Persistent High Rates of HBsAg Loss With IFN-α in Chronic HBV Suggestive of Its Potential

Blood sample with sexually transmitted diseases: HIV, HBV, HCV, Syphilis
Researchers conducted a retrospective medical chart review to analyze relationship between HBsAg loss and a single PEG(IFN-α) course.

In chronic hepatitis B virus (HBV) infection, persistent high rates of hepatitis B surface antigen (HBsAg) loss were observed from a single course of pegylated/conventional interferon-alpha (IFN-α) treatment, according to a long-term follow-up study published in Clinical Gastroenterology and Hepatology. The persistent effects of IFN-α suggest its potential role in novel combination therapies in search of a functional cure.

In this retrospective medical chart review, researchers analyzed data from hepatitis B e antigen (HBeAg)-positive patients with chronic hepatitis B treated with pegylated or conventional IFN-α between 2002 and 2014 at the Erasmus University Medical Center in Rotterdam, Netherlands. The primary endpoint was HBsAg loss with or without HBsAg antibody seroconversion, defined as a negative qualitative HBsAg result after the first dose of IFN-α treatment. Secondary endpoints included HBeAg loss with or without seroconversion and clinical outcomes (all-cause mortality, liver-related morbidity). Patients were censored upon retreatment.

The study included 267 patients, 67.4% men, median age of 32 years at baseline (interquartile range [IQR], 24-42), 57.7% White and 33.7% Asians. The median follow-up duration was 11.5 years (IQR, 6.6-19.0), and 57.7% of the patients were followed for more than 10 years.

The annual incidence of HBsAg loss remained elevated over the years, reaching 5.3% HBsAg loss per year by 6 years from the start of treatment, leading to a cumulative incidence of 13.9% in 5 years and 31.5% in 10 years. Incidence rates of HBeAg and HBsAg loss were 14.7 and 3.2 per 100 person-years (PY) of follow-up, respectively, censored at retreatment.

Baseline factors associated with a higher rate of HBsAg loss were White, at least forty years of age, HBV genotype A, and presence of cirrhosis at the start of IFN-