U.S. flag An official website of the United States government
  1. Home
  2. Vaccines, Blood & Biologics
  3. Vaccines
  4. Statistical Review and Evaluation - Fluarix
  1. Vaccines

Statistical Review and Evaluation - Fluarix

FDA #: STN 125127/319

SPONSOR: GSK Biologicals

NAME of PRODUCT: Influenza Virus Vaccine (Fluarix®)

DOCUMENT SUBMITTED:  BLA supplement seeking indication for active immunization of person 3 years of age and older

FROM: Sang Ahnn (HFM-217)

THROUGH:  A. Dale Horne, Chief, Vaccine Evaluation Branch (HFM-217)

TO:     Sara Gagneten (HFM-478)
Melisse Baylor (HFM-475)

CC: HFM-217/A. Dale Horne
HFM-215/Henry Hsu
HFM-215/Chronological File

EXECUTIVE SUMMARY

This submission contains the results of GSK’s clinical immunogenicity and safety study, Fluarix-US-005. Supportive data from 2 non-IND pediatric studies, Fluarix-056 and Fluarix-062, are also included in this submission.  Based on the results included in this submission, the applicant is seeking indication for active immunization of persons 3 years of age and older. The applicant’s immunogenicity objective was not met.

RECOMMENDATION: Because 40% of the data supporting the sought indication were excluded from analysis, and because of the post hoc nature of the 3-5 year old subgroup  analysis, these data are less than ideal for recommending approval of the proposed indication.  I recommend that OVRR consider whether there are other relevant factors that would support approval of the applicant’s requested expanded age indication.

BACKGROUND

Fluarix® was approved on August 31, 2005 under the accelerated approval regulations.  On April 1, 2009, Fluarix® was approved for its clinical efficacy for persons 18 years of age and older.  This submission contains the results of GSK’s clinical immunogenicity and safety study, Fluarix-US-005, and results from 2 non-IND studies.  The purpose of this submission is to extend the indication down to 3 years of age.

Fluarix-US-005

This is a multi-center (~25 centers in US), single-blind, active-controlled study to demonstrate the immunological non-inferiority of Fluarix to Fluzone in children 6 months to 5 years of age.  A total of 3327 subjects (6 months to 18 years of age) were enrolled and randomized into two treatment groups (Fluarix or Fluzone).  Primary immunogenicity analysis was based on subjects 6 months to 5 years of age (N=1498 enrolled and randomized at 1:1 ratio to either Fluarix or Fluzone).  Subjects 5 to 18 years of age were used only for safety analyses.

Immunogenicity

Primary analysis of immunogenicity was based on the ATP cohort, which included all evaluable subjects for whom data concerning immunogenicity endpoint measures were available (see Table 1 below).

Table 1. Number of subjects enrolled and eligible for immunogenicity analysis

 

Fluarix

Fluzone

n

%

n

%

Enrolled

748

100%

750

100%

Vaccinated

748

100%

748

99.7%

ATP immunogenicity cohort

426

57.0%

445

59.3%

ATP immunogenicity cohort
with both pre- and post-vaccination results available

 

425

 

56.8%

 

443

 

59.1%

Over 40% of subjects (323 from Fluarix group and 307 from Fluzone group) were excluded from the ATP immunogenicity cohort, as indicated in the above table.  The main reasons for exclusion were non-compliance of vaccination and/or blood-sampling schedules, and missing serological data.  Based on ATP immunogenicity cohort, Table 2 shows the primary immunogenicity results.

Table 2. Primary Immunogenicity Results (ATP Cohort of 6 months to 5 years of age, N=425 for Fluarix and N=443 for Fluzone)

 

GMT ratio
(Fluzone/Fluarix)

Non-inferiority

SCR* difference
(Fluzone-Fluarix)

Non-inferiority

Strain

Point Estimate

95% CI
Lower    Upper Limit      Upper

UL<>

Point Estimate

95% CI
Lower       Upper
Limit         Limit

UL<>

A/New Caledonia

1.50

1.27

1.77

NO

14.0

7.8

20.0

NO

A/Wisconsin

1.65

1.40

1.95

NO

13.6

7.6

19.6

NO

B/Malaysia

1.48

1.23

1.80

NO

14.5

7.8

20.9

NO

* SCR: seroconversion rate

As shown in Table 2, non-inferior immunogenicity of Fluarix to Fluzone was not established, based on pre-specified non-inferiority criteria. 

The 3-5 year stratum had better immunogenicity results than did the 6 month - 3 year stratum.  The applicant is seeking a Fluarix indication for 3 years and older based on the following post-hoc analysis result (Table 3).

Table 3. Immunogenicity Results (ATP Cohort of 3 years to 5 years of age, N=220 for Fluarix and N=220 for Fluzone)

 

GMT ratio
(Fluzone/Fluarix)

Non-inferiority

SCR* difference
(Fluzone-Fluarix)

Non-inferiority

Strain

Point Estimate

       95% CI
Lower     Upper Limit     Upper

UL<>

Point Estimate

       95% CI
Lower          Upper
Limit            Limit

UL<>

A/New Caledonia

1.22

0.97

1.53

NO

-0.5

-8.8

7.9

YES

A/Wisconsin

1.08

0.86

1.34

YES

-0.5

-9.0

8.1

YES

B/Malaysia

1.14

0.86

1.52

NO

2.3

-7.0

11.5

NO

* SCR: seroconversion rate

Safety

As discussed and agreed with the clinical reviewer, I fully depend on the clinical reviewer’s expertise in this area.  The clinical reviewer has requested no additional analyses of the safety data by me, and thus none are presented here.

Reviewer’s comments

  1. Over 40% of subjects were excluded from the ATP immunogenicity cohort.  This may reflect poor execution of this study.  However, among the subjects excluded from the immunogenicity analysis, I did not find any significant imbalance between the two groups with respect to important baseline variables.
  2. The applicant is seeking an indication based on post-hoc subgroup analysis.  Even though the 3-5 year age group was used as a randomization stratum, it was not pre-specified for separate immunogenicity analysis.  The post-hoc nature of any analysis increases the likelihood that a positive finding is spurious. 
  3. As shown in Table 3 above, the non-inferior immunogenicity of Fluarix to Fluzone was technically not established in this post-hoc subgroup for all strains.  However, the B strain commonly demonstrates weaker results among licensed influenza vaccines, so the results here for the B strain are not surprising.  Also, the non-inferiority margins that were missed were nonetheless close to meeting the non-inferiority criteria – i.e., they were near misses.

ResourcesForYou

 
Back to Top