabolic parameters. PBPK models are also called biologically based tissue dosimetry
models.
Prevalence—The number of cases of a disease or condition in a population at one point in time.
Prospective Study—A type of cohort study in which a group is followed over time and the pertinent
observations are made on events occurring after the start of the study.
Recommended Exposure Limit (REL)—A National Institute for Occupational Safety and Health
(NIOSH) time-weighted average (TWA) concentration for up to a 10-hour workday during a 40-hour
workweek.
Reference Concentration (RfC)—An estimate (with uncertainty spanning perhaps an order of
magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups)
that is likely to be without an appreciable risk of deleterious noncancer health effects during a lifetime.
The inhalation RfC is expressed in units of mg/m3 or ppm.
Reference Dose (RfD)—An estimate (with uncertainty spanning perhaps an order of magnitude) of the
daily oral exposure of the human population to a potential hazard that is likely to be without risk of
deleterious noncancer health effects during a lifetime. The oral RfD is expressed in units of mg/kg/day.
Reportable Quantity (RQ)—The quantity of a hazardous substance that is considered reportable under
the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). RQs are
(1) ≥1 pound or (2) for selected substances, an amount established by regulation either under CERCLA or
under Section 311 of the Clean Water Act. Quantities are measured over a 24-hour period.
Reproductive Toxicity—The occurrence of adverse effects on the reproductive system that may result
from exposure to a hazardous substance. The toxicity may be directed to the reproductive organs and/or
the related endocrine system. The manifestation of such toxicity may be noted as alterations in sexual
behavior, fertility, pregnancy outcomes, or modifications in other functions that are dependent on the
integrity of this system.
Retrospective Study—A type of cohort study based on a group of persons known to have been exposed
at some time in the past. Data are collected from routinely recorded events, up to the time the study is
undertaken. Retrospective studies are limited to causal factors that can be ascertained from existing
records and/or examining survivors of the cohort.
Risk—The possibility or chance that some adverse effect will result from a given exposure to a hazardous
substance.
Risk Factor—An aspect of personal behavior or lifestyle, an environmental exposure, existing health
condition, or an inborn or inherited characteristic that is associated with an increased occurrence of
disease or other health-related event or condition.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

E-6
APPENDIX E

Risk Ratio/Relative Risk—The ratio of the risk among persons with specific risk factors compared to the
risk among persons without risk factors. A risk ratio that is greater than 1 indicates greater risk of disease
in the exposed group compared to the unexposed group.
Short-Term Exposure Limit (STEL)—A STEL is a 15-minute TWA exposure that should not be
exceeded at any time during a workday.
Standardized Mortality Ratio (SMR)—A ratio of the observed number of deaths and the expected
number of deaths in a specific standard population.
Target Organ Toxicity—This term covers a broad range of adverse effects on target organs or
physiological systems (e.g., renal, cardiovascular) extending from those arising through a single limited
exposure to those assumed over a lifetime of exposure to a chemical.
Teratogen—A chemical that causes structural defects that affect the development of an organism.
Threshold Limit Value (TLV)—An American Conference of Governmental Industrial Hygienists
(ACGIH) concentration of a substance to which it is believed that nearly all workers may be repeatedly
exposed, day after day, for a working lifetime without adverse effect. The TLV may be expressed as a
Time-Weighted Average (TLV-TWA), as a Short-Term Exposure Limit (TLV-STEL), or as a ceiling
limit (TLV-C).
Time-Weighted Average (TWA)—An average exposure within a given time period.
Toxicokinetic—The absorption, distribution, metabolism, and elimination of toxic compounds in the
living organism.
Toxics Release Inventory (TRI)—The TRI is an EPA program that tracks toxic chemical releases and
pollution prevention activities reported by industrial and federal facilities.
Uncertainty Factor (UF)—A factor used in operationally deriving the Minimal Risk Level (MRL),
Reference Dose (RfD), or Reference Concentration (RfC) from experimental data. UFs are intended to
account for (1) the variation in sensitivity among the members of the human population, (2) the
uncertainty in extrapolating animal data to the case of human, (3) the uncertainty in extrapolating from
data obtained in a study that is of less than lifetime exposure, and (4) the uncertainty in using lowestobserved-adverse-effect level (LOAEL) data rather than no-observed-adverse-effect level (NOAEL) data.
A default for each individual UF is 10; if complete certainty in data exists, a value of 1 can be used;
however, a reduced UF of 3 may be used on a case-by-case basis (3 being the approximate logarithmic
average of 10 and 1).
Xenobiotic—Any substance that is foreign to the biological system.

