A Study on Predictors for Progression-Free Survival in the First-Line Treatment of Recurrent/Unresectable Esophageal Cancer

Background In Japan, combination chemotherapy with 5-FU + CDDP (FP) and immune checkpoint inhibitors (ICIs) and dual ICI combination therapy have been approved for use in first-line treatment of recurrent/unresectable esophageal cancer since November 2021, and have become widely used in clinical practice. Systemic inflammatory nutritional biomarkers have been reported as potential predictors of chemotherapy in various carcinomas, including esophageal cancer, but none have been firmly established. Methods We retrospectively examined the predictors of progression-free survival (PFS) in 60 patients who underwent the first-line treatment for recurrent/unresectable esophageal cancer in our hospital from June 2022 to December 2024. The breakdown of the first-line treatment regimen was as follows: FP + ICI (Pembrolizumab/Nivolumab<Nivo>) in 34 patients (19/15) and Nivo+Ipilimumab<Ipi> in 26 patients. We assessed systemic inflammatory nutritional biomarkers, including neutrophil lymphocyte ratio (NLR), prognostic nutrition Index (PNI), and CRP albumin ratio (CAR). Results The median PFS was 140 days in the FP + ICI group and 105 days in the Nivo+Ipi group. In a multivariate analysis of predictors in the FP + ICI group, CAR (HR 3.019; 95% CI 1.128–8.083; P = 0.027) and smoking (HR 7.886; 95% CI 1.045–59.481; P = 0.045) were extracted as independent predictors of PFS. In the Nivo+Ipi group, CAR (HR 3.051; 95% CI 1.173–7.931; P = 0.022) and immune-related adverse events (irAEs) (HR 0.349; 95% CI 0.129–0.945; P = 0.038) were extracted as independent predictors of PFS. Conclusion This study suggested the usefulness of CAR and smoking as a predictor of PFS in FP + ICI patients in the first-line treatment of esophageal cancer, and the usefulness of CAR and irAE as predictors of PFS in Nivo+Ipi patients in the first-line treatment of esophageal cancer. Future research should focus on validating CAR in broader populations.