Control of FWER Using Close Testing for 3-Arm PK Similarity Trials

• Activity Number: 335
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Steven Y. Hua*+ and Ray Li and Kerry Barker and Siyan Xu and Christy Chuang-Stein and Shan Mei Liao
• Companies: Pfizer and Pfizer and Pfizer and and Pfizer and Pfizer
• Keywords: Biosimilarity ; Closed test ; FWER ; 3-arm pairwise comparisons ; Multiplicity ; PK similarity

Abstract

A typical pharmacokinetic (PK) similarity trial consists of a biosimilar product and two regionally sourced versions of the same innovator biologic drug. PK similarity is established if AUC (area under the concentration-time curve) and Cmax (maximum concentration) are equivalent across three products under the same acceptance criteria for the bioequivalence of two drugs.

Control of family wise error rate (FWER) arises in pairwise comparisons for each of the 3 parameters (AUC0-T, AUC0-inf and Cmax) as well as across these 3 correlated parameters. Although no currently available standards to address this issue, there are different ways to consider. For example, control of FWER can be made only for the 3 pairwise comparisons within each parameter, or for the 6 pairwise comparisons for the two parameters: AUC0-T (or AUC0-inf) and Cmax.

In this paper, we consider a p-value based closed test involving multiple pairwise comparisons within each parameter and of multiple parameters in a 3-arm PK similarity trial, for which we control the FWER using closure testing principle. Operating characteristics of this and other methods are compared using a clinical trial data set.