Evaluating Food and Drug Administration approved gastrointestinal cancer drugs: clinical benefit and trial endpoints over the past two decades.

BACKGROUND: The increasing reliance on accelerated approvals and surrogate endpoints for Food and Drug Administration (FDA) approvals of gastrointestinal (GI) cancer therapies raises concerns about their clinical benefit and long-term patient outcomes. The shift toward single-arm trials in regulatory decisions further complicates treatment evaluation. MATERIAL AND METHODS: This retrospective observational study evaluated all FDA approvals for GI cancer therapies from January 2006 through January 2025. Data were extracted from FDA archives, ClinicalTrials.gov, and PubMed. Approvals were categorized by regulatory pathway (accelerated vs regular), trial design (single-arm vs randomized), and primary endpoint (surrogate vs overall survival [OS]). Clinical benefit was assessed based on OS improvement and the European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS). The primary outcome was the proportion of approvals based on surrogate versus OS endpoints. Secondary outcomes included use of single-arm designs, frequency of accelerated approvals, OS gain, and the proportion meeting ESMO-MCBS substantial benefit (score ≥ 4). RESULTS: The FDA granted 60 GI cancer drug approvals from 67 trials. Approvals rose from 15 (25%) in 2006-2014 to 45 (75%) in 2015-2025. Single-arm trials increased to 24%, and surrogate endpoints were used in 41.8% (ORR 20.9%, PFS 19.4%). Median OS improvement was 2.1 months (IQR: 1.6-2.65). Only 24.3% of trials met ESMO-MCBS substantial benefit. Of 15 accelerated approvals, 66.7% remained pending, 13.3% received full approval, and 20% were withdrawn. CONCLUSION: Expedited approvals have improved drug access in GI oncology, but modest benefits highlight the need to balance speed with outcomes that truly matter to patients.