Dose-Seeking Phase I Trials for Currently Approved Molecular-Targeted Therapies in the USA: The Dose-Limiting Toxicity Definition Issue

Background: There is substantial evidence that classically used toxicity-driven dose-escalating phase I trials are not optimal for defining the recommended phase II dose for molecular-targeted therapies. Objective: This study aimed to assess the actual methodology used for phase I trials of approved molecular-targeted therapies for solid tumors in the USA. Methods: We evaluated the designs and endpoints used in 53 single-agent dose-seeking phase I trials that were published between 2001 and November 2015, investigating US Food and Drug Administration (FDA)-approved molecular-targeted therapies for solid tumors (n = 28). Results: In all but three cases, the trials used dose-escalating designs that were toxicity-driven (50 trials, 94 %). The “3+3 design” was used in 25 trials (47 %). In 47 trials (89 %), dose-limiting toxicities were assessed during the first 28 days; the definitions of dose-limiting toxicities were similar to those used in cytotoxic drug trials (Grade 4 neutropenia, febrile neutropenia, thrombocytopenia with hemorrhage). Conclusions: In most trials, the dose-limiting toxicity definition did not specifically address the expected side effects related to the mechanisms of action of the molecular-targeted therapy, the expected side effects of which differ significantly from cytotoxic agents. © 2016, Springer International Publishing Switzerland.