BACKGROUND: Combination immune checkpoint inhibition with ipilimumab plus nivolumab (NIVO + IPI) has shown promising efficacy in advanced esophageal squamous cell carcinoma (ESCC) in the CheckMate648 trial. However, real-world evidence regarding its safety, efficacy as first-line therapy, and host-related biomarkers relevant to immunotherapy remains limited. METHODS: This multicenter retrospective study evaluated a large cohort of 111 patients with unresectable advanced or recurrent ESCC who received first-line NIVO + IPI therapy. Treatment response, treatment-related adverse events, and prognostic factors were analyzed. RESULTS: The objective response and disease control rates in cases with target lesions were 44.0% and 70.7%, respectively. Treatment-related adverse events ≥ Grade 2 occurred in 58 (52.3%) patients, including one Grade 4 event (type 1 diabetes) and two Grade 5 events (biliary infection and myocarditis). The median overall survival (OS) and progression-free survival were 22 months (95% confidence interval CI: 13-not reached) and 5 months (95% CI 3-8), respectively. OS was significantly affected by lymph node metastasis in unresectable advanced disease and by liver metastasis in recurrent disease. Multivariate analysis of OS identified the C-reactive protein-to-albumin ratio (CAR), a marker of host immune-inflammatory status, as the only independent prognostic parameter (hazard ratio = 2.99, 95% CI 1.35-6.63, P = 0.0071). CONCLUSIONS: In this large real-world cohort, first-line NIVO + IPI therapy demonstrated meaningful clinical activity and an acceptable safety profile in advanced ESCC. Treatment outcomes varied according to metastatic patterns, suggesting an influence of organ-specific immune microenvironments, and CAR emerged as a simple and robust prognostic biomarker. These findings support the real-world applicability of dual immune checkpoint blockade and highlight the importance of host immune context in patient selection.
Hara et al. (Sun,) studied this question.