A basket trial design for dose optimization based on a partial pooling hierarchical Bayesian model using model averaging

Music City Center 

The FDA initiated the project Optimus and issued the draft guidance for dose optimization. The FDA recommends randomized parallel dose-response cohorts to generate additional data at promising dose levels and implies that different dosages may be needed for different indications. In addition to dose optimization, with the recent advancements in precision medicine and cancer biology, the development of cancer treatments has shifted toward the search for agents targeted to specific molecular profiles that may appear in more than one type of cancer. The basket trial is a novel clinical trial design that enables the simultaneous assessment of a new treatment in multiple indications, defined as baskets. Concerning the dose optimization perspective by the FDA and the recent trend of basket trials in early-phase clinical trials, this paper proposes a dose-ranging basket trial design based on a partial pooling hierarchical Bayesian model considering efficacy and toxicity outcomes, where the indications and dose level define baskets. Our simulation study shows that the proposed approach correctly selected the optimal doses and declared the efficacy at each basket compared to the other approaches.

dose optimization

basket trial

hierarchical Bayesian model

model averaging

heterogeneity