BOIN-ET: Bayesian Optimal Interval Design for Dose Finding Based on Both Efficacy and Toxicity Outcomes

• Activity Number: 674 - Adaptive Design - 3
• Type: Contributed
• Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Kentaro Takeda* and Masataka Taguri and Satoshi Morita
• Companies: Astellas Pharma Global Development, Inc. and Yokohama City University and Kyoto University
• Keywords: Bayesian adaptive design; dose finding; interval design; efficacy-toxicity; optimal dose

Abstract

One of the main purposes of a phase I dose-finding trial in oncology is to identify an optimal dose (OD) that is both tolerable and has an indication of therapeutic benefit for subjects in subsequent phase II and III trials. Many phase I dose-finding methods based solely on toxicity considerations have been proposed under the assumption that both toxicity and efficacy monotonically increase with the dose level. Such an assumption may not be necessarily the case, however, when evaluating the optimal dose for molecular targeted, cytostatic, and biological agents, as well as immune-oncology therapy. To address this issue, we extend the Bayesian optimal interval (BOIN) design, which is nonparametric and thus does not require the assumption used in model-based designs, in order to identify an OD based on both efficacy and toxicity outcomes. The new design is named "BOIN-ET" (Takeda, Taguri, and Morita, 2018). A simulation study shows that BOIN-ET has advantages than the model-based approaches in both the percentages of correct ODs selected and the average number of patients allocated to the ODs across a variety of realistic settings.