IOL: Statistical and Design Considerations for Dose Optimization

The US FDA initiated Project Optimus to reform dose optimization and shift the dose selection paradigm from identifying the maximum tolerated dose to the optimal biological dose in oncology drug development. This shift presents significant challenges in the design and implementation of dose-finding trials. This Introductory Overview Lecture will review statistical considerations and design strategies to address these challenges. Efficacy-integrated and two-stage design strategies will be discussed, illustrated with model-based and model-assisted designs. Practical tools, such as backfill, utility for efficacy/toxicity tradeoff, and more complex settings, such as combination trials, will also be discussed. Real-world examples and easy-to-use software will be provided to empower participants to grasp the underlying statistical principles and design efficient dose optimization trials.