Adaptive Design and Personalized Medicine: The Future Is Now

• Activity Number: 97
• Type: Other
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: ASA
• Author(s): Donald Berry*+
• Companies: MD Anderson Cancer Center
• Keywords: Innovations in clinical trial design ; Biomarkers ; Longitudinal modeling ; Simulations

Abstract

Modern clinical trials have served society well. They tend to be simple, with rigid designs and usually two arms. A disadvantage is that progress in developing therapies is slow and expensive. The success rate of phase III trials remains unacceptably low. And addressing a single question makes it difficult to learn about which patients benefit from which therapies. In some diseases we overtreat the majority to benefit the few . the problem is that we don't know which few.

Adaptive designs make trials more informative by enabling asking many questions. Adaptive designs use the accumulating data to modify the trial's course. For example, a single trial might identify the appropriate patient population, dose and regimen, and therapeutic combinations, and then switch seamlessly into a phase III confirmatory mode. Of special importance for personalized medicine is considering many therapies in the same trial, using accruing data to gradually matching patient subtypes with their most appropriate therapies. I'll describe building adaptive designs, including the role of the Bayesian predictive probabilities and the roles of simulations and longitudinal modeling for complicated designs.