Evaluation of Multiplicity Control Strategies for a New Study with Multiple Endpoints and Two Doses

• Activity Number: 371
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Hua Guo* and Kaifeng Lu
• Companies: Actavis and
• Keywords: Clinical trials ; Multiplicity ; Gatekeeping ; Family-wise error rate

Abstract

In phase III confirmatory trials, it is typical to have both a primary endpoint and a secondary endpoint to characterize the treatment effect for a regulatory claim with two active doses of study drug compared to placebo. Rigorous statistical methods are available to provide strong control -of the study wise type I error rate. Typically, the secondary endpoint is under powered compared with the primary endpoint. We conducted a set of simulations in this setting to compare the performance of several popular methods (including graphic approach, truncated Hochberg method, matched gate keeping method, and gatekeeping testing via adaptive alpha allocation etc.), and then provide recommendations for this type of study design. Other eligible designs are evaluated as well to evaluate the potential savings in sample size.