Review Committee Report: Inadvertent Shipment
of Live Bacillus anthracis Spores by DoD

Committee for Comprehensive Review of DoD Laboratory Procedures, Processes, and
Protocols Associated with Inactivating Bacillus anthracis Spores

July 13, 2015

 Table of Contents
Executive Summary

Page 2

Background

Page 7

Findings

Page 12

Recommendations

Page 18

Appendix A: Laboratory Visits and Discussions

Page 23

Appendix B: Commonalities and Differences in Laboratory Procedures

Page 35

Appendix C: Review Committee Membership

Page 37

Page   1

 Executive Summary
On May 29, 2015, due to the inadvertent shipment of live anthrax from a Department of Defense
(DoD) laboratory, the Deputy Secretary of Defense (DSD) directed the Under Secretary of
Defense (USD) for Acquisition, Technology & Logistics (AT&L) to conduct a 30-day review of
the Department’s safety practices for generating and handling inactivated Bacillus anthracis
(BA). This independent review was in addition to the concurrent investigation initiated by the
Centers for Disease Control and Prevention (CDC). For more than a decade, DoD and CDC
have been working closely on optimal oversight of and compliance with the select agent
regulations. These regulations impact all U.S. laboratories that work with biological select
agents including those supported by the DoD's Critical Reagents Program (CRP), which
primarily serves the interagency and the biodefense research community.
USD (AT&L) established a review committee to conduct the comprehensive review mandated
by DSD. This document is the Review Committee’s report; membership of the committee is
provided in Appendix C. The committee was asked to address the following critical areas:
•

The root cause for the incomplete inactivation of BA samples at DoD laboratories;

•

Why post inactivation viability testing did not detect the presence of live BA;

•

Existing DoD laboratory biohazard safety protocols and procedures;

•

DoD laboratory adherence to established procedures and protocols; and

•

Identification of systemic problems and what steps should be taken to fix those problems.

The United States Department of Defense Chemical and Biological Defense Program (CBDP)
develops medical and physical countermeasures to protect the warfighter from chemical and
biological threats. BA is an exceptionally resilient organism. This resilience combined with its
infectious nature makes BA ideally suited for potential adversaries’ biological weapons
programs. Therefore it is critical for the Department to have a strong countermeasures program
to protect our warfighters against this dangerous organism.
As part of the above process, the DoD routinely inactivates many select agents such as Bacillus
anthracis (the causative agent of anthrax), Francisella tularensis (the causative agent of
tularemia), and Burkholderia mallei (the causative agent of Glanders disease in horses) through a
number of chemical, thermal, or radiation procedures. There are many federal partners and
private sector laboratories outside the DoD that depend on DoD’s ability to provide them with
inactivated pathogens for national security and public safety missions. The four DoD
laboratories whose mission involves inactivation of BA are: Edgewood Chemical Biological
Center (ECBC) in Maryland, the United States Army Medical Research Institute of Infectious
Diseases (USAMRIID) in Maryland, the Naval Medical Research Center (NMRC) in Maryland,
and Dugway Proving Ground (DPG) in Utah.
Page   2

 What Happened?
The DoD regularly ships inactivated biological materials for research, development, testing, and
evaluation to industry, academia, and other Federal laboratories. On May 22, 2015, a private
company notified the CDC that inactivated BA spores in its possession were live. Live BA has
the potential to pose a threat to human health and it is only distributed to facilities regulated by
the Federal Select Agent Program.
The CDC began an investigation and determined that a BA sample originating from DPG on
April 20, 2015, was not fully inactivated and contained viable BA spores. We now know that,
over the course of the last decade, 86 facilities in the United States and seven other countries
have received low concentrations of live BA spore samples from DPG thought to be completely
inactivated. The low numbers of live spores found in inactivated DoD samples did not pose a
risk to the general public. Nonetheless, the shipment of live BA samples outside of the select
agent program restrictions (at any concentration) is a serious breach of regulations.
What Has DoD Done Since The Incident?
The DoD issued a moratorium on the shipping of inactivated BA from DPG, ECBC, and
USAMRIID. NMRC does not currently inactivate or ship BA.
DSD advised any laboratories that may have received inactivated BA from the DoD that they
should stop working with it until further instructions are received from DoD and CDC. The
CDC promulgated disposition instructions for laboratories who received samples of inactivated
BA spore preparations from DPG. The office of the Assistant Secretary of Defense for Nuclear,
Chemical, and Biological programs subsequently issued a directive to all DoD laboratories with
inactivated BA spores to immediately determine if any inactivated material at their laboratory
contained live BA.
The DoD created a website that summarizes current information pertaining to the BA situation at
http://www.defense.gov/home/features/2015/0615_lab-stats/.
The DoD opened a hotline that is available for telephone calls 24 hours a day, 7 days a week, and
posted the hotline number to the website as a resource for any questions that laboratories or the
public may have regarding this incident.
On June 1, the Office of the Assistant Secretary of Defense for Nuclear, Chemical, and
Biological Defense Programs established a Task Force to coordinate the DoD response.
On June 5, 2015, the CDC issued updated guidance for testing inactivated BA spores for growth.
All DoD laboratories that began testing using in-house protocols restarted testing in conformance
with the updated CDC protocol to verify that inactivated samples contain no live BA spores.
Page   3

 The DoD has validated that there are no additional irradiated BA samples of concern that are
demonstrating growth using currently available DoD and CDC viability testing protocols.

Root Cause
To perform root cause analysis, the committee visited all four DoD laboratories with capabilities
for BA irradiation to interview members of the scientific staff and management. The
observations and recommendations in this report apply to all four DoD laboratories conducting
this work. The committee also focused on the question of why materials originating from the
Dugway Proving Ground (DPG) exclusively exhibited live BA spores in irradiated samples.
For this report, the committee reviewed existing procedures and protocols from all four DoD
laboratories that produce and work with gamma irradiated BA spores and the scientific literature
pertaining to bacterial spore inactivation.
A key finding by the committee is that there is insufficient technical information in the broader
scientific community to guide the development of thoroughly effective protocols for inactivation
of spores and viability testing of BA. This has contributed to the creation of protocols that do
not completely or permanently sterilize BA with gamma irradiation. The absence of this critical
information is a scientific community-wide problem and needs to be addressed with irradiation
standards and viability testing procedures. Pending the publication of standards, the Centers for
Disease Control and Prevention (CDC) has agreed to initiate a review of the Federal Select
Agent Program Authorities to oversee inactivation procedures and consider rule making or other
actions to expand the requirements for assurance of adequate inactivation. 1
The DPG is a production facility and its output of inactivated BA materials far exceeds other
DoD laboratories that inactivate BA. Several potential causes requiring additional scientific
research are suggested. Two additional factors specific to DPG, which may have contributed to
the presence of live BA in inactivated samples, include low sampling volume for viability testing
and a very short time period between the completed irradiation cycle and start of the viability
testing. It is clear that BA spores are particularly difficult to kill 2 and live spores injured by
irradiation may be able to repair their injuries over time. 3,4

1

As described in the Department of Defense and Health and Human Services Memorandum (subject: briefing on
DoD Anthrax Sample Investigation) for the President, June 2, 2015.
2
Spotts Whitney, Ellen A., et. al. “Inactivation of Bacillus anthracis Spores” Emerging Infectious Diseases 9
(2003): 623-627.
3
Chowdhury, M. S. U., et. al. “Influence of Postirradiation Incubation Temperatures on Recovery of RadiationInjured Clostridium botulinum 62A Spores.” Applied and Environmental Microbiology 32 (1976): 172-178.
4
Abshire, Robert L., et. al. “Resistance and Recovery Studies on Ultraviolet-Irradiated Spores of Bacillus pumilus”
Applied and Environmental Microbiology 39 (1980): 695-701

Page   4

 Findings
The root cause for the incomplete inactivation of BA samples at DoD laboratories:
A single root cause for shipping viable BA samples could not be identified. DoD personnel
appear to have followed their own protocols correctly. However, the committee found inherent
deficiencies in protocols for three phases in the production of inactive spores that could lead to
non-sterile products: 1) radiation dosing, 2) viability testing, and 3) aseptic operations
(contamination prevention). These deficiencies and other factors 5 contributed to the
establishment of protocols that do not completely or permanently sterilize these samples.
Why post inactivation viability testing did not detect the presence of live BA:
There is no single root cause to explain why the BA samples were incompletely inactivated, or
why viability testing did not detect live BA spores. Contributing factors that may have resulted
in undetected live BA spores during viability testing in DPG’s samples include deficiencies in
sample sizes and inadequate incubation periods after irradiation.
Existing DoD laboratory biological safety protocols and procedures:
The committee identified existing DoD laboratory safety protocols and procedures in each
location. However, these procedures are not standardized amongst the laboratories.
DoD laboratory adherence to established procedures and protocols:
In most cases, the committee observed that DoD laboratories followed their own established
procedures and protocols.
Identification of systemic problems and what steps should be taken to fix those problems:
The primary systemic issue responsible for failures in the preparation of inactivated BA spores is
the lack of specific validated standards to guide the development of protocols, processes, and
quality assurance measures.
Abridged Recommendations
Quality Assurance: Enhance quality control programs at DoD laboratories working with
hazardous select agents and other pathogens.
Standardize BA Inactivation Protocols Across Laboratories: All DoD laboratories should
follow a common Standard Operating Procedure (SOP) for such practices as irradiation and
viability testing.
Institute More Rigorous Quality Procedures: Establish quality assurance (QA) and quality
control (QC) procedures for inactivation and viability testing of BA spores.

