Validation of inflammation-based prognostic models in patients with hepatitis B-associated hepatocellular carcinoma: a retrospective observational study.

ABSTRACT

BACKGROUND AND OBJECTIVE: The objective of this study was to investigate the prognostic significance of several inflammation-based models in hepatitis B-associated hepatocellular carcinoma (HCC). PATIENTS AND METHODS: We retrospectively reviewed 470 cases of hepatitis B-associated HCC. Preoperative data were collected to calculate the inflammation-based markers, including systemic immune-inflammation index (neutrophil×platelets/lymphocyte), platelets-to-lymphocyte ratio, and neutrophil-to-lymphocyte ratio (NLR). Overall survival and recurrence-free survival were estimated by the Kaplan-Meier method and Cox analysis. RESULTS: During a median follow-up time of 29 months, 34.0% (160/470) of patients died and 36.0% (169/470) experienced recurrence. Compared with patients with lower scores of inflammation models, patients in the higher group had larger tumor diameter and higher risk of vascular invasion (both P<0.05). Multivariate analysis revealed that age, tumor size, platelets-to-lymphocyte ratio, NLR, and systemic immune-inflammation index were the independent predictors for both overall survival and recurrence-free survival. Furthermore, the combination of tumor size and NLR showed a significantly better discrimination ability for survival (C-index=0.716, 95% confidence interval 0.664-0.768) than both Barcelona Clinic Liver Cancer and Cancer of Liver Italian Program. CONCLUSION: The inflammation-based markers, in particular the combination of NLR with tumor size, are effective tools for assessing prognosis in hepatitis B-associated HCC.