Development and validation of a prognostic model for first-line immunotherapy for metastatic esophageal squamous cell...

Esophageal cancer (EC) is a highly prevalent malignant tumor worldwide, with patients in Asia mainly suffering from esophageal squamous cell carcinoma (ESCC). Most patients are already in the metastatic stage when diagnosed, with a poor prognosis. For patients with advanced distant metastasis, immunotherapy combined with chemotherapy has gradually become an important treatment method, and the survival time of some patients has significantly improved. However, the prognostic factors for this group of patients after treatment have not yet been determined. Therefore, constructing a relevant nomogram is of great significance for accurately evaluating patient survival and guiding personalized treatment.

A retrospective analysis was conducted on the clinical data of 224 patients with advanced-stage ESCC who received immunotherapy combined with chemotherapy and developed distant metastasis. Establish a predictive nomogram for the survival outcomes of the training cohort using univariate and multivariate Cox regression analysis and least absolute shrinkage and selection operator (LASSO). The performance of the model is further measured in the validation cohort and overall queue by examining the concordance index (C-index), calibration curves, decision curve analysis (DCA), accuracy (receiver operating characteristic curve), and utility (patient stratification into low-risk vs high-risk groups).

Patient-related inflammatory indicators and clinical pathological characteristics were analyzed using LASSO, univariate, and multivariate Cox regression analyses, and a nomogram model was established for the five factors. The nomogram model exhibited acceptable discrimination, with C-indices of 0.818, 0.804, and 0.815 for the training cohort, validation cohort, and entire cohort, respectively. The calibration chart confirmed a good consistency between the model prediction results and the actual observations. DCA showed that the model had good clinical application value. The predictive performance of the model was evaluated using the time-dependent ROC curves. According to the total score of the nomogram in the training cohort, a cutoff value of 116.84 points was taken to divide the patients into high-risk and low-risk groups. Among them, the median overall survival (OS) of high-risk group patients was significantly shorter than that of low-risk group patients in the training cohort, validation cohort, and entire cohort.

This study constructed a prognostic nomogram for metastatic ESCC patients receiving first-line immunotherapy combined with chemotherapy. The model integrates five independent prognostic factors and demonstrates good discriminatory power, calibration, and clinical utility in training, validation, and the entire cohort. It can effectively divide patients into high-risk and low-risk groups to distinguish differences in OS, providing a basis for clinicians to accurately assess patient survival prognosis and develop personalized treatment strategies.