Futility Boundary Design Based on Probability of Success Under New Drug Development Paradigm

• Activity Number: 113
• Type: Topic Contributed
• Date/Time: Monday, August 4, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Yijie Zhou*+ and Ruji Yao and Bo Yang and Ram Suresh
• Companies: AbbVie and Merck and AbbVie and GlaxoSmithKline
• Keywords: futility ; magnitude of point estimate ; probability of success

Abstract

Statistical significance has been the traditional focus of clinical trial design. However, an increasing emphasis from the medical and payer perspective arise on the magnitude of treatment effect based on point estimates. The magnitude of point estimates to demonstrate sufficient medical value is typically larger than that to demonstrate statistical significance. Therefore, a new clinical trial design needs to take into account the magnitude of point estimates.

Futility analysis is commonly used in clinical trial design, and traditional futility boundaries are designed based on power or conditional power to preserve the probability of achieving statistical significance at the end of a trial. With the additional trial objective for a sufficiently large point estimate, we propose a new futility analysis design approach where futility boundaries are selected to preserve the probability of observing a sufficiently large point estimate of treatment effect. We denote such probability using the terminology "probability of success" and the relative preservation of this probability as "pseudo-power". Simulation is conducted to evaluate the operational characteristics of this approach.