Surv-CRM-12: A Bayesian Phase I/II Survival CRM for right-censored toxicity endpoints with competing disease progression

Abstract

The growing interest in new classes of anti-cancer agents, such as molecularly-targeted therapies (MTAs) and immunotherapies with modes of action different from those of cytotoxic chemotherapies, has changed the dose-finding paradigm. In this setting, the observation of late-onset toxicity endpoints could likely be precluded by trial discontinuation due to disease progression, defining a competing event to toxicity. Trial designs where dose-finding is modeled in the framework of a survival competing risks model appear particularly suited. We aimed to provide a phase I/II dose-finding design that allows the dose-limiting toxicity (DLT) outcomes to be delayed or unobserved due to competing progression within the possibly long observation window. The proposed design named the Survival-12-continual reassessment method (Surv-CRM-12), based on survival models for right-censored DLT and progression endpoints, aimed at identifying a dose that optimized the progression outcome among a set of acceptable doses. In this competing risks framework, cause-specific hazards for DLT and progression-free of DLT were considered, with model parameters estimated using Bayesian inference. The optimal dose is that dose level that minimizes the cumulative incidence of progression among an acceptable set of doses with a cumulative incidence of DLT below a target threshold. In a simulation study, design operating characteristics were evaluated and compared to the TITE-BOIN-ET design and a nonparametric benchmark approach. The performances of the proposed method were consistent with the complexity of scenarios as assessed by the nonparametric benchmark. We found that the proposed design presents desirable operating characteristics in selecting the optimal dose and safety. We propose a design for phase I/II trials targeting the subdistribution function of toxicity and progression for dose-finding, using working models for censored data. It allows prolonged observation windows resulting in administrative censoring and competing risks. Dose-finding; Phase I/II; Competing risks; Survival data; Oncology Les auteurs n'ont pas précisé leurs éventuels liens d'intérêts.