DEPARTMENT OF HEALTH & HUMAN SERVICES

US Food and Drug Administration

CBER/OVRR/DVRPA

1401 Rockville Pike

Rockville, MD 20852-1448

 

MEMORANDUM

 

To: Committee Members, VRBPAC

From: Office of Vaccine Research and Review and the Division of Vaccines and Related Product Applications

Re: February 27, 2007 Meeting Topic 1: Safety and Immunogenicity of H5N1 Influenza Virus Vaccine, A/Vietnam/1203/2004 (clade 1), 90g/mL, Manufactured by Sanofi Pasteur

 

I.      Introduction

 

Pandemic influenza outbreaks have occurred when a new subtype of an influenza A virus emerges to which the population has not been exposed and has little or no immunity. During the twentieth century, three pandemic influenza outbreaks occurred. Pandemic influenza viruses can evolve following genetic reassortment of two co-circulating viruses, one of which originates from an animal reservoir and one from human origin. Such a reassortment led to the emergence of the 1957 H2N2 subtype and the 1968 H3N2 subtype. Recent research suggests that the 1918-1919 H1N1 subtype likely resulted from a series of genetic mutations in multiple genes in an influenza virus of avian origin. These mutations appear to have allowed the virus to adapt to and spread among humans.3,5,7 The 1918-19 H1N1 pandemic virus, the most lethal of the twentieth century, resulted in about 50 million deaths worldwide.2 The genetic sequencing, phylogenetic analysis and reconstruction of the 1918-19 H1N1 pandemic virus have provided important insights into virulence factors of influenza viruses.5,7

 

Scientific experts have been monitoring the H5N1 influenza virus subtype for over nine years, and have concluded that it has great potential to be the next pandemic virus. The first H5N1 virus known to have infected humans occurred in Hong Kong in 1997, causing 18 cases, including six deaths. Since mid-2003, this virus has caused the largest and most severe outbreaks in poultry on record. In December 2003, 4 people who were exposed to birds infected with the H5N1 virus were themselves infected with the H5N1 virus, and they all died9. Recent research suggests that the evolution and other characteristics of the H5N1 virus bear similarity to the highly lethal 1918-1919 pandemic influenza virus.5,7 Confirmed infection with H5NI viruses has been associated with approximately 60% mortality and the most recent data (as of January 29, 2007) indicate that a total of 270 people in 10 countries have been infected with H5N1 viruses resulting in 164 fatalities (60.7%). For the most recent information on the number of confirmed clinical cases due to H5N1 viruses, please refer to the World Health Organizations website: www.who.int/csr/disease/avian_influenza/country/en/index.html.

 

As with many other communicable diseases, vaccines are considered the first line of defense against influenza viruses including pandemic strains. Vaccines were available for the 1957 and 1968 pandemic viruses, but arrived too late to have an impact. As a result, great social and economic disruption, as well as loss of life, accompanied the three pandemics of the previous century. Thus, International efforts are underway to begin to address the production and licensure of influenza vaccines for prevention of influenza caused by pandemic strains. Regulatory authorities in Canada and the European Union, the World Health Organization, the U.S. Centers for Disease Control, the U.S. National Institutes of Health (NIH), and the U.S. FDA/CBER are working on guidance and programs for evaluation and use of influenza vaccines for prevention of pandemic influenza strains.

 

II.      Regulatory Background

 

Presently, there are no vaccines licensed in the U.S. for avian influenza viruses such as H5N1 that have the potential to be the next pandemic strain. FDA is working with other government partners such as the National Institutes of Health (NIH), the Centers for Disease Control (CDC) and the Department of Health and Human Services (DHHS) as well as the vaccine industry to diversify and strengthen influenza vaccine manufacturing, and provide flexible and rapid regulatory pathways which will lead to the licensure of pandemic influenza vaccines. FDA published two draft guidance documents in 2006 outlining the clinical pathway to license seasonal as well as pandemic influenza vaccines.11,12 Further, FDA has expedited review of new applications and supplements in order to achieve the goal of having a safe and effective vaccine available for use during a pandemic influenza outbreak. Finally, FDA is also encouraging pre-pandemic development of vaccines for use in the period prior to a pandemic, such as for priming the general population. In session 2 of the VRBPAC on February 27, 2007 (see the briefing document for Topic 2: Clinical Development of Influenza Vaccines for Pre-pandemic Uses), FDA will be discussing what data should be required and obtained to support these uses.

 

In 2004 the National Institute of Allergy and Infectious Diseases (NIAID) awarded contracts to Aventis Pasteur (now sanofi pasteur) and Chiron Corporation to support the production of an investigational vaccine based on the A/Vietnam strain of H5N1 avian influenza virus. In the same year DHHS issued a contract to Aventis Pasteur to manufacture and store what are now 20 million doses of avian influenza H5N1 vaccine. FDA facilitated the rapid initiation of clinical trials by NIHs Vaccine and Treatment Evaluation Units (VTEU) with the H5N1 A/Vietnam/1203/2004 investigational vaccine, and in March, 2005, recruitment began for the clinical trial to investigate the safety and immunogenicity of the H5N1 A/Vietnam/1203/2004 avian influenza vaccine produced by sanofi pasteur. The trial was initiated in April 2005 and the results were published in the NEJM in March of 2006.6 In general, the conclusion of the study was that the higher the dosage of vaccine, the greater the antibody response produced. Moreover, the dose response study showed that more HA antigen than the universally accepted seasonal dose of 15 mcg was needed to increase immune response.

