FDA–AACR Strategies for Optimizing Dosages for Oncology Drug Products: Selecting Dosages for First-in-Human Trials

Dosage optimization has become a focus in oncology drug development, as highlighted by recent FDA initiatives, including Project Optimus. Traditionally, most oncology drug development programs identify a maximum tolerated dosage and advance this dose in subsequent clinical trials and premarket applications. This approach has been routinely applied for cytotoxic chemotherapeutics, in which higher dosages generally yield more efficacy as well as toxicity. However, it is less suited for the more targeted pharmacology of modern oncology drugs, in which excessive escalation may only add additional toxicity. Instead, paradigms that utilize the totality of data accumulated throughout drug development can better determine optimized dosages that minimize the risk of underdosing, leading to exposure to subtherapeutic dosages, and overdosing, leading to unnecessary toxicities. Appropriate selection of dosing in first-in-human (FIH) trials is crucial, as it facilitates the efficient identification of optimized doses for subsequent trials. Nonclinical research and clinical data from previous trials can inform both FIH dosage selection and trial design. When background data are lacking, modeling and simulation techniques have been developed to integrate information to determine a rational starting dose. Additionally, innovative model-informed clinical trial designs allow for statistically guided dose escalation and recommendations and can be updated in real time to maximize potential patient benefit within the FIH trial. Unfortunately, these techniques remain underutilized. In this article, in a series of three discussing innovative strategies for dosage optimization, we highlight expectations and provide suggestions for the future of dosage selection and optimization in FIH oncology trials.