5

The absence of specific scientific community standards for inactivated BA, and continuing scientific uncertainty
regarding the survival, injury, and repair of spores exposed to gamma radiation.

Page   5

 Clarify The Conditions Of The Material Transfer Agreement (MTA): Material transfer
agreements enable DoD laboratories to communicate potential hazards to the customers and
maintain a positive inventory tracking for potential recalls on all select agent inactivated
materials.
Perform Preventive Maintenance: All reusable mechanical equipment employed throughout
the process should be routinely maintained and calibrated, from mechanical pipettes to the
irradiator.
Establish And Manage An Environmental Surface Sampling Program: Some laboratories
lack written procedures to document, investigate, and report contamination found outside
primary containment areas during environmental persistent agent sampling.
Establish Validated Dose Curves: Radiation dose curves should be generated for each BA
strain used for production, at the same concentrations used in production and performed on the
same irradiator as used for inactivation of spore preparations.
Understand The End Users’ Needs: The Chief Science Officer of the lab, Army or Command,
should work with individual customers to understand sample requirements and determine the
appropriate strains and material needed to support the objectives.
Quantitate Spores Before Irradiation: Spore preparations intended for irradiation should be
quantified as precisely as possible to maximize the likelihood of achieving the inactivation levels
predicted by kill curves produced with the same strain in the same irradiator.
Peer Review: Establish BA spore inactivation and viability testing protocols that are based on
relevant scientific data, standards, and studies conducted to fill knowledge gaps. All protocols
and subsequent protocol modifications should be subject to a peer review process, validated and
implemented uniformly across similar operations.
Program Management: Program managers should provide adequate laboratory space,
equipment and time to conduct relevant safety and surety research for select agents and other
pathogens. Program managers should develop a plan to track and document the implementation
and long-term sustainability of the recommended corrective actions identified through this
review panel as well as all internal and external audits.

Page   6

 Background
On May 22, 2015, the Centers for Disease Control and Prevention (CDC) was contacted by a
private company regarding the growth of live Bacillus anthracis (BA) from a sample that was
purported to be inactivated (i.e., killed bacterial spores). The CDC began an investigation
working with the Department of Defense (DoD) laboratories, state officials, and the Federal
Bureau of Investigation (FBI). This investigation revealed that on April 20, 2015, inactivated BA
samples (Ames strain) originating at the Dugway Proving Ground (DPG) was routed through the
Edgewood Chemical Biological Center (ECBC). ECBC sent the inactivated BA material to six
different private sector companies on April 29, 2015. This shipment was sent to support a DoD
effort to develop a new rapid field-based test to identify biological threats in the environment.
One company in Maryland forwarded the BA sample to a subcontractor in Maryland. It was this
subcontractor that notified the CDC Select Agent Program on May 22, 2015 of the live BA
finding.
The CDC and State health departments recommended post-exposure prophylaxis (PEP) for eight
laboratory workers based on aerosol-generating procedures they had performed with the BA
sample. Other individuals who were not in the area during the aerosol-generating procedures
were deemed not at risk. The CDC also recommended that the three receiving laboratories that
worked with the BA sample outside of a biosafety cabinet (BSC) in Biosafety Levels (BSL) -1
and -2 be closed until decontamination procedures were completed.
Over the following days, the DoD identified additional facilities that had received samples
containing a small number of viable BA spores from DPG, instead of fully inactivated samples.
By May 29, 2015, the extent of this problem, and serious concerns about the DoD laboratory
practices performing work with BA, resulted in establishment of a committee for comprehensive
review of DoD laboratory procedures, processes, and protocols associated with inactivating
spore-forming BA. A more detailed time-line of the events and activities associated with this
incident is provided in Figure 1.

Page   7

  

 

 

 

 

 

 

 

 

DPG ships inactivated a ECBC ships samples to
samples to ECBC on 4/20 six private companies on 4/29
I
DPG and Labs ?lIl'lEI?lI
ECBC stop CDC began 
shipments of onsite procedures En
inactivated investigation an Open bench
CDC poti?ed biological at DPG and be Closed until
of live BA select agents ECBC cleaned
May 2015
CD5 initiated expanded Meeting on D5D
dailv calls the_ the Eageral Establishes
 DOD moratorlum roles review
and State ihe l" committee
Three new public Two new :3h'pP'mg Three new 
batches Health Labs batches '"aml?late batches eal 
from DPG from DPG agents to all from DPG amhmmes
. . . . . . labs . . .
veri?ed llve veri?ed llve veri?ed llve
June 2015
noti?ed Review Review
all labs that committee committee
received any meets to visits ECBC
BA from DPG develop a
REVEW to consider strategy
committee it live Two new
meets to batches Review
consider key from DPG committee
?ndings veri?ed live visits 

6 

Figure 1: Events and activities time-line for the Bacillus anthracis (BA) incident.

Two new
labs
iden??ed
with live
samples
CDC found
two new
labs with hot
Rem?? samples
committee
visits NMRC
and
USAMRIID



One new
batch from
DPG veri?ed
live
Review
committee
starts draft
report

Page 8

Why Is This Work Done At DoD?
The U.S. Department of Defense Chemical and Biological Defense Program (CBDP) develop
medical and physical countermeasures to protect the warfighter from chemical and biological
threats. To support the development of detection assays/technologies reagents must be tested
against the target hazardous agents. These tests are used to evaluate the efficacy of the
countermeasures, to verify the accuracy of testing equipment and effectiveness of protective
gears. Inactivated agents are used whenever possible in these activities to minimize hazards to
the workforce without sacrificing reliability. The DoD routinely inactivates biological agents
through a number of chemical, thermal, or radiation procedures. There are many federal partners
and private sector laboratories outside the DoD that depend on DoD’s ability to provide them
with inactivated pathogens for national security and public safety missions. These include the
FBI, Department of Homeland Security, the Environmental Protection Agency, and private
sector companies, as well as many internal DoD components.
Why Is Irradiation The Method Of Choice?
The use of gamma radiation to sterilize is a common practice for medical devices, parenteral
(administration by other than oral route) pharmaceuticals, and raw foods to increase shelf life.
This particular form of radiation is high energy, with deep penetration into a wide variety of
materials. For biodefense applications, preserving cellular components such as proteins and
nucleic acids is necessary for the inactivated material to be used to develop and validate
detection assays, and for quality assurance activities such as proficiency testing. One of the
challenges faced by the biodefense community is that BA detection methods must accurately and
sensitively detect active endospores, yet the detection methods must have safe, non-infectious
positive controls for training and test validation. Although the most effective spore
decontamination/disinfection methods (chemicals and high heat) do render them inactive, they
also degrade/denature the surface antigens which are critical for antibody generation as well as
for binding-sites for antibody-based assays.
During empirical research on the effect of inactivation methods on detection, it was found that all
methods can affect detection assays 6, but that gamma irradiation consistently allowed antibodybased tests to remain effective. This was further reinforced in studies by the CDC 7 and Defense
Research and Development Canada Suffield. 8 Gamma irradiation continues to be the preferred
method to inactivate BA endospores in order to retain immunoassay functionality. However,
excess radiation does results in dose-dependent and irreversible damage to proteins and nucleic
acids. Therefore, the target radiation dose for inactivating spores for research is high enough to
6

Dang, Jessica L., et al. “Bacillus Spore Inactivation Methods Affect Detection Assays.” Applied and
Environmental Microbiology 67(2001): 3665–3670.
7
Dauphin, Leslie A., et al. “Gamma Irradiation Can Be Used To Inactivate Bacillus anthracis Spores without
Compromising the Sensitivity of Diagnostic Assays.” Applied and Environmental Microbiology 74(2008): 4427–
4433.
8
Technical Memorandum DRDC, Suffield TM 2005-236, December 2005.

