Dose-finding approach for dose escalation with overdose control considering incomplete observations

Abstract

We propose a hybrid design, the time-to-event dose-escalation method with overdose control (TITE-EWOC), introducing the time-to-event approach, developed by Cheungit et al., in the EWOC method, developed by Babb et al. The aim of this new design is to decrease the dose-finding trial duration, without impairing the characteristics of the EWOC design, especially the overdose control ability. We conducted a simulation study, exploring four dose–toxicity relationships and three mean inter-patient arrival times. Performances of TITE-EWOC were compared with those of the EWOC method. This study shows that the trial duration can be greatly decreased with the TITE-EWOC, without impacting the proportion of overdosed patients or the number of dose-limiting toxicities by trial, for all explored dose–toxicity relationships, except for very short inter-patient arrival times. The ability of the method to find the true maximum tolerated dose remains unchanged.