Validation of a Nomogram Predicting Survival After Trimodality Therapy for Esophageal Cancer

Abstract

Background: Recently, a nomogram was developed for the prediction of overall survival (OS) after treatment with neoadjuvant chemoradiotherapy (nCRT) combined with surgery for esophageal or junctional cancer. The nomogram included clinical nodal category, pathologic tumor category, and number of positive lymph nodes in the resection specimen. The aim of this study was to externally validate the nomogram in an international multiinstitutional cohort of patients, and to explore the prognostic use of the nomogram for the prediction of progression-free survival (PFS) after nCRT plus surgery. Methods: Patients with potentially resectable esophageal or junctional carcinoma that underwent nCRT plus surgery between 1998 and 2015 at 3 academic centers were included. The discriminative ability of the nomogram for the prediction of OS and PFS was quantified by Harrell's C-statistic. Calibration of the nomogram was visually assessed by plotting actual OS and PFS probabilities against predicted probabilities. Results: Some 975 patients were included. The discriminative ability of the nomogram for OS was moderate (C-statistic, 0.61) and comparable to that of the initial cohort (C-statistic, 0.63). The nomogram was also useful for the prediction of PFS (C-statistic, 0.64). Calibration of the nomogram was accurate for both OS and PFS, with predicted estimates corresponding closely with the actual observed estimates. Conclusions: The nomogram accurately predicted OS and PFS after nCRT plus surgery in an independent international cohort of esophageal cancer patients. The current validated model may enable risk-stratified adjuvant treatment allocation and identify patients in need of routine surveillance after treatment.