A Consistency-Adjusted Strategy for Testing Alternative Endpoints in a Clinical Trial

• Activity Number: 132
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Mohamed A. Alosh*+
• Companies: U.S. Food and Drug Administration
• Keywords: Alternative endpoints ; subgroups ; consistency ; dependency ; study power

Abstract

A clinical trial might involve more than one clinically important endpoint (subgroup) each of which can characterize the treatment effect of the experimental drug under investigation. For prespecifeid alternative endpoints (subgroups) there are several approaches which can be used for testing for efficacy for the alternative endpoints or the subgroup and total study population. Traditional multiplicity approaches use constant significance levels for these alternative endpoints. However, some recent multiplicity strategies allow the alpha-level allocated to testing subsequent endpoint to be dependent on the results of previous endpoint. In this presentation we discuss the need for establishing a minimum level of efficacy for the previous endpoint before proceeding to test for the subsequent alternative endpoint (subgroup) so that potential problems in interpreting study findings can be avoided. We consider implementing such requirements, called consistency criteria, along with adaptation of the significance level for subsequent endpoints at the study design stage and investigate its impact on study power. In addition, we consider its application to actual clinical trial data.