 

Of note, the data in the published study differ from FDAs presentation of the data. Moreover, in the NEJM paper, the investigators performed the microneutralization assay (MN) and the hemagglutination-inhibition assay (HI) on serum samples in an attempt to correlate the two assays. In doing so, microneutralization assays were performed at an initial dilution of 1:20, and those that were negative were assigned a titer of 10. In the HI assays after treatment with receptor-destroying enzyme to remove nonspecific inhibitors of agglutination, the serum samples were tested at an initial dilution considered to be 1:20. CBER reviewers characterized the receptor destroying enzyme treated human serum in the HI assay as a dilution factor of 1:10 which is considered standard practice instead of 1:20. The HI assay data was re-characterized from the samples already read and the results of the re-characterization were submitted to the BLA.

 

Pathways exist to allow consideration of benefits and risks of early intervention against virulent potential pandemic influenza strains, including integration into public health preparedness, as is done for seasonal influenza strains. For seasonal influenza vaccines, once a new inactivated influenza vaccine manufactured using a certain process is licensed, no additional clinical data are required to substitute new vaccine strains into the vaccine. However, the NIH study showed that clinical data may be needed not only for this vaccine, but perhaps for vaccines against other pandemic influenza strains or subtypes to determine the optimum dose. Given the benefit versus risk of having a licensed vaccine against a pandemic strain of influenza, after consultation with NIAID and sanofi pasteur, the FDA determined that the data from the H5N1 clinical study could be submitted to the FDA for evaluation to support licensure of a H5N1 vaccine. Moreover, our rationale at the time was that having clinical data about the optimum dose and immune response will make us better informed about the number of doses needed to assist in preparing for a real influenza pandemic.

 

This BLA for the H5N1 A/Vietnam/1203/2004 vaccine manufactured by sanofi pasteur is the first U.S. license application for a vaccine against H5N1 influenza virus strain. FDA continues to work with its partners in government and industry to facilitate the development of other vaccines against potential pandemic influenza viruses. The US DHHS is also pursuing other approaches to an H5N1 vaccine, such as adjuvanted vaccines and vaccines made in cell cultures. If licensed, this vaccine will be the first and only vaccine available against H5N1 strain in the interim until other influenza vaccines against H5N1 are developed and licensed.

 

 

III.      Clinical Study FUG01

 

FUG01, a randomized, double-blinded, placebo-controlled, dose-ranging, Phase I/II study in healthy adults, ages 18 to 64 years was designed to investigate the safety, reactogenicity, and immunogenicity of the investigational influenza A/H5N1 virus vaccine, (A/Vietnam/1203/04), in healthy adults.

 

The primary objectives of the study were:

  • To determine the dose-related safety of subvirion inactivated H5N1vaccine in healthy adults.
  • To determine the dose-related immunogenicity of subvirion inactivated H5N1vaccine in healthy adults approximately 1 month following receipt of 2 doses of vaccine.
  • To provide information for the selection of the best dose levels for further studies.

 

The secondary objective of the study was:

  • To evaluate dose-related immunogenicity and the percent of subjects responding

approximately 1 and 7 months after the first vaccination.

 

The influenza vaccine used in this study was provided by sanofi pasteur, Inc. and is a monovalent subvirion H5N1 vaccine (HA of A/Vietnam/1203/04) packaged in

single-dose vials each containing 1-mL of vaccine at concentrations of either 30 g/mL and 90 g/mL. Subjects were randomized in a 1:2:2:2:2 ratio to receive either saline placebo, or doses of 7.5 g, 15 g, 45 g, or 90 g of influenza A/H5N1 virus vaccine on Day 0 and Day 28. Placebo or vaccine was administered intramuscularly into the deltoid of choice using a 1-inch, 23-gauge needle. The vaccine doses were delivered in volumes according to the following table:

 

Table 1: Vaccine Dose Volumes

 

Dose

Volume

Placebo

0.5 mL of 0.9% NaCl

7.5 g

0.25 mL from 30 g/mL vaccine vial

15 g

0.5 mL from 30 g/mL vaccine vial

45 g

0.5 mL from 90 g/mL vaccine vial

90 g

1.0 mL from 90 g/mL vaccine vial

 

The vaccine and placebo were administered by an unblinded vaccine administrator. The subjects and the study personnel who performed any study assessments were blinded to the treatment assignments.

The National Institutes of Health (NIH) conducted the study at three U.S. sites (UCLA, U. of Rochester, and U. of Maryland) in two stages: Stage I and Stage II. The inclusion of the two stages, was to allow for an early evaluation of safety in the first cohort to receive vaccine, prior to enrolling the entire cohort. Overall 452 subjects were enrolled in the study. Stage I consisted of the baseline laboratory testing and initial vaccination of 118 subjects: 12 placebo, 28, 25, 25 and 28, in the 7.5 g, 15 g, 45 g, and 90 g, respectively, of the influenza A/H5N1 virus vaccine groups followed by a 7-day safety assessment period. The 7-day safety assessment did not meet safety criteria for stopping the study; therefore, the NIH proceeded with enrolling the remaining 334 subjects (Stage II).

