Studies of equivalence in clinical vaccine research

The development of combination vaccines as well as the improved manufacture of vaccines have resulted in the need for clinical studies seeking equivalence or non-inferiority as investigators seek to demonstrate that combination vaccines achieve the same levels of efficacy, immunogenicity, and safety as their individual counterparts. Given the nature of the statistical analysis, studies of equivalence require particular attention to sample size. The current double-significance method of Neyman-Pearson attempts to address problems with equivalence testing, but it leads to very large sample sizes and illogical results. With such large samples, one can find a clinically trivial difference that is still statistically significant. The late Professor Alvan R. Feinstein proposed a more logical approach that would call for smaller, more workable sample sizes. Understanding the issues involved in sample size calculations for such studies is important to those who design clinical vaccine studies. The implications of the calculations will have far-reaching effects on the feasibility of the study such as availability of subjects, the success with recruitment, and the overall expenses. In fact, the feasibility issues may prevent the study being undertaken at all.