|
Abstract:
|
We present a two-stage dose finding phase I/II design of a combination of two drugs with continuous dose levels in 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 or continual reassessment method. 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 a parametric model 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 some simulations and application to a real trial.
|
