ABSTRACT
Objective To explore the effects of preoperative serum cancer antigen 19-9 (CA19-9) and neuron-specific enolase (NSE) on the prognosis of patients with esophageal squamous cell carcinoma (ESCC). Methods This prospective study enrolled 176 patients with ESCC. 2 test was used to analyze the relationship between CA19-9, NSE and general clinical features. Survival curves were estimated using Kaplan-Meier method and comparisons were performed using the log-rank test. The Cox proportional hazards model was performed for multivariate analyses of overall survival (OS) and disease free survival (DFS). Results The patients with both high CA19-9 and NSE had the poor prognosis compared with those had both low CA19-9 and NSE (OS: HR=2.310, 95% CI: 1.208-4.418; DFS:HR=2.354, 95% CI：1.265-4.381). Compared to the separate detection of the two markers, the combined detection of CA19-9 and NSE was more accurate in the prognosis prediction of patients with ESCC (OS:C-index=0.686; DFS:C-index=0.684). Conclusions Preoperative serum CA19-9 and NSE were risk factors for the prognosis of patients with ESCC. Combined detection had higher accuracy of prediction of prognosis in patients with ESCC.