Background
Phase III RCT (long-term analysis) included 493 patients with unresectable advanced or recurrent gastric or gastroesophageal junction adenocarcinoma refractory to or intolerant of two or more prior chemotherapy regimens. Third-line or later setting.
Interventions and follow up
Arm A: Nivolumab 3 mg/kg IV q2wk
Arm B: Placebo q2wk
Primary endpoint: OS
mFollow up: 8.87 vs 8.59 mo, arm A vs B (minimum 2-year follow-up analysis)
Arm B: Placebo q2wk
Primary endpoint: OS
mFollow up: 8.87 vs 8.59 mo, arm A vs B (minimum 2-year follow-up analysis)
Results
mOS: 5.26 vs 4.14 mo, arm A vs B (HR 0.63; 95% CI 0.51-0.78; P<.0001)
12-mo OS: 27.3% vs 11.6%
ORR: 11.9% vs 0%
12-mo OS: 27.3% vs 11.6%
ORR: 11.9% vs 0%
Adverse events
Overall: Treatment-related events any-grade 43% vs 27%, arm A vs B
Grade ≥3: 10% vs 4%
Common: pruritus, diarrhea, fatigue; immune-related events generally manageable
Grade ≥3: 10% vs 4%
Common: pruritus, diarrhea, fatigue; immune-related events generally manageable
Conclusions
Nivolumab improved OS versus placebo in heavily pretreated advanced gastric/GEJ adenocarcinoma, with durable benefit and a manageable safety profile.
Key Limitations
Absolute OS gain modest in a heavily pretreated population. Conducted in Asian patients (Japan, Korea, Taiwan), limiting generalizability. PD-L1 status not required for benefit.
Clinical Context
ATTRACTION-2 supported nivolumab approval in Asia for third-line gastric/GEJ cancer. First-line chemoimmunotherapy (CheckMate 649) has since moved PD-1 inhibition earlier. ESMO recognizes checkpoint inhibition in advanced gastric cancer.
References