Background
Perioperative FLOT (fluorouracil, leucovorin, oxaliplatin, docetaxel) is a standard of care for resectable gastric/gastro-oesophageal junction (GEJ) adenocarcinoma. MATTERHORN is a global, randomized, double-blind, placebo-controlled phase 3 trial (147 centers, 20 countries; N=948) in treatment-naive, resectable stage II–IVa gastric/GEJ adenocarcinoma. The primary EFS analysis was previously positive; this report presents the key secondary endpoint of overall survival.
Results
Interventions and follow up: Arm A: Durvalumab 1500 mg IV q4w + FLOT (2 neoadjuvant + 2 adjuvant cycles), then durvalumab q4w for 10 additional cycles.
Arm B: Placebo + FLOT (identical schedule).
Primary endpoint: Event-free survival (EFS; previously reported).
Key secondary (reported here): Overall survival (OS).
mFollow up: 31.5 months.
Results: OS: HR 0.78 (95% CI 0.63–0.96); P=.021 (significance threshold P<.0499) — met.
EFS (prior): 2-yr 67.4% vs 58.5%; HR 0.71 (95% CI 0.58–0.86); P<.001.
pCR: 19.2% vs 7.2% (relative risk 2.69, 95% CI 1.86–3.90).
Arm B: Placebo + FLOT (identical schedule).
Primary endpoint: Event-free survival (EFS; previously reported).
Key secondary (reported here): Overall survival (OS).
mFollow up: 31.5 months.
Results: OS: HR 0.78 (95% CI 0.63–0.96); P=.021 (significance threshold P<.0499) — met.
EFS (prior): 2-yr 67.4% vs 58.5%; HR 0.71 (95% CI 0.58–0.86); P<.001.
pCR: 19.2% vs 7.2% (relative risk 2.69, 95% CI 1.86–3.90).
Adverse events
Grade 3/4 (any): 71.6% vs 71.2%.
Treatment-related death: 6 (1%) vs 2 (<1%).
Delayed surgery: 10.1% vs 10.8%; delayed adjuvant initiation 2.3% vs 4.6%.
Treatment-related death: 6 (1%) vs 2 (<1%).
Delayed surgery: 10.1% vs 10.8%; delayed adjuvant initiation 2.3% vs 4.6%.
Conclusions
Perioperative durvalumab + FLOT significantly improved overall survival (HR 0.78) over FLOT alone, confirming the earlier EFS and pCR benefit. MATTERHORN establishes perioperative durvalumab–FLOT as a new standard of care for resectable gastric/GEJ adenocarcinoma, with benefit reported regardless of PD-L1 status.
Key Limitations
The OS benefit, though significant at final analysis, was modest (HR 0.78) and the piecewise early hazard (months 0–12) showed no separation. The contribution of the neoadjuvant versus adjuvant immunotherapy components cannot be isolated, and roughly a quarter of enrolled patients were not randomized. Durability of benefit in PD-L1–low disease and across surgical/chemotherapy practice patterns warrants continued follow-up.
Clinical Context
On Nov 25, 2025 the FDA approved perioperative durvalumab + FLOT for resectable gastric/GEJ adenocarcinoma — the first perioperative immunotherapy in this setting. It builds on FLOT4 as the chemotherapy backbone; NCCN and ESMO endorse perioperative FLOT as standard, with immunotherapy now added on the strength of MATTERHORN. This contrasts with KEYNOTE-585 (perioperative pembrolizumab), which did not meet its EFS threshold.