 

Subjects were eligible if they were in good health as determined by vital signs, medical history and a targeted physical examination based on medical history. Additionally, women of child-bearing potential had to agree to practice adequate contraception for the entire study period, and Stage I subjects needed to have normal screening laboratory values. Concomitant medications were recorded and included all medications taken in the past 30 days and through 28 days after second vaccination (approximately Day 56) or early termination, whichever occurs first. Medications not allowed during the study period included, but were not limited to, oral and parenteral steroids, high-dose inhaled steroids, and immunosuppressive or cytotoxic drugs.

 

All Stage I subjects were screened for eligibility laboratory evaluations including hemoglobin, white blood cells, platelets, alanine aminotransferase, and creatinine levels up to 14 days before the first and second vaccinations and again 7 days after each vaccination.

 

Venous blood samples were collected from each subject prior to each vaccination and at Days 28 and 56, and 6 months after the second vaccination for immunogenicity assays. Serum immunogenicity testing was conducted by a central laboratory, Southern Research

Institute in Birmingham, AL. Venous blood samples were assayed for HAI antibodies.

The H5N1 HAI was based on the format described in the WHO Training Manual on

Animal Influenza Diagnosis and Surveillance,8 but modified to use horse erythrocytes according to the method of Dr. Iain Stephenson.4

 

Additionally, for all female subjects of childbearing potential in Stage I and II of this study, urine pregnancy tests were performed within 24 hours prior to the first and second vaccination.

 

Safety assessments included recording all observed or reported adverse events (AEs) that occurred during the study, regardless of the relationship to study product. Initial vaccine reactions were assessed for 15 to 30 minutes after vaccination following both the first and second vaccine doses. Subjects were asked to record both solicited vaccine reactions and any unsolicited AEs on a memory aid for 7 days following both vaccinations. Study personnel contacted subjects by telephone one to three days after each vaccination to review any AEs and SAEs. Adverse events were collected through 28 days following the second dose of vaccine (approximately Day 56). Serious adverse events were collected throughout the study through Day 208.

 

The Safety Monitoring Committee (SMC) was responsible for reviewing the 7-day laboratory results, adverse events and reactogenicity data to determine whether or not the study should progress to Stage II based on the pre-defined halting rules. The SMC performed another review 7 days after Stage I subjects received their second vaccination.

 

Pre-defined stopping rules were in place. If any of the following criteria were met the study was not allowed to enroll, vaccinate or revaccinate any subjects without SMC review and recommendation:

 

The following study safety halting rules were in place:

 

Table 2: Study Safety Halting Rules

 

 

 

A.                Population Description

 

A total of 452 subjects were enrolled at 3 U.S. sites (UCLA, Rochester and University of Maryland). The first subject was enrolled and randomized on April 4, 2005. The final visit for the last subject in the study was on January 25, 2006. Demographics and baseline characteristics of enrolled subjects are summarized in the table below. The majority of the subjects were Caucasian (80.8%), females (53.5%) with a mean age of 40.5 years. Additionally, the majority of subjects (58.4%) had not received the 2004-2005 influenza vaccine.


 

Table 3: Demographics and Baseline Characteristics of Subjects in FUG01

 

 

Placebo

N=48

7.5 g

N=102

15 g

N=101

45 g

N=98

90 g

N=103

All

N=452

Gender n (%)

Male

Female

19 (39.6)

29 (60.4)

51 (50.0)

51 (50.0)

36 (35.6)

65 (64.4)

56(57.1)

42(42.9)

48 (46.6)

55 (53.4)

210 (46.5)

242 (53.5)

Age (years)

Mean

Median

SD

Min;Max

40.4

38.1

12.81

21; 63

41.2

40.0

12.71

19; 65

41.3

40.3

11.84

22; 64

40.4

38.6

12.28

19; 64

39.4

38.1

12.12

18; 64

40.5

39.5

12.27

18; 65

Race n (%)a

White

41 (85.4)

84 (82.4)

80 (79.2)

76 (77.6)

84 (81.6)

365 (80.8)

Asian

5 (10.4)

14 (13.7)

14 (13.9)

11 (11.2)

8 (7.8)

52 (11.5)

Black or African American

4 (8.3)

5 (4.9)

6 (5.9)

12 (12.2)

11 (10.7)

38 (8.4)

American Indian/Alaskan Native

0

1 (1.0)

2 (2.0)

0

0

3 (0.7)

Native Hawaiian or other Pacific Islander

0

0

1 (1.0)

0

0

1 (0.2)

Ethnicity (%)

Hispanic or Latino

4 (8.3)

13 (12.7)

11 (10.9)

6 (6.1)

13 (12.6)

47 (10.4)

Non-Hispanic or Non-Latino

44 (91.7)

89 (87.3)

90 (89.1)

92 (93.9)

90 (87.4)

405 (89.6)

Prior Influenza Vaccine (%)b

Yes

20 (41.7)

40 (39.2)

42 (41.6)

42 (42.9)

44 (42.7)

188 (41.6)

No

28 (58.3)

62 (60.8)

59 (58.4)

56 (57.1)

59 (57.3)

264 (58.4)

a More than 1 race can be checked on the CRF, therefore, N > 452 and (%) > 100%

b Influenza vaccine administered in the 2004 2005 season

Source: CSR volume 8.1, page 70, Table 5.3

CRT ENROLL

 

