Clinical usefulness of noninvasive fibrosis indices for predicting hepatocellular carcinoma in treatment-naïve patients with chronic hepatitis B following entecavir therapy.

Abstract

This study aimed to evaluate the clinical usefulness of the aminotransferase to platelet ratio index (APRI), fibrosis-4 (FIB-4), and modified FIB-4 (mFIB-4) indices in predicting hepatocellular carcinoma (HCC) in patients receiving entecavir (ETV) treatment. Among 1955 patients treated with ETV, a total of 857 treatment-naive chronic hepatitis B patients (424 with liver cirrhosis [LC], 433 without cirrhosis) treated with ETV for more than 1 year were analyzed. Of the 857 patients, 85 (9.9%) patients (77 in the LC group and 8 in the non-LC group) developed HCC during the follow-up period. The median observation period was 6.9 years. Multivariate regression analysis of HCC incidence revealed that the initial mFIB-4 index (hazard ratio [HR] 1.058; 95% confidence interval [CI], 1.007-1.112; p=0.027) and improvement in the FIB-4 index after 1 year of ETV treatment (HR 0.531; 95% CI, 0.339-0.831; p=0.006) were independent prognostic factors in the entire cohort. In the LC group, the improvement of the FIB-4 index following ETV treatment (HR 0.491; 95% CI, 0.280-0.861; p=0.013) was negatively correlated with incidence of HCC. However, the area under the receiver operating characteristic curve of specific cut-off values of the FIB-4 index at baseline and 1 year after ETV treatment were 0.572 (95% CI, 0.504-0.640) and 0.615 (95% CI, 0.546-0.684), respectively. In the non-LC group, none of the invasive fibrosis indices could predict HCC incidence. The specific cut-off value of the FIB-4 index was not suitable for predicting HCC. However, the improvement in the FIB-4 index after 1year of ETV therapy could be a predictor of HCC development in cirrhotic patients.