• Thomas M Braun, Peter F Thall, Hoang Nguyen, and Marcos de Lima.
• Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, MI, USA. tombraun@umich.edu
• Clin Trials. 2007 Jan 1; 4 (2): 113-24.

BackgroundTraditionally, phase I clinical trial designs are based upon one predefined course of treatment while varying among patients the dose given at each administration. In actual medical practice, patients receive a schedule comprised of several courses of treatment, and some patients may receive one or more dose reductions or delays during treatment. Consequently, the overall risk of toxicity for each patient is a function of both actual schedule of treatment and the differing doses used at each adminstration.PurposeOur goal is to provide a practical phase I clinical trial design that more accurately reflects actual medical practice by accounting for both dose per administration and schedule.MethodsWe propose an outcome-adaptive Bayesian design that simultaneously optimizes both dose and schedule in terms of the overall risk of toxicity, based on time-to-toxicity outcomes. We use computer simulation as a tool to calibrate design parameters.ResultsWe describe a phase I trial in allogeneic bone marrow transplantation that was designed and is currently being conducted using our new method. Our computer simulations demonstrate that our method outperforms any method that searches for an optimal dose but does not allow schedule to vary, both in terms of the probability of identifying optimal (dose, schedule) combinations, and the numbers of patients assigned to those combinations in the trial.LimitationsOur design requires greater sample sizes than those seen in traditional phase I studies due to the larger number of treatment combinations examined. Our design also assumes that the effects of multiple administrations are independent of each other and that the hazard of toxicity is the same for all administrations.ConclusionsOur design is the first for phase I clinical trials that is sufficiently flexible and practical to truly reflect clinical practice by varying both dose and the timing and number of administrations given to each patient.

18 keywords...

hide…