Clinical utility of pretreatment Glasgow prognostic score in non-small-cell lung cancer patients treated with immune checkpoint inhibitors

Objectives: Immune checkpoint inhibitors (ICIs) have become one of the standard therapies in non-small-cell lung cancer (NSCLC). Although inflammatory indices, including Glasgow prognostic score (GPS), modified Glasgow prognostic score (mGPS), and C-reactive protein/albumin ratio (CAR) were reported to be reliable predictors for survival in cancer patients, their clinical utility in NSCLC patients treated with ICIs is unknown. Materials and methods: Advanced or recurrent NSCLC patients (n = 304) treated with ICI monotherapy at the National Hospital Organization Kyushu Cancer Center and Kyushu University Hospital between January 2016 and December 2019 were analyzed. Information on patient demographics, GPS, mGPS, and CAR at diagnosis were collected. The time-dependent area under curves (AUCs) of receiver operating characteristic curves for the prediction of overall survival (OS) for each factor were compared. Results: Of the three indices, GPS was the most significantly correlated with the degree of disease control rate (DCR) (DCR of GPS of 0, 1, and 2: 63.6 %, 49.4 %, and 41.4 %, respectively). The time-dependent AUC values of GPS for the prediction of OS were superior to those of mGPS and CAR (time-dependent AUC values of GPS, mGPS, and CAR for the prediction of 1-year OS: 0.7005, 0.6736, and 0.6565, respectively). GPS was significantly correlated with performance status (PS) (P < 0.0001) and clinical stage (P = 0.0139). GPS in combination with PS effectively predicted survival at 1 year ranging from 83.5 % (GPS = 0, PS = 0) to 25.0 % (GPS = 2, PS = 2, 3). A multivariable analysis revealed that GPS was an independent predictor of PFS and OS (P = 0.0009 and P = 0.0100, respectively). Conclusions: We report for the first time that GPS represents a simple and useful prognostic factor in NSCLC patients treated with ICIs and should be validated prospectively.