Background
Phase III, double-blind RCT in 698 patients with previously untreated, locally advanced unresectable or metastatic HER2-positive gastric or GEJ adenocarcinoma. Randomized to pembrolizumab vs placebo, each added to trastuzumab plus chemotherapy. The Nature 2021 report was the prespecified first interim analysis.
Interventions and follow up
Arm A: Pembrolizumab 200 mg Q3W + trastuzumab (8 mg/kg load, 6 mg/kg thereafter) Q3W + investigator's choice chemotherapy (5-FU 800 mg/m² Days 1-5 + cisplatin 80 mg/m² Q3W, or capecitabine 1000 mg/m² BID Days 1-14 + oxaliplatin 130 mg/m² Q3W)
Arm B: Placebo + trastuzumab + same chemotherapy
Primary endpoints: OS and PFS
Secondary endpoints: ORR and DOR
Median follow up: 12.0 mo (interim analysis)
Arm B: Placebo + trastuzumab + same chemotherapy
Primary endpoints: OS and PFS
Secondary endpoints: ORR and DOR
Median follow up: 12.0 mo (interim analysis)
Results
ORR (first 264 pts): 74.4% vs 51.9%, difference 22.7% (95% CI 11.2-33.7), P<.001 (Arm A vs B)
CR: 11.3% vs 3.1%
PR: 63.2% vs 48.9%
OS and PFS: Immature at interim
Final OS update (ASCO 2024): 20.0 mo vs 16.8 mo, HR 0.80, 95% CI 0.67-0.94, P=.0040; benefit confined to PD-L1 CPS ≥1 (HR 0.79, 95% CI 0.66-0.95)
CR: 11.3% vs 3.1%
PR: 63.2% vs 48.9%
OS and PFS: Immature at interim
Final OS update (ASCO 2024): 20.0 mo vs 16.8 mo, HR 0.80, 95% CI 0.67-0.94, P=.0040; benefit confined to PD-L1 CPS ≥1 (HR 0.79, 95% CI 0.66-0.95)
Adverse events
GI: Diarrhea 53% vs 44%, nausea common in both arms (Arm A vs B)
Neurologic: Peripheral neuropathy 23% vs 19%
Hepatic: Increased AST 21% vs 13%
Overall: Grade 3-4 events similar between arms (~57%); immune-related events more frequent with pembrolizumab
Neurologic: Peripheral neuropathy 23% vs 19%
Hepatic: Increased AST 21% vs 13%
Overall: Grade 3-4 events similar between arms (~57%); immune-related events more frequent with pembrolizumab
Conclusions
Adding pembrolizumab to trastuzumab and chemotherapy in advanced HER2-positive gastric/GEJ adenocarcinoma markedly increased ORR and CR at interim; final analysis confirmed an OS benefit limited to PD-L1 CPS ≥1 tumors.
Key Limitations
Initial report based on early interim ORR with immature OS/PFS. OS benefit not observed in PD-L1 CPS <1 subgroup, prompting label restriction. Investigator's-choice chemotherapy backbone introduces heterogeneity.
Clinical Context
FDA granted accelerated approval (2021) then full approval (2023) for pembrolizumab plus trastuzumab and chemotherapy in first-line HER2-positive gastric/GEJ adenocarcinoma, restricted to PD-L1 CPS ≥1. ESMO endorses this regimen as a first-line standard for PD-L1 CPS ≥1 HER2-positive disease.
References
Janjigian YY et al, Nature, 2021, PMID: 3491212
Final OS: Janjigian YY et al, Ann Oncol, 2024 (ESMO LBA 1400O)
Final OS: Janjigian YY et al, Ann Oncol, 2024 (ESMO LBA 1400O)