Whether interferon (IFN)-α therapy is better than nucleos(t)ide analogs (NAs) in the prevention of adverse outcomes, including hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB) is still uncertain or controversial. This study aimed to compare the cumulative incidence of adverse outcomes in patients with CHB on IFN-α- and NA-based therapies.

This was a retrospective study of patients with CHB on antivirals. Patients treated with IFN-α (IFN-α or peginterferon-α) with or without NAs were defined as the IFN-α group, and those only receiving NAs were defined as the NAs group. Propensity score matching (PSM) was used to minimize baseline bias. Cox regression models were performed to select possible factors related to adverse outcomes development.

All 1247 patients were divided into the IFN-α (n = 877) and NAs (n = 370) groups. 26patients (20 and 6 in the NAs and IFN-α groups) developed adverse outcomes (decompensated cirrhosis, liver failure, HCC, liver transplantation and deaths) during a median follow-up of 5.2 years. The cumulative adverse outcomes occurrence at 10 years was significantly lower in the IFN-α group than in the NAs group in all (1.1% vs. 11.9%, P <0.001) and treatment-naïve (1.1% vs. 12.4%, P <0.001) patients. Similar trends were observed after PSM and differentiation of cirrhosis. Multivariate analysis before and after PSM showed that IFN-α-based treatment was independently associated with a lower adverse outcomes incidence (before/after PSM: P = 0.001/P = 0.002). HCC risk stratification analyses revealed that the superiority of IFN-α in preventing HCC was more significant in patients with high-risk HCC.

IFN-α-based therapy was superior to NAs in preventing adverse outcomes in patients with CHB regardless of cirrhosis, and in reducing HCC in those with a high risk of HCC.