Backgrounds: ALT ≥ 80 U/L and HBV DNA ≥ 2000 IU/ml are treatment criteria of APASL guidelines for chronic hepatitis B (CHB) patients. The need of antiviral therapy for patients in gray zone (ALT < 80 U/L or HBV DNA < 2000 IU/ml) is controversial. This study aimed to develop a scoring system to predict hepatocellular carcinoma (HCC) and evaluate the benefit of antiviral therapy in these patients. Methods: Seven hundred and forty-nine patients were analyzed. Significant variables were weighted to develop a scoring system for HCC prediction. The area under receiver operating curves (AUROC) were estimated and validated by REVEAL-HBV cohort (n = 3527). Results: Older age (p < 0.001), male sex (p = 0.036), family history of HCC (p = 0.002) and HBV DNA ≥ 2000 IU/ml (p = 0.045) were independently associated with HCC. A 14-point risk score system predicts 3 and 5-years HCC risk to be 0.866 and 0.868 of AUROC, respectively in the derivation cohort; 0.821 and 0.820, in the REVEAL-HBV cohort. The cumulative HCC incidence was higher in the high risk (score ≥ 8) group both in derivation and validation cohorts (p < 0.001). Patients with antiviral therapy had lower HCC incidence compared to those without (p = 0.016). Of note, antiviral therapy significantly decreased HCC in the high risk group (p = 0.005), but not in the low risk group (p = 0.705). Conclusions: A risk scoring system is established and validated. Of CHB patients in gray zone of APASL guidelines, those with risk scores ≥ 8 had higher risk of HCC, but the risk could be significantly reduced by antiviral therapy.
All Science Journal Classification (ASJC) codes