Summary Basis for Regulatory Action

Date: October 7, 2016

From: Timothy A. Fritz, PhD, Review Committee Chair

BLA/ STN#: 125285/194

Applicant Name: Protein Sciences Corporation

Date of Submission: December 8, 2015

PDUFA Goal Date: October 7, 2016

Proprietary Name/ Established Name: Flublok Quadrivalent, Influenza Vaccine

Indication: Prevention of influenza disease in persons 18 years of age and older caused by
influenza virus subtypes A and types B contained in the vaccine.

Recommended Action: Approval

Signatory Authorities Action: Approval

Offices Signatory Authority: Wellington Sun, MD, Director, Division of Vaccines and Related
Products Applications, Office of Vaccines Research and Review
□ I concur with the summary review.
□ I concur with the summary review and include a separate review to add further analysis.
□ I do not concur with the summary review and include a separate review.

Specific documentation used Reviewer – Date of Review

in developing the SBRA
Clinical Review Cynthia Nolletti, MD – 04 October 2016
Statistical Review Rong Fu, PhD – 01 September 2016
Bioassay Statistical Review Rong Fu, PhD – 09 September 2016
Pharmacovigilance Review Emily Jane Woo, MD, MPH – 02 June 2016
CMC/Product Review Maryna Eichelberger, PhD – 04 October 2016
CMC/Facilities and Equipment Lori Peters, MS – 03 October 2016
Labeling Reviews Cynthia Nolletti, MD – 04 October 2016
Emily Jane Woo, MD, MPH – 02 June 2016
Sonny Saini, PharmD – 23 August 2016
Daphne Stewart – 15 September 2016
Testing Methods and Analytical Manju Joshi, PhD – 23 August 2016
Chemistry Reviews Alfred Del Grosso, PhD – 07 September 2016
Hyesuk Kong, PhD – 15 June 2016, 12 August 2016
Bioresearch Monitoring Review Colonious King – 07 September 2016

1
1. Introduction
Flublok® Quadrivalent is a vaccine manufactured by Protein Sciences Corporation (PSC) for the
active immunization of persons 18 years and older for the prevention of influenza disease caused
by influenza virus subtypes A and types B contained in the vaccine. Compared to Flublok®,
Flublok Quadrivalent contains an additional rHA antigen from a type B influenza virus. The
presence of the two B-type virus antigens in quadrivalent influenza vaccines is intended to
provide protection against both B/Yamagata and B/Victoria lineage influenza viruses. The
vaccine is formulated as a sterile, aqueous, buffered solution of four purified, recombinant
influenza hemagglutinins (rHAs) and contains no egg proteins. The four rHAs are produced in
Spodoptera frugiperda insect cells using a Baculovirus Expression Vector System (BEVS) in
which the cells are infected with a baculovirus engineered to contain the gene for the
corresponding influenza HA antigen. Each 0.5 mL dose of Flublok Quadrivalent contains 180
mcg of rHA antigens (45 mcg each of H1, H3, B/Victoria and B/Yamagata rHAs) and may
contain residual amounts (≤ 19 mcg) of baculovirus and insect cell proteins.

CBER received supplement STN 125285/194 on 08 December 2015 with efficacy,

immunogenicity and safety data from a required postmarketing study (PSC12) to verify clinical
benefit, a condition of accelerated approval specified in the 29 October 2014 approval letter
granting licensure for the use of the Flublok trivalent formulation in persons aged 50 years and
older (STN 125285/78). Data from PSC12 were also used to support licensure of Flublok
Quadrivalent in this population. Immunogenicity and safety data from a second study, PSC16,
were included to support the use of Flublok Quadrivalent in persons 18-49 years of age. Data
were also provided to support licensure of a new contract facility (b) (4)
used to formulate and fill Flublok Quadrivalent into single-dose syringes.

2. Background
Flublok (trivalent formulation) was licensed in the United States on 16 January 2013 (STN
125285/0) for persons 18-49 years of age. Accelerated approval was granted on 29 October
2014 (STN 125285/78) to extend the use of Flublok to persons 50 years of age and older and was
based on safety and non-inferior immunogenicity results from three clinical studies. At CBER’s
request, PSC agreed to conduct a clinical endpoint study in persons 50 years and older to support
traditional approval due to more limited data relating hemagglutination inhibition (HI) antibody
titers from non-egg based vaccines with protective immunity. To fulfill the accelerated approval
requirement to confirm clinical benefit, it was agreed that PSC could use a quadrivalent
formulation of Flublok (Flublok Quadrivalent) in the study in persons 50 years of age and older
(PSC12). PSC12 was a Phase 3, relative vaccine efficacy, immunogenicity and safety clinical
trial. PSC also conducted a Phase 3, non-inferior immunogenicity and safety study (PSC16) to
support the licensure of Flublok Quadrivalent in persons 18-49 years of age.

3. Chemistry Manufacturing and Controls (CMC)

a) Product Quality

2
The four Flublok Quadrivalent rHA monovalent bulk drug substances used to produce
Flublok Quadrivalent Drug Product are manufactured using the same process licensed for the
manufacture of Flublok (STN 125285/0).

