CENTER FOR DRUG EVALUATION AND
RESEARCH

APPLICATION NUMBER:

2056920rig15000

SUMMARY REVIEW

 

Division Director Review

Summary Review for Regulatory Action

 

 

 

 

 

 

 

 

 

 

 

 

 

Date (electronic stamp)

From Jean-Marc Guettier, MD

Subject Division Director Summary Review

205692

Supplement 

Applicant Name Eli Lilly and Company

Date of Submission 10/18/2013

PDUFA Goal Date 08/18/2014

Proprietary Name Basaglar (insulin glargine injection)

Established (USAN) Name

Dosage Forms Strength Injection 100 units/mL

Proposed Indication(s) Improve glycemic control in Type 1 (adults and
children) and Type 2 (adults) diabetes mellitus

Action/ Recommended Action: Tentative Approval

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Material Reviewed/Consulted

OND Action Package, including: Names of discipline reviewers
Medical Officer Review Lisa Yanoff, MD

Statistical Review Lee Ping Pian, 
Pharmacology Toxicology Review Miyun Tsai-Turton, 

CMC Review/OBP Review Xavier Ysern, and Muthukumar Ramaswamy, 
Microbiology Review Jessica Cole, 

Clinical Pharmacology Review Manoj Khurana, 

DDMAC Ankur Kalola

DSI Jong Hoon Lee

CDTL Review Lisa Yanoff, MD

Sarah Vee, PharmD

 

 

0ND=0ffice of New Drugs

DDMAC=Division of Drug Marketing, Advertising and Communication
Office of Surveillance and Epidemiology

DMEPA=Division of Medication Error Prevention and Analysis
DSI=Division of Scientific Investigations

Division of Drug Risk Evaluation

DRISK=Division of Risk Management

CDTL=Cross?Discipline Team Leader

Page 1 of 14

Reference ID: 3612365

Division Director Review

Signatory Authority Review Template
1. Introduction
Eli Lilly and Company submitted a new drug application pursuant to section 505(b)(2) of the
Federal Food Drug and Cosmetic Act (FD&C Act) for Basaglar. Basaglar injection is a solution
containing 100 units of insulin glargine per mL filled in a 3 mL glass cartridge pre-assembled
in an auto-injector pen-device (Basaglar KwikPen). The applicant is seeking to indicate
Basaglar to improve glycemic control in adults and pediatric patients with type 1 diabetes
mellitus and in adults with type 2 diabetes mellitus.

2. Background
The applicant relies, in part, on FDA’s finding of safety and effectiveness for the listed drug
Lantus1 (insulin glargine [rDNA origin] injection; NDA# 021081) to support approval of
Basaglar. The applicant established a “bridge” between Basaglar and Lantus to demonstrate
that Basaglar was sufficiently similar to Lantus such that reliance is scientifically justified.
These data, together with product-specific data (including product-specific data
demonstrating safety and effectiveness), establish Basaglar’s safety and effectiveness for its
proposed conditions of use.
The composition, strength, and presentation of Basaglar are similar to the composition,
strength and presentation of the US-approved listed drug (refer to Drs. Ysern and
Ramaswamy’s review for details). The applicant, in a series of analytical studies, compared
the identity, purity, potency and stability of Basaglar to Lantus. The applicant evaluated the
impact of potential differences between the two products on safety and efficacy using
toxicology bridging studies, clinical pharmacology bridging studies and clinical studies. These
bridging studies support the scientific appropriateness of reliance on FDA’s finding of safety
and effectiveness for Lantus to support approval of Basaglar, and the clinical studies also
provide data on the safety and effectiveness of Basaglar.
The toxicology bridging studies allows for an abbreviated non-clinical development program
and provides scientific justification for reliance on FDA’s finding of safety for Lantus (as
reflected in product labeling that describes, among other things, reproduction and early
development, carcinogenicity and chronic toxicology studies) to support approval of Basaglar.
Additionally, these toxicology studies qualify, from a toxicological perspective, any
differences in impurity/degradant profiles that may result from a difference in manufacturing
processes between Basaglar and Lantus.

1

Note: The applicant refers to Lantus (NDA# 021081) as ‘US-approved Lantus’ in the application to distinguish it
from the Lantus product approved for use in the European Union. In my review Lantus refers to the USapproved listed drug.

