A0644
Title: Posterior predictive design for phase I clinical trials
Authors: Shouhao Zhou - Penn State University (United States) [presenting]
Abstract: Model-assisted designs are cutting-edge adaptive designs to find the maximum tolerated dose (MTD) in phase I clinical trials. They enjoy superior performance compared to more complicated, model-based adaptive designs, but with their decision rule pretabulated, they can be implemented as simply as the conventional algorithmic designs. We propose the posterior predictive (PoP) design, a novel model-assisted design based on Bayesian interval hypothesis testing for dose escalation and de-escalation. The work moves beyond the previous model-assisted designs by theoretically achieving consistency in selecting the true MTD and global optimality in dose transition. The simulation studies demonstrate that the PoP design can achieve significant improvement in operating characteristics to identify the MTD, thereby providing a useful upgrade to the prevalent model-assisted designs.