Drug Substance
PSC’s Meriden, CT and Pearl River, NY facilities are licensed to manufacture Flublok
monovalent bulk Drug Substance (DS) under STN 125285/0 and 125285/127. Working
cell banks of expresSF+ Spodoptera frugiperda insect cells are grown in serum-free
medium and expanded in (b) (4) culture. To express the rHA, the (b) (4) culture is
infected with a baculovirus engineered to contain the gene for the viral hemagglutinin
protein. Following expression, the cells are harvested by (b) (4) and solubilized
with a detergent solution to extract the rHA. (b) (4)

chromatography columns to purify the rHA. (b) (4)

and the purified rHA is (b) (4) to produce monovalent
bulk rHA DS. Monovalent bulk DS release testing is performed at the Meriden, CT
facility and at (b) (4) .

Drug Product
Monovalent bulk DS is shipped from Pearl River, NY or Meriden, CT to (b) (4)
to formulate and fill Flublok Quadrivalent Drug Product (DP)
into single-dose syringes. The DS used in Flublok Quadrivalent manufacture validation
studies were rHAs from viruses recommended for 2015/16 quadrivalent influenza
vaccines: A/California/7/2009 (H1N1), A/Switzerland/9715293/2013 (H3N2),
B/Phuket/3073/2013 (a B/Yamagata-lineage virus) and B/Brisbane/60/2008 (a
B/Victoria-lineage virus). Formulation and filling data from 3 validation lots ((b) (4)
) demonstrated that the viruses are adequately mixed within (b) (4) and the
formulated bulk can be held for up to (b) (4) . A (b) (4)
. The quadrivalent formulated bulk is (b) (4)
and
then filled into single dose glass syringes in the filling area. Data were provided to
demonstrate the product is consistently filled into the syringes.

Drug Product is shipped from (b) (4) to Meriden, CT where most release testing is
performed. Sterility release testing is contracted to (b) (4)
. Minor changes were made to some DP release specifications to account
for the addition of the second B-type influenza antigen. Stability data provided in the
supplement support a 6 month shelf-life for Flublok Quadrivalent when stored at 2-8°C.
The date of manufacture is designated as the date the formulated bulk is sterile filtered
and filled into the final container.

The presence of two B-type influenza antigens in quadrivalent influenza vaccines

required a modification to the Single Radial Immunodiffusion (SRID) assay to ensure
proper measurement of the potency of the individual B-type antigens. CBER’s review
concluded that PSC appropriately revised and validated their SRID assay.

3
 Three Flublok Quadrivalent DP process validation lots ((b) (4) ) were
submitted for CBER testing in support of the supplement. CBER’s potency test results
for the B/Brisbane/60/2008 antigen of lots (b) (4) and (b) (4) were more than 20% lower than
PSC’s results prompting CBER to request that PSC conduct a root cause investigation.
The discrepancy was determined to have several causes, including differences in the
operating parameters of the SRID assays between the CBER and PSC laboratories, and
the loss in DP potency that occurred between testing by PSC and CBER. To prevent the
recurrence of discrepant results, PSC agreed to align their assay with CBER’s assay and
potency testing will be coordinated to ensure that CBER testing occurs within 1 week of
PSC testing. CBER confirmed that implementing these measures will reduce SRID assay
result differences between CBER and PSC.

b) CBER Lot Release

A review of Product Release Branch records indicate that there are no pending lots or issues
that would affect approval of the submission. Revisions to the Lot Release Protocol template
submitted by PSC were found to be acceptable.

c) Facilities Review/Inspection
Facility information and data provided in the BLA were reviewed by CBER and found to be
sufficient and acceptable. The facilities involved in the manufacture of Flublok Quadrivalent
are listed in the table below. The activities performed and inspectional histories are noted in
the table and are further described in the paragraphs that follow.

Manufacturing Facilities Table for Flublok Quadrivalent

FEI Inspection/ Results/
Name/address
number waiver Justification
Drug substance
manufacturing, release and
stability testing of drug
product, lot release of drug
product

Protein Sciences Corporation 1221506 Waived Team Biologics

1000 Research Parkway June 2016
Meriden, CT 06450 VAI
Drug substance
manufacturing

Protein Sciences Corporation 3011286996 Waived CBER/DMPQ

401 North Middletown Road March 2015
Pearl River, NY 10965 VAI

4
 FEI Inspection/ Results/
Name/address
number waiver Justification
Drug product manufacturing
including formulation, syringe
fill/finish, in-process testing,
labeling, and packaging

(b) (4) (b) (4) PAI (b) (4)

VAI

General safety and sterility

release testing for drug
product

(b) (4) (b) (4) Waived (b) (4)

NAI

(b) (4) (b) (4) Waived (b) (4)

NAI
VAI – Voluntary Action Indicated, NAI – No Action Indicated

Team Biologics performed a surveillance inspection from June 8 – 15, 2016 of Protein
Sciences Corporation’s drug substance manufacturing facility in Meriden, CT. The
corrective actions to the observations were deemed satisfactory by Team Biologics and the
inspection was classified as voluntary action indicated (VAI).

