BACKGROUND: The effect of nucleos(t)ide analogs (NAs) versus interferon (IFN) on the occurrence of hepatocellular carcinoma (HCC) in chronic hepatitis B (CHB) is controversial. We assessed whether antiviral strategy affected HCC development in CHB patients at different HCC risks. METHODS: 1112 CHB patients with antiviral therapy were included in this retrospective study. Patients treated with NAs only were classified into NAs group (n = 682) while those received IFN treatment with or without NAs were defined as IFN group (n = 430). Propensity score matching (PSM) was applied to minimize baseline differences. RESULTS: Totally, 31 patients developed HCC during follow-up (median 5.41 years). The cumulative HCC incidence at 10 years was significantly lower in the IFN group than NAs group (2.7% vs 8.0%, p < 0.001). Similar results were obtained in the PSM-cohort. Patients with IFN-based treatment were less likely to develop HCC than those with NAs (Hazard ratio = 0.15; 95% CI 0.04-0.66; p = 0.012). Subgroup analyses demonstrated that this superiority of IFN in reducing HCC development was obvious in patients at high- but not low-risk of HCC. CONCLUSIONS: Reduction of HCC development was more significant in CHB patients at higher HCC risk with IFN-based therapy than NAs treatment.
BACKGROUND: The effect of nucleos(t)ide analogs (NAs) versus interferon (IFN) on the occurrence of hepatocellular carcinoma (HCC) in chronic hepatitis B (CHB) is controversial. We assessed whether antiviral strategy affected HCC development in CHB patients at different HCC risks. METHODS: 1112 CHB patients with antiviral therapy were included in this retrospective study. Patients treated with NAs only were classified into NAs group (n = 682) while those received IFN treatment with or without NAs were defined as IFN group (n = 430). Propensity score matching (PSM) was applied to minimize baseline differences. RESULTS: Totally, 31 patients developed HCC during follow-up (median 5.41 years). The cumulative HCC incidence at 10 years was significantly lower in the IFN group than NAs group (2.7% vs 8.0%, p < 0.001). Similar results were obtained in the PSM-cohort. Patients with IFN-based treatment were less likely to develop HCC than those with NAs (Hazard ratio = 0.15; 95% CI 0.04-0.66; p = 0.012). Subgroup analyses demonstrated that this superiority of IFN in reducing HCC development was obvious in patients at high- but not low-risk of HCC. CONCLUSIONS: Reduction of HCC development was more significant in CHB patients at higher HCC risk with IFN-based therapy than NAs treatment.