Page 2 of 14

Reference ID: 3612365

 Division Director Review

The clinical pharmacology bridging study (ABEO) allow for an abbreviated clinical
pharmacology program and provides scientific justification for reliance on FDA’s finding of
safety and effectiveness for Lantus (as reflected in product labeling that describes, among
other things, clinical pharmacology studies such as special populations studies, timing of
administration studies, drug interaction studies) to support approval of Basaglar.
The clinical studies compared the impact of product-related differences on clinical efficacy
outcomes (i.e., HbA1c), provide comparative clinical safety data in the chronic use setting and
comparative clinical immunogenicity data after repeated and chronic dosing. In the clinical
program, long term safety for Basaglar and Lantus were compared in the two adult
populations for whom Lantus is indicated (i.e., type 1 and type 2 diabetes populations).
Safety concerns including but not limited to those related to interactions with coadministered drugs, risk of hypoglycemia, risk of immune mediated disorders, risk of weight
gain, risk of fluid retention and risk of cardiovascular complications (i.e., heart failure; CVD)
are different between the two diseased populations studies and both studies were needed to
provide robust comparative safety data relevant to the intended uses.
The two clinical studies allow for an abbreviated clinical development program and provide
the scientific justification needed to allow reliance on FDA’s finding of safety and
effectiveness for Lantus to support approval of Basaglar across different patient populations
(adult and pediatric patients with type 1 diabetes mellitus and type 2 diabetes mellitus,
patients with type 2 diabetes at high risk of cardiovascular events) and across multiple clinical
use scenarios (e.g., as basal bolus therapy, as basal therapy added to background oral antidiabetic agents).
To support approval of a completely novel insulin molecule (i.e., a new analog) clinical data to
establish efficacy and safety of the new product across the two distinct adult patient
populations (type 1 and type 2 diabetes) and across the most common clinical use scenarios
(basal bolus regimen, added to background oral drugs) generally is required. In general, at
least two clinical studies in type 1 diabetes and two clinical studies in type 2 diabetes are
used to support an adult indication for each of these two diseases. Additional studies could
be required pre-marketing to qualify the clinical impact of a novel identified risk or to assess
the efficacy and safety of a novel dosing regimen if this is proposed by the applicant. We also
would need to consider whether pediatric clinical studies would be required to support a
pediatric indication for patients with type 1 diabetes and type 2 diabetes.
In phase 3 clinical trials, the applicant also compares Basaglar to European Union approved
Lantus (EU-approved Lantus); the applicant included sites in phase 3 trials outside the US.
The applicant provides analytical data, toxicology data and clinical pharmacology data
comparing Lantus to EU-approved Lantus that provide an adequate scientific bridge to the
U.S.-approved listed drug and justify the relevance of this supportive comparative data.
Studies that were used for this bridging purpose will not be discussed in this review and are
detailed in individual reviews and in Dr. Yanoff’s CDTL memorandum.

Page 3 of 14

Reference ID: 3612365

 Division Director Review

3. CMC/Device
CMC data were reviewed by Drs. Ysern and Ramaswamy and summarized in Dr. Yanoff’s CDTL
memorandum. The applicant characterized the chemistry, manufacturing processes and
controls for the drug substance, excipients and drug product in Basaglar.
The drug substance in Basaglar is manufactured by recombinant DNA technology using a
(b) (4)
specific E.Coli production strain.
The composition, strength, and presentation of Basaglar are similar to the composition,
strength and presentation of the listed drug (refer to Drs. Ysern and Ramaswamy’s review for
details).
Basaglar is a sterile, clear, and colorless aqueous solution containing 600 nmol2 of insulin
glargine, 17 mg of glycerol3, 2.7 mg of metacresol4, 30.0 μg of zinc5 and water for injection.
Hydrochloric acid or sodium hydroxide is added to adjust the product pH to a final desired pH
of approximately 4.0. The strength of the proposed formulation is 100 Units/mL. Dr.
Ramaswamy notes that excipients used in Basaglar injection are similar to those used in
(b) (4)
Lantus except for the following minor differences.

The application contains analytical data comparing the identity, purity, potency and stability
of Basaglar to Lantus. Comparative studies included structural characterization to assess the
primary, secondary, tertiary and quaternary protein structure (e.g., intact and reduced mass,
peptide mapping, amino acid analysis, N-terminal sequencing, IEF, Far UV, tertiary near-UV,
2D-NMR and light scattering techniques), physicochemical characterization, biological
potency characterization (e.g., in vitro receptor binding and functional assays), impurity
profile characterization, and stability testing comparisons using multiple batches of Lantus
and Basaglar.
(b) (4)
Two minor differences between Basaglar and Lantus were (b)noted.
Basaglar contains
(4)
process-related impurity not
(b) (4)
found to be present in Lantus. A slightly higher content of
were observed in Basaglar compared to
Lantus in accelerated stability studies. Dr. Ramaswamy states that this may not translate into
significant difference during actual long-term storage condition. The Applicant suggests that

2

Drug product concentration of 600 nmol = 3.6378 mg of insulin glargine per mL of product

