SESSION L

Master protocol trials can markedly improve the efficiency of oncology and rare disease drug development by evaluating a therapy across multiple molecularly defined cohorts or/and evaluating multiple therapies for the same disease within a single protocol. While their benefits are well established conceptually, the next frontier is scaling these advantages to broader, high-impact scenarios through rigorously designed trials in both exploratory and confirmatory settings with the right context. This tutorial will translate conceptual frameworks into practical know-how, covering design choices, statistical operating characteristics, operational execution, the trial results interpretations, and the considerations for maximizing regulatory acceptance. Both frequentist and Bayesian frameworks will be elaborated in different design settings.

What you will learn:

• Core concepts and taxonomy: Understand the differences between exploratory versus confirmatory basket and platform trials, histology-agnostic versus histology-restricted cohorts, and single-arm versus randomized designs.
• Design elements for master protocol trials: Learn how to select cohorts, the recommended number of cohorts in the baskets, and set early stopping rules; compare tumor-agnostic decision frameworks, Bayesian versus frequentist approaches in information borrowing, multiplicity control for platform trials and usage of non-concurrent controls for platform trials.
• Statistical rigor for decision-making: Understand methods to ensure robust inference: type I error control, operating characteristics, hierarchical modeling and partial pooling, pre-specification of decision rules to fit different decision-making scenarios.
• Path to confirmatory evidence and regulatory considerations: Explore how to transition from signal detection to confirmatory trials. Identify the right development context for randomized confirmatory master protocol trials, the potential concerns from regulatory and mitigations to the potential regulatory risks.
• Case studies and pitfalls: Examine case studies illustrating efficient signal detection and a simulated case study in the confirmatory setting, successful real-world examples and discuss common pitfalls and how to avoid them. Provide tips and techniques for practical considerations regarding the case studies.

By the end of the session, participants will be able to design fit-for-purpose master protocol trials, articulate when and how to introduce randomization and controls for both basket and platform settings and implement statistical and operational practices that maximize patient benefit while generating robust evidence for meaningful decision-making.

Instructors’ Biography:

Robert A. Beckman, MD is Professor of Oncology and of Biostatistics, Bioinformatics, and Biomathematics at Georgetown University Medical Center and Scientific Advisor to the Vice President for Research. He is an oncology clinical researcher and mathematical biologist specializing in cancer evolution, precision medicine, and innovative clinical trial design.  Dr. Beckman has held leadership roles in oncology clinical research across multiple pharmaceutical companies, contributing to bringing numerous therapies into development and to market, and has co‑invented novel biomarker‑driven and adaptive trial strategies.

Cong Chen, PhD is Scientific Associate Vice President (AVP) in Early Development Statistics at Merck & Co., Inc., leading statistical strategy for oncology early and transitional development programs.He specializes in innovative clinical trial design and decision‑making strategies and has supported major oncology programs, including playing a key role in the development and regulatory acceleration of pembrolizumab (KEYTRUDA).  Dr. Chen is a Fellow of the American Statistical Association, a leader in industry working groups on innovative design, and has published extensively on clinical trial methodology with over 100 papers and book chapters