Background
Phase 3 open-label RCT (CheckMate 227, part 1). Multiple cohorts; primary OS endpoint reported for 793 chemotherapy-naive patients with stage IV or recurrent NSCLC. Cohort 1a (PD-L1 ≥1%): nivolumab + ipilimumab vs chemotherapy. Cohort 1b (PD-L1 <1%): nivolumab + ipilimumab vs nivolumab + chemotherapy vs chemotherapy. No prior systemic therapy. EGFR/ALK mutations not excluded.
Interventions and follow up
Arm A: Nivolumab 3 mg/kg IV q2wks + ipilimumab 1 mg/kg q6wks (dual checkpoint blockade)
Arm B: Platinum-based chemotherapy per investigator choice (carboplatin/paclitaxel, carboplatin/pemetrexed, or cisplatin/pemetrexed)
Primary endpoint: OS in PD-L1 ≥1% population (cohort 1a)
mFollow up: Minimum 29.3 month
Arm B: Platinum-based chemotherapy per investigator choice (carboplatin/paclitaxel, carboplatin/pemetrexed, or cisplatin/pemetrexed)
Primary endpoint: OS in PD-L1 ≥1% population (cohort 1a)
mFollow up: Minimum 29.3 month
Results
OS (PD-L1 ≥1%): 17.1 vs 14.9 months, HR 0.79 (95% CI 0.67–0.93), P=.007
2-yr OS rate: 40.0% vs 32.8%
Median duration of response: 23.2 months (nivo+ipi) vs 6.2 months (chemo)
OS (PD-L1 <1%, cohort 1b): 17.2 vs 12.2 months, HR 0.62 (95% CI 0.48–0.78)
2-yr OS rate: 40.0% vs 32.8%
Median duration of response: 23.2 months (nivo+ipi) vs 6.2 months (chemo)
OS (PD-L1 <1%, cohort 1b): 17.2 vs 12.2 months, HR 0.62 (95% CI 0.48–0.78)
Adverse events
Main adverse events: Grade ≥3 treatment-related AEs: 32.8% (nivo+ipi) vs 36.0% (chemo). Any-grade immune-mediated AEs: 24.5% vs 3.3%. Grade ≥3 immune-mediated AEs: 9.6% vs 0.9%. Treatment-related deaths: 1.7% (nivo+ipi) vs 0.9% (chemo). Most common immune-mediated: skin (34%), endocrine (19%), hepatic (10%), pulmonary (5%).
Conclusions
Nivolumab + ipilimumab significantly improved OS vs chemotherapy in PD-L1 ≥1% advanced NSCLC (HR 0.79, p=0.007) with markedly longer response durability. Activity was also observed in PD-L1 <1% patients, establishing dual checkpoint blockade as a chemotherapy-free first-line option across a broad NSCLC population.
Key Limitations
Key Limitations: Open-label design. EGFR/ALK-mutant patients (enrolled but not separately powered) appear to derive minimal or no benefit from IO combinations — an important exclusion in clinical practice. TMB (tumor mutational burden) showed initial promise as a biomarker in early analyses but did not demonstrate robust OS correlation in the FDA-submitted dataset, and TMB testing is not required for approval. Dual IO carries significantly higher immune toxicity vs single-agent pembrolizumab. The absolute OS benefit (2.2 months median) is modest, and long-term tail of OS curve (durable responses) is the key differentiator vs chemotherapy.
Clinical Context
FDA approved in 2020 for first-line metastatic NSCLC with PD-L1 ≥1% and no EGFR/ALK alterations. Radiation oncology relevance: (1) Stage III CRT + durvalumab (PACIFIC) vs stage IV dual IO (CheckMate 227) represent parallel immunotherapy paradigms; (2) RT-IO combinations exploring abscopal effect and synergy with checkpoint blockade are active research areas. Competing first-line options include pembrolizumab monotherapy (KEYNOTE-024, PD-L1 ≥50%) and pembrolizumab + chemotherapy (KEYNOTE-189/407). Patient and tumor selection among these options remains an active clinical and research question.
References