31 – Power and Sample Size
Simulations on Sample Sizes and Powers Calculations for a Registry Vaccine Study
John Liang
Pfizer
Scott Patterson
P�zer
World Health Organization (WHO) guidance was issued in 2009 covering Recommendations to assure the quality, safety and efficacy of pneumococcal conjugate vaccines. From the new WHO guidance (at section C2.2.3, page 27), two primary co-endpoints are recommended: "The endpoints used in the primary analysis should be the percentage of subjects with IgG =0.35 �g/mL AND the serotype-specific IgG GMC (Geometric Mean Concentration) ratios". The purpose of this work is to show simulations to define sample size and power calculation for a study in support of WHO registration requirements accounting for multiplicity across serotypes and across endpoints. A phase 3, parallel-group, randomized, active-controlled, double-blinded trial was to be designed to evaluate immunogenicity of Test Vaccine as compared with Control Vaccine. There were 3 infant series vaccinations and 1 toddler dose vaccination. The IgG data collected after 3 infant series vaccinations are in interest for primary endpoints. Multiplicity across endpoints was controlled by Bonferroni adjustment (i.e. 5% type 1 error was divided equally across endpoints.