CBER/DMPQ performed a pre-approval inspection from March 2 – 6, 2015 of Protein

Sciences Corporation’s drug substance manufacturing facility in Pearl River, NY under Prior
Approval Supplement STN 125285/127. All 483 issues were resolved and the inspection
was classified as VAI.

CBER/DMPQ performed a pre-approval inspection from (b) (4) of

(b) (4) in (b) (4) , the contract manufacturer responsible for the drug
product manufacturing activities, including filling of Flublok Quadrivalent drug product into
syringes. The inspection was classified as VAI and all inspectional observations have been
satisfactorily resolved.

ORA performed surveillance inspections of (b) (4) drug product testing

locations in (b) (4) and (b) (4) in (b) (4) and (b) (4) , respectively.
The inspections were both classified as no action indicated (NAI).

5
 d) Environmental Assessment
The BLA included a request for categorical exclusion from an environmental assessment
under 21 CFR 25.31 (c). The FDA concluded that this request is justified as the
manufacturing of Flublok Quadrivalent will not alter significantly the concentration and
distribution of naturally occurring substances and no extraordinary circumstances exist that
would require an environmental assessment.

e) Container Closure System

The drug product is filled into a 1mL (b) (4) glass syringe supplied by (b) (4)
which is configured with a Luer-lock for needle attachment and has a
(b) (4) . The syringes are provided needleless and
the needle size is selected by the healthcare provider administering the vaccine. The tip caps
are an (b) (4) and are supplied by (b) (4)
. The bromobutyl stopper is supplied by (b) (4) . The plunger
(b) (4) (b) (4)
rods are manufactured by and are composed of (b) (4) . is responsible
for (b) (4) , and sterilizing the
components of the container closure system.

Container closure integrity testing of the syringes was performed by (b) (4)
using the (b) (4) test method; all acceptance criteria were met.

4. Nonclinical Pharmacology/Toxicology
A developmental toxicity study was conducted for Flublok (trivalent formulation) and reviewed
under STN 125285/0. Nonclinical pharmacology/toxicology data were not provided in this
supplement because CBER advised PSC under IND 15784 that this data would not be needed
due to the similarity of Flublok Quadrivalent to Flublok.

5. Clinical Pharmacology
No new clinical pharmacology data was required in support of this supplement.

6. Clinical/ Statistical
a) Clinical and Statistical Summary of Efficacy and Immunogenicity Results

The safety, immunogenicity and efficacy of Flublok Quadrivalent were evaluated in persons
≥ 18 years old in two clinical studies, PSC12 and PSC16. A U.S.-licensed, quadrivalent
inactivated influenza vaccine, Fluarix® Quadrivalent (IIV4, GlaxoSmithKline) was used as
the active comparator in each study.

Study PSC12

Study PSC12 was a Phase 3, observer-blinded, randomized, active controlled, relative

efficacy trial conducted at 40 U.S. sites during the 2014-2015 influenza season. The

6
predominant influenza virus in circulation during this season in the U.S. was an A/H3N2
strain that was poorly matched to the A/H3N2 antigen present in the influenza vaccine
formulations. The study was designed to evaluate the relative efficacy, immunogenicity and
safety of Flublok Quadrivalent as compared to IIV4 in ambulatory, medically stable adults 50
years and older. A total of 8963 subjects were enrolled and randomized 1:1 to receive a
single dose of Flublok Quadrivalent (180 mcg) or IIV4 (60 mcg) administered
intramuscularly on study Day 0. Females and blacks/African Americans were somewhat
overrepresented while Asians and Hispanics/Latinos were underrepresented relative to the
U.S. population.

The primary endpoint of PSC12 was protocol-defined, reverse transcriptase polymerase

chain reaction (rtPCR)-confirmed, influenza-like illness (ILI) beginning at least 14 days post-
vaccination and caused by any influenza strain. To demonstrate non-inferior (NI) relative
vaccine efficacy (rVE) of Flublok Quadrivalent as compared to IIV4, the success criteria for
the primary endpoint was defined as:

• The lower bound of the two-sided 95% confidence interval (CI) of the rVE > -20%,
where rVE = 1- Relative Risk (RR) = 1 - (Attack Rate FlublokQuadrivalent / Attack Rate IIV4).

The population (efficacy population) used to analyze the rVE primary objective for study
PSC12 comprised 8604 subjects and was defined as all randomized subjects who received
study vaccine and provided any follow-up documentation for ILI beginning at least 14 days
after vaccination. Subjects with significant protocol deviations that could adversely impact
efficacy were excluded from the efficacy population. The data in Table 1 show that the
success criterion for the primary endpoint was met.

