Adaptive Biomarker Population Selection in Phase III Confirmatory Trials

• Activity Number: 62
• Type: Topic Contributed
• Date/Time: Sunday, August 9, 2015 : 4:00 PM to 5:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Xiaoyun Li* and Cong Chen
• Companies: Merck and Merck Research Laboratories
• Keywords: Enrichment design ; biomarker design ; oncology clinical trials ; personalized medicine ; subgroup analysis ; time-to-event endpoints

Abstract

In contemporary cancer drug development, it is critical to find the right biomarker population for an experimental treatment. In this paper, we propose an enrichment design for confirmatory Phase III trials where non-performing biomarker subpopulation(s) are excluded from the final analysis. Unlike the previous work in literature whereas the same endpoint is used for population selection and final analysis, we propose the use of a sensitive intermediate endpoint (whenever available) for population selection. We demonstrate that the overall type I error can be well controlled in this case, and we will further show through a worked example that the use of a sensitive intermediate endpoint for population selection helps increase the study power.