Table 4 presents the disposition of the study subjects. The majority of the 452 enrolled subjects completed the study (range: 92.9 99.0%). All except one randomized subject received at least one vaccination of either vaccine, at one of the four dose ranges studied, or placebo. The 45 g group had the largest number of discontinuations with 7 (7.1%) subjects prematurely discontinued followed by the 7.5 g group (6 subjects, 5.9%), the 90 g group (3 subjects, 2.9%), the placebo group (1 subject, 2.1%) and the 15 g group (1 subject, 1.0%). The most common reason for discontinuation was lost to follow-up. Overall, 168 subjects committed at least one protocol violation: 18 subjects (37.5%) in the placebo group, 33 subjects (32.4%) in the 7.5 μg study group, 37 subjects (36.6%) in the 15 μg study group, 37 subjects (37.8%) in the 45 μg study group, and 43 subjects (41.7%) in the 90 μg study group. The University of Rochester site (n = 67, 39.9%) had more protocol violations than the other two sites (UCLA site, n = 51, 30.5% and University of Maryland n = 50, 29.8%). The majority of protocol violations were missed visits, or visits outside of pre-specified windows.

 

A total of 47 subjects were excluded from the per protocol population for immunogenicity due to protocol violations and study non-completion reducing the total immunogenicity population to 405 subjects.

 

Table 4: Summary of Subject Disposition

 

 

Placebo

N=48

n (%)a

7.5 g

N=102

n (%)

15 g

N=101

n (%)

45 g

N=98

n (%)

90 g

N=103

n (%)

All

N=452

n (%)

All Randomized

48 (100)

102 (100)

101(100)

98(100)

103(100)

452 (100)

Safety Population

(Safety Analysis Set)b

48 (100)

101 (99)

101 (100)

98 (100)

103 (100)

451 (99.8)

Received Vaccine #1

48 (100)

101 (99)

101(100)

98 (100)

103 (100)

451 (99.8)

Received Vaccine #2

46 (95.8)

99 (97.1)

99 (98)

92 (93.9)

100 (97.1)

436 (96.5)

Vaccine 1 Reactogenicity Data

48 (100)

101 (99)

101 (100)

98 (100)

103 (100)

451 (99.8)

Vaccine 2 Reactogenicity Data

46 (95.8)

99 (97.1)

99 (98)

92 (93.9)

100 (97.1)

436 (96.5)

Completed Study

47(97.9)

96 (94.1)

100 (99)

91 (92.9)

100 ( 97.1)

434 (96)

Subjects Discontinued

1 (2.1)

6 (5.9)

1 (1.0)

7 (7.1)

3 (2.9)

18 (4.0)

Reasons for Discontinuation

Randomized but not Vaccinated

SAEs (Other than Death)

Adverse Events other than SAE

Lost to follow-up

Non-compliance/Protocol deviation

Termination by site or Sponsor

Voluntary Withdrawal by Subject

Death

 

0

0

1 (2.1)c

1 (2.1)

0

0

0

0

 

1 (1.0)

0

0

3 (2.9)

0

0

2 (2.0)

0

 

0

0

0

1 (1.0)

0

0

0

0

 

0

0

0

4 (4.1)

1 (1.0)

0

1 (1.0)

1 (1.0

 

0

0

1 (1.0)d

2 (1.9)

0

0

0

0

 

1 (0.22)

0

1 (0.22)

11 (2.4)

1 (0.22)

0

3 (0.66)

0

FAS Population for Immunogenicityb

48 (100)

100 (99)

101 (100)

98 (100)

102 (99)

449 (99.3)

Subjects Excluded from PP population for Immunogenicity

Protocol violatorse/Did not meet entry criteria

Visit out of window/Study non-completers

6 (12.5)

 

0

 

6 (12.5)

9 (8.8)

 

2 (0.98)

 

7 (6.9)

7 (6.9)

 

1 (0.99)

 

6 (5.9)

13 (13.3)

 

0

 

13 (13.3)

12 (11.7)

 

0

 

12 (11.7)

47 (10.4)

 

3 (0.66)

 

44 (9.7)

PP Population for Immunogenicityb

42(87.5)

93(91.2)

94(93.1)

85(86.7)

91(88.3)

405(89.6)

a. Percentages are based on the total number of randomized subjects enrolled in each treatment group.

b. Study populations are defined in the Immnunogenicity and Safety Results sections. FAS = full analysis set; PP = per protocol

c Subject 06FRO176 (Placebo group) as reported on the CRF, got only the first vaccination, and discontinued vaccination due to adverse reaction to previous vaccination. However, subject completed the protocol, including the Day 208 bleed.

d. Subject 06FLA119 (90 μg group) got only the first vaccination, and terminated early. According to the CRF, subject was a voluntary withdrawal due to 'absence from work following AE'. However, the source document indicated that the reason was subject decision following grade 3 AE resulting in one day absence from work.

e. Protocol violators are counted only once according to their first violation.

Source: CSR FUG01 volume 8.1, page 72,

B.                 Immunogenicity ASSESSMENTS and Results

 

In the original protocol (Version 1.0) dated January 21, 2005 the primary and secondary endpoints for the FUG01 trial were as follows:

 

Primary endpoints

  • Adverse event or SAE information (solicited in-clinic and via memory aids,

concomitant medications, and periodic targeted physical assessments).

