Design and Comparison of Adaptive Clinical Trials Using Simulation

• Activity Number: 58
• Type: Topic Contributed
• Date/Time: Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): William Prucka*+
• Companies: Eli Lilly and Company
• Keywords: clinical trial design ; clinical trial simulation ; adaptive designs

Abstract

In drug development, clinical trials are designed to address many questions. For example, does the experimental drug have a benefit? If so, what dose and patient population provides a meaningful benefit that outweighs risk? Depending on the context of the trial in the overall clinical plan, a trial may address only a subset of these questions, building upon learning from previous trials. To determine the best trial design for the context, large-scale simulation studies are used to compare the performance of multiple alternative designs across different scenarios of drug effect. Such simulation studies evaluate the benefits of adaptive designs; which often perform better than fixed trials across scenarios, since they learn from and adapt in response to emerging data. This paper focuses on simulation methods used to simultaneously optimize each alternative design option as well as enable meaningful quantitative comparisons across designs. Specific areas of focus are, 1) designing and assessing reliable success criteria (critical success factors), and 2) developing of performance metrics for comparing multiple designs. Illustrative examples from drug development are provided.