Page   9

 make the product safe to handle, but low enough to preserve essential antigenic properties and
nucleic acid integrity.
Which DoD Laboratories Do This Work?
Work with live and inactivated BA material is focused in four DoD laboratories: The Army
Dugway Proving Ground (DPG) in Utah, the Army Edgewood Chemical Biological Center
(ECBC) in Maryland, the United States Army Medical Research Institute of Infectious Diseases
(USAMRIID) in Maryland, and the Naval Medical Research Center (NMRC) in Maryland. Over
the last few years, the funding for BA research has declined and the volume of BA work in many
of these laboratories has dropped precipitously. Today, pathogen activities at NMRC are colocated with USAMRIID within the National Interagency Biodefense Campus (NIBC), at Fort
Detrick Maryland, to cross leverage capabilities. NMRC has no current BA activities but
USAMRIID remains active in BA research. ECBC has a low volume of activities associated
with BA, while DPG continues to be the largest producer of BA and other pathogens, largely to
support the Critical Reagents Program (CRP) and its customers.
Methods Used In Reviewing This Incident.
The committee for comprehensive review of DoD laboratory procedures, processes, and
protocols associated with inactivating spore-forming BA was created by order of the Deputy
Secretary of Defense on May 29, 2015 by the Under Secretary of Defense for Acquisition,
Technology and Logistics (AT&L). Within AT&L, the Deputy Assistant Secretary of Defense
for Nuclear Matters was appointed as the committee chair, to provide an impartial review of the
processes used within DoD laboratories.
The committee chair engaged with subject matter experts from interagency partners, academia,
and industry. Based on the nature of this incident and relevant experiential or academic
credentials needed for this review, a team of experts with specialties in gamma irradiation, select
agents, BA, microbiology, biosafety, biosecurity, quality assurance and good laboratory practices
was assembled. The committee initially met on June 4, 2015, and was briefed on the inadvertent
shipment of live BA samples and the committee’s charter. During the initial meeting, the
committee was tasked with developing a series of detailed probing questions based on the
available facts to identify potential gaps and the likely root cause(s) of the incident. Furthermore,
the committee members were asked to implement the charter directed by DSD when constructing
their questions.
The committee chair reviewed the questions proposed by individual team members and selected
a set of standard questions to be used at every site visit. The committee traveled to and met with
the scientific staff and management for ECBC, NMRC & USAMRIID, and DPG on June 8, 9,
and 17, 2015, respectively. After site visits, additional questions were sent to each laboratory for
further clarification, if needed.
Page   10

 The results presented in this report consolidates the information gathered during the site visits,
and presents a review of each lab’s procedures, the national/international standards documents,
and a review of relevant scientific literature.2-4, 6-18 Detailed discussions of site visits are outlined
in Appendix A. 910111213141516

9

Cook, A. M., et. al. “Gamma Irradiation of Bacillus subtilis Spores in the Presence of Sugars.” Journal gen.
Microbiology 34 (1964): 185-198.
10
Hoile, Rebecca, et. al. “Gamma Irradiation as a Biological Decontaminant and its Effect on Common Fingermark
Detection Technique and DNA profiling” Journal of Forensic Sciences 66 (2010): 171-177.
11
Heninger, Sara J., et. al. “Decontamination of Bacillus anthracis spores: Evaluation of Various Disinfectants”
Applied Biosafety 14 (2009): 7-10.
12
Setlow, P. “Spores of Bacillus subtilis: Their Resistance to and Killing by Radiation, Heat and Chemicals”
Journal of Applied Microbiology 101(2006): 514-525.
13
Ghosh, S., et. al. “Superdormant Spores of Bacillus Species Have Elevated Wet-Heat Resistance and Temperature
Requirements for Heat Activation.” Journal of Bacteriology 191(2009): 5584-5591.
14
Nelson, F. E. “Factors Which Influence the Growth of Heat-Treated Bacteria” Journal of Bacteriology 45(1942):
395-403.
15
Roberts. Richardet. E. and Aldous, Elaine “Recovery from Ultraviolet Irradiation in Escherichia coli” Journal of
Bacteriology 57(1949): 363-375.
16
Stringer, Sandra C., et. al. “Contrasting Effects of Heat Treatment and Incubation Temperature on Germination
and Outgrowth of Individual Spores of Nonproteolytic Clostridium botulinum Bacteria” Applied and Environmental
Microbiology 75(2009): 2712–2719.

Page   11

 Findings
The Deputy Secretary of Defense tasked the committee to conduct a comprehensive review of
DoD laboratory procedures, processes, and protocols associated with inactivating BA spores and
to address the following critical areas:
1.
2.
3.
4.
5.

The root cause for the incomplete inactivation of BA samples at DoD laboratories;
Why post inactivation viability testing did not detect the presence of live BA;
Existing DoD laboratory biohazard safety protocols and procedures;
DoD laboratory adherence to established procedures and protocols; and
Identification of systemic problems and what steps should be taken to fix those problems.
Why Only Samples At DPG Contained Live BA?

As of this writing, only materials originating from the DPG have contained live BA spores in
irradiated samples. Other DoD laboratories that inactivate BA have been able to routinely
identify live BA in their inactivated samples; subsequently mitigating issues arising from
incomplete inactivation. Based on statistics alone, we can predict that sterilization by gamma
irradiation will occasionally fail to completely inactivate BA samples. Therefore, it is critical for
the post irradiation viability tests to accurately determine if live spores exist in inactivated
samples. The key issue with the DPG is not that the irradiation procedures failed; it is that the
viability testing did not detect live BA spores in inactivated samples containing live spores.
It is important to emphasize DPG’s distinctive production role in the defense biosecurity arena.
DPG’s primary mission is to provide developmental, operational, and production testing to
support the Nation's chemical and biological defense programs. DPG's Chamber and Field Test
capabilities provide DoD with the ability to conduct testing of individual protective equipment,
contamination avoidance detection systems, and decontamination programs. In order to conduct
this mission, DPG produces and uses live and inactivated biological agents. Moreover, DPG is
the Nation’s primary producer of inactivated select agents and toxins for use in other Chemical
and Biological Defense Program research and development efforts. DPG is the largest producer
of inactivated BA which is used in numerous research, development, testing and evaluation
programs within the U.S. interagency, industry, and academic communities.
The committee studied the irradiation and viability testing procedures used at DPG and noted
that the amount of solution extracted from the inactivated sample, to be tested for viability, was
the lowest amongst DoD laboratories (5% of the sample is used for viability testing).
Additionally, within this production facility, DPG samples experienced a very short time period
between the completed irradiation cycle and start of the viability testing. The confluence of large
production quantities associated with the DPG, low sampling volume of the inactivated material
for viability testing, and a very short time period between the completed irradiation cycle and
start of the viability testing may have exacerbated the likelihood of not properly identifying live
BA spores in inactivated samples. The development and implementation of ineffective
Page   12

 irradiation and viability testing procedures took place over the last decade; this represents an
institutional problem at DPG and does not necessarily reflect on any one individual. Additional
details for potential root causes are provided in the following sections.
1. The root cause for the incomplete inactivation of BA samples at DoD laboratories:
A single root cause for shipping viable BA samples could not be identified. DoD personnel
appear to have followed their own protocols correctly. However, the committee found inherent
deficiencies in protocols for three phases in the production of inactive spores that could lead to
non-sterile products: 1) radiation dosing, 2) viability testing, and 3) aseptic operations
(contamination prevention). These deficiencies and other factors5 contributed to the
establishment of protocols that do not completely or permanently sterilize these samples.
Why spores may have survived the irradiation process:
Laboratory workers rely on BA kill curves that are internally generated or published in
the scientific literature. However, the conditions used to inactivate the BA samples in
DoD laboratories are often not validated by these kill curves. Furthermore, the kill curve
studies performed by various laboratories7,17,18,19,20 are limited in nature due to
inconsistent experimental parameters. There is evidence in the literature that different
strains of BA show varying degrees of radiation resistance. This problem is further
complicated as various studies are done using different spore concentrations, different
sample volumes, different radiation arrangements (e.g., cooled/uncooled samples, single
dose/multiple dose treatment), and different measurement units (e.g., MRad/kGy,
%survived/cfu). 21 To illustrate the variability of this data on a single strain of BA, the
committee has converted several DoD kill curves and open literature information to a
unified set of variables as depicted in Figure 2.
The figure specifically plots the total number of remaining viable spores (to remove
variations in volumes and concentrations within irradiation chamber) as a function of
gamma radiation dose. To illustrate more clearly some of the landmarks in this figure
they have been color coded. The red box denotes the radiation dose region often used by
DoD laboratories. The yellow box denotes the limit of detection of the sampling
procedures. The red-dot at the top of the ordinate (y-axis) denotes the number of spores
17

Bowen, Jane E., et al. “Inactivation of Bacillus anthracis Using a Cobalt Source.” Salisbury Medical Bulletin
(Special Supplement), NO. 87 (1989), 70-72.
18
Watson, S. and Patience, B. “Irradiation of Bacillus Spore Inactivation Methods Affect Detection Assays.”
Salisbury Medical Bulletin (Special Supplement), NO. 87 (1989), 103.
19
RIID-1994, Bacillus anthracis Kill-Curves Established at U.S. Army Medical Research Institute of Infectious
Diseases, 1994.
20
RIID-2003, Bacillus anthracis Kill-Curves Established at U.S. Army Medical Research Institute of Infectious
Diseases, 2003.
21
Mega Rad (MRad) and kilo Grays (kGy) are units of absorbed radiation. Colony forming unit (cfu) is used to
estimate the number of viable microorganism in a given sample volume.