Table 1: Relative Vaccine Efficacy of rtPCR-Confirmed ILI Due to All Influenza Virus
Strains – PSC12 (Efficacy Population)
Flublok Flublok IIV4 IIV4 Relative Risk rVE (95% CI)
Quadrivalent Quadrivalent (RR)
N=4303 N=4303 N=4301 N=4301
N (# of cases) Attack Rate N (# of Attack ARFlublok Quadrivalent (1 - RR) x 100
(AR) cases) Rate /ARIIV4
96 2.2 138 3.2 0.70 30% (10%, 47%)
Source: STN 125285/194.9, Module 5, PSC12 CSR, Table 14.2.1.1 (03Mar2016)
Attack Rate (AR) = # of cases of ILI / # subjects in the treatment group
Relative Risk (RR) = ARFlublok Quadrivalent / ARIIV4
Relative Vaccine Efficacy (rVE) of Flublok Quadrivalent versus IIV4 = (1 – RR) x 100

The PSC12 protocol stated that rVE would be tested for superiority as an exploratory
analysis if non-inferior rVE was demonstrated. Superior VE of Flublok Quadrivalent relative
to IIV4 was pre-specified as a LB of the two-sided 95% CI of rVE > 9%. Though the
success criterion of superiority was met (Table 1), the analysis of superiority was not pre-
specified in the study’s Statistical Analysis Plan and PSC was advised that a claim of
superiority would not be included in the Flublok Quadrivalent package insert.

Post hoc, exploratory analyses of rVE according to influenza type A or B were performed by
PSC but were not powered for statistical significance. The results show a trend towards non-

7
inferior rVE for Flublok Quadrivalent against influenza A but not for influenza B where the
number of cases were fewer and 95% CIs wider: the rVE for influenza A (all A/H3N2) was
36% (95% CI: 14, 53) and for influenza B 4% (95% CI: -72, 46).

Immunogenicity data were collected from 614 subjects (314 Flublok Quadrivalent and 300
IIV4 recipients) at 5 pre-selected study sites in PSC12. Flublok Quadrivalent
immunogenicity as compared to IIV4 immunogenicity was evaluated as a secondary study
endpoint. Serum was collected from subjects prior to vaccination (study Day 0) and 28 days
following vaccination (study Day 28) to measure antibody seroconversion rates (SCRs) and
geometric mean titers (GMTs). Seroconversion was defined as either a pre-vaccination
hemagglutination inhibition (HI) titer of < 1:10 and a post-vaccination HI titer of ≥ 1:40, or a
pre-vaccination HI titer of ≥ 1:10 and a minimum 4-fold rise in post-vaccination HI titer at
Day 28.

The immunogenicity of the four Flublok Quadrivalent antigens was compared to the
corresponding IIV4 antigens using Day 28 post-vaccination HI GMT ratios and SCR
differences for a total of 8 secondary immunogenicity endpoints. The pre-specified success
criteria for establishing the non-inferior immunogenicity of Flublok Quadrivalent as
compared to IIV4 were as follows for all four vaccine antigens:

1. The upper bound of the 2-sided 95% confidence interval (CI) for the GMT ratio
(GMTIIV4 / GMTFlublok Quadrivalent) ≤ 1.5, AND
2. The upper bound of the 2-sided 95% confidence interval for the SCR difference
(SCRIIV4 – SCRFlublok Quadrivalent) ≤ 10%.

The study results showed that Flublok Quadrivalent met the GMT ratio success criterion for
the A/H1N1, A/H3N2 and B/Yamagata antigens but failed to meet the GMT ratio success
criterion for the B/Victoria antigen. The SCR success criterion was met for the A/H3N2 and
B/Yamagata antigens but was not met for the A/H1N1 and B/Victoria antigens.

Study PSC16

Study PSC16 was a Phase 3, observer-blind, randomized, comparator-controlled trial

conducted at ten U.S. sites during the 2014-2015 influenza season. The study was designed
to evaluate the safety, reactogenicity, and immunogenicity of Flublok Quadrivalent compared
to IIV4. The study was conducted in healthy, medically stable adults 18 to 49 years of age.
A total of 1350 subjects were enrolled and randomized 3:1 to receive a single, 0.5 mL dose
of Flublok Quadrivalent (180 mcg) or IIV4 (60 mcg) administered intramuscularly. The
distribution of subject characteristics was generally balanced but males and Asians were
underrepresented while females and Blacks/African Americans were overrepresented relative
to the total U.S. population.

Serum was collected from subjects prior to vaccination (study Day 0) and 28 days following
vaccination (study Day 28) to measure antibody SCRs and GMTs. Seroconversion was
defined as for study PSC12.

8
The immunogenicity of the four Flublok Quadrivalent antigens were compared to the
corresponding antigens in IIV4 using the CBER-defined criteria of Day 28 post-vaccination
HI GMT ratios and SCR differences for a total of 8 co-primary endpoints. The pre-specified
success criteria for establishing the non-inferior immunogenicity of Flublok Quadrivalent as
compared to IIV4 were as follows for all four vaccine antigens:

1. The upper bound of the 2-sided 95% confidence interval (CI) for the GMT ratio
(GMTIIV4 / GMTFlublok Quadrivalent) ≤ 1.5, AND
2. The upper bound of the 2-sided 95% confidence interval for the SCR difference
(SCRIIV4 – SCRFlublok Quadrivalent) ≤ 10%.