(b) (4)

Page 4 of 14

Reference ID: 3612365

 Division Director Review

the presence of

(b) (4)

in the Lantus formulation

(b) (4)

Basaglar does not contain
and the acceptance criteria set in the product specification for these impurities were
found to be acceptable.
(b) (4)

Dr. Ramaswamy concludes that Basaglar is similar to Lantus with respect to product
composition, strength, presentation, physicochemical, structural, biological properties, and
stability profile under long term storage.
I concur with the conclusions reached by the chemistry reviewers regarding the acceptability
of the manufacturing of the drug substance and drug product. Manufacturing site
inspections were acceptable. Stability testing supports a shelf-life of 24 months when the
(b) (4)
pre-filled pen is stored at
away from excessive light. In-use stability data support
an in-use period of 28 days at up to 30°C.

4. Nonclinical Pharmacology/Toxicology
Dr. Tsai-Turton notes that no biologically important differences between Basaglar and Lantus
were detected on receptor binding assays or receptor functional assays at either the insulin
or IGF-1 receptors (refer to Study No DBT93 and Study No DBT149). Data from these studies
demonstrate that both products contain potent, functional, insulin molecules and suggest in
vitro insulin receptor or IGF-1 receptor affinity and potency are similar between Basaglar and
Lantus products.
The applicant conducted a 4 week toxicity study using Basaglar and Lantus (Study No
8229488). This study did not identify major differences in pharmacokinetic, glucodynamics,
local tolerability, and toxicity profile between Basaglar and Lantus, and supports the scientific
appropriateness of reliance on FDA’s finding of safety for Lantus.
I concur with the conclusions reached by the pharmacology/toxicology reviewer that there
are no outstanding pharm/tox issues that preclude approval.

5.

Clinical Pharmacology/Biopharmaceutics

The clinical pharmacology study ABEO allows for an abbreviated clinical pharmacology
program and contributes to the scientific justification for reliance on FDA’s finding of safety
and effectiveness for Lantus to support marketing approval of Basaglar. This study was a
comparative PK/PD, single-dose (0.5 U/kg), 24-hour, glucose clamp study carried out in
healthy volunteers. To measure serum glargine levels following drug administration, the
applicant relied on a validated radioimmunoassay with cross-reactivity to endogenous insulin.
Slightly lower maximum and overall serum insulin exposure were observed after a single dose
of Basaglar compared to a single dose of Lantus. However the results of the study support a
conclusion that maximum (i.e.., Cmax) and overall exposure (AUC0-24h) to serum insulin were

Page 5 of 14

Reference ID: 3612365

 Division Director Review

sufficiently similar. The 90% confidence interval for the geometric mean ratios for Cmax and
AUC0-24h were within 0.8 and 1.25 between Basaglar and Lantus. Dr. Khurana also notes in his
review that time to achievement of maximum insulin concentration was similar between
Basaglar and Lantus. The study results also support a conclusion that maximum glucose
lowering after a single dose (Glucose Infusion Ratemax) and overall glucose lowering after a
single dose (Glucose Infusion Rate AUC0-24h) were similar between Basaglar and Lantus; the
90% confidence interval for the geometric mean ratios for the two PD parameters were
within 0.8 and 1.25. Refer to Table 4 in Dr. Khurana’s review.
Dr. Yanoff has summarized all clinical pharmacology studies submitted with the Basaglar new
drug application in Table 1 of her review. It is important to note that studies ABEE, ABEI and
ABEA in the Table were neither required to support Basaglar approval in the US nor used to
establish a scientific bridge to justify the appropriateness of reliance on FDA’s finding of
safety and effectiveness for Lantus. Studies ABEM (i.e., Basaglar versus EU-approved Lantus
single dose PK/PD dose response study) and ABEN (Lantus versus EU-approved Lantus single
dose PK/PD comparative study) provide supportive data and a PK/PD bridge between the EUapproved Lantus and the U.S.-approved listed drug (Lantus) to justify the relevance of this
supportive comparative data in the application.
I concur with the conclusions reached by the clinical pharmacology/biopharmaceutics review
team that there are no outstanding clinical pharmacology issues that preclude approval.

6. Clinical Microbiology
I concur with the conclusions reached by the clinical microbiology reviewer that there are no
outstanding clinical microbiology or sterility issues that preclude approval.

