Objective:To assess the role of Glasgow prognostic score (GPS) in the prognostic evaluation of nasopharyngeal carcinoma patients.Methods:Clinical data of 129 nasopharyngeal carcinoma patients who received radical radiotherapy in Affiliated Hospital of Jiangnan University from January 2012 to December 2013 were retrospectively analyzed. Clinicopathological characteristics of the patients were collected, including gender, age, TNM staging, pathological type and treatment regimen, etc. The GPS before and at 3 months after radiotherapy were calculated. The survival curve was drawn by the Kaplan- Meier method. Cox regression model was used for analysis of prognostic factors. The area under the receiver operating characteristic (ROC) curve (AUC) was utilized to evaluate the predictive capability of clinical parameters on prognosis. Results:With a median follow-up of 89.0 months (range: 5.1-104.6 months), the 5-year progression-free survival (PFS) of 129 patients was 79.8% and 84.5% for the 5-year overall survival (OS). At 3 months after radiotherapy, the 5-year PFS were 85.6%, 61.1% and 33.3% in the GPS 0, 1 and 2 groups, and 90.4%, 66.7% and 33.3% for the 5-year OS, respectively (all P<0.01). At 3 months after radiotherapy, the GPS, clinical staging (Ⅰ-Ⅲ vs. Ⅳ A) and concurrent chemotherapy were significantly correlated with PFS and OS (all P<0.01). ROC curve showed that at 3 months after radiotherapy, the AUC values of GPS, clinical staging and two combined in predicting OS were 0.694, 0.815 and 0.860, respectively. Conclusions:At 3 months after radiotherapy, higher GPS is an independent poor prognostic factor for nasopharyngeal carcinoma patients. The combination of GPS and clinical staging yields high accuracy in the prognostic evaluation of nasopharyngeal carcinoma patients.