Abstract
The prognostic impact of the combination of the modified Glasgow prognostic score (mGPS) and C-reactive protein/albumin ratio (CAR) in colorectal cancer (CRC) is unclear. We aimed to investigate the clinical usefulness of this combination as a predictor of survival in CRC patients. We retrospectively evaluated 769 CRC patients who had undergone surgery between January 2006 and March 2014. The CAR and mGPS within 1 month postoperation were examined. The integrated area under the curve (iAUC) was compared among mGPS, CAR, and the combined classification (CC). The optimal CAR cut-off for discriminating overall survival was 0.14. Based on this cut-off, the mGPS 0 group was divided into the mGPS 0 with low CAR and the mGPS 0 with high CAR groups, whereas all mGPS 1 and 2 patients were classified into the high CAR group. CC was an independent prognostic factor, and its iAUC value (0.587, 95% CI 0.553–0.624) was superior to those of the mGPS (0.544, 95% CI 0.516–0.576) (bootstrap iAUC mean difference = 0.043; 95% CI = 0.015–0.072) and CAR (0.578, 95% CI 0.545–0.613) (bootstrap iAUC mean difference = 0.009; 95% CI = 0.002–0.017), respectively. In conclusion, the combination of mGPS and CAR has a synergistic effect and has a higher prognostic accuracy than mGPS or CAR alone in patients with CRC.