7. Clinical/Statistical-Efficacy
In this section, I summarize the pivotal efficacy findings for the Basaglar program. For full
details on efficacy analyses refer to Drs. Pian and Yanoff’s reviews.
A clinical study in type 1 diabetes (Study ABEB) and another in type 2 diabetes (study ABEC)
provide the scientific justification needed to allow reliance on FDA’s finding of effectiveness
for Lantus to support an indication of Basaglar for the treatment of type 1 diabetes mellitus
and the treatment of type-2 diabetes mellitus (refer to Background Section).
Study ABEB; Type 1 Diabetes Mellitus; Basal-Bolus Use
Study ABEB was a randomized (1:1) multinational, multicenter, active-controlled, open-label,
24-week trial in patients with type 1 diabetes mellitus (T1DM). Patients in this trial were
adult patients diagnosed with type-1 diabetes for at least one year, inadequately controlled

Page 6 of 14

Reference ID: 3612365

 Division Director Review

on an insulin regimen that included a ‘basal’ insulin and a ‘meal time’ insulin. Patients were
randomized to Basaglar or Comparator (Lantus or EU-approved Lantus depending on region).
The intervention period was divided into a titration phase where insulin doses were adjusted
to target HbA1c, fasting plasma glucose and post prandial glucose levels and a maintenance
phase where insulin doses were to be relatively stable.
Demographic and disease characteristics across the two intervention groups and across the
US and European regions were similar (refer to Tables 6 and 7 in Dr. Pian’s review). Patients
in the US region were on average older, more overweight per BMI criteria, and had had
diabetes for longer than patients randomized in the European region. More subjects
randomized to Basaglar in the US-region were male (~68%) compared to the overall (58%) or
European region (~55%) respectively.
The primary efficacy endpoint was the difference in the change in HbA1c from baseline
between intervention groups at Week 24. The study was powered to exclude the possibility
that glycemic control, captured using HbA1c change from baseline to Week 24, on Basaglar
was worse by 0.4%8 or more (non-inferiority margin) compared to the glycemic control
observed on comparator (Lantus + EU-Lantus). A non-inferiority trial design using a noninferiority margin of 0.4% was regarded as an appropriate design to compare the glucose
lowering efficacy of the two products and establish the efficacy of Basaglar in this setting.
The primary efficacy results from Study ABEB are shown below. The applicant demonstrates
that the glucose lowering efficacy of Basaglar co-administered with insulin Lispro at the end
of 24 weeks is similar to the glucose lowering efficacy of Lantus + EU-approved Lantus coadministered with insulin Lispro in patients with type 1 diabetes.
Table 1: Primary Efficacy Results Type 1 DM Trial-Study ABEB
Treatment Arm
n* Baseline
Adjusted Mean
HbA1c
Change From
[% (±SEM)]
Baseline HbA1c
[% (±SEM)]
Basaglar
Lantus and EU-approved
Lantus

267 7.9 (0.09)
267 7.9 (0.09)

-0.35 (0.05)
-0.46 (0.05)

Adjusted Between
Group† Difference in
HbA1c Change from
Baseline [% (95%
CI)]
+0.11 (-0.002, 0.22)
(p-value = 0.055)

Source: Table 9 in Dr. Pian’s Review.
*Subjects randomized with at least one post-baseline HbA1c value.
†Primary comparison on the full analysis set popula on with data up to me of discon nua on used and missing data imputed using LOCF. Estimates
are based on an analysis of covariance model (ANCOVA) with treatment, country, time of basal insulin injection (daytime, evening/bedtime) as fixed
effects and baseline HbA1c as a covariate.
8

A non-inferiority margin of 0.4% is routinely used to compare the relative efficacy of two insulin products in situations where a placebocontrol trial would be unethical. For ABEB the applicant calculated the sample size [refer to the applicant’s statistical analysis plan (SAP)] for
two non-inferiority margin’s (i.e., 0.4% and 0.3%). In the SAP, the number to be randomized was calculated based on the number of
completers (N=384) needed to exclude an NI margin of 0.4% with 90% power and the number of completers (N=432) needed to exclude an
NI margin of 0.3% with 90% power. Both assume no difference in effect, a .05 two-sided alpha and a 15% discontinuation rate. In the end,
the applicant randomized more patients than required for efficacy objectives (N=536) which provides additional exposure to assess for rarer
safety outcomes.

Page 7 of 14

Reference ID: 3612365

 Division Director Review

Results for subgroup analyses by regions separating US-/Puerto Rico sites (Lantus
comparator) from non- US sites (EU-approved Lantus comparator) are shown below.

Table 2: Primary Ef?cacy Results Type 1 DM Trial (Study ABEB) by Regional Subgroups
comparing US/Puerto Rico sites (Lantus used as comparator) versus sites (EU-

Lantus used as comparator).