Page   13

 present in the 1667 lot irradiated at DPG and that was found to contain live spores, after
shipping (this is the inactivated BA lot that started the CDC investigation). The bluehashed circle surrounding the red dot outlines the spore count range irradiated in all DoD
laboratories. All other depicted data points, in the gray box, are actual experimental data
collected to establish kill-curves. As evident from the figure, it can be seen that: 1) the
DoD routinely operates outside validated experimental data for kill curves 2) starting
with larger number of spores requires a significantly higher total radiation dose for
complete inactivation (i.e., the majority of spores are killed by a small radiation dose but
remaining survivors require very high doses for inactivation), 3) the signal-to-noise ratio
for low spore counts (for up to 10% sampling) cannot guarantee verification for complete
inactivation of the spores, and 4) the results are non-linear.

Figure 2: Consolidated kill curve data from multiple sources.
Coupled with published observations that injured spores are capable of germinating and
repairing damage, it can be concluded that the end-to-end inactivation process for BA has
not been adequately characterized and validated by the scientific community.
2. Why post inactivation viability testing did not detect the presence of live BA:
There are a variety of factors that contribute to the inability of post inactivation viability testing
to detect live BA spores. According to the DoD laboratories, gamma irradiation with an
absorbed dose of 40 kGy is expected to result in an 8 to 10 log reduction in viable spores in the
suspension. If the concentration of spores is 109 / ml, an 8 log reduction would result in 10
Page   14

 viable spores/ml. One possibility is that sampling 5% of an irradiated batch may randomly result
in no viable spores extracted for viability testing. An alternative explanation is that radiation
damaged spores may exhibit a significantly slower growth rate relative to un-injured spores.
There is information in the literature indicating that injured, but not inactivated spores of
Bacillus species as well as other spore forming bacteria have delayed germination times. When
these injured bacteria are able to germinate, they will repair the injury. The small sample sizes
taken for viability testing and short incubation periods (based on viable spores not injured), may
have resulted in a viability testing protocol that is prone to error.
In a preparation of spores that had not been highly purified, cellular debris and inactivated spores
could serve as a source of nutrients for damaged, but not inactivated spores in the population.
These damaged spores could, given time, begin to germinate, repair damage, and develop into
live cells in the preparation. This hypothesis needs to be carefully tested, but results from the
CDC on one tube of lot AGD0001667 that was held at ambient temperature over a weekend in a
warehouse lends some credence to this hypothesis. This sample, when tested, showed growth
that was deemed “too numerous to count”, whereas other tubes from the same lot that were either
frozen or refrigerated showed counts of 200-300 cfu/ml (130 cfu/ml, when retested at DPG) ,
suggesting that allowing the sample to rest at ambient temperature promotes recovery of the
injured spores. These types of experiments have not been reported in the literature and therefore
represent a significant knowledge gap in DoD laboratories’ understanding of the optimal
conditions for viability testing. It is recommended that a series of experiments be performed to
ensure that all DoD laboratories working with BA use conditions that allow for the capture and
growth of any injured spore.
Knowledge gaps regarding BA spore recovery mechanisms post irradiation along with
insufficient sampling are thought to be major factors contributing to the failure of the viability
testing.
Why cross-contamination of sterile inactivated spores cannot be ruled out:
After irradiation (see Appendix A, DPG discussion), the spores were returned to the same
room in which they were produced and in which positive control samples were handled.
Common equipment was used for handling both viable and nonviable materials. These
practices create openings for contaminating sterile irradiated spores with viable spores. In
addition, because all select agents are handled in the same room, any sterile product is
susceptible to contamination with any other organism in the program. The only
significant contamination control practice was the use of separate biological safety
cabinets (BSC) for work with viable and nonviable agent. However, BSCs do not offer
full containment, if used improperly, and the cross-contamination potential is similar to
the open bench when using common pipettes for sterile and non-sterile products. Ideally,
viability tests should be conducted in facilities separate from production and positive

Page   15

 control areas. If space is inadequate, engineering controls should be adopted to prevent
cross-contamination.
3. Existing DoD laboratory biological safety protocols and procedures:
The committee identified existing DoD laboratory safety protocols and procedures in each
location. However, these procedures are not standardized amongst the laboratories. The
committee observed some deviations from protocols related to the inactivation process existed
between researchers, and these deviations did not appear to undergo a peer review process.
These procedures should be standardized within a reasonable timeframe for similar operations.
4. DoD laboratory adherence to established procedures and protocols:
In most cases, the committee observed that DoD laboratories followed their own established
procedures and protocols. Some procedures lacked technical rigor and documented pedigree in
some laboratories. The consequence of potential procedural deviations is not always apparent to
the researcher.
5. Identification of systemic problems and what steps should be taken to fix those
problems:
The primary systemic issue responsible for failures in the preparation of inactivated BA spores is
the lack of specific validated standards to guide the development of protocols, processes, and
quality assurance measures. Standards for irradiation sterilization (not specific to BA) as well as
additional relevant standards exist, as delineated below. Ensuring DoD laboratories adopt some
of these guidelines will significantly address many of the existing quality assurance (QA) and
quality control (QC) deficits:
•

ISO Guide 34:2009 (E) General requirements for the competence of reference material
producers

•

ISO 17025:2005 General requirements for the competence of testing and calibration
laboratories

•

ANSI/AAMI/ISO 11137-1:2006/(R)2010 Sterilization of health care products—
Radiation—Part 1: Requirements for development, validation, and routine control of a
sterilization process for medical devices

•

ANSI/AAMI/ISO 11137-2:2013 Sterilization of health care products— Radiation—Part
2: Establishing the sterilization dose

•

ANSI/AAMI/ISO 11137-3:2006/(R) 2010 Sterilization of health care products—
Radiation—Part 3: Guidance on dosimetric aspects

•

ANSI/AAMI/ISO 11737-2: 2009/(R)2014 Sterilization of medical devices –
Microbiological methods – Part 2: Tests of viability performed in the definition,
validation and maintenance of a sterilization process
Page   16

 •

United States Pharmacopeia 71 Sterility Tests

All laboratories relied more on historical practices than validating the processes used. This
resulted in development of independent protocols with fundamental differences, as well as
potential weaknesses. BA is an agent with high radiation resistance requiring rigorous validation
through all protocols and processes. Standards cited above, specifically the ANSI/AAMI/ISO
11137 series prescribe how to validate the irradiation process and develop a sterility assurance
level for each product type treated. Although these standards were developed for medical
devices, many elements of them are applicable and should serve as a roadmap for DoD
laboratories.
The committee acknowledges the finite probability of some small number of spores surviving an
irradiation treatment. These numbers may be so low that the demonstration of sterility can only
be proven by sampling the entire lot (which is clearly not an acceptable solution). However,
proper validation of the processes can minimize the inherent risks. Rigorous testing and
performance monitoring on a routine basis, including all equipment used, will ensure best
practices are consistently used. This includes incorporation of preventive maintenance on all
equipment and performing calibrations on a routine basis. The general standard for certification
frequency is yearly, with more frequent certifications required if local conditions dictate a greater
need (extremely high use, aged equipment, etc.).
Another potential systemic source of error is that the irradiation process has too many unknowns.
Issues such as volume, temperature and titer (concentration in term of colony forming unit per
milliliter or cfu/ml) of spores subjected to irradiation vary between runs. Titers ranged from 106
cfu/ml to 1010 cfu/ml, without an understanding of how factors such as concentration will affect
the process. Use of a common radiation dose without demonstration of its effectiveness across
this 10,000-fold range provides opportunity for error.
Viability testing also has systemic issues. The sample size (5-10%), taken for testing post
irradiation, may be too low to allow for detection of low numbers of viable or injured spores. In
addition, the specific culture media used has to enable the growth of stressed spores. Two
standards cited above, ANSI/AAMI/ISO 11737 and the United States Pharmacopeia title 71
Sterility Test, provide specific guidance on conducting viability testing on sterilized medical
devices, parenteral drugs, and other products and materials. In addition, CFR 21 Food and Drug
Administration require the viability test method be validated, and culture media used subjected to
growth promotion testing prior to use. Although these documents are not specific to BA, they
can serve as a starting point for DoD laboratories and can be tailored to meet the specific
requirements.

