A scoring system based on inflammatory and nutritional indicators to predict the long-term survival of patients with non-metastatic nasopharyngeal carcinoma

To develop a simple scoring system based on baseline inflammatory and nutritional markers to predict the long-term prognosis of patients with nasopharyngeal carcinoma (NPC). Conducted a retrospective analysis of clinical data from 1024 newly diagnosed non-metastatic NPC patients. A total of 15 pre-treatment inflammatory and nutritional markers were collected as candidate variables. Receiver operating characteristic (ROC) curves were used to determine the optimal cutoff points for each parameter. Survival analysis was performed using Kaplan-Meier method and Cox regression analysis. Besides, the Inflammation Nutrition Risk Score (INRS) was calculated for each patient by assigning each independent prognostic factor a score of 1. Multivariate Cox regression analysis showed that serum albumin (ALB), systemic immune-inflammation index, and monocyte count (M) were independent prognostic factors for OS (P < 0.05). Survival analysis showed that higher INRS was associated with a worsened prognosis. Patients in the high-risk group had shorter OS than in the low-risk group. In the training group, the 3-, 5-, and 8-years OS rates for the low-risk group versus high-risk group were 92.5% versus 87.8%, 87.4% versus 75.1%, and 84.6% versus 62.2%, respectively (P < 0.05). In the validation group, the 3-, 5-, and 8-years OS rates for the low-risk group vs. high-risk group were 95.0% versus 86.4%, 92.1% versus 82.2%, and 89.5% versus 74.3%, respectively (P < 0.05). Further subgroup analysis showed a significant difference in the OS between the high-risk group and low-risk group in patients with locally advanced disease (P < 0.05). The ROC curve demonstrated that INRS had a similar predictive value for long-term survival in NPC patients compared to TNM staging and serum EBV-DNA levels. Pretreatment ALB, M, and SIRI are independent prognostic factors for long-term survival in patients with NPC. INRS constructed based on these three factors can serve as a long-term prognostic indicator for NPC.