On the Use of Historical Information in Selecting Design Type for Phase II Oncology Trials

• Activity Number: 470
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Yiyi Chen*+ and Rongwei (Rochelle) Fu and Zunqiu Chen
• Companies: Oregon Health & Science University and Oregon Health & Science University and Oregon Health & Science University
• Keywords: phase II clinical trials ; design ; historical data

Abstract

Some phase II oncology trials are single-arm trials while some other are randomized trials. Both the single-arm and the randomized trials have advantages and disadvantages. In practice, it is generally the clinical investigators who decide whether the trial should be a single-arm or a randomized trial. We argue that this is a critical design decision that should be made with a quantitative evaluation of the available trial resources, pre-trial information and uncertainty on unknown parameters. We propose a Bayesian method in finding the optimal design type that yields the most efficient resource allocation for a phase II oncology trial from three candidates: (1) single-arm design (2) 1:1 randomization design (3) 1:2 randomization design. The application of this method is illustrated by a real clinical trial.