Page   17

 Recommendations
The body of this report contains detailed recommended corrective actions for implementation to
be considered by the Army, in collaboration with the Navy. In broad terms, the committee’s
recommendations can be distilled into three categories: enhancement of quality assurance, a
more extensive scientific peer review process, and improvements in program management for
inactivation and viability testing of BA.
Because of significant differences in irradiation procedures and viability testing for inactivated
BA samples, it is possible that this level of inconsistency exists in other high hazard operations.
The DoD should review all select agent procedures at the DoD laboratories. The committee also
found variance in each of the chains of command between the laboratories. This resulted in
inconsistent promulgation of safety best practices. It is prudent that the Army, Navy and Office
of the Secretary of Defense review the chain of command and pursue consistent safety practices.
The Chemical, Biological Defense Programs (CBDP) should ensure sufficient funding is
available for research related to the development of standardized irradiation and viability testing
protocols. The Office of Secretary of Defense (OSD) should oversee Army and Navy adherence
to the implementation plan and timelines.
Prior to implementing the committee’s other recommendations, DoD should continue its
moratorium on the production and shipment of inactivated BA, except as required for
development of standardized irradiation and viability testing protocols. Production of inactivated
BA spores should resume only after implementation of peer reviewed and validated procedures
for irradiation and viability testing.
The committee appreciates differences in operations between research and development facilities
and production facilities. Although research facilities should be provided flexibility to explore
the boundaries of science, the production facilities must operate in a repeatable and reliable
fashion to ensure quality products. For some common operations such as inactivation and
viability testing of BA, common procedures must be utilized. The committee’s findings are
divided into three categories and for each category, detailed recommendations are provided.
A) Quality Assurance: Enhance quality control programs at DoD laboratories working with
hazardous select agents and other pathogens. Establish consistent expectations regarding
equipment maintenance, calibration, record keeping, experimental controls,
environmental monitoring, and sample preparation. Conduct root cause analyses for any
procedures that result in anomalous outcomes. Procedures must stay within the validated
experimental design parameters. Provide accurate descriptions of sample contents and
storage and usage instructions for samples shipped to outside institutions.
a. Standardize BA Inactivation Protocols Across Laboratories: All DoD
laboratories should follow a common Standard Operating Procedure (SOP) for
such practices as irradiation and viability testing. This should be defined by the
Page   18

 highest standard necessary for any one of the laboratories using the most complex
sample.
i. Institute requirements to destroy any sample that deviates significantly
(e.g., 10, 20, 30, 50% etc.) from standard irradiation conditions. If the
sample fails the viability test, it should be logged, any anomalies of the
sample recorded, and the sample should be destroyed. A root-cause
analysis should be initiated to help understand what factors contributed to
the failure.
ii. Empirically identify optimal media and additives for recovery of injured
BA spores. Determine whether spores injured by gamma radiation can
recover (i.e., repair damage) in the formulation in which it is shipped as
well as the optimal temperature for recovery, and whether or not a shock is
required to induce germination (e.g., heat, freeze-thaw cycle). When
performing positive growth controls on media, inoculate with low spore
counts (10 to 100 cfu) to simulate low level spore survival in irradiated
samples.
iii. Adopt rigorous procedures to obviate cross-contamination of sterile
product with viable agent.
b. Institute More Rigorous Quality Procedures: Quality assurance (QA) and
quality control (QC) procedures for inactivation and viability testing of BA spores
were lacking across all DoD laboratories. Information on what should be routine,
such as noting how often a sample failed viability testing, was often absent within
the laboratories. The lack of a mature QA/QC program, and corresponding
incomplete documentation, makes determining the root cause of the problem
impractical.
c. Clarify The Conditions Of The Material Transfer Agreement (MTA): The
MTA for inactivated BA spore preparation should clearly prescribe the risks
associated with use of the product, such as indicating that the sample passed
viability testing at the limits of detection. Further qualifications within the MTA
could include sample storage, number of freeze-thaw cycles recommended and
requirements in the event the CRP needs to recall the lot. The samples should
have a declared shelf-life or use-by date. At a minimum, the MTA should be
signed prior to shipment and be required for each lot.
d. Enhance Preventive Maintenance: Reusable mechanical equipment used
throughout the process should be routinely maintained and calibrated, from
mechanical pipettes to the irradiator. Given the complete dependence of the
processes on key instruments, it was surprising that no systematic preventive
maintenance is performed. For example, ensuring the irradiator and dosimeter are
performing to specification is vital to operations. Both these instruments are
calibrated to National Institute of Standards and Technology (NIST)
specifications at installation, indicating the importance of the calibration process.
Even minor issues such as changes in humidity can affect these instruments.
Page   19

 Performing yearly certifications by qualified professionals should be mandated
for everything from pipettes to the dosimeter.
e. Establish And Manage An Environmental Surface Sampling Program: Some
laboratories lack written procedures to document, investigate, and report
contamination found outside primary containment areas during environmental
persistent agent sampling. The procedures, at a minimum, should include
notification protocols, including notification of the Safety Manager of the sample
locations, results and follow-up actions taken.
f. Establish Validated Dose Curves: Radiation dose curves should be generated
for each BA strain used for production, at the same concentrations used in
production and performed on the same irradiator as used for inactivation of spore
preparations. In order to accurately assess the correct amount of radiation to kill
BA spores effectively at various concentrations, radiation kill curves should be
generated for the specific BA spore strain being produced.
g. Understand The End Users’ Needs: The Chief Science Officer of the lab, Army
or Command, should work with individual customers to understand sample
requirements and define the appropriate strains and material needed to support the
objectives. For example, avirulent or attenuated strains should be used where
applicable to meet the intended use and goal. Many of the intended uses of these
materials can be easily fulfilled by such a substitution. In some cases there might
still be a need to inactivate virulent strains to fulfil a legitimate use.
h. Quantitate Spore Before Irradiation: Spore preparations intended for
irradiation should be quantified as precisely as possible to maximize the
likelihood of achieving the inactivation levels predicted by kill curves produced
with the same strain in the same irradiator. Spore preparations can be accurately
enumerated only if they contain negligible numbers of viable vegetative cells and
do not exhibit significant clumping. Therefore, preparations should be examined
microscopically prior to spore enumeration to determine if they meet
predetermined standards for the content of phase-bright and phase-dark spores,
vegetative cells, and debris. If preparations do meet standards, clumping should
be reduced by the use of surfactants and disruptive techniques (e.g., sonication) to
maximize the accuracy of counts.
B) Peer Review: Establish BA spore inactivation and viability testing protocols that are
based on relevant scientific data, standards, and studies conducted to fill knowledge gaps.
All protocols and subsequent protocol modifications should be subject to a peer review
process, validated and implemented uniformly across similar operations. Ensure internal
and external audits have a scientific component to assess the validity of proposed
procedures and assure a science based foundation for proposed methodologies, changes
and deviations. Improve collaboration between DoD laboratories and with the broader
BA research communities to promote information sharing, peer review, and promulgation
of best practices and lessons learned.

Page   20

 a. Periodic Scientific Reviews Of Protocols And Procedures: The rate of
advancement in science is accelerating and often these advancements will have
material impact on existing protocols and procedures. To ensure that the DoD
laboratories are taking advantage of the most up to date knowledge, external
Subject Matter Experts (SMEs) should be engaged periodically to review current
best practices and determine if new information requires updating of protocols
and procedures.
b. Assemble A Group of (Internal Or External) Irradiation SMEs: To identify
and determine the studies that needs to be done to better understand the effects
and limitations of irradiation for inactivation of BA spores. This should include
an investigation of the following variables: irradiator model variations; dose of
radiation required to inactivate BA spores; inter-strain susceptibility to radiation
exposure; concentration of spores and dispersion (evenly dispersed versus
clumped etc.) when exposed to radiation; and broth types for spore suspensions.
c. Assemble A Group Of (Internal Or External) Viability SMEs: To determine
the best approach for implementation of viability tests for inactivated BA spores.
This should include: when to subject the irradiated material for viability testing;
what type of broth should be used to support and monitor recovery of damage
spores; what temperature should be used for incubation of these samples and for
how long; how much of the irradiated material should be subjected to viability
tests; what type of agar plates should be used for growth monitoring; what
temperature should be used for incubation of plates and for how long; what
positive controls and negative controls should be implemented to support viability
testing; and remedial action notification processes for viability test failures.
d. Audits And Inspections: DoD laboratories that utilize select agents should be
inspected by audit teams comprised of competent microbiologists, scientists, and
credentialed subject matter safety experts. Inspections should be meaningful with
a focus on technically reviewing the processes and procedures for handling select
agents. These audit teams should be led by the major command surety program
managers and comprised of subject matter experts with operational experience in
select agents and toxins.
e. Assure Material Integrity: The materials subjected to irradiation must maintain
their integrity for the specific intended use and application. Overexposure of
materials to radiation may compromise the sample integrity and reduce its
potency to support the specific intended use.
f. Provide Independent Viability Verification: Employ an independent laboratory
to verify a sample is inactivated prior to shipment. Implement a process for
performing a secondary viability test prior to shipping or distributing materials
from the production laboratories to customers. This independent confirmation
will reduce anomalies or limitations associated with the production or viability
testing process.

Page   21

 C) Program Management: Program managers should provide adequate laboratory space,
equipment and time to conduct relevant safety and surety research for select agents and
other pathogens. Program managers should develop a plan to track and document the
implementation and long-term sustainability of the recommended corrective actions
identified by this review committee as well as all internal and external audits.
a. Prioritize R&D For Those Pathogens Where Information is Lacking: A
problem that was endemic to all DoD laboratories visited was the failure to
recognize the importance of knowledge gaps. Unknown variables throughout the
irradiation and viability processes make a root cause analysis problematic.
Performing key experiments that can define effects of variables such as; spore
concentration, irradiation temperature, type of media, sampling size for validity
testing should be researched to define more definitively appropriate parameters
for the protocols needed.
b. Minimize Cross-Contamination: Ensure laboratory space associated with
production of inactivated spores is separate from laboratory space that performs
viability tests in order to avoid cross-contamination. Live materials and
inactivated materials should be handled in separate laboratory space or in space
with sufficient engineering controls and practices to obviate cross-contamination.
c. Revise Death Certificates: Instead of using "Kill or Death Certificate", rename
the certification of inactivation to "Viability Test Results” with information about
the sampling approach and the length of incubation period. An express statement
of the detection limit of the viability testing protocol should be included, as well
as a statement that it is the responsibility of the receiving laboratory to ensure that
the work conducted using this material is executed in a safe and contained manner
minimizing any potential risk to their laboratory scientists. Additionally, if any
growth is observed by the receiving laboratory, it is the responsibility of the
receiving party to notify the sender and other appropriate parties immediately, and
secure the sample.
d. Track Inactivated Select Agents With The Same Procedure Used For Live
Select Agents: This electronic tracking mechanism will enable a better sample
custody in future for potential recalls if necessary.
e. Establish An Oversight Group To Insure Implementation Of Corrective
Actions: For the corrective actions to be effective and persistent, an internal
group should actively engage in tracking standards, processes and procedures
throughout the laboratories.

