A new Bayesian dose-finding design for drug combination trials

摘要

Combining drugs based on dosage limits has many advantages for curing malignant tumors. The problem of finding the maximum tolerated dose with targeted dose-limiting toxicity rate in drug combination trials is considered. A new Bayesian adaptive dose-finding design is proposed for such drug combination trials. The proposed Bayesian approach first adaptively models the local behavior of the dose response surface along the search path so that the model gains more efficiency after weighting response by distance. The method uses noninformative prior rather than informative ones. Methodology and algorithmic representation of the Bayesian adaptive modeling are presented in detail. In order to demonstrate the superiority of the proposed approach over to that of some existing methods, a simulation study is conducted. Based on the results it is observed that the proposed Bayesian learning procedure is more robust and efficient.