The population (immunogenicity population) used to analyze the non-inferior

immunogenicity primary objective for study PSC16 was defined as all randomized subjects
who received a dose of study vaccine, provided serum samples for Day 0 and Day 28 HI
antibody titers within specified windows, and subjects had no major protocol deviations that
might adversely impact the immune response. The immunogenicity population comprised
1292 subjects of whom 969 received Flublok Quadrivalent and 323 received IIV4 (3:1
randomization). The results for SCRs, SCR differences, GMTs, GMT ratios and 95% CIs are
shown in Tables 2 and 3.

Table 2: Baseline and Day 28 Post-Vaccination HI GMTs and GMT Ratios for Flublok
Quadrivalent Relative to IIV4 in Adults 18 through 49 Years of Age – PSC16
(Immunogenicity Population)
Strain Day Flublok Quadrivalent IIV4 GMT Met
GMT GMT Ratio GMT
(95% CI) (95% CI) (95% CI) Criteria?*
N=969 N=323
A/H1N1 0 59 53 -- --
(54,65) (45,63)
A/H1N1 28 493 397 0.81 Yes
(460,527) (358,441) (0.71,0.92)
A/H3N2 0 74 70 -- --
(68,82) (60,81)
A/H3N2 28 748 377 0.50 Yes
(700,800) (341,417) (0.44,0.57)
B/Yamagata 0 26 24 -- --
(24,29) (21,28)
B/Yamagata 28 156 134 0.86 Yes
(145,168) (119,151) (0.74,0.99)
B/Victoria 0 12 11 -- --
(11,13) (10,12)
B/Victoria 28 43 64 1.49 No
(40,46) (57,71) (1.29,1.71)
Source: STN 125285/194.9, Module 5, PSC16 CSR, Table 14.2.1.1.1 (07Mar2016).
Abbreviations: HI=hemagglutinin inhibition; IIV4=Fluarix Quadrivalent; GMT=geometric mean titer.
*Success criteria for the GMT ratio (GMTIIV4 / GMTFlublok Quadrivalent): UB of the 95% CI must be ≤ 1.5.

9
Table 3: Day 28 Post-vaccination HI SCRs and SCR differences between Flublok
Quadrivalent and IIV4 in Adults 18 through 49 Years of Age – PSC16 (Immunogenicity
Population)
Strain Flublok Quadrivalent IIV4 SCR Met SCR
SCR SCR Difference Success
N=969 N=323 % (95% CI) Criteria?*
% (95% CI) % (95% CI)
A/H1N1 66.7 63.5 -3.2 Yes
(63.6,69.6) (58.0,68.7) (-9.2,2.8)
A/H3N2 72.1 57.0 -15.2 Yes
(69.2,74.9) (51.4,62.4) (-21.3,-9.1)
B/Yamagata 59.6 60.4 0.7 Yes
(56.5,62.8) (54.8,65.7) (-5.4,6.9)
B/Victoria 40.6 58.2 17.6 No
(37.4,43.7) (52.6,63.6) (11.4,23.9)
Source: STN 125285/194.9, Module 5, PSC16 CSR, Tables 14.2.1.2 (07Mar2016).
Abbreviations: HI=hemagglutinin inhibition; IIV4=Fluarix Quadrivalent; SCR=seroconversion rate.
*Success criteria for the SCR difference (SCRIIV4 - SCRFlublok Quadrivalent): UB of the 95% CI must be ≤ 10%.

The data in Tables 2 and 3 show that the antibody response to Flublok Quadrivalent met the
non-inferior success criteria for the GMT ratio and SCR difference co-primary endpoints for
the A/H1N1, AH3/N2 and B/Yamagata vaccine antigens but failed to meet the success
criteria for the B/Victoria antigen. Lower antibody responses to B-type as compared to A-
type viruses are not uncommon and the low baseline antibody titers measured for the B
antigens (Table 2, Day 0) suggest that the study population was immunologically naïve to the
B-type (particularly B/Victoria) viruses which may have contributed to the low Day 28 titers.
The use of whole, inactivated B-type viruses in place of the more typically used “split” B-
type viruses in the HI assay used for PSC16 also likely contributed to the lower GMTs
observed for the B versus the A antigens. The statistically significantly lower B/Victoria
GMTs and SCRs observed for Flublok Quadrivalent recipients as compared to IIV4
recipients may also be due, in part, to antigenic differences between the B/Victoria rHA
present in Flublok Quadrivalent and the egg-based B/Victoria antigen used in the HI assay.
The sequence of the B/Brisbane rHA component in Flublok used in PSC12 and PSC16
differed from the egg-grown rHA at a glycosylation site, and therefore antigenic differences
between rHA and egg-grown antigen used in the HI assay may result in lower antibody titers
measured in sera from Flublok recipients as compared to recipients of the egg-grown, split
vaccine IIV4 comparator. This is also suggested by ferret studies.

b) Pediatrics

Under STN 125285/0, PSC requested and was granted a waiver from studies of Flublok in
children less than 3 years of age because data provided from a randomized, controlled study
(PSC02) strongly suggested that Flublok would not be effective in children younger than 3
years of age. In this supplement (STN 125285/194) PSC also requested a waiver of studies
of Flublok Quadrivalent in children less than 3 years of age. The waiver request was

10
 presented to the FDA Pediatric Review Committee. Waiver of Flublok Quadrivalent studies
in children less than 3 years was granted due to the similarity of composition and
manufacturing process between Flublok and Flublok Quadrivalent.

