Stakeholders’ Reactions to the 2025 FDA Draft Guidance on Overall Survival in Oncology Drug Approvals

Abstract

Overall survival (OS) and quality of life (QoL) are widely recognised as the most robust endpoints for capturing meaningful clinical benefit in oncology. Yet the heterogeneity of oncology settings, characterised by long post-progression survival, multiple subsequent therapies, crossover practices and rare or indolent diseases, often makes timely assessment of these endpoints infeasible. As a result, reliance on surrogate endpoints has increased in recent years, raising concerns about the rigour and reliability of the evidence supporting oncology drug approvals.

On 18 August 2025, the U.S. Food and Drug Administration (FDA) released a draft guidance proposing a strengthened role for OS in oncology trials, signalling a potential shift in regulatory expectations. The draft was then opened for a 60-day public consultation, during which stakeholders could provide feedback. Here, we examine 24 formal and informal stakeholder submissions obtained through the FDA consultation website and professional networks, using qualitative thematic analysis to understand how the oncology community perceived the feasibility of the draft guidance. We found broad agreement that OS is the most clinically meaningful endpoint; however, stakeholders widely argue that powering trials for OS is impractical in many modern oncology settings. Challenges related to trial design, statistical complexity (including non-proportional hazards and intercurrent events) and the risk of misleading interim OS signals were identified as major barriers to implementation.

These findings suggest that, while the draft guidance is perceived as clinically well-motivated, its operational feasibility remains uncertain. The wide range of concerns raised by different stakeholders, alongside the diversity of oncology settings, underscores the need for clearer, context-specific recommendations. Continued engagement between the FDA and the oncology community will be essential to ensure that an eventual final guidance reinforces the central role of OS without compromising feasibility, ethical standards or timely patient access to innovation.