Development of a nomogram for overall survival in patients with esophageal carcinoma: A prospective cohort study in China

Figure 1 Flow diagram of study design. PLR: Platelet-to-lymphocyte ratio; ALB/GLB: Albumin-to-globulin ratio; LASSO: Least absolute shrinkage and selection operator; DCA: Decision curve analysis; ROC: Receiver operating characteristic.

Figure 2 Variable selection by least absolute shrinkage and selection operator Cox regression model. A coefficient profile plot was produced against the log (lambda) sequence. A: 22 variables with nonzero coefficients were selected by optimal lambda; B: By verifying the optimal parameter (lambda) in the least absolute shrinkage and selection operator model, the partial likelihood deviance (binomial deviance) curve was plotted vs log (lambda) and dotted vertical lines were drawn based on 1 standard error criteria.

Figure 3 Prediction model for predicting 1-, 3- and 5-year overall survival of patients with Esophageal carcinoma. A: Nomogram model; B: The interface of the web-based nomogram. PLR: Platelet-to-lymphocyte ratio; ALB/GLB: Albumin-to-globulin ratio; BMI: Body mass index; KPS: Karnofsky performance status.

Figure 4 Receiver operating characteristic and calibration curves of the prediction model for 1-, 3- and 5-year overall survival prediction. A: Receiver operating characteristic (ROC) in the training cohort; B: ROC in the validation cohort; C: Calibration plot in the training cohort; D: Calibration plot in the validation cohort. AUC: Area under the receiver operating characteristic curve; OS: Overall survival.

Figure 5 Decision curve analysis for the prediction model’s ability to predict overall survival in Esophageal carcinoma patients and the prediction model distinguished the risk of Esophageal carcinoma patients. A: Decision curve analysis (DCA) in the training cohort; B: DCA validation cohort; C: The prediction model in the training cohort; D: The prediction model in the validation cohort.