Testing Multiple Endpoints in Multiple Prespecified Subgroups Using Group Sequential Procedures

• Activity Number: 44
• Type: Contributed
• Date/Time: Sunday, August 9, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Christine Gause* and Keaven Anderson
• Companies: Merck and Merck
• Keywords: Multiplicity ; Subgroup analysis ; Nested hypotheses ; group seqeuential ; information-based

Abstract

Testing for a pre-specified subgroup in clinical trials is often of interest if it is possible that only a portion of the study population may benefit from the therapy under study. We consider the problem of testing multiple endpoints in multiple pre-specified subgroups using graphical methods for multiplicity adjustment extended to group sequential designs by Maurer and Bretz (2013). We will address the problem of an unknown amount of statistical information for some of the endpoints at interim or final analyses. This work will be extended to the case of early stopping in a pre-specified subgroup of an overall population where testing in the overall population remains of interest. A separate approach considering group sequential methods to account for the correlation between a pre-specified subgroup and overall population proposed by Spiessens and Debois (2010) will be extended to include interim analyses. A simulation study comparing the approaches will be presented, and illustrative examples will be provided.