Abstract: Objective: The inflammatory response score (IRS) was constructed based on the ratio of neutrophils to lymphocytes, the ratio of platelets to lymphocytes and the ratio of lymphocytes to monocytes to predict the survival rate of HCC patients after hepatectomy.Methods: 284 patients with HCC were selected from Beijing Chaoyang emergency rescue center and air force characteristic medical center from May 2013 to January 2019. The best truncation values of NLR, PLR and LMR were obtained by ROC curve and NPL-IRS was constructed.The validity of NPL-IRS was evaluated by time ROC.Results: 284 patients were followed up for 3~43 months, the follow-up time was (21.52 ±7.98) months. At the end of follow-up, 157 patients died, with one-year mortality of 24.6% and three-year mortality of 55.3%. The AUC of NLR, PLR and LMR were 0.889, 0.895 and 0.861 respectively, and the best truncation values were 2.34, 98.74 and 3.43 respectively. The overall survival rate of NLR≤2.34, PLR≤98.74, LMR>3.43 group was higher than that of the corresponding group (P < 0.05). Cirrhosis (yes), tumor number (multiple), vascular invasion (yes), tumor stage (Ⅲ~Ⅳ), NPL-IRS-1, NPL-IRS-2 and NPL-IRS-3 were the independent risk factors of death in HCC patients (P<0.05). The best truncation value of NPL-IRS was 1 and AUC was 0.794 (95% CI=0.742~0.839). The results of time ROC curve show that AUC of NPL-IRS is higher than that of NLR, PLR and LMR at any time.Conclusion: Based on the best truncation values of NLR, PLR and LMR, npl-irs was successfully constructed in this study, which can predict the prognosis of HCC patients more accurately than a single inflammatory response standard.

Key words: Hepatocellular carcinoma, Inflammation, Prognosis