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

F-1

APPENDIX F. ACRONYMS, ABBREVIATIONS, AND SYMBOLS
AAPCC
ACGIH
ACOEM
ACMT
ADI
ADME
AEGL
AIC
AIHA
ALT
AOEC
AP
APFO
AST
atm
ATSDR
AWQC
BCF
BMD/C
BMDX
BMDLX
BMDS
BMR
BUN
C
CAA
CAS
CDC
CEL
CERCLA
CFR
Ci
CI
cm
CPSC
CWA
DHHS
DNA
DOD
DOE
DWEL
EAFUS
ECG/EKG
EEG
EPA
ERPG
Et-PFOSAAcOH
F

American Association of Poison Control Centers
American Conference of Governmental Industrial Hygienists
American College of Occupational and Environmental Medicine
American College of Medical Toxicology
acceptable daily intake
absorption, distribution, metabolism, and excretion
Acute Exposure Guideline Level
Akaike’s information criterion
American Industrial Hygiene Association
alanine aminotransferase
Association of Occupational and Environmental Clinics
alkaline phosphatase
ammonium perfluorooctanoate
aspartate aminotransferase
atmosphere
Agency for Toxic Substances and Disease Registry
Ambient Water Quality Criteria
bioconcentration factor
benchmark dose or benchmark concentration
dose that produces a X% change in response rate of an adverse effect
95% lower confidence limit on the BMDX
Benchmark Dose Software
benchmark response
blood urea nitrogen
centigrade
Clean Air Act
Chemical Abstract Services
Centers for Disease Control and Prevention
cancer effect level
Comprehensive Environmental Response, Compensation, and Liability Act
Code of Federal Regulations
curie
confidence interval
centimeter
Consumer Products Safety Commission
Clean Water Act
Department of Health and Human Services
deoxyribonucleic acid
Department of Defense
Department of Energy
drinking water exposure level
Everything Added to Food in the United States
electrocardiogram
electroencephalogram
Environmental Protection Agency
emergency response planning guidelines
2-(N-ethyl-perfluorooctane sulfonamide) acetic acid
Fahrenheit

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

F-2
APPENDIX F

F1
FDA
FIFRA
FR
FSH
g
GC
gd
GGT
GRAS
HEC
HED
HHS
HPLC
HSDB
IARC
IDLH
IRIS
Kd
kg
kkg
Koc
Kow
L
LC
LC50
LCLo
LD50
LDLo
LDH
LH
LOAEL
LSE
LT50
m
mCi
MCL
MCLG
Me-PFOSAAcOH
MF
mg
mL
mm
mmHg
mmol
MRL
MS
MSHA
Mt
NAAQS

