Precision Dose-finding Cancer Clinical Trials in the Setting of Broadened Eligibility
Broadening eligibility criteria in cancer trials has been advocated to represent the true patient population more accurately. While the advantages are clear in terms of generalizability and recruitment, novel dose-finding designs are needed to ensure patient safety. These designs should be able to recommend precise doses for subpopulations if such subpopulations with different toxicity profiles exist. While dose-finding designs accounting for patient heterogeneity have been proposed, all existing methods assume the source of heterogeneity is known and thus pre-specify the subpopulations or only allow inclusion of a few patient characteristics. We propose a precision dose-finding design to address the setting of unknown patient heterogeneity in phase I cancer clinical trials where eligibility is expanded, and multiple eligibility criteria could potentially lead to different optimal doses for patient subgroups. The design offers a two-in-one approach to dose-finding by simultaneously selecting patient criteria that differentiate the maximum tolerated dose (MTD) and recommending the subpopulation-specific MTD if needed, using marginal models to sequentially incorporate patient covariates. Our simulation study compares the proposed design to the naive approach of assuming patient homogeneity and our design recommends multiple doses when heterogeneity exists and a single dose when no heterogeneity exists. The proposed dose-finding design addresses the challenges of broadening eligibility criteria in cancer trials and the desire for a more precise dose in the context of early phase clinical trials.