  • Proportion of subjects in each dose group achieving a serum neutralizing antibody

titer ratio of 1:40 against the influenza A/H5N1 virus on Day 56.

 

Secondary endpoints

  • Geometric mean titer and the frequency of 4-fold or greater increases in neutralizing antibody titers in each group 1 month after receipt of each dose, and 7 months after receipt of the first dose of vaccine.
  • Geometric mean titer and the frequency of 4-fold or greater increases in serum HAI antibody titers 1 month after receipt of each dose, and 7 months after receipt of the first dose of vaccine.
  • Development of serum antibody responses against antigenically drifted variants of H5 influenza virus.

 

After the initial review by FDA and IRB and approval of the original protocol

(version 1.0, dated 21 January 2005), changes were introduced into subsequent versions

with full consultation and agreement of the sponsor and investigators as follows:

 

Amendment 1: Protocol version 2.0, dated March 15, 2005. The changes are summarized as follows:

  • The name of trial vaccine was updated as subvirion inactivated H5N1 vaccine
  • Primary end point was clarified to indicate the period of evaluation as 28 days after receipt of the second dose of vaccine (approximately 56 days).
  • Administrative changes to clarify: the study design, sponsors name, background

information, potential risks, exclusion criteria, concomitant medication, study schedule, and assessment of safety sections of the protocol.

 

Amendment 2: Protocol version 3.0, dated June 13, 2005. The changes are summarized as follows:

  • Added the GMT and frequency of 4-fold or greater increases in serum HAI antibody titers as a primary endpoint and removed it as a secondary endpoint
  • Deleted the second bulleted secondary endpoint and added it as the second bulleted primary endpoint.
  • Added the serum HAI as a primary outcome measure.
  • Updated protocol summary that describe the SMC review of safety data and the

halting rules for the study.

 

Amendment 3: Protocol version 4.0, dated July 14, 2005. The changes are summarized as follows:

  • Two end-points were moved from secondary to primary endpoints:

a) Geometric mean titer (GMT) and frequency of 4-fold or greater increases in

neutralizing antibody titers in each group 1 month after receipt of each dose, and

7 months after receipt of the first dose of vaccine.

b) Geometric mean titer and frequency of 4-fold or greater increases in serum

hemagglutination inhibition (HAI) antibody titers in each group 1 month after

receipt of each dose, and 7 months after receipt of the first dose of vaccine.

 

  • Administrative change to reflect Dr Linda Lamberts new position as Chief of the

Respiratory Disease Branch.

 

 

In the final version of the protocol, version 4.0, dated July 14, 2005 the study endpoints were stated as follows:

 

The primary endpoints were:

         Proportion of subjects in each dose group achieving a serum neutralizing antibody titer ratio of 1:40 against the influenza A/H5N1 virus 28 days following second dose of vaccine (approximately Day 56).

         Geometric mean titer and the frequency of 4-fold or greater increases in neutralizing antibody titers in each group 1 month after receipt of each dose, and 7 months after receipt of the first dose of vaccine.

         Geometric mean titer and the frequency of 4-fold or greater increases in serum HAI antibody titers 1 month after receipt of each dose, and 7 months after receipt of the first dose of vaccine.

 

Of note: Neutralizing antibody assay data were not included in the BLA submission due to an agreement between the Sponsor, NIH/NIAID/DMID, CBER, and sanofi pasteur (CBER minutes of pre-BLA meeting of April 21, 2006 and DMID minutes of May 10, 2006 teleconference between CBER and DMID).

 

The secondary endpoint was:

         Development of serum antibody responses against antigenically drifted variants of H5N1 influenza virus.

 

Of note: Data on the development of serum antibody responses against antigenically drifted variants of H5N1 influenza virus were not included in the BLA submission.

 

Serum specimens for serologic analysis were obtained before vaccination 1 (baseline,

Day 0), before vaccination 2 at visit 3 (Day 28 post-vaccination 1), and post-vaccination 2 at visit 5 (Day 56; Day 28 post-vaccination 2), and visit 6 (Day 208; 6 months post-vaccination 2).

 

Sera from these blood samples were used for immunogenicity testing. Test results from the per-protocol population for immunogenicity were used for the analyses of the antibody responses.

 

Of note: The immunogenicity data from study FUG01 presented in the BLA are different than the immunogenicity data presented in the Treanor, et al publication6 because the BLA:

         is based on the final data, whereas the publication was based on interim data

         uses an initial dilution factor of 1:10 (see Section 4.2.2), whereas the publication used an initial dilution factor of 1:20.

         for baseline less than lower limit of quantitation (< LLOQ), uses a fold-rise calculation that considers LLOQ as baseline (see Section 4.3.2.1), whereas the publication considered 0.5 LLOQ as baseline.

         considers a subject with a < 1:10 baseline titer needed to have a ≥ 1:40 postvaccination titer to be classified as having a four-fold rise; whereas in the publication, a subject with a < 1:20 baseline titer needed to have the same post-vaccination titer to be classified as having a four-fold rise, despite having a higher baseline titer.

 

The FDA received an assay validation package for HAI antibody; however, despite requests no such assay validation package was received for the microneutralization assay. Therefore, the FDA reviewed the HAI antibody assay data as presented in the BLA and not the data derived from the unvalidated microneutralization assay.