Two postmarketing required pediatric studies were established in the January 16, 2013
approval letter for the original Flublok license application to fulfill the requirements of the
Pediatric Research Equity Act (PREA). One study was in children 3-5 years old and the
second study was in children 6-17 years old. In the Pediatric Study Plan included in this
supplement, PSC proposed to conduct these studies using Flublok Quadrivalent. CBER
agreed with PSC’s plan which was presented to the FDA Pediatric Review Committee who
also agreed that the proposal was acceptable. Subsequently, PSC proposed to combine the
two PREA studies into a single, relative efficacy study (PSC17) in children 3-17 years of
age. CBER agreed that the combined study could be used to fulfill the PREA requirements.

c) Bioresearch Monitoring

CBER Bioresearch Monitoring inspected two clinical study sites. Each site enrolled subjects
for both study PSC12 and study PSC16. The inspections did not reveal significant problems
that would impact the data submitted in the supplement.

d) Other Special Populations

Flublok has not been studied in pregnant/lactating women or immunocompromised

individuals. During negotiations preceding the January 16, 2013 approval of the original
Flublok license application, PSC agreed to establish a prospective pregnancy registry to
monitor pregnant women immunized with Flublok. This commitment remains to be
completed and PSC plans to revise the protocol for this study (PSC15) to include Flublok
Quadrivalent recipients.

7. Safety
The safety population for studies PSC12 and PSC16 comprised 10,002 subjects and was defined
as all subjects who received a dose of study vaccine and for whom any safety data (PSC16) or
any evaluable safety data (PSC12) were available after vaccination. There were 1330 subjects
(of whom 998 received Flublok Quadrivalent) from study PSC16 (18-49 years) and 8672
subjects (of whom 4328 received Flublok Quadrivalent) from study PSC12 (≥50 years), for a
total of 5326 subjects who received a single 180 mcg dose of Flublok Quadrivalent and 4676
subjects who received a single 60 mcg dose of IIV4. The safety population was used for the
analyses of unsolicited adverse events (AEs), serious adverse events (SAEs), and medically
attended adverse events (MAEs). Among all subjects, 13.3% were 18-49 years, 51.8% 50-64
years, and 34.8% ≥65 years. To evaluate safety, both studies actively solicited local and
systemic reactogenicity events for 7 days, collected unsolicited AEs for 28 days, and collected
both SAEs and MAEs for 6 months post-vaccination. Safety was summarized using descriptive
statistics. Overall, the safety of Flublok Quadrivalent was acceptable and comparable to IIV4 in
adults ≥18 years of age.

11
Deaths and Discontinuations
No deaths or discontinuations due to AEs occurred in PSC16 (adults 18-49 years). Twenty
subjects died in PSC12 (adults ≥50 years) during the six month post-vaccination study period,
Flublok Quadrivalent n=8, IIV4 n=12. The clinical reviewer agreed with the investigator and
Applicant’s assessments that all deaths were unrelated to study vaccine.

SAEs and MAEs

In adults 18-49 years (PSC16), SAEs occurred in ten (1.0%) Flublok Quadrivalent and two
(0.6%) IIV4 recipients during the six months post-vaccination and, of these, three (0.6%)
Flublok Quadrivalent recipients had three SAEs while no IIV4 recipients had SAEs during the 28
days post-vaccination. The clinical reviewer agreed with the investigators and Applicant’s
assessments that none of the SAEs appeared related to study vaccines.

In subjects ≥50 years (PSC12), a total of 145 (3.4%) and 132 (3.0%) subjects in the Flublok
Quadrivalent and IIV4 treatment groups, respectively, experienced SAEs over the six month
safety follow-up period. Of these subjects, 25 (0.6%) and 22 (0.5%) Flublok Quadrivalent and
IIV4 recipients, respectively, reported SAEs in the 28 days post-vaccination. The types and
frequencies of SAEs were balanced between treatment groups. Most SAEs were events that
occur commonly in an older adult and elderly population and none appeared clearly related to
study vaccines. Other than an imbalance of ILIs (more in IIV4 recipients), MAEs were balanced
between treatment groups.

