The Glasgow Prognostic Score (GPS), which is calculated with C-reactive protein (CRP) and albumin (Alb) values, is a prognostic indicator for various types of cancers. However, its role in lung cancer still remains unclear, and its optimal cut-off values are controversial. Here, we evaluated the significance of the GPS and adjusted GPS (a-GPS) using our institution's cut-off values in patients undergoing resection for primary lung cancer.

We analysed 1043 lung cancer patients who underwent resection between 1998 and 2012. The overall survival (OS) probabilities of the GPS subgroups were estimated using the Kaplan–Meier method and were compared using the log-rank test. The prognostic significance of the GPS and the a-GPS was assessed by the Cox proportional hazards model with clinicopathological variables and inflammation markers, such as the neutrophil-to-lymphocyte ratio (NLR) and the platelet-to-lymphocyte ratio (PLR). The GPS was calculated based on cut-off values of 1.0 mg/dl for CRP and 3.5 g/dl for Alb, as previously reported. The a-GPS was calculated based on cut-off values 0.3 mg/dl for CRP and 3.9 g/dl for Alb, which are the standard thresholds used by our institution.

The GPS and the a-GPS were correlated with preoperative factors, such as age, sex, smoking status, the NLR and the PLR, and oncological factors, including the pathological stage, histological type and level of lymphovascular invasion. The 5-year OS rates were 82, 55 and 55% with GPS 0, 1 and 2 (1 vs 0: P < 0.01; 2 vs 1: P = 0.66), respectively, and 88, 67 and 59% with a-GPS 0, 1 and 2 (1 vs 0: P < 0.01; 2 vs 1: P = 0.04), respectively. Multivariable analysis revealed that the GPS [1 vs 0, hazard ratio (HR): 1.63, 2 vs 0, HR: 1.44] and the a-GPS (1 vs 0, HR: 2.00, 2 vs 0, HR: 2.10) were independent prognostic factors. The a-GPS classification showed a clearer prognostic distribution than the GPS classification.

The GPS is a useful prognostic indicator of the OS in lung cancer surgery. The optimal cut-off values for GPS estimation may need to be re-evaluated.