The primary analysis of the primary endpoints was performed on the Per Protocol (PP) population defined as the subset of those subjects who meet the following conditions:

  • Satisfied inclusion and exclusion criteria
  • Received vaccines correctly according to randomization
  • Blood samples taken pre- and 28 days following each vaccination with a valid pre-vaccination 1, post-vaccination 1, and post-vaccination 2 serology result,
  • Blood sampling visits occurred within the specified time window.

 

The secondary analysis of the primary endpoints was performed on the full analysis set (FAS). The FAS is defined as the subset of subjects who meet the following conditions:

  • Randomized and received a vaccine.
  • Blood sample taken post-vaccination and at least 1 valid post-vaccination serology result.

Table 5 presents the immunogenicity results for the PP population subjects.

 

Table 5: Summary of Geometric Mean Titers (GMTs), > 4-Fold Rise, and > 1:40 Hemagglutinin Inhibition (HAI) titers -- H5N1 (Per Protocol Population)

 

 

Placebo

Na=42

7.5g

N=93

 

15 g

N=94

45 g

N=85

90 g

N=91

GMT

Mb

Baseline GMT

95% CI

42

5.4

(4.8,5.9)

93

5.8

(5.1,6.6)

94

5.1

(4.9,5.4)

85

5.5

(5.1,6.0)

91

5.2

(4.9,5.5)

M

28 Days post Vaccination 1 GMT

95% CI

42

5.5

(4.8,6.2)

93

6.5

(5.6,7.6)

94

6.8

(5.7,8.2)

85

12.0

(8.6,16.7)

91

13.7

(9.8,19.2)

M

28 Days post Vaccination 2 GMT

95% CI

42

5.5

(4.8,6.4)

93

7.3

(6.2,8.7)

94

9.7

(7.8,12.2)

85

17.8

(12.7,24.9)

91

30.6

(22.1,42.2)

M

6 Months post Vaccination 2

95% CI

41

5.6

(4.8;6.6)

92

6.1

(5.3;7.0)

93

6.6

(5.7;7.7)

83

10.6

(8.0;14.2)

91

11.8

(8.9;15.7)

4-Fold Rise in HAI titer

28 Days post Vaccination 1

%

95% CI

 

0

0

(0.0;8.4)

2

2.2

(0.3;7.6)

7

7.4

(3.0;14.7)

19

22.4

(14.0;32.7)

21

23.1

(14.9;33.1)

28 Days post Vaccination 2

%

95% CI

 

0

4

4.3

(1.2;10.6)

15

16.0

(9.2;25.0)

29

34.1

(24.2;45.2)

41

45.1

(34.6;55.8)

6 Months post Vaccination 2

%

95% CI

 

1

2.4

(0.1;12.9)

2

2.2

(0.3;7.3)

5

5.3

(1.8;12.1)

19

22.6

(14.4;33.4)

16

17.6

(10.4;27.0)

Age/Prior Flu Vaccine Strata

< 40, - vaccine

<40, + vaccine

>40, - vaccine

>40, + vaccine

 

1 (2.5)

0

0

0

 

1 (1.1)

0

0

4 (4.3)

 

6 (6.4)

3 (3.2)

8 (8.5)

2 (2.1)

 

13 (15.3)

5 (5.9)

10 (11.3)

3 (3.5)

 

24 (26.4)

6 (6.6)

7 (7.7)

10 (11.0)

a. N=number of subjects in dose group

b. M=number of subjects with available data

Source: IMMUN0 CRT; CSR FUG01 Volume 8.1, pp. 116 - 117


Table 5: Summary of Geometric Mean Titers (GMTs), > 4-Fold Rise, and > 1:40Hemagglutinin Inhibition (HAI) titers _H5N1 (Per Protocol Population) - continued

 

 

Placebo

Na=42

7.5g

N=93

 

15 g

N=94

45 g

N=85

90 g

N=91

> 1:40 HAI titer

Pre-vaccination

%

95% CI

 

0

0

(0.0;3.8)

3

3.2

(0.7;9.1)

0

0

(0.0;3.8)

2

2.4

(0.3;8.2)

1

1.0

(0.0;6.0)

28 Days post Vaccination 1

%

95% CI

 

1

2.4

(0.1;12.6)

6

6.5

(2.4;13.5)

8

8.5

(3.7;16.1)

19

22.4

(14.0;32.7)

22

24.2

(15.8;34.3)

28 Days post Vaccination 2

%

95% CI

 

1

2.4

(0.1;12.6)

6

6.5

(2.4;13.5)

16

17.0

(10.1;26.2)

29

34.1

(24.2;45.2)

42

46

(35.6;56.9)

6 Months post Vaccination 2

%

95% CI

 

2

4.8

(0.6;16.5)

5

5.4

(1.8;12.2)

6

6.4

(2.4;13.5)

19

22.4

(14.4;33.4)

17

18.7

(17.3;28.2)

Age/Prior Flu Vaccine Strata

< 40, - vaccine

<40, + vaccine

>40, - vaccine

>40, + vaccine

 

1 (2.5)

0

0

1 (2.5)

 

1 (1.1)

2 (2.2)

0

7 (7.5)

 

7 (7.4)

5 (5.3)

9 (9.6)

2 (2.1)

 

14 (16.5)

5 (5.9)

10 (11.8)

6 (7.1)

 

26 (28.6)

6 (6.6)

7 (7.7)

11 (12.1)

a. N=number of subjects in dose group

b. M=number of subjects with available data

Source: IMMUN0 CRT; CSR FUG01 Volume 8.1, pp. 116 - 117

 

A total of six subjects had preexisting antibody at levels > 1:40 (Table 6.) These subjects differed from the remainder of the study subjects in that the majority of them were male (83.3%), one-third were Asian, the mean age was 50.2 years, and most subjects had received the 2004-2005 influenza vaccine (83.3%).


