Preoperative NUn score serves as a robust predictor of overall and disease-specific survivals following radical surgery for gastric cancer.

Abstract

Methods to preoperatively stratify oncological risks associated with gastric cancer (GC) are limited. Host inflammatory parameters, i.e., serum C-reactive protein (CRP) and albumin levels, are known to be associated with outcomes. We examined the relationships between disease-specific mortality and four CRP-albumin-based indices (CRP-albumin ratio [CAR], modified Glasgow prognostic score [mGPS], Osaka prognostic score [OPS], and NUn score) preoperatively measured in cases with resectable GC. Survival outcomes of 1290 consecutive GC patients with oncological gastrectomy were reviewed. Predictive significances of preoperative CAR, mGPS, OPS, and NUn scores were assessed with time-dependent receiver operating characteristic curves and Cox regression analyses. Median follow-up was 107months. Area under the curve for predicting overall and disease-specific survivals (OS/DSS) for the preoperative NUn score was clearly superior to those of the other parameters. On univariate Cox regression analysis, preoperative CAR, mGPS, OPS, and the NUn score all correlated significantly with OS/DSS. On multivariate Cox regression analysis, the preoperative NUn score, as a continuous variable, showed an independent relationship with OS (hazard ratio [HR] 1.32, 95% confidence interval [CI] 1.16-1.50, per 1-unit increase, P < 0.001) and even DSS (HR 1.23, 95% CI 1.02-1.49, P = 0.032). The other three markers failed to maintain independence for DSS. Preoperative NUn scores are stably associated with outcomes, including disease-specific mortality, possibly serving as a simple measure to define the likelihood of progression to systemic disease after meticulous surgery for GC, which may contribute to identifying patients who would benefit from additional modalities.