Statistical Review and Evaluation, Amendment 12 - MenHibrix
BLA STN 125363
Active immunization to prevent invasive meningococcal disease caused by Neisseria meningitidis serogroups A, C, Y and W-135.
Submission Date: 08/12/2009
Amendment 12, dated 15 April 2011 (response to CR letter issued 11 June 2011)
Vaccine Evaluation Branch (VEB)
Barbara Krasnicka, Ph.D.
Tammy Massie, Ph.D.
A. Dale Horne, Dr. P.H.
Meghan Ferris, M.D., MPH
Joseph Temenak, Ph.D., Primary Reviewer,
1. Executive Summary
Biologics License Application (BLA) STN 125363 was submitted by GlaxoSmithKline (GSK) Biologicals on August 12th, 2009, for licensing the MenHibrix® (Haemophilus influenzae type b and Neisseria meningitidis serogroups C and Y-tetanus toxoid conjugate vaccine; originally designated Hib-MenCY-TT) vaccine. Clinical development of this vaccine was conducted under US IND (b)(4). The applicant seeks licensure of MenHibrix® for active immunization of infants and toddlers aged 6 weeks through 15 months for the prevention of invasive diseases caused by Haemophilus influenzae type b (Hib) virus and Neisseria meningitides serogroups C (MenC) and Y (MenY) bacteria. Hib-MenCY-TT vaccine is to be administered as a 4-dose series (0.5-mL per dose) by intramuscular injections at 2, 4, 6, and 12 through 15 months of age. The first dose may be given as early as at 6 weeks of age.
2. Review of responses to CR letter from June 11th 2010
On June 11th, 2010, a CR letter covering 88 items was issued to GSK (GlaxoSmithKline). Among other items, 9 deficiencies specified by the statistical reviewer were included in this CR. In this memo, the applicant’s responses to these statistical concerns are reviewed.
Background of the statistical CR questions/comments
The BLA 125363 submission reported results from eleven studies. Six of them investigated effects of the primary vaccine doses (primary vaccination phase). The remaining five studies assessed effects of the fourth dose (booster) vaccination phase. Two studies also evaluated antibody persistence.
The immunogenicity data based on the Hib-MenCY-TT-009/010 pivotal study did not support the pre-specified criteria for the study success. The first co-primary objective (the lot-to-lot consistency of 3 manufacturing lots of Hib-MenCY-TT vaccine) was not met. The estimated values of the geometric mean titers (GMTs) for three lots were comparable for the C serogroup, but not for the Y serogroup. Statistical analyses showed differences in GMTs, especially for lots A and B, for the Y serogroup. There could be several reasons for the objectives related to lot-to-lot consistency not being met. For instance, the small number of subjects included in these analyses, manufacturing inconsistencies, and/or variability between assay runs used for measuring titers could be the reasons. In the CR letter, the statistical reviewer requested more information on assay runs, in particular identification numbers of assay runs. Such indicators were not included in the submitted by the applicant SAS datasets despite that the Agency requested such indicators during the pre-BLA meeting.
There could be one more reason influencing the immunogenicity hypotheses results, namely, over 20% of the immunogenicity data were missing. This could introduce biases into the study results.
As per the applicant’s pre-specified assumptions in the study protocol, the objectives of study Hib-MenCY-TT-009/010 should be assessed in a hierarchical manner according to the order presented in the protocol. Due to this presumption and the fact that the first co-primary objective (the lot-to-lot consistency) was not met, the pre-specified criteria, based purely on statistical principles and without consideration of other subject-matter disciplines, for the study success were not fulfilled,.
For safety assessment of the pivotal studies Hib-MenCY-TT-009/010 and Hib-MenCY-TT-011/012 results, the applicant presented only descriptive analyses (sometimes adjusting for the factor “country”) demonstrating that there were no meaningful differences between HibMenCY-TT and Hib vaccines with regard to safety. The statistical analyses were performed for each study separately and then for the pooled data from different studies. However, issues regarding comparability of studies (due to the use of different protocols and different study populations) should be considered.
A major concern relating to the applicant’s safety assessment was that no clear, unbiased, comprehensive evaluation of the safety profile of four doses of the Hib-MenCY-TT (MenHibrix) vaccine for infants and toddlers was presented. Statistical analyses taking into account the longitudinal structure of safety data, missing data, and influence of some covariates like “country” and “medication used” could reduce possible biases and improve the precision of the results. Therefore, such statistical analyses should be applied to infants’ safety data, and the applicant was asked to perform this type of statistical safety analysis and supply the results obtained.
Summary of the applicant’s responses evaluation
The applicant provided a Complete Response on April 15, 2011. All the reviewer’s statistical requests were addressed. The following materials and information were provided:
- The updated immunogenicity datasets for the pivotal clinical trial(s)
- The main SAS programs and macros
- The statistical results on the:
- influence of assay variability on statistical results
- influence of the factors “race,” “center,” and “gender” on the immune responses to MenHibrix vaccination.
- A sensitivity analysis related to the missing immunogenicity data after the fourth dose of MenHibrix vaccine
- An analysis checking influence of factors “country” and “use of concomitant medication” on occurrence of adverse events
- An explanation of the enrollment structure by center.
According to the discussions with CBER CMC and the assay statistical reviewer, the applicant so far has not addressed a number of important issues, in particular issues related to the performance of serum bactericidal assay using human complement (hSBA assay). Until the applicant resolves those issues, it is not possible to evaluate completely the immune responses to MenHibrix vaccine.
As per the assay statistical reviewer, Dr. Tsai-Lien Lin, the hSBA assay stability issue has not been addressed adequately. Since there are no clinical efficacy data, the hSBA assay measurements constitute the basis for the effectiveness evaluations, and the quality of assay performance can have influence on the conclusion drawn from immunogenicity data, the statistical reviewer defers stating final recommendations regarding this BLA until the applicant’s response (related to the hSBA assay) to the CR2 letter has been evaluated.