Key points are not available for this paper at this time.
BACKGROUND: Multiple perioperative inflammatory markers are considered important factors affecting the long-term survival of esophageal cancer (EC) patients. Hematological parameters, whether single or combined, have high predictive value. AIM: To investigate the inflammatory status of patients with preoperative EC using blood inflammatory markers, and to establish and validate competing risk nomogram prediction models for overall survival (OS) and progression-free survival (PFS) in EC patients. METHODS: A total of 508 EC patients who received radical surgery (RS) treatment in The First Affiliated Hospital of Zhengzhou University from August 5, 2013, to May 1, 2019, were enrolled and randomly divided into a training cohort (356 cases) and a validation cohort (152 cases). We performed least absolute shrinkage and selection operator (LASSO)-univariate Cox- multivariate Cox regression analyses to establish nomogram models. The index of concordance (C-index), time-dependent receiver operating characteristic (ROC) curves, time-dependent area under curve (AUC) and calibration curves were used to evaluate the discrimination and calibration of the nomograms, and decision curve analysis (DCA) was used to evaluate the net benefit of the nomograms. The relative integrated discrimination improvement (IDI) and net reclassification improvement (NRI) were calculated to evaluate the improvement in predictive accuracy of our new model compared with the AJCC staging system and another traditional model. Finally, the relationship between systemic inflammatory response markers and prognostic survival was explored according to risk plot, time-dependent AUC, Kaplan-Meier and restricted cubic spline (RCS). RESULTS: Based on the multivariate analysis for overall survival (OS) in the training cohort, nomograms with 10 variables, including the aggregate index of systemic inflammation (AISI) and lymphocyte-to-monocyte ratio (LMR), were established. Time-dependent ROC, time-dependent AUC, calibration curves, and DCA showed that the 1-, 3-, and 5 year OS and PFS probabilities predicted by the nomograms were consistent with the actual observations. The C-index, NRI, and IDI of the nomograms showed better performance than the AJCC staging system and another prediction model. Moreover, risk plot, time-dependent AUC, and Kaplan-Meier showed that higher AISI scores and lower LMR were associated with poorer prognosis, and there was a nonlinear relationship between them and survival risk. CONCLUSION: AISI and LMR are easy to obtain, reproducible and minimally invasive prognostic tools that can be used as markers to guide the clinical treatment and prognosis of patients with EC.
Wang et al. (Fri,) studied this question.