Probability of Statistical Success of Two Trials Given Observed Data

• Activity Number: 473
• Type: Contributed
• Date/Time: Wednesday, August 3, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Jianliang Zhang*+ and Jenny J. Zhang
• Companies: MedImmune, Inc. and U.S. Food and Drug Administration
• Keywords: Predictive power ; probability of statistical success ; phase III trials

Abstract

In drug development, after completion of phase II proof-of-concept trials, the sponsor needs to make a go/no-go decision to start expensive phase III trials. The probability of statistical success (PoSS) of the phase III trials based on available data from earlier studies is an important factor in that decision-making process. Instead of statistical power, the predictive power of a phase III trial, which takes into account the uncertainty in the estimation of treatment effect from earlier studies, has been proposed to evaluate the PoSS. While the predictive power for a single trial has been extensively studied, statistical methods for the evaluation of the joint predictive power as a measure of the joint PoSS of two or more Phase III trials has not yet been seen in published literature to the best of our knowledge. We show that the predictive powers of two future trials are not statistically independent. Thus, the joint predictive power should be evaluated based on the joint distribution of the predictive statistics, and not as a simplistic product of the predictive powers of the individual trials. Relevant formulae are developed and numerical examples are provided.