Adverse Events of Special Interest (AESIs)

During the six months post-vaccination, no subjects 18-49 years (PSC16) or ≥50 years (PSC12)
experienced AESIs (potential risks associated with influenza vaccines and defined in the
sponsor’s pharmacovigilance plan), other than possible hypersensitivity events. Collection of
potential hypersensitivity events were not pre-specified but were evaluated post hoc in both
studies. Events were mostly mild in severity and non-serious, and, for many, causality uncertain.
Rates were low and very small imbalances may have been due to chance alone. No severe or
serious allergic reactions, including anaphylaxis, were reported following administration of
Flublok Quadrivalent or IIV4 in either study although, in the reviewer’s opinion, one non-serious
case of bronchospasm three days following Flublok Quadrivalent in PSC16 might have more
appropriately been categorized as severe rather than moderate in intensity. Overall, Flublok
Quadrivalent was not associated with a greater risk of clinically significant acute hypersensitivity
in the safety database of 5326 adults ≥18 years participating in these two studies.

Solicited Local and Systemic AEs

In both PSC16 and PSC12, the incidence and severity grades of solicited local and systemic
reactogenicity events were generally similar between treatment groups and were consistent with
what is described in the current Package Inserts. Among adults 18-49 years (PSC16), the most
common local reactogenicity events were injection site tenderness (Flublok Quadrivalent 47.9%,
IIV4 46.7%) and pain (Flublok Quadrivalent 36.8%, IIV4 36.4%). The rates of injection site
redness were low but occurred more frequently among Flublok Quadrivalent recipients as
compared to IIV4 (4.2% versus 0.9%). The most common systemic symptoms were headache
(Flublok Quadrivalent 20.3%, IIV4 21.1%), fatigue (Flublok Quadrivalent 16.5%, IIV4 16.6%),

12
muscle pain (Flublok Quadrivalent 12.8, IIV4 11.7%), and joint pain (Flublok Quadrivalent
9.5%, IIV4 10.2%). Among adults ≥50 years, the most common local reactogenicity events were
injection site tenderness (Flublok Quadrivalent 34.3%, IIV4 37.1%) and pain (Flublok
Quadrivalent 18.9%, IIV4 22.0%). The most common solicited systemic symptoms were
headache (Flublok Quadrivalent 12.7%, IIV4 13.5%), fatigue (Flublok Quadrivalent 12.2%, IIV4
12.2%), muscle pain (Flublok Quadrivalent 8.5%, IIV4 8.8%), and joint pain (Flublok
Quadrivalent 7.5%, IIV4 8.0%). In both studies most events were mild to moderate (Grade 1 to
Grade 2) in severity and short in duration. Severe (Grade 3) reactions were uncommon.

Postmarketing AEs
CBER review of the Vaccine Adverse Event Reporting System (VAERS) identified anaphylaxis
and other severe allergic reactions after Flublok trivalent vaccine, particularly among individuals
with a self-reported history of egg allergy or allergy to influenza vaccines. No deaths have been
reported and no safety signals have been identified, but many of the reports describe life-
threatening reactions that necessitated emergency treatment. Some patients experienced
persistent wheezing and swelling—even after receiving epinephrine, nebulizers, and
antihistamines. Thus far there are no reports of positive rechallenge, i.e., similar reaction after
subsequent doses of Flublok.

VAERS is a passive surveillance system with potential for reporting bias and is lacking in
denominator data. The number and variety of cases reported for Flublok did not allow for
conclusions regarding a causal relationship or for an estimate of relative risk.

8. Advisory Committee Meeting

A Vaccines and Related Biologics Products Advisory Committee (VRBPAC) meeting was not
held for this supplement. A VRBPAC meeting was held on November 19, 2009, for the original
Flublok licensing application (STN 125285/0) and there were no issues associated with this
supplement that required a new Advisory Committee meeting.

9. Other Relevant Regulatory Issues

PSC committed to establishing a pregnancy registry (study PSC15) for Flublok under the
original Flublok licensure application (STN 125285/0). PSC stated that the protocol for PSC15
will be revised to include Flublok Quadrivalent recipients. PSC also committed to conducting an
observational Phase 4 safety study (PSC13) under STN 125285/0. This study will include
recipients of Flublok but not Flublok Quadrivalent. If PSC13, postmarketing safety surveillance,
or other sources of data suggest a signal of serious risk or potential for serious risk, then the
CBER Office of Biostatistics and Epidemiology, Division of Epidemiology may recommend a
phase 4 study to evaluate the safety of Flublok Quadrivalent.

10. Labeling
In addition to the Flublok Quadrivalent package insert (PI), a revised Flublok (trivalent) PI was
also provided because the relative vaccine efficacy data from PSC12 was used to support
traditional approval of Flublok in persons 50 years and older as well as approval of Flublok

13
Quadrivalent in this population. The Flublok Quadrivalent package insert (PI) included safety
and efficacy data from studies PSC12 and PSC16. The Flublok PI was revised to include
efficacy data from PSC12. To comply with the 2014 draft FDA Guidance for Industry,
“Pregnancy, Lactation, and Reproductive Potential:
Labeling for Human Prescription Drug and Biological Products - Content and Format”
significant revisions were made to Section 8, Use in Specific Populations, of both PIs. The PIs
were primarily reviewed by the Clinical, Pharmacovigilance and Advertising and Promotional
Labeling Branch Reviewers.

