Background
Phase III open-label RCT, N=970, previously untreated unresectable advanced, recurrent, or metastatic esophageal squamous cell carcinoma. Randomized 1:1:1 to nivolumab+chemotherapy, nivolumab+ipilimumab, or chemotherapy alone. Dual primary endpoints: OS and PFS in tumor-cell PD-L1 ≥1% (TC ≥1%) and in all randomized patients, tested hierarchically for both nivo+chemo and nivo+ipi vs chemo.
Interventions and follow up
Arm A (nivo+chemo): Nivolumab 240 mg q2wk + cisplatin 80 mg/m² d1 + 5-FU 800 mg/m²/day d1–5 q4wk
Arm B (nivo+ipi): Nivolumab 3 mg/kg q2wk + ipilimumab 1 mg/kg q6wk (chemo-free IO doublet)
Arm C (control): Cisplatin 80 mg/m² + 5-FU 800 mg/m²/day (chemotherapy alone)
Primary endpoint: OS and PFS (dual primary: TC ≥1% and all patients)
Follow-up: 13 mo (minimum)
Arm B (nivo+ipi): Nivolumab 3 mg/kg q2wk + ipilimumab 1 mg/kg q6wk (chemo-free IO doublet)
Arm C (control): Cisplatin 80 mg/m² + 5-FU 800 mg/m²/day (chemotherapy alone)
Primary endpoint: OS and PFS (dual primary: TC ≥1% and all patients)
Follow-up: 13 mo (minimum)
Results
mOS, nivo+chemo vs chemo (TC ≥1%): 15.4 vs 9.1 mo, HR 0.54, P<.001
mOS, nivo+chemo vs chemo (all): 13.2 vs 10.7 mo, HR 0.74, P=.002
mOS, nivo+ipi vs chemo (TC ≥1%): 13.7 vs 9.1 mo, HR 0.64, P=.001
mOS, nivo+ipi vs chemo (all): 12.7 vs 10.7 mo, HR 0.78, P=.01
mOS, nivo+chemo vs chemo (all): 13.2 vs 10.7 mo, HR 0.74, P=.002
mOS, nivo+ipi vs chemo (TC ≥1%): 13.7 vs 9.1 mo, HR 0.64, P=.001
mOS, nivo+ipi vs chemo (all): 12.7 vs 10.7 mo, HR 0.78, P=.01
Adverse events
Grade ≥3 treatment-related: nivo+chemo 47%, nivo+ipi 32%, chemo 36%.
Immune-mediated (any grade): nivo+ipi 42% vs nivo+chemo 24% vs chemo 20%.
Treatment-related deaths: 5 (nivo+chemo), 3 (nivo+ipi), 6 (chemo).
Immune-mediated (any grade): nivo+ipi 42% vs nivo+chemo 24% vs chemo 20%.
Treatment-related deaths: 5 (nivo+chemo), 3 (nivo+ipi), 6 (chemo).
Conclusions
Both nivolumab combinations significantly improved OS vs chemotherapy in advanced esophageal SCC. Nivolumab + chemotherapy demonstrated a particularly large OS benefit in TC ≥1% (mOS 15.4 vs 9.1 mo, HR 0.54), establishing it as a new first-line standard.
Key Limitations
TC PD-L1 ≥1% used (not CPS), which may select differently than in gastric trials. Minimum follow-up of only 13 months is short for OS analysis. Chemo backbone is cisplatin/5-FU, not carboplatin/paclitaxel. KEYNOTE-590 (pembro + chemo) and RATIONALE-306 also show benefit, creating competing first-line options.
Clinical Context
FDA approved nivolumab + chemotherapy and nivolumab + ipilimumab (chemo-free for TC ≥1%) as first-line options for advanced esophageal SCC. KEYNOTE-590 (pembrolizumab + chemo) is an alternative FDA-approved first-line regimen. The OS benefit (HR 0.54 in TC ≥1%) is among the largest seen in IO trials in this setting. ESMO guidelines endorse IO + chemo first-line.
References