Interim OS Project
Interim Overall Survival (OS) Project
Can data-driven models improve the design and interpretation of early clinical trial results?
What is the unmet need and why does it matter?
Overall survival (OS) serves as the gold standard endpoint in oncology drug development. New therapies that extend survival remain a key goal to improve patient outcomes. Although longer survival represents critical progress for patients, clinical trials measuring OS often take longer to complete and, therefore, longer to generate data needed to support regulatory approval and patient access to new therapies.
To address this, endpoints that can be measured earlier, such as response rate and progression free survival (PFS), are used to assess treatments in oncology clinical trials, often through the Accelerated Approval pathway. These endpoints play an increasingly important role in expediting approval of promising therapies; however, when these earlier endpoints are ready to be assessed, early (“interim”) OS results may also be evaluated. Interpreting interim OS data can be challenging due to small event numbers, limited follow-up, and trial dynamics such as treatment crossover, creating uncertainty about a drug’s true benefit-risk profile.
How are we helping to find solutions?
Friends is addressing the challenges in interpreting interim OS data by bringing together experts from across the oncology community to develop frameworks and simulation-based approaches that enable careful design and comprehensive analysis of interim OS data. This work aims to help ensure that oncology trials can better balance early efficacy signals with expected long-term survival outcomes.
How does this impact patients?
Rigorous interim OS evaluations can support earlier access to effective drugs and help when a drug may be causing harm or when there is uncertainty around its benefit. However, when interim OS evaluations are not well-planned and carefully executed, results may be inconclusive, limiting their ability to inform timely and confident decisions. By addressing challenges in interpreting interim OS data and establishing clear frameworks for assessment, we can ensure that interim OS results are actionable, ultimately supporting safer, faster progress for patients.
Project Outcomes
2025
- Friends launched an initiative to explore how modeling and simulation can provide evidence-based frameworks for interpreting interim OS data and develop best practices to support consistent interpretation of interim OS in regulatory decision-making.
2024
- Friends 2024 Annual Meeting included a white paper and panel discussion that outlined key considerations for improving the analysis and interpretation of interim OS data in oncology clinical trials.
Project Partners
Amgen, AstraZeneca, Bayer, Crossbow Therapeutics, Daiichi Sankyo, Eli Lilly & Co., EMD Serono, Fred Hutchinson Cancer Center, Genentech, GSK, Johnson & Johnson Innovative Medicine, MMS Holdings, NMD Group, Novartis, Pfizer, U.S. Food and Drug Administration (FDA).