 

 

 

 

 

 

 

 

 

 

 

Treatment Arm n* Baseline Adjusted Mean Adjusted Between Groupf
HbA1c Change From Difference in HbA1c
Baseline HbA1c Change from Baseline 
(95% Cl)]
Basaglar 98 7.8 (0.12) -0.22 (0.06) +0.19 (0.02, 0.36)
Lantus? 96 7.7 (0.12) -0.41 (0.06) (p-value 0.028)
Basaglar 169 7.9 (0.11) -0.46 (0.07) +0.07 (-0.08, 0.21)
EU-approved Lantus 171 7.9 (0.12) -0.53 (0.08) (p-value 0.345)
Source: Table 9 in Dr. Pian?s Review.
?Subjects randomized with at least one post-baseline HbA1c value.
Comparison are based on the full analysis set population with data up to time of discontinuation used and missing data imputed using LOCF.
Estimates are based on an analysis of covariance model (ANCOVA) with treatment, country, time of basal insulin injection (daytime,
evening/bedtime) as ?xed effects and baseline HbA1c as a covariate.

 

The results of these analyses are consistent with the overall analysis and based on these
analyses one would conclude that Basaglar is effective. These results also support a
conclusion that Basaglar is sufficiently similar to Lantus from a glucose lowering standpoint
in type 1 diabetes to justify reliance, in part, on FDA's finding of effectiveness for Lantus in
this disease (as reflected in product labeling that describes, among other things, efficacy
studies performed in type 1 diabetes). Minor differences in effect size across subgroups are
likely the result of chance, small sample size, or minor differences in trial conduct across the
two regions rather than due to differences in comparator response driven by product-related
differences between Lantus and EU-approved Lantus. This is supported by the fact that
differences across regions appear predominantly driven by differences in Basaglar response
between the two regions (same drug across the two regions) and not by large differences in
comparator response across the two regions. Potential differences in trial conduct across the
two regions is supported by the observation of a somewhat lower completion rate in the US
region (refer to Table 5 in Dr. Pian?s review) and the fact that patients randomized to Basaglar
in the US region, on average, had a comparatively larger decrease in their baseline short
acting co-administered insulin dose lispro insulin) immediately following randomization
compared to patients randomized to Basaglar in Europe or randomized to Lantus and EU-
approved Lantus (refer to figure 9 in Dr. Pian?s review). Finally, subgroup analyses across
regions in the type 2 diabetes trial (ABEC) did not confirm the presence of a difference
between Basaglar/Lantus and Basaglar/EU-approved Lantus comparisons and this trial, in
contrast to trial ABEB, is better suited to detect differences due to comparator products
because it does not have the issue of confounding by a co-administered short-acting insulin
product.

Page 8 of 14

Reference ID: 3612365

 

Division Director Review

Results for supportive secondary outcomes (i.e., Self-monitored blood glucose, HbA1c
responder rates, and insulin dose)9 were consistent with results based on HbA1c and support
a conclusion that Basaglar is effective in type 1 diabetes mellitus.
Study ABEC; Type 2 Diabetes Mellitus; Basal added to Oral Anti-Diabetic Use
Study ABEC was a randomized (1:1) multinational, multicenter, active-controlled, doubleblind, 24-week trial in patients with type 2 diabetes mellitus (T2DM). Patients in this trial
were adult patients diagnosed with type-2 diabetes, inadequately controlled on two or more
oral anti-diabetic agents on a stable dose for at least 12-weeks prior to screening. Patients
were randomized to Basaglar or Comparator (Lantus or EU-approved Lantus depending on
region). The intervention phase was divided into a titration phase where intervention insulin
doses were adjusted to target a fasting plasma glucose of 100 mg/dL and a maintenance
phase where insulin doses were to have remained relatively stable.
Demographic and disease characteristics across the two intervention groups and across the
US and European regions were similar (refer to Tables 18 and 19 in Dr. Pian’s review).
The primary objective and efficacy endpoint were similar to those of trial ABEB described (i.e,
excluding a non-inferiority margin of 0.3%-0.4%). Results for the primary analysis across the
entire trial population and subgrouped by regions are shown in Tables 3 and 4.
Table 3: Primary Efficacy Results Type 2 DM Trial-Study ABEC
Treatment Arm
n* Baseline
Adjusted Mean
HbA1c
Change From
[% (±SEM)]
Baseline HbA1c
[% (±SEM)]
Basaglar
Lantus and EU-approved
Lantus

369 8.3 (0.08)
375 8.3 (0.08)

-1.29 (0.06)
-1.34 (0.06)

Adjusted Between
Group† Difference in
HbA1c Change from
Baseline [% (95%
CI)]
+0.05 (-0.07, 0.17)
(p-value = 0.40)

Source: Table 21 in Dr. Pian’s Review.
*Subjects randomized with at least one post-baseline HbA1c value.
†Primary comparison on the full analysis set popula on with data up to me of discon nua on used and missing data imputed using LOCF. Estimates
are based on an analysis of covariance model (ANCOVA) with treatment, country, sulfonylurea use, time of basal insulin injection (daytime,
evening/bedtime) as fixed effects and baseline HbA1c as a covariate.

