A Bayesian Oncology Dose-Ranging Study Design

• Activity Number: 293
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Bo Ma*+ and Ohad Amit and Yuehui Wu
• Companies: GlaxoSmithKline and GlaxoSmithKline and GlaxoSmithKline
• Keywords: dose ranging study ; oncology ; bayesian

Abstract

In traditional oncology dose ranging studies, frequentist statistics are routinely used, and the most common primary efficacy endpoint is the binary response. We designed a Bayesian dose ranging study with decision rules using posterior probability based on both efficacy and safety endpoints. Prior study information was incorporated through prior specification. Innovative co-primary efficacy endpoints, binary response along with continuous tumor size change, were used to increase statistical efficiency of the efficacy endpoint and maximize the opportunity to discriminate the doses. Interim analysis was built into the study to enable the study stop if there is sufficient discrimination between doses. This randomized, double-blind study will be conducted in approximately 150 subjects. Eligible subjects will be randomized in a 1:1 ratio to receive one of the 2 dose levels. An interim analysis will be performed among the first 50 subjects. Bayesian decision rules will be used to evaluate the relative risk-benefit of the 2 dose levels at both the interim and final analysis. Statistical simulation was conducted to evaluate the operating characteristics of the study design.