Development of a prognostic model based on nutritional and inflammatory indicators for predicting postoperative survival in esophageal cancer: a retrospective study.
Systemic inflammation, immunity, and nutritional status are integral to tumor biology, shaping the microenvironment and influencing esophageal cancer (EC) outcomes. Yet, their integration into pragmatic prognostic tools-and potential implications for immunotherapy stratification-remain limited. This retrospective study assessed the prognostic value of the inflammation-immunity-nutrition score (IINS) and red cell distribution width-to-lymphocyte ratio (RLR), indicators reflecting host immunity, systemic inflammation, and nutritional reserve, in EC patients.
Clinical data from 660 EC patients who underwent radical surgery (2012-2018) were retrospectively analyzed and randomly assigned to training (n = 459) and validation (n = 201) cohorts. Candidate predictors were screened using LASSO and entered into multivariable Cox models. A nomogram incorporating IINS, RLR, and clinical covariates was constructed and validated with the C-index, calibration, and time-dependent AUC; clinical utility was evaluated with decision curve analysis (DCA), Integrated Discrimination Improvement (IDI), and Net Reclassification Index (NRI).
IINS, RLR, and eight additional factors were independent prognostic variables. The nomogram showed good calibration and superior discrimination versus AJCC staging, with a higher C-index and AUC in both cohorts. DCA, IDI, and NRI confirmed greater net benefit and improved risk reclassification.
This study proposes and internally validates a nomogram linking immune-nutritional surrogates with survival in EC. By reflecting systemic inflammation and host immunity, the model supports individualized risk stratification, perioperative optimization, and may inform patient selection for immunotherapy. External multicenter validation is warranted.
Clinical data from 660 EC patients who underwent radical surgery (2012-2018) were retrospectively analyzed and randomly assigned to training (n = 459) and validation (n = 201) cohorts. Candidate predictors were screened using LASSO and entered into multivariable Cox models. A nomogram incorporating IINS, RLR, and clinical covariates was constructed and validated with the C-index, calibration, and time-dependent AUC; clinical utility was evaluated with decision curve analysis (DCA), Integrated Discrimination Improvement (IDI), and Net Reclassification Index (NRI).
IINS, RLR, and eight additional factors were independent prognostic variables. The nomogram showed good calibration and superior discrimination versus AJCC staging, with a higher C-index and AUC in both cohorts. DCA, IDI, and NRI confirmed greater net benefit and improved risk reclassification.
This study proposes and internally validates a nomogram linking immune-nutritional surrogates with survival in EC. By reflecting systemic inflammation and host immunity, the model supports individualized risk stratification, perioperative optimization, and may inform patient selection for immunotherapy. External multicenter validation is warranted.