9

Refer to Tables 10 and 11 in Dr. Yanoff’s CDTL memorandum.

Page 9 of 14

Reference ID: 3612365

 Division Director Review

Table 4: Primary Efficacy Results Type 2 DM Trial (Study ABEC) by Regional Subgroups
comparing US/Puerto Rico sites (Lantus used as comparator) versus sites (EU-
approved Lantus used as comparator).

 

 

 

 

 

 

 

 

 

 

 

Treatment Arm n* Baseline Adjusted Mean Adjusted Between Groupf
HbA1c Change From Difference in HbA1c
Baseline HbA1c Change from Baseline 
(iSEMll (95% Cl)]
Basaglar 205 8.4 (0.10) -1.29 (0.06) +0.01 (-0.15, 0.18)
Lantus 213 8.3 (0.10) -1.30 (0.06) (p-va/ue 0.88)
Basaglar 164 8.3 (0.10) -1.25 (0.07) +0.11 (-0.07, 0.29)
EU-approved Lantus 162 8.3 (0.10) -1.36 (0.08) (p-value 0.23)
Source: Table 21 in Dr. Pian?s Review.
?Subjects randomized with at least one post-baseline HbA1c value.
Comparison are based on the full analysis set population with data up to time of discontinuation used and missing data imputed using LOCF.
Estimates are based on an analysis of covariance model (ANCOVA) with treatment, country, time of basal insulin injection (daytime,
evening/bedtime) as ?xed effects and baseline HbA1c as a covariate.

 

The results establish that Basaglar administered daily over 24 weeks is effective in patients
with type 2 diabetes inadequately controlled on two oral agents at baseline compared to
comparator (Lantus or EU-approved Lantus), Lantus alone, and EU-approved Lantus alone.
These results also support that Basaglar is sufficiently similar to Lantus from a glucose
lowering standpoint in type 2 diabetes to justify reliance, in part, on finding of
effectiveness for Lantus in this disease (as reflected in product labeling that describes, among
other things, efficacy studies performed in type 2 diabetes).

Results for supportive secondary outcomes self-monitored blood glucose, HbA1c
responder rates, and insulin doselo) were consistent with results based on HbA1c and further
support a conclusion that Basaglar is effective patients with type 2 diabetes mellitus.

8. Safety

The main safety dataset comprises all subjects who participated in trials ABEB and ABEC. In
these two trials, 644 subjects were exposed to Basaglar and the mean exposure time to
Basaglar was 34 weeks 7.8 months). Five hundred subjects were exposed to Basaglar
for six months or greater. Relatively few patients were exposed to Basaglar for 2 12 months
and 2 18 months. Although the diabetes guidance recommends that ?1300-1500 patients
and 300-500 patients be exposed to a novel insulin product for 2 12 months and 18 months
respectively, the size of the population exposed to Basaglar and exposure duration to
Basaglar was regarded as adequate to provide product specific safety and justify reliance, in
part, on finding of safety for Lantus as reflected in product labeling.

The applicant presents comparative safety data in three ways. The main safety analyses
compare safety data generated for all patients exposed to Basaglar to safety data generated

 

1? Refer to Tables 25, 26, 27 in Dr. Pian?s review.

Page 10 of 14

Reference ID: 3612365

 

