Bringing Innovation in Bioequivalence Studies: Evaluating Multiple Test Formulations in a Single Crossover Pivotal Bioequivalence Study Using Multiplicity Correction

Abstract

ICH M13A recognizes higher‐order bioequivalence (BE) studies involving multiple test or comparator products. When any one of several test formulations may independently demonstrate BE and support subsequent selection, multiplicity adjustment is required to control the family‐wise type I error rate. This requirement introduces a practical penalty through increased sample‐size requirements, wider confidence intervals, and a narrower range of test‐to‐reference geometric mean ratios that would result in BE conclusions. Although the statistical foundations of multiplicity correction are well established, the available literature has focused predominantly on methodological procedures, with limited consideration of the practical questions encountered during pharmaceutical product development. This review briefly summarizes the regulatory and statistical background and presents a scenario‐based evaluation of the consequences of multiplicity in multi‐formulation crossover BE studies. Analytical assessments were used to quantify sample‐size inflation, confidence interval widening, and contraction of the T/R range that would result in a BE conclusion. Simulated BE study datasets complemented these evaluations, analyzed using full‐data pooled‐MSE (mean squared error) and separate pairwise ANOVA (analysis of variance) approaches and four multiplicity strategies: no adjustment, Bonferroni, Holm, and Hochberg for multiplicity adjustment. The review provides a practical framework for balancing statistical penalty against the value of preserving the ability to retain more than one formulation candidate within a single pivotal study or development. The present evaluation is restricted to non‐replicate crossover BE studies.