Robust Optimal Interval Design with an Application to Dose-Finding in Drug-Combination Trials

• Activity Number: 546
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Ruitao Lin* and Guosheng Yin
• Companies: The University of Hong Kong and The University of Hong Kong
• Keywords: Dose finding ; drug combination ; interval design ; maximum tolerated dose ; random walk ; robustness

Abstract

Many statistical methods have been developed for dose finding in phase I clinical trials, which can mainly be classified as model- and algorithm-based approaches. Most of the methods require that practitioners prespecify several trial parameters, which, however, could be arbitrary and thus result in sensitiveness of the trial performance. To overcome the burden of design prespecification, we propose a robust optimal interval (ROIN) design to locate the maximum tolerated dose (MTD) in dose-finding trials. The optimal interval is determined by minimizing the probability of the average incorrect decisions under the Bayesian paradigm. Our method requires the minimum design specification of the target toxicity rate only; neither does it impose any parametric assumption on the underlying distribution of the toxicity curve, nor it needs prespecification of any other design parameters. In the application to drug-combination trials, we develop random-walk and parallel-crossing ROIN designs to identify the MTD combination. The proposed designs are simpler and much easier to implement, while their performances are competitive and robust.