Page   22

 Appendix A: Laboratory Visits and Discussions
ECBC Visit
The review Committee visited ECBC on June 8, 2015. During this visit, the committee chair
provided an overview of the live BA shipment incident and the purpose of the visit and openly
discussed aspects of the work done at the laboratory with committee members. Further, the
committee received a demonstration on how samples are loaded in the ECBC cobalt-60
irradiator. During the demonstration, dose curves (mapping of the exposure rate within the
chamber) were shown with an emphasis placed on the relevance and importance of location of
the sample. The following sections provide a synopsis of what was discussed during the tour and
visit.
Preparations of Inactivated BA Spores:
ECBC produced three batches of inactivated BA spores. Two preparations of spores, produced
in 2006 and 2007, passed viability testing. ECBC indicated that these two batches were created
for the National Bioforensic Analysis Center (NBFAC), to support an electron microscopy
research study. The remaining spore preparation, which was made in 2014, was destroyed
following viability testing that demonstrated viable spores were present. This batch differed
from the others in that the spores were centrifuged prior to irradiation with the supernatant
(excess liquid on top of sample pellet) removed. The resulting wet pellet was then irradiated (as
opposed to spores held in suspension). The assumed inactivated sample was successfully
identified as still viable by the viability testing and it was destroyed.
Irradiation Methodology:
ECBC uses a Cobalt-60 gamma irradiator for inactivation. The primary function of the irradiator
is to treat unknown samples prior to analysis for chemicals; it is not used for routine production
of inactivated spores. The inactivation gamma radiation dose was determined from work
published by the United Kingdom Ministry of Defense Chemical Defense Establishment as well
as kill curve experiments conducted at ECBC using Bacillus globigii, an organism that is
claimed to be more difficult to kill with radiation than BA. Both studies showed that a complete
kill was achieved with a dose of 40 kilo Grays (kGy), ECBC made it policy to use a dose that is
30% higher, or 54 kGy. This decision was made to increase confidence in the treatment of the
unknown samples prior to chemical testing.
To conduct the procedure, the radiation safety officer (RSO) calculates the position of the
container of spores to be irradiated in the chamber, as well as the time of exposure to achieve the
necessary dose, factoring in the age of the radiation source (Cobalt-60). The particular model of
irradiator used contains a turntable to rotate materials undergoing irradiation in front of three
gamma radiation sources; however, validation of the dose received is not performed. ECBC
scientists observed cellular debris and clumping of the spore suspensions after irradiation was
Page   23

 performed. ECBC does not maintain control charts to identify the frequency of off-normal
operations by their irradiation instruments.
Viability Testing:
ECBC takes 10% of the irradiated lot of spore material for viability testing. This sample size
was determined by a statistician factoring in infectious doses, including consideration of
organisms that have a low infectious dose. According to ECBC, to obtain a range of 6-7 log
reduction (e.g., 1 log reduction is 90% killed, 2 log is 99%, 3 log is 99.9%, etc.), this sample size
is considered statistically relevant. Viability testing is conducted as soon as possible after
irradiation, which is not always within the same workday, but is always within 24 hours. The
microbial culture medium used for the test is Trypticase Soy Agar (TSA), a solid growth medium
used in petri dishes. The viability tests were incubated for five days at 37oC. The viability test
procedure does not utilize methods to recover stressed organisms such as culturing through broth
media followed by plating on solid media, or richer media containing higher levels of nutrients.
Additional Comments:
The personnel at ECBC were highly cooperative and willing to answer questions from the
committee. The Standard Operating Procedures (SOP) addressed safety, and incorporated
thorough risk assessments. The committee noted that ECBC used statistical analysis to
determine the size of sample (10%) to be taken for viability testing. However, the assumptions
for this analysis focused on infectious dose, and possibly did not take into account stressed
spores or cells. Additionally, 10% sampling provides 95% confidence, which may be
statistically relevant, but insufficient for verification of complete inactivation. Direct plating was
performed for viability testing, whereas enrichment through broth culture followed by plating
may have been more optimal. The incubation time of five days also presents an opportunity for
improvement. In the pharmaceutical industry, a standard practice is to incubate viability test
broths and plates for 14 days. In addition, growth promotion testing of the culture media prior to
use in (or in parallel with) viability testing is required to ensure that growth of small numbers of
viable organisms are detected. Use of positive and negative controls for the viability test is
another opportunity for improvement.
In discussions, the committee pointed out that kill curves, in different volumes/concentrations
and various suspension matrices, should be performed with all strains of BA that could be
subject to inactivation by irradiation. It has been observed that different strains have varying
resistance to radiation, and as part of process validation these experiments need to be performed.7
It was also pointed out that germination curves for injured spores show a longer time to
germinate compared to non-injured spores. In an irradiated spore preparation, low numbers of
surviving spores could use residual cellular debris and broken spore debris as a nutrition source.
Continuing discussions on nutrients revealed that Trypticase Soy Agar is not a rich medium as

Page   24

 compared to Sheep Blood Agar or Brain Heart Infusion Agar and it may not provide the
optimum growth conditions.
When provided opportunity to comment on improvements, ECBC identified three areas:
perform independent verification of radiation dose by using alanine test strips, utilize broth
enrichment as a component of viability testing, and lastly, incorporate a dwell (rest) time post
irradiation before performing viability testing. Research papers, using different irradiation
techniques, have shown that irradiated spores that are injured but not killed may recover and
become viable after a period of rest.3,4,15

Page   25

 USAMRIID Visit
The review committee visited USAMRIID and NMRC on June 9, 2015. Due to the co-location
of these two facilities, it was more efficient to meet with both teams at the same time. The
committee chair provided an overview of the live BA shipment incident and the purpose of the
visit. The committee members were allowed to freely discuss various aspects of the work done
at the laboratory. The following sections provide a synopsis of what was discussed during the
committee’s visit with USAMRIID staff.
Preparations of Inactivated BA Spores:
The original number of inactivated batches reported by USAMRIID was overestimated due to a
misunderstanding of the request for information. USAMRIID provided a count of any
preparation containing BA, including both vegetative cells and spores. As a research institute,
USAMRIID does not routinely irradiate BA spores. The irradiated BA spore preparations were
used to create antibodies or to support activities in the Diagnostic Systems Division. The spore
preparations used to produce antibodies were sent to an outside vendor to inject into rabbits (for
antibody production). None of the rabbits injected with these BA spore preparations became ill
or died from infection (thus illustrating that if any live spores remained, they were present at a
concentration below infectivity to the rabbit). The current estimate of inactivated BA spore
preparations produced at USAMRIID is about 26 (<10 samples since 2009). During the
discussion with the technical staff, there were inconsistent answers about how often irradiated
samples test positive. One researcher used a generic answer of 2-3% (no documentation was
provided); others pointed out that at USAMRIID some samples are purposefully subjected to a
sub-lethal dose for research purposes and are expected to fail the viability tests. In general,
USAMRIID uses exceptionally purified spores with minimal contaminants (typically >99%
pure).
Irradiation Methodology:
USAMRIID conducted kill curve studies on Ames and Colorado strains of BA, more than a
decade ago. The study subjected samples to gamma radiation doses up to 40 kGy with the
results indicating spore inactivation at 20 kGy. For inactivation purposes, the decision was made
to use 40 kGy as the dose for spores. Earlier in this report we showed that the current
operational procedures used by USAMRIID (and all other DoD laboratories) are outside of the
historical experimental kill curves produced at USAMRIID and captured in published
research.7,17-20
To conduct the procedure, the radiation safety officer (RSO) calculates the position of the
container of spores to be irradiated in the chamber, as well as the time of exposure to achieve the
necessary dose, factoring in the age of the radiation source (Cobalt-60). USAMRIID routinely
freezes their samples and maintains this state by employing an ice/dry-ice bath in which the
samples sit throughout the irradiation. Independent verification of the dose received is not
Page   26