Due to continuing postmarketing VAERS reports of anaphylaxis-like reactions in Flublok

recipients with known allergies (including to eggs) or reactions to previous influenza
vaccinations, the Pharmacovigilance and Clinical Reviewers recommended revising a portion of
Section 6.2, Post-marketing Experience, of the Flublok Quadrivalent package insert from:

Immune system disorders: anaphylaxis, anaphylactoid reactions, allergic reactions, and

other forms of hypersensitivity.

to:

Immune system disorders: anaphylaxis, anaphylactoid reactions, allergic reactions, and

other forms of hypersensitivity including reactions among people with a self-reported history
of egg allergy or previous allergic reaction to influenza vaccine.

The rationale for the revision was to clarify that the risk of anaphylaxis and other
hypersensitivity reactions following influenza vaccination is not necessarily related to egg
proteins, and might help providers and patients to make a more informed decision regarding the
use Flublok or Flublok Quadrivalent. The proposed revision was discussed by management
within the CBER Office of Vaccines Research and Review (OVRR) who considered the
presentation of post-marketing reports of anaphylaxis and other allergic reactions in the current
approved Flublok (trivalent) package insert to be appropriate, sufficient, and consistent with this
section of labeling for other vaccines. OVRR did not concur with the proposed qualification so
it was not included in the PI.

All other labeling issues (including those for carton and container labeling) were satisfactorily
resolved through communication with PSC.

11. Recommendations and Risk/ Benefit Assessment

a) Recommended Regulatory Action
The safety, efficacy and immunogenicity data provided in this supplement support the
use of Flublok Quadrivalent for the prevention of influenza disease caused by influenza
virus subtypes A and types B contained in the vaccine. The review committee
recommends traditional approval of Flublok Quadrivalent in persons 18 years of age and
older. The data provided also confirm the clinical benefit of Flublok (trivalent

14
 formulation) and the review committee recommends traditional approval of Flublok in
persons 50 years and older.

b) Risk/Benefit Assessment
Overall, the safety of Flublok Quadrivalent was acceptable and comparable to IIV4 in
adults ≥18 years of age and no safety signals were identified. In persons ≥ 50 years,
Flublok Quadrivalent demonstrated greater vaccine efficacy relative to a U.S.-licensed
quadrivalent influenza vaccine (IIV4, Fluarix Quadrivalent) during a season in which an
antigenically mismatched influenza A/H3N2 predominated. Flublok Quadrivalent also
demonstrated non-inferior immunogenicity in persons 18-49 years as compared to IIV4
against 3 of the 4 antigens present in the vaccine. Non-inferior immunogenicity against
the B/Victoria antigen was not demonstrated in this population. Thus, effectiveness of
Flublok Quadrivalent against influenza B is less certain due to fewer cases in the clinical
endpoint study and a rVE of 4% with wide CIs (95% CI: -72%, 46%) in adults ≥50 years
as well as lower immunogenicity against both B virus strains not only in older adults but
also in adults 18-49 years. However, the lower B/Brisbane antibody titers elicited by
Flublok Quadrivalent as compared to those elicited by IIV4 may be due to antigenic
differences between the B/Brisbane rHA and the egg-grown B/Brisbane antigen used in
the HI assay. Because an accurate assessment of VE depends on many changing
variables and requires multiple years of study, there is some inherent uncertainly in
estimating the effectiveness of influenza vaccines in any particular year.

Potential advantages of Flublok Quadrivalent relative to egg-based influenza vaccines

include closer antigenic matching to circulating strains due to recombinant technology
which preserves the protein sequence of the HA antigen in contrast to propagation in eggs
which requires adaptation or reassortant mutations to increase yield. Manufacture is not
dependent on availability of eggs and, in the event of a pandemic, has the potential to be
increased more quickly than egg-based methods to meet demand. Regarding the
potential advantages of Flublok Quadrivalent in persons with egg allergy, an increasing
body of published evidence and recommendations from the Advisory Committee on
Immunization Practices (ACIP) support the safety of egg-based influenza vaccines in
persons with egg allergy, even in those with a history of anaphylaxis to egg protein.
Therefore the absence of egg proteins in Flublok Quadrivalent may not confer significant
additional benefit over egg-based influenza vaccines in most persons with egg allergy.

Overall, the potential benefits of Flublok Quadrivalent outweigh potential risks and favor
approval.

c) Recommendation for Postmarketing Risk Management Activities

There were no recommendations for a Risk Evaluation and Mitigation Strategy (REMS)
or a Postmarketing Requirement.

d) Recommendation for Postmarketing Activities

The relative vaccine efficacy study (PSC17) of Flublok Quadrivalent in children 3-17
years is a deferred study required by the Pediatric Research Equity Act. An agreement
was reached with PSC on the general plan and timing of this study.

15