Abstract
PURPOSE: This study was designed to investigate the prognostic value of the combination of high-sensitivity C-reactive protein, lymphocyte, and albumin in patients with resectable colorectal cancer. PATIENTS AND METHODS: Seven-hundred-and-nineteen patients who underwent colorectal cancer resection in Hubei Cancer Hospital were included. Inflammation-Immunity-Nutrition score (0-6) was constructed based on preoperative high-sensitivity C-reactive protein, lymphocyte, and albumin. Time-dependent receiver operating characteristic curve, decision curve, Kaplan-Meier survival curve, Cox regression, and C-index were conducted to detect the prognostic values of inflammation-immunity-nutrition score. The prognostic values of inflammation-immunity-nutrition score in different subgroups by sex, location of tumor, pathologic stage, and KRAS mutation were also explored. The prognostic performance of inflammation-immunity-nutrition score was further compared with that of other traditional prognostic indicators. RESULTS: The median follow-up time was 40 months. High inflammation-immunity-nutrition score (>2 scores) presented worse survival, with the adjusted hazard ratios (95% confidence intervals) of 3.106 (2.202-4.380) for overall survival and 2.105 (1.604-2.764) for disease-free survival. Besides, the associations of high inflammation-immunity-nutrition score with overall survival were even stronger in cases with wild type KRAS, with the adjusted hazard ratios (95% confidence intervals) of 4.018 (2.355-6.854). Considering the AUCs, C-indices, and hazard ratios estimates, inflammation-immunity-nutrition score presented better prognostic performance than high-sensitivity modified Glasgow prognostic score, high-sensitivity C-reactive protein to albumin ratio, prognostic nutrition index, carcinoembryonic antigen, and carbohydrate antigen 19-9 for overall survival. CONCLUSION: Inflammation-immunity-nutrition score might serve as a powerful prognostic score in patients with colorectal cancer for overall survival, particularly in patients with wild type KRAS.