 

Table 6: Summary of subjects with preexisting H5 HAI antibody titer levels > 1:40

 

Patient ID

Vaccine dose

(g)

Sex

Ethnicity/

Racea

Age

(yrs)

Prior Flu Vaccine

Baseline visit

Visit 3 (D 28)

Visit 5 (D 56)

Visit 6

(6 mos)

06FLA012

7.5

M

H, C

35.3

Y

160

160

113

--

06FLA155

7.5

M

A

35.2

Y

160

80

40

113

06FLA157

7.5

M

C

57.5

Y

56

113

226b

113

06FLA160

45

F

A

55.5

N

40

28

28

20

06FUM172

45

M

C

61.1

Y

40

905b

1280b

640b

06FUM111

90

M

C

56.6

Y

80

134

113

113

a = A=Asian, C=Caucasian, H = Hispanic

b = four-fold increase in baseline titers

Source: IMMUN0 CRT

 

C.                Safety Assessment and Results

 

The safety analysis set population consisted on the 451 subjects who received at least one dose of placebo or vaccine.

 

NIH assessed the following primary safety endpoint:

         Adverse event (AE) or serious AE (SAE) information (solicited and unsolicited)

 

Deaths and Other Serious Adverse Events (SAEs)

 

A 52-year-old male subject died approximately 23 days after receiving the first dose of randomized 45 g Inactivated Influenza A/H5N1. The cause of death was chronic alcoholism with hepatomegaly (2265 grams) and marked hepatic steatosis. The event was classified as not related to the study product by the investigator and the sponsor. The subject denied alcohol and drug abuse on his screening questionnaire and because he was a Stage II subject he did not have entry laboratory evaluations.

 

Three other subjects experienced serious adverse events after any vaccination (Table 7). None of these SAEs were considered related to study product.

 


Table 7: Serious Adverse Events after Any Vaccination Safety Analysis Set

 

 

Subject

Number

Treatment Group

Age

(Yrs)

Preferred

Term

Severity/

Grade

Relatedness to Vaccination (as Assessed by Investigator)

Last Vaccination and Date

Days to Onset

Duration

(in days)

Outcome

06FRO073

Placebo

50.6

Menorrhagia

Severe/

Grade 3

Not Related

2

06/06/2005

95

1

Resolved without sequelae

06FRO180

15 g

62.7

Breast Cancer

Severe/

Grade 3

Not Related

2

06/16/2005

29

224

Resolved without sequelae

06FUM082

90 g

57.5

Cerebrovascular

accident

Severe/

Grade 3

Not Related

2

06/16/2005

113

11

Resolved without sequelae

Source: AE0 CRT

 

Reactogencity Events

 

Solicited local and systemic reactogenicity events were captured throughout the study period, and presented in the following table.

 

Solicited injection site (local) AEs included: Pain, tenderness, redness and swelling.

 

Solicited systemic AEs included: Feverishness, malaise, body aches (exclusive of the injection site), nausea and headache.

 

Of note the Investigators did not pre-specify criteria for grading the severity of redness and swelling, which resulted in 116 subjects with measurable redness and swelling without an assigned severity level. Post-hoc, the FDA reviewer assigned grades to these 116 subjects according to the Draft Guidance for Industry: Toxicity Grading Scale for Healthy Adult and Adolescent Volunteers Enrolled in Preventive Vaccine Clinical Trials.10


 

Table 8: Frequency of Local and Systemic Reactogenicity Events in Study FUG01

 

 

Placebo

N=48

n (%)

7.5g

N=101

n (%)

15 g

N=101

n (%)

45 g

N=98

n (%)

90 g

N=103

n (%)

Subjects with > 1

Injection Site AE(s)

22 (45.8)

60 (59.4)

69 (68.3)

77 (78.6)

87 (84.5)

Subjects with Moderate (Gr 2) Injection Site AEs

2 (4.2)

1

 

8

10

16 (15.5)

Subjects with Severe (Gr 3) Injection Site AEs

0 (0)

2

3

6

6 (5.8)

# of Injection Site AEs

89

237

266

424

537

Pain

20 (22.5)

52 (21.9)

89

149

194 (36.1)

Tenderness

19 (21.3)

69 (29.1)

91

150

189 (35.2)

Erythema/Redness

31 (34.8)

85 (35.9)

62

95

98 (18.2)

Induration/Swelling

19 (21.3)

31 (13.1)

24

30

56 (10.4)

Subjects with > 1

Systemic AE(s)

28 (58.3)

40 (39.6)

48 (47.5)

35 (35.7)

49 (47.6)

Subjects with Moderate (Gr 2) Systemic AEs

10 (20.8)

15 (14.9)

21 (20.8)

6 (6.1)

12 (11.7)

 

 

 

 

 

 

Subject with Severe (Gr 3) Systemic AEs

0

1 (1.0)

