Development of a predictive model for pathological complete response following neoadjuvant immunotherapy and chemotherapy in locally advanced resectable esophageal squamous cell carcinoma

Neoadjuvant immunotherapy and chemotherapy (nICT) has demonstrated substantial therapeutic efficacy in patients with locally advanced esophageal squamous cell carcinoma (LA-ESCC). This study aimed to develop a model to predict the pathological complete response (pCR) following neoadjuvant therapy for esophageal squamous cell carcinoma (ESCC). In total, 517 patients with LA-ESCC from two hospitals who underwent nICT or chemotherapy were enrolled between July 2018 and June 2025. The patients were subsequently divided into training, internal validation, and external validation cohorts. All patients received chemotherapy or the same chemotherapy regimen in combination with immunotherapy; both regimens were administered every 3 weeks for 2–4 cycles. Univariate and multivariate logistic regression analyses were used to identify independent predictors of pCR. The predictive performance of the nomograms was further assessed using receiver operating characteristic (ROC) curve analysis, decision curve analysis curves, and clinical impact curves. Among the patients included in the study, 29.21% achieved pCR. Multivariate regression analysis revealed that independent risk factors for pCR included differentiation, treatment type, preoperative albumin level, pretreatment neutrophil-to-lymphocyte ratio, and pretreatment lymphocyte-to-monocyte ratio. The numbers 1 after the abbreviations represent pretreatment, and 2 represent preoperative values, respectively. The AUC was 0.809 for the training group, 0.786 for the internal validation group, and 0.806 for the external validation group. This model demonstrated good predictive performance for pCR following neoadjuvant therapy in patients with ESCC based on a multicenter database, supporting clinical decision-making and personalized treatment strategies.