first-filial generation
Food and Drug Administration
Federal Insecticide, Fungicide, and Rodenticide Act
Federal Register
follicle stimulating hormone
gram
gas chromatography
gestational day
γ-glutamyl transferase
generally recognized as safe
human equivalent concentration
human equivalent dose
Department of Health and Human Services
high-performance liquid chromatography
Hazardous Substance Data Bank
International Agency for Research on Cancer
immediately dangerous to life and health
Integrated Risk Information System
adsorption ratio
kilogram
kilokilogram; 1 kilokilogram is equivalent to 1,000 kilograms and 1 metric ton
organic carbon partition coefficient
octanol-water partition coefficient
liter
liquid chromatography
lethal concentration, 50% kill
lethal concentration, low
lethal dose, 50% kill
lethal dose, low
lactic dehydrogenase
luteinizing hormone
lowest-observed-adverse-effect level
Level of Significant Exposure
lethal time, 50% kill
meter
millicurie
maximum contaminant level
maximum contaminant level goal
2-(N-methyl-perfluorooctane sulfonamide) acetic acid
modifying factor
milligram
milliliter
millimeter
millimeters of mercury
millimole
Minimal Risk Level
mass spectrometry
Mine Safety and Health Administration
metric ton
National Ambient Air Quality Standard

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

F-3
APPENDIX F

NAS
NCEH
ND
ng
NHANES
NIEHS
NIOSH
NLM
nm
nmol
NOAEL
NPL
NR
NRC
NS
NTP
OR
OSHA
PAC
PAH
PBPD
PBPK
PEHSU
PEL
PEL-C
PFBA
PFBuS
PFDeA
PFDoA
PFHpA
PFHxA
PFHxS
PFNA
PFOA
PFOSA
PFOS
PFUA
pg
PND
POD
ppb
ppbv
ppm
ppt
REL
REL-C
RfC
RfD
RNA
SARA
SCE

National Academy of Science
National Center for Environmental Health
not detected
nanogram
National Health and Nutrition Examination Survey
National Institute of Environmental Health Sciences
National Institute for Occupational Safety and Health
National Library of Medicine
nanometer
nanomole
no-observed-adverse-effect level
National Priorities List
not reported
National Research Council
not specified
National Toxicology Program
odds ratio
Occupational Safety and Health Administration
Protective Action Criteria
polycyclic aromatic hydrocarbon
physiologically based pharmacodynamic
physiologically based pharmacokinetic
Pediatric Environmental Health Specialty Unit
permissible exposure limit
permissible exposure limit-ceiling value
perfluorobutyric acid
Perfluorobutane sulfonic acid
perfluorodecanoic acid
perfluorododecanoic acid
perfluoroheptanoic acid
perfluorohexanoic acid
perfluorohexane sulfonic acid
perfluorononanoic acid
perfluorooctanoic acid
perfluorooctane sulfonamide
perfluorooctane sulfonic acid
perfluoroundecanoic acid
picogram
postnatal day
point of departure
parts per billion
parts per billion by volume
parts per million
parts per trillion
recommended exposure level/limit
recommended exposure level-ceiling value
reference concentration
reference dose
ribonucleic acid
Superfund Amendments and Reauthorization Act
sister chromatid exchange

***DRAFT FOR PUBLIC COMMENT***

 PERFLUOROALKYLS

F-4
APPENDIX F

SD
SE
SGOT
SGPT
SIC
SMR
sRBC
STEL
T1/2
TLV
TLV-C
TRI
TSCA
TWA
UF
U.S.
USDA
USGS
USNRC
VOC
WBC
WHO

standard deviation
standard error
serum glutamic oxaloacetic transaminase (same as aspartate aminotransferase or AST)
serum glutamic pyruvic transaminase (same as alanine aminotransferase or ALT)
standard industrial classification
standardized mortality ratio
sheep red blood cell
short term exposure limit
Half-life
threshold limit value
threshold limit value-ceiling value
Toxics Release Inventory
Toxic Substances Control Act
time-weighted average
uncertainty factor
United States
United States Department of Agriculture
United States Geological Survey
U.S. Nuclear Regulatory Commission
volatile organic compound
white blood cell
World Health Organization

>
≥
=
<
≤
%
α
β
γ
δ
μm
μg
q1 *
–
+
(+)
(–)

greater than
greater than or equal to
equal to
less than
less than or equal to
percent
alpha
beta
gamma
delta
micrometer
microgram
cancer slope factor
negative
positive
weakly positive result
weakly negative result

***DRAFT FOR PUBLIC COMMENT***