2 (2.0)

0

1 (1.0)

# of Systemic AEs

125

207

349

109

171

Headache

45 (36)

70 (33.8)

107 (30.7)

44 (40.4)

65 (38.0)

Malaise

43 (34.4)

57 (27.5)

98 (28.0)

31 (28.4)

51 (29.8)

Body aches

28 (22.4)

44 (21.3)

82 (23.5)

21 (19.3)

29 (17.0)

Nausea

4 (3.2)

20 (9.7)

42 (12.0)

7 (6.4)

15 (8.8)

Feverishness

5 (4.0)

16 (7.7)

20 (5.7)

6 (5.5)

11 (6.4)

Source: LOCAL0 and SYSTEM0 CRTs

 

Other Adverse Events

 

The most frequently reported AEs, regardless of causality, were in the system organ class (SOC) infections and infestations (21.5%) followed by, respiratory, thoracic and mediastinal disorders (14%) and gastrointestinal disorders (10%). Adverse event frequency was similar or slightly higher in the placebo arm as compared to the vaccine arms except in the Respiratory, thoracic and mediastinal disorders SOC where the AE rates in the vaccine arms (10.9%, 13.9%, 7.1% and 9.7% in the 7.5 g, 15 g, 45 g, 90 g arms respectively)were slightly higher than the placebo arm (6.25%). The most commonly reported preferred terms in this SOC were pharyngolaryngeal pain (namely. sore throat), 33.3%; cough, 22.2%; nasal congestion, 20.0% and sinus congestion, 11.1%.

 


IV.      SUMMARY

 

The BLA contained the results of an NIH-sponsored phase I/II dose-ranging study of inactivated influenza A/H5N1 vaccine manufactured by Sanofi-Pasteur, Inc. All of the manufacturing information in this BLA was cross-referenced to the manufacturing information contained in the existing BLA for Fluzone held by Sanofi-Pasteur, Inc. Among the study participants, there were no significant safety signals that would preclude administration of this vaccine to additional persons. The immunogenicity data, as assessed by the HAI antibody assay, indicate that 90 mcg dose elicited a better immune response than the lower doses group following two vaccines doses administered approximately one month apart. Previous studies of influenza viruses have suggested that a titer of ≥1:40 is associated with up to 50% protection against influenza illness.1 Whether these data can be extrapolated to the effectiveness of an A/H5N1 influenza virus is not clear.

 

The questions to the committee will focus on the adequacy of the data to support a licensed indication active immunization for the prevention of illness causes by influenza A/Vietnam/2004. If the data are not deemed adequate, the committee will be asked what additional data would be needed to support approval. If the data are adequate, the committee will be asked what studies might be conducted in the post-approval setting to assess the benefit and risk of the product.

.


 

V.      References

 

  1. de Jong JC, Palache AM, Beyer WEP, Rimmelzwaan GF, Boon ACM, Osterhaus ADME. Haemagglutination-inhibiting antibody to influenza virus. Developmental Biology (Basel). 2003;115:6373.

 

  1. Johnson NPAS, Mueller J. Updating the accounts: Global mortality of the 1918-1920 Spanish influenza pandemic. Bulletin of the History of Medicine. 2002;76:105-115.

 

  1. Reid AH, Fanning TG, Janczewski TA, Lourens RM, Taubenberger JK. Novel origin of the 1918 pandemic influenza virus nucleoprotein gene. Journal of Virology. 2004;78:12462-12470.

 

  1. Stephenson I, Wood JM, Nicholson KG, Charlett A, Zambon, MC. Detection of anti-H5 responses in human sera by HI using horse erythrocytes following MF59-adjuvanted influenza A/Duck/Singapore/97 vaccine.
    Virus Research. 2004;103(1-2):91-5.

 

  1. Taubenberger JK, Reid AH, Lourens RM, Wang R, Jin G, Fanning TG. Characterization of the 1918 influenza virus polymerase genes. Nature. 2005;437:889-893.

 

  1. Treanor J, et al. Safety and Immunogenicity of an Inactivated Subvirion Influenza A (H5N1) Vaccine. The New England Journal of Medicine. 2006; 354 (13):1343-51.

 

  1. Tumpey TM, Basler CF, Aguilar PV, Zeng H, Solorzano A, Swayne DE, Cox NJ, Katz JM, Taubenberger JK, Palese P, Garcia-Sastre A. Characterization of the reconstructed 1918 Spanish influenza pandemic virus. Science. 2005;310:77-80

 

  1. Serologic diagnosis of influenza virus by hemagglutination inhibition. In: WHO

Manual on Animal Influenza Diagnosis and Surveillance 2002.5Rev.1 Pages 37-39.

 

  1. Avian Influenza Fact Sheet. February 2006. http://www.who.int/mediacentre/factsheets/avian_influenza

 

  1. Draft Guidance for Industry: Toxicity Grading Scale for Healthy Adult and Adolescent Volunteers Enrolled in Preventive Vaccine Clinical Trials. 4/9/2005.

 

  1. Draft Guidance for Industry: Clinical Data Needed to Support the Licensure of Trivalent Inactivated Influenza Vaccines. 3/2/2006.

 

  1. Draft Guidance for Industry: Clinical Data Needed to Support the Licensure of Pandemic Influenza Vaccines - 3/2/2006.