Division Director Review

in all patients exposed to comparator (Lantus and EU-approved Lantus) across the two trials.
Supportive analyses compare specific safety data generated for the subgroup of patients
exposed to Basaglar in the US region to safety data generated in patients exposed to Lantus.
A third supportive analyses compares specific safety data generated for the subgroup of
patients exposed to Basaglar in the EU region to safety data generated in patients exposed to
EU-approved Lantus. Dr. Yanoff has reviewed safety data across the three safety groupings
and the three analyses groupings are sufficiently granular to establish Basaglar’s safety.
Deaths, Serious Adverse Events, Discontinuations due to Adverse Events and Common Adverse
Events
There were three deaths in the program, one on Basaglar (lung adenocarcinoma after 5
months of exposure) and two on Lantus or EU-approved Lantus (hypertrophic
cardiomyopathy and myocardial infarction). The three deaths were not judged to be product
related due to the presence of known pre-existing risk factors in the case histories and in the
case of the lung adenocarcinoma the relatively short exposure to product prior to the cancer
diagnosis.
No imbalance in non-fatal serious adverse events between Basaglar (5.4%) and Lantus or EUapproved Lantus (6.5%) were noted (refer to Table 15 in Dr. Yanoff’s review). No large
imbalance in any specific adverse event terms was noted across system organ class to suggest
a Basaglar related issue. Hypoglycemia (2.3%) coronary artery disease (~0.3%) and cellulitis
(~0.2%) accounted for the majority of events in both groups.
Discontinuations due to adverse events were infrequent (<3%) but more subjects randomized
to Lantus or EU-approved Lantus discontinued due to an adverse event (1.2% versus 2.6% for
Basaglar versus Lantus comparators). Review of individual adverse events preferred terms do
not raise concern with regard to a product specific issue related to Basaglar (Refer to Table 16
in Dr. Yanoff’s review).
The proportion of patients reporting at least one common adverse event was similar
between Basaglar and Lantus or EU-approved Lantus (refer to Table 20 in Dr. Yanoff’s
review). Imbalances in specific events were in general due to small differences between
groups (i.e., 2-7 additional cases responsible for any specific imbalance). Imbalances for
adverse event terms denoting similar issues were on a whole balanced. For example
abnormal weight gain events were reported more frequently in Basaglar treated patients (10
cases versus 3) but the terms adverse events of weight increased and oedema peripheral
were reported more frequently in the Lantus or EU-approved Lantus groups. Analyses based
on directly measured weight did not reveal a difference between Basaglar and Lantus or EUapproved Lantus. No imbalance in the proportion of reported events for any one specific
adverse event term raises a concern that would suggest an issue related to Basaglar.
Hypoglycemia

Page 11 of 14

Reference ID: 3612365

 Division Director Review

Drs. Yanoff and Pian have reviewed comparative data for severe hypoglycemia across the
main pool and across the two regional subgroup pools for the Type 1 diabetes trial (ABEB)
and the Type 2 Diabetes (ABEC) trial individually. In the type 1 diabetes trial, no differences
in the rate of severe hypoglycemia was observed between Basaglar and Lantus and EUapproved Lantus, Basaglar and Lantus in the US-region, and Basaglar and EU-approved Lantus
in the EU-region (Refer to Table 28 in Dr. Pian’s review). In the type 2 diabetes trial, no
differences in the rate of severe hypoglycemia was observed between Basaglar and Lantus
and EU-approved Lantus, Basaglar and Lantus in the US-region, and Basaglar and EUapproved Lantus in the EU-region (Refer to Table 29 in Dr. Pian’s review).
Analyses based on severe hypoglycemia are clinically important and informative because
these events compare the occurrence of specific life-threatening events between
interventions. These analyses are limited however by the infrequent occurrence of these
events in clinical trials. To provide supportive data for analyses based on severe events,
analyses based on less specific but more frequent hypoglycemic events were also carried out.
These analyses were based on various definitions of ‘hypoglycemia’ (e.g., defined for example
by the presence of a self-measured blood glucose < 54 mg/dL and presence of symptoms
compatible with hypoglycemia) and the results of these analyses were consistent with
analyses based on severe hypoglycemic events (refer to Table 17 in Dr. Yanoff’s review) and
did not suggest a difference in the risk of hypoglycemia between Basaglar and Lantus.
Immunogenicity
Immunogenicity data were reviewed by Drs. Sheikh (Office of Biotechnology Products) and
Yanoff. No significant differences in immune response between Basaglar and Lantus or EUapproved Lantus in either the type 1 diabetes or type 2 diabetes patient populations were
noted with respect to anti-body development at any visit. The applicant reviewed the impact
of developing ‘treatment emergent antibodies’ on specific efficacy (HbA1c response, insulin
dose) and safety outcomes (hypoglycemia) across the entire safety dataset and in the
regional Lantus subgroup datasets (refer to Summary of Clinical Safety pages 71-80). Overall,
development of treatment emergent antibodies in either the Basaglar, Lantus or EUapproved Lantus groups did not impact these efficacy or safety outcomes (Source: figure
2.7.4.2 in the summary of clinical safety).
Dr. Yanoff reviewed data for outliers (defined as patients with a percent binding antibodies of
>20%). Three patients met this definition in the Basaglar group and two in the Lantus or EUapproved Lantus group. The presence of > 20% binding antibodies did not appear to
correlate to HbA1c response, insulin dose or risk of hypoglycemia.
These data establish the immunogenicity profile of Basaglar and support a conclusion that
Basaglar has an acceptable immunogenicity profile.
Allergic and Injection Site Reactions