 performed. USAMRIID does not maintain control charts to identify the frequency of off-normal
operations by their irradiation instruments.
Viability Testing:
Inactivated spores are prepared in small batches (up to 15 ml) based on experimental needs. Post
irradiation, 10% of the batch is plated on Sheep Blood Agar (SBA). A 48-hour incubation time
was chosen due to BA being a fast growing organism. Broth cultures are used if greater volumes
than the small batch are prepared.
USAMRIID described a small batch as 3-4 ml of inactivated spores per rabbit for antibody
production. Up to 15 ml is produced to support testing on up to 5 rabbits per test. Spore titer
(concentration) for these small batches is determined by direct plate count, most of the time on
SBA. This is the same lot of SBA used for titer determination during viability testing.
USAMRIID conducts viability testing as soon as possible post irradiation, within 24 hours. It is
noteworthy that the researcher doing the spore preparation work for antibody production chose to
conduct viability testing by taking 50% of the preparation as the sample.
Thorough contamination mitigation is implemented prior to conducting viability testing. The
biological safety cabinet (BSC) is decontaminated with 10% stabilized bleach, or Hype-Wipe®
disinfecting towels with bleach, or freshly prepared bleach. As part of this contamination
mitigation, positive controls, media checks and other steps where viable BA is grown are
conducted in areas that are physically separated from where viability testing is conducted.
Additional Comments:
USAMRIID viability testing on irradiated spore preparation demonstrated an estimated failure
rate of <3% (i.e., >97% of samples showed complete gamma inactivation). The preparations
showing growth during viability testing were associated with the transition to a new irradiator.
When a committee member asked for the result of a root cause analysis, it was stated that no root
cause analysis was performed. During subsequent discussions, it appears that a potential source
of irradiation failure was identified by USAMRIID personnel. The new irradiator used a
turntable to expose the material to the cobalt source. The RSO stated that JL Shepard, the
manufacturer of this irradiator (484R), has a problem with the turntable stopping during the
irradiation run. There is no real-time mechanism to identify when during the irradiation cycle
the turntable stops functioning. In one instance, the RSO advised the researcher of a turntable
failure (hence the sample was not uniformly irradiated), and when viability testing was
performed, growth was observed. The spore preparation was subjected to repeat irradiation in an
Atomic Energy of Canada Ltd Gammacell 220 irradiator that drops the material into a chamber
surrounded by radiation sources. Viability testing showed no growth after this treatment. The
two other preparations showing growth during viability testing were also associated with the
transition; improper placement of sample in the chamber of the model 484R was suggested as the
cause.
Page   27

 In an open discussion, the committee discussed various aspects of spores under pH stress,
germination time and the influence of heat stressing. The U.S. Department of Agriculture has
evaluated this phenomenon and recommends extending incubation times when testing for
viability. The CDC DSAT is now requiring all second checks for viability to be conducted in
enriched media for 14 days (based on June 5, 2015 updated guidance for testing inactivated BA
spores).
USAMRIID staff recommended post irradiation microscopic evaluation of spore preparations,
and it was pointed out that spore preparations from Dugway Proving Ground are “rough preps”,
with cellular and spore debris present. It was hypothesized (in the absence of experimental data)
that this debris can degrade over time to possibly form a nutritional supplement that would
support germination and growth. In a follow up note to the committee chair, USAMRIID
officials elaborated that:
“Recent interagency discussions has identified that clumping noted in Dugway
spore preparations may be a significant contributing factor to both failed
inactivation and delayed growth during viability testing. Samples held at room
temperature have shown a 3 log increase in colony counts and we know that
clumps dissociate over time releasing spores from their center into solution.
Micrographs have also shown very large spore clusters mixed with debris in these
preparations. This could result in an apparent increase in viable spores when it is
actually an increase in availability of previously existing spores that were trapped
within a clump.”
Concern was also expressed regarding low titers (concentration) of viable spores in an irradiated
preparation, and the probability of their detection through sampling.
In response to a question from a committee member as to the feasibility of using an attenuated
BA strain rather than inactivated BA spores, USAMRIID identified a number of possible
alternatives. During discussions with USAMRIID’s staff, it became clear that recipients of
inactivated spores should be informed that there is no guarantee of zero germinating spores in the
preparation.

Page   28

 NMRC Visit
The review committee visited NMRC on June 9, 2015. Due to the co-location of NMRC with
USAMRIID, it was more efficient to meet with both teams at the same time. The committee
chair provided an overview of the live BA shipment incident and the purpose of the visit. The
committee members were allowed to freely discuss various aspects of the work done at the
laboratory. It is important to note that NMRC does not have current funding to pursue BA
research activities and accordingly, they are responsible for verification of legacy samples
prepared as early as 2005. In its current location, NMRC has BSL-3 laboratories at USAMRIID
and operate under USAMRIID’s Select Agent Registration. NMRC adheres to USAMRIID's
BSL-3 standard operating procedures. The following sections provide a synopsis of what was
discussed during the committee’s visit with NMRC staff.
Preparations of Inactivated BA Spores:
The initial number of inactivated BA spore preparations reported by NMRC represented double
counting; the actual number of preparations is less than 40. These preparations were used to
develop antibodies at both an outside vendor as well as the USAMRIID Large Animal Research
Facility, and to conduct testing of immunological assays under development. As part of Defense
Base Realignment and Closure (BRAC), all NMRC BA work was moved to USAMRIID in
2011. There is a memorandum of agreement in place and NMRC works under USAMRIID’s
operational and safety protocols.
Irradiation Methodology:
Spore preparations under the NMRC protocol used the JL Shepherd & Associates Co-60 Model
109 Irradiator, which surrounds the material being treated with source rods. The dosage used by
NMRC was a minimum of 32.5 kGy. The NMRC protocol used multiple runs at lower doses but
achieved a final absorbed dose of 40-50 kGy. The spore suspensions were frozen solid for each
irradiation run. Up to 400 ml spore suspensions were irradiated. NMRC currently plans to use
the facility at USAMRIID for their sample irradiation. In their current location, USAMRIID
does not maintain instrument quality control and performance charts. In their previous location,
NMRC did not maintain control charts to identify the frequency of off-normal operations by their
instruments. At the time, the irradiator used by NMRC was under the control of Walter Reed
Army Medical Center (WRAMC). Any records relating to this irradiator would have been held
by WRAMC rather than NMRC.
Viability Testing:
The NMRC protocol for viability testing had two different test routines detailed, one for gross
contamination and another for low levels of contamination, with sample sizes defined based on
the amount of material to be tested. In tests for gross contamination, the maximum sample to be
taken was defined as 1-5%. Inoculation was on plate media, chosen from a list of media in the
Page   29

 protocol (the protocol states not to use Tryptic Soy Agar-TSA). A noteworthy element of the
protocol is that a media growth promotion test is conducted with the viability test. In tests for
low levels of contamination, a 10% sample of the material is inoculated into broth and incubated
followed by plating. The brain heart infusion broth used for BA is supplemented with 10%
serum to create a rich medium.
Additional Comments:
When NMRC produced inactivated BA spores, the documentation of viability testing provided to
the recipient did not state that the preparation was sterile, but that viable spores were below the
limit of detection. Viability tests had to be read by two people who would sign off on the
documentation of the procedure.
NMRC routinely used multiple radiation doses to preserve antigen structure. Up to a week
between runs could occur, in part due to scheduling time in the irradiator. Similar to
USAMRIID’s observation, when inactivated BA spores were sent for antibody production there
was no report of inoculated animals becoming ill or dying.

Page   30

 DPG Visit
The review committee visited DPG on June 17, 2015. The committee chair provided an
overview of the live BA shipment incident and the purpose of the visit. During the visit, the
committee noted that the staff was very concerned about the recent events. The committee
members were allowed to freely discuss various aspects of the work done at the laboratory. The
Biological Safety Officer for the Life Sciences Division (LSD) at DPG opened the meeting with
a presentation on Post safety and the agenda of the laboratory visit.
The committee visited Building 2032 to observe the two gamma cell irradiators utilized by LSD.
Irradiator 484R has been in operation since 2012. This irradiator has 3 cobalt source rods in the
rear of chamber and a rotating turntable. Samples are irradiated at chamber temperature without
ice (DPG’s in-house research has shown no significant changes in sample temperature during
irradiation process). Alanine stripes are placed on opposite sides of the vial holding the sample
(primary container). Alanine forms a very stable free radical when subjected to gamma
radiation. These free radicals yield electron paramagnetic resonance signals that are dose
dependent, thus allowing an accurate measurement of the radiation dose. Irradiation failures are
communicated electronically to laboratory personnel (usually the principal investigator).
However, control charting or monitoring of these failures is not performed.
Next, the committee visited Baker Laboratory which now houses a state of the art biological
aerosol test chamber. Lastly, the committee visited the Life Sciences Test Facility (LSTF) in
Building 2029. LSTF houses the Critical Reagents Program (CRP) Antigen Repository. As of
this reporting, all positive BA spore lots were produced on behalf of the CRP. The CRP Antigen
Repository consists of a sample packaging and shipping lab, -70oC sample storage freezers, two
BSL-2 laboratories for production and characterization of non-Select Agent material, and a BSL3 sample preparation and viability testing lab. The following sections provide a synopsis of what
was discussed during the committee’s tour and visit.
Preparations of Inactivated BA Spores:
DPG produces inactivated biological agents in support of the DoD Critical Reagents Program.
These materials are provided to customers to support research and development, quality
assurance, and test and evaluation activities. When the committee asked “how often do
irradiated samples fail the viability test?” the answer was about 2-3%. DPG communicated that
no root cause analysis was performed on these anomalies. At the end of our discussion, the
committee members asked for documentation on failed viability tests of irradiated samples. The
documentation provided shows that since 2012, there have been a total of 19 BA inactivations.
From these, four samples failed the viability test (21% failure rate). About 10% were attributed
to a failing turntable in the irradiator. Lack of supporting documents for many of the answers to
critical questions during the committee’s visit pointed to poor record keeping in a critical
production laboratory.
Page   31

