ABSTRACT
ABSTRACT Bioequivalence (BE) of two drugs is usually demonstrated by rejecting two one-sided null hypotheses using the two one-sided tests (TOSTs) for the primary metrics: area under the concentration-time curve (AUC) and maximum concentration (Cmax). The decision rule for BE often requires equivalence to be achieved on both metrics that contain four one-sided null hypotheses together; without adjusting for multiplicity, the family-wise error rate (FWER) could rise above the nominal type I error rate α. In this chapter, we propose two multiplicity adjustments in testing for BE, including a closed-test procedure that controls FWER by treating the two
CONTENTS
8.1 Introduction................................................................................................ 156 8.1.1 Motivation....................................................................................... 156 8.1.2 Outline............................................................................................. 157
8.2 Closed Test Procedure in Bioequivalence Using Two One-Sided Test............................................................................................ 158
8.3 Analysis of Two Datasets.......................................................................... 161 8.3.1 Test of Equivalence for Each Parameter Using
Two One-Sided Test ....................................................................... 162 8.3.2 Bayesian Analysis .......................................................................... 162 8.3.3 Closed Test ...................................................................................... 163
8.4 Alpha-Adaptive Sequential Testing Procedure Using Two One-Sided Test ................................................................................... 166
8.5 Control of Family-Wise Error Rate in Alpha-Adaptive Sequential Procedure ................................................................................ 172 8.5.1 Null Hypotheses Are True for Both ln(AUC) and ln(Cmax) ...... 173 8.5.2 Null Hypotheses Are True for ln(AUC) but False for ln(Cmax).... 173 8.5.3 Null Hypotheses Are False for ln(AUC) but True for ln(Cmax).....174