Page 12 of 14

Reference ID: 3612365

 Division Director Review

Allergic reactions were captured using a list of preferred terms (Refer to Table APP 2.7.4.7.13
in the summary of clinical safety appendix). No differences in the occurrence of allergic
reactions were noted (41 cases in Basaglar versus 38 cases for Lantus + EU-approved Lantus).
The most frequently reported terms across the entire safety pool were arthralgia (1.6%
versus 2.0% for Basaglar versus Lantus + EU-approved Lantus respectively), pruritus (1.1%
versus 0.8% for Basaglar versus Lantus + EU-approved Lantus respectively), injection site
reaction (0.9% versus 0.8% for Basaglar versus Lantus + EU-approved Lantus respectively) and
rash (0.8% and 0.8% for Basaglar versus Lantus + EU-approved Lantus respectively).
Imbalances between groups were due to small numerical differences. These data do not
reveal large differences between Basaglar and Lantus with regard to occurrence of allergic
reactions and establish that Basaglar has an acceptable safety profile with regard to
occurrence of allergic reactions.
Adverse events related to the administration site (e.g., injection site reactions, induration,
pruritus, nodule, local swelling) were more frequently reported in the Basaglar group (n=11
or 1.7%) compared to the Lantus + EU-approved Lantus group (n=6 or 0.9%). Differences in
specific terms are shown in Table 18 in Dr. Yanoff’s review. Overall imbalances were driven
by numerically few cases. Small differences in reported incident injection site reactions, did
not lead to more frequent discontinuation or translate to clinically important differences in
efficacy. These data establish that Basaglar has an acceptable safety profile with regard to
occurrence of administration site reactions.
Vital Signs, Laboratory and ECG Abnormalities:
Dr. Yanoff has reviewed these outcomes and notes that no clinically significant differences
across these safety parameters between Basaglar and Lantus in the safety pool were
observed (refer to Section 7.4.2-7.4.4 in her review).
The results of these safety analyses establish the safety Basaglar administered daily over ≥ 24
weeks in patients with type 1 and type 2 diabetes mellitus and provide scientific justification
for reliance, in part, on FDA’s finding of safety for Lantus (as reflected in product labeling that
describes, among other things safety outcomes from clinical studies performed in type 1 and
type 2 diabetes).

9. Advisory Committee Meeting
No new efficacy or safety issue rose to the level of requiring the input from an advisory panel.
Therefore no advisory committee was convened.

10.

Page 13 of 14

Reference ID: 3612365

Labeling

 Division Director Review

As noted above, this 505(b)(2) application for Basaglar relies, in part, on FDA’s finding of
safety and effectiveness for Lantus (insulin glargine [rDNA origin] injection). We are
tentatively approving Basaglar with the established name “insulin glargine injection,”
consistent with current nomenclature practices for products approved under the FD&C Act.
The nonproprietary name of Basaglar and Lantus reflects certain scientific characteristics of
these products. A deviation from current nonproprietary naming practices for products
approved under the FD&C Act is not warranted for Basaglar at this time. We note, however,
that nomenclature practices for biological products continue to be under review within FDA,
and we will consider this issue again at such time as Lilly requests final approval of Basaglar.
Refer to Dr. Yanoff’s review for additional comments regarding labeling.

11.


Decision/Action/Risk Benefit Assessment

Regulatory Action

I recommend a tentative approval.


Risk Benefit Assessment

In this application Eli Lilly, established the efficacy and safety of Basaglar in adults with type 1
and type 2 diabetes in two clinical trials (i.e., ABEB, ABEC). Use of Basaglar was shown to be
effective at improving glycemic control in adults with type 1 and type 2 diabetes. Basaglar
specific risks were consistent with known risks for the product class (i.e., insulin) and no novel
risks were identified in safety analyses across the two disease populations. The impact of
potential product related differences between Basaglar and Lantus on safety and efficacy was
evaluated using a toxicology bridging study, a clinical pharmacology bridging study (ABEO)
and the two pivotal clinical studies (ABEB, ABEC). These studies were used to support a
determination of ‘sufficient similarity’ between Basaglar and Lantus to justify reliance, in
part, on FDA’s finding of safety and effectiveness for Lantus (as reflected in specifics of
product labeling that describe, non-clinical studies, clinical pharmacology studies and clinical
efficacy studies and clinical safety studies).


Recommendation for Postmarketing Risk Evaluation and Mitigation Strategies

No new safety findings from this clinical development program prompt the need for a
postmarketing risk evaluation and management strategies.


Recommendation for other Postmarketing Requirements and Commitments

No new safety findings from this clinical development program prompt the need for a
postmarketing requirements and commitments.

Page 14 of 14

Reference ID: 3612365

 --------------------------------------------------------------------------------------------------------This is a representation of an electronic record that was signed
electronically and this page is the manifestation of the electronic
signature.
--------------------------------------------------------------------------------------------------------/s/
---------------------------------------------------JEAN-MARC P GUETTIER
08/18/2014

Reference ID: 3612365