 Irradiation Methodology:
Two different irradiators have been used at DPG. The first, a Gammacell 220 was taken out of
service in 2012, and a JL Shepherd & Associates Co-60 Model 484 with a two inch turntable was
placed in service in the same year.
Irradiation dosage was based on literature and historical information. The dosage received
during runs was determined by taping two alanine dosimetry strips, on opposite sides, to the
outside of the tube containing the spore suspension. The tube(s) is then placed inside a
polypropylene secondary container with an O-ring sealed lid. Upon loading, the turntable is
turned on via a switch on the control panel, the door is closed and the irradiation process is
initiated. The run is performed at ambient temperature, and logged in a notebook. When the run
is over, the door is opened, and turntable operation is confirmed visually; however, this is not
often documented. If any anomalies such as a turntable failure are observed, then the anomaly is
communicated to the researcher. The alanine strips are subsequently read in a Bruker e-scan
alanine dosimeter reader and any anomalies are recorded in the instrument log and reported to
the principal investigator. The DPG technician observed about 10% of the irradiated samples
had some form of anomaly.
The time of exposure is derived annually by a technician taping five test strip inside a glass
beaker and running it through hourly intervals in the irradiator. At the end of each time point, a
single test strip is removed and analyzed on the dosimeter. When all time points are completed,
the doses are plotted against time. This plot is used to determine time of exposure to achieve
dose. This plot is never compared to the expected dose from the Co-60 source(s) based on the
known decay kinetics. This is a built-in control that should be used to verify the dosimeter
calculations. The dosimeter is periodically checked, once in 2012, and again recently. When
asked if the e-scan dosimeter requires routine calibration, the DPG technician responded that
calibration was not required to his knowledge. It was not determined if the Bruker e-scan was
calibrated and if it met the National Institute of Standards and Technology (NIST) traceability
criteria upon initial installation.
Empirically, DPG observed that the difference in radiation exposure of the outside of the
secondary container and the inside was 10-20 kGy which was consistent with the irradiator isodose curves. Consequently, DPG discontinued taping alanine strips to the secondary container
and began using strips taped to opposite sides of the primary container.
Approximately 50 irradiation runs are performed a year including virus preparations for internal
research use and certain culture media; it was estimated that approximately 500 irradiation
procedures conducted to inactivate bacterial agents have been performed in the history of the
Critical Reagents Program.

Page   32

 Viability Testing:
Irradiated samples are often tested immediately after irradiation, typically within 30 minutes.
The protocol initially provided to the committee, WDL-BIO-147 was not the protocol used to
prepare CRP inactivated antigens. A hard copy of the protocol CRPAR-WI-007 was provided to
the committee during the meeting. This protocol calls for a 5% sample of the irradiated material
to be inoculated into 2X nutrient broth (the concentration of nutrients are twice the normal),
incubated at 340 C for 48 hours. Two hundred microliters of inoculated broth is then plated
across 10 TSA plates and the plates are incubated for a minimum of 48 hours. It was noted that
for the pre-2012 irradiation samples, an average incubation time for the samples was 12.9 days
and for the new instrument, the average was 13.3 days. Positive (undiluted) and negative
controls are performed at the same time in different biological safety cabinets in the same room.
DPG procures the culture media from an ISO accredited source, and does not perform growth
promotion testing on new lots. Based on what the committee learned at other DoD laboratories,
this procedure should be revisited to include best-in-class practices.
Additional Comments:
At the request of the committee, DPG reviewed past records and found that lots of inactivated
BA spores that contained viable material were produced using both the decommissioned and
currently active irradiator.
From the laboratory walk through the committee learned that with the exception of the irradiator
facility, all other elements of the select agent production process occurs in the same BSL-3
laboratory. Agent propagation, sporulation, and viability testing procedures are performed in the
same laboratory space that is shared with other inactivated biological agent work, without
dedicated equipment (aside from two biological safety cabinets). The committee noted that this
approach could introduce cross-contamination. Furthermore, key reagent transfer equipment
such as mechanical pipets is also shared. Although it is unknown at this time if crosscontamination occurred with DPG samples, it cannot be ruled out. DPG acknowledges this issue
and plans to have dedicated laboratory space for viability testing when a new laboratory building
is commissioned.
A committee member noted the discrepancy on the death certificates for two lots of inactivated
spores (AGD0001151 and 1667) where protocol WDL-BIO-147 Rev1 Ch1 is cited, when the
actual protocol in use was CRPAR-WI-007 Ver 5. DPG acknowledged that they recently had
this error pointed out and that it would be corrected.
DPG personnel stated that the typical irradiation time to achieve 40 kGy exposures is 240
minutes at the current decay of the Co-60 source in the model 484 irradiator. It was pointed out
that lot AGD0001667 had a radiation dose of 124.02 kGy noted on the death certificate. Review
of the records showed that the lot had been irradiated twice after growth occurred on the viability
test after the first run. The time for the first run was noted as 220 minutes, and the second run
Page   33

 was 240 minutes. The alanine test strips showed a cumulative dose of 124.02 kGy which is what
was reported. Upon showing no growth on the second viability test, the lot was approved for
release March 18, 2014. No effort was made to investigate the anomaly of the dosimeter
showing 124.02 kGy under exposure conditions that are expected to produce a total exposure
around 80 kGy.
DPG personnel noted in the retesting experiments currently underway, multiple Bacillus species
including B. anthracis strains that were inactivated and archived have come up positive (i.e.,
viable spores are present in inactivated samples).
The review committee was unable to identify if there had been a similar event with non-CRP
program samples. Typically, 100% of non-CRP batches are sent to customers, or used for
internal R&D purposes, with nothing archived, thus no samples remain at DPG to retest.
Archived samples (specifically data tied to variations in production methods) could help gain
insight in achieving best laboratory practices when striving for complete inactivation results.
After the visit, the committee received some additional information from DPG regarding their
spore preparation, reiterating that all laboratory equipment is surface decontaminated before and
after each use to minimize cross-contamination.

Page   34

 Appendix B: Commonalities and Differences in Procedures
The following table outlines the variability in irradiation procedure used at DoD laboratories.

The following table outlines the variability in viability testing used at DoD laboratories.
Page   35

 Comparison of viability testing procedures for gamma

irradiated B. anthracis spores

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sterility Testing
Sampling Culture Media Incubation
Laboratory St 
38:1? Broth 801i Media Broth to Direct Incubation Incubation Positive Negative
 Plate? Plating Temp (0C) Time (hrs) control? control?
. 25-37
ECBC 10% None ?3 ?3pm No 100 ?appropriate 72 No No
Agar ulfplate 
growth temp
. . A I None Blood Agar I None I None i 
10 0? 0 1dent1i1ed 1dent1t1ed 1dent1t1ed 37 48 NO 3 es
Blood Agar,
Brain Heart Bra?? Heart Broth:
Small vol. I . Inrusmn 
140 Inius1on A HI 1 pi and >?72
NIVTRC Broth g. 200ul are 35-37 72 Yes Yes
Large vol. Nutr1ent 
-10% I plated Plates.
1 ?5 9?0 serum Agar? 7' 2
Mueller-
Hinton agar
25 ml 2X .
- 
Dugway 5% nutr1ent soy 200 Agar plates
broth
Speci?ed in 6f8f15 meeting at ECBC
Speci?ed in 6.911 5 meeting at 

(C)

Speci?ed in 6H 7.115 meeting at DPG

Page 36

Appendix C: Primary and Support Member of the Review Committee
*

Committee Members Visited DoD Laboratories

NAME

ORGANIZATION

Dr. Vahid Majidi*, Committee Chair

OASD(NCB)/NM

Dr. Douglas Beecher*

FBI

Dr. Gigi Gronvall*

UPMC

Ms. Donna Hudson*

Consultant/SAIC

Dr. Alan Jacobson*

DTRA

CAPT Malcolm Johns USPHS*

DHS

Dr. Richard Meyer*

Consultant/Tauri Group

Dr. William Palmisano*

ATEC

Mr. Geoffrey Phillips*

USAMRMC

Mr. Scott Winter*

DTRA

SSA John Woodill*

FBI

Dr. Malin Young*

SNL

Ms. Lisa Foddrill

DHS

Dr. John Jensen

USDA

Mr. Joseph Kozlovak

USDA

Dr. James Lindsay

USDA

Dr. Brendan Niemira

USDA

Dr. Segaran Pillai

DHS

Dr. James Rogers

USDA

Ms. Susan Weekly

Consultant

Dr. David White

USDA

Page   37