Nomogram predicted overall survival and cancer-specific survival of patients with cardia gastric adenocarcinoma (CGA): A Population- Based Study

Abstract

Abstract Background: This study aims to develop and validate two nomograms to predict the overall survival (OS) and cancer-specific survival (CSS) of cardia gastric adenocarcinoma (CGA) patients. Methods: A total of 6069 patients diagnosed with CGA were selected from the SEER database. They were further randomized in a 7:3 percentage into training and validation cohorts. Univariate and multivariate Cox proportional hazards regression were conducted to evaluate the prognostic factors of OS and CSS. Based on this, two nomograms were constructed to predict the prognosis of CGA patients. We used the area under the ROC curve (AUC), concordance index (C-index), and calibration curve to determine the predictive accuracy and discriminability of the nomograms. The decision curve analysis (DCA) was employed to confirm the clinical effectiveness of the nomograms further. Patients were risk-stratified according to nomogram scores, and Kaplan–Meier curves were plotted to compare survival outcomes among risk subgroups. Results: COX regression analysis showed eight independent risk factors associated with OS and nine independent risk factors associated with CSS. Based on the above results, two nomograms were constructed in the training cohorts for predicting OS and CSS in CGA patients. The results showed that the OS nomogram C-index of training cohorts was 0.711 (95%CI: 0.702-0.719) and the CSS nomogram C-index was 0.731 (95%CI: 0.722-0.740). The OS nomogram C-index of validation cohorts was 0.723 (95%CI: 0.710-0.736) and the CSS nomogram C-index was 0.746 (95%CI: 0.732-0.759). The calibration curve and ROC indicated that the nomogram prediction agreed well with the actual survival. AUC(>0.75) and DCA indicated that the model had good clinical application value. In addition, survival results between different subgroups according to the risk of the Kaplan-Meier curve has obvious differences. Conclusions: Two prognostic nomograms for CGA patients were developed to help clinicians judge the prognosis of patients and make clinical decisions.