Background
Phase III double-blind RCT, N=755, pathological stage IIIA–C gastric or GEJ adenocarcinoma after curative gastrectomy with D2 or more extensive lymph-node dissection. 96 hospitals in Japan, South Korea, Taiwan, and China. ATTRACTION-5 tested adjuvant nivolumab + chemotherapy vs placebo + chemotherapy in the post-D2 gastrectomy setting in Asia.
Interventions and follow up
Arm A: Nivolumab 360 mg IV q3wk + investigator-choice chemotherapy: S-1 (40 mg/m² bid d1–28 q6wk) or capecitabine + oxaliplatin (CAPOX) q3wk
Arm B: Placebo + same chemotherapy backbone
Primary endpoint: Relapse-free survival (central assessment)
Follow-up: 49.1 mo
Arm B: Placebo + same chemotherapy backbone
Primary endpoint: Relapse-free survival (central assessment)
Follow-up: 49.1 mo
Results
3-yr RFS: 68.4% vs 65.3%, HR 0.90, 95% CI 0.69–1.18, P=.44 (not significant)
OS: NR (not reported as primary endpoint)
OS: NR (not reported as primary endpoint)
Adverse events
Overall TRAEs: 99% (nivo) vs 98% (placebo); most common were decreased appetite, nausea, diarrhea, neutropenia, peripheral neuropathy.
Discontinuation: 9% (nivo) vs 4% (placebo); no new safety signals.
Discontinuation: 9% (nivo) vs 4% (placebo); no new safety signals.
Conclusions
Adjuvant nivolumab + chemotherapy did not improve relapse-free survival vs placebo + chemotherapy in stage III gastric/GEJ cancer after D2 gastrectomy (HR 0.90, P=.44). Adjuvant nivolumab is NOT supported in this setting.
Key Limitations
Asian-only population; D2 dissection may leave less residual disease, reducing IO benefit potential. Mixed chemotherapy backbones (S-1 and CAPOX). PD-L1 CPS was not used for patient selection — a recognized limitation given that IO benefits appear CPS-dependent in advanced disease. Mature OS data not yet available. Contrast: in the adjuvant setting after neoadjuvant chemoradiation (esophageal, CheckMate 577), nivolumab showed benefit — the biology may differ in gastric cancer post-primary surgery.
Clinical Context
ATTRACTION-5 joins KEYNOTE-585 (perioperative) as a negative IO trial in resectable/adjuvant gastric cancer. Only MATTERHORN (durvalumab + FLOT, perioperative) showed benefit. S-1 or CAPOX adjuvant monotherapy remains standard in Asian patients. ESMO does not endorse adjuvant nivolumab in this setting; IO benefit in gastric cancer is currently limited to the advanced/metastatic and esophageal chemoradiation settings.