658 – Statistical Issues Specific to Therapeutic Areas II
Dose Finding in Early Phase I/II Cancer Clinical Trial Using Drug Combinations of Cytotoxic Agents
Mourad Tighiouart
Cedars Sinai Medical Center
Quanlin Li
Cedars Sinai Medical Center
André Rogatko
Cedars Sinai Medical Center
We present a two-stage dose finding phase I/II design of a combination of two drugs with continuous dose levels in early phase cancer clinical trials. The goal is to estimate dose combination regions that are tolerable and with a desired level of efficacy. In the first stage of the design, the relationship between doses and the probability of dose limiting toxicity (DLT) is modeled parametrically and the design proceeds using conditional escalation with overdose control. At the end of stage I, the maximum tolerated dose (MTD) curve is estimated as a function of Bayes estimates of the model parameters. In the second stage of the design, we investigate some parametric models to link the probability of treatment efficacy with dose combinations along the MTD curve. We propose a Bayesian adaptive design for conducting the phase II trial with the goal of determining dose combination regions along the MTD curve with a desired level of efficacy. The methodology is evaluated by presenting the operating characteristics under different scenarios for the true probability of treatment efficacy as a function of dose combinations on the MTD curve and trial sample size.