Survival and prognostic factors in patients with synchronous multiple primary esophageal squamous cell carcinoma receiving definitive radiotherapy: A propensity score-matched analysis.

Abstract

To compare the lesion characteristics and radiotherapy efficacy of patients with single and multiple esophageal squamous cell carcinoma (ESCC), to evaluate the effect of multiple lesions on ESCC, and establish a nomogram survival prediction model for patients with synchronous multiple primary esophageal squamous cell carcinoma (SMPESCC) who received definitive radiotherapy. The study enrolled 1,034 patients with ESCC who underwent definitive radiotherapy between 2010 and 2020. The efficacy of radiotherapy was compared between 101 patients with SMPESCC and 933 patients with single ESCC. Propensity score matching was used to control for potential confounders. For patients with SMPESCC, a nomogram prediction model was established based on the Cox regression model. The median OS was 30.00 (95% CI = 25.08-34.92) months for the single lesion group and 19.00 (95% CI = 15.51-22.48) months for the multiple cancer group respectively. Multivariate COX regression analysis showed that multiple cancer was an independent prognostic factor for ESCC patients (HR=1.89, 95%CI=1.49-2.38, P<0.001). Cox multivariate analysis of SMPESCC patients showed that T stage (P =0.002), chemotherapy (P =0.006), and lesion spacing (P =0.004) were independent prognostic factors associated with OS. The nomogram was established by combining T stage, chemotherapy, and lesion spacing, and Harrell's C index was 0.711 after internal cross-validation. The calibration curve and decision curve analysis confirmed that the nomogram survival prediction model had a good predictive value for individual survival. The survival rate of single esophageal cancer is significantly better than that of multiple lesions. Patients with SMPESCC exhibit worse survival than patients with single ESCC. Multiple lesions have a significant impact on the survival of patients with ESCC. The nomogram model established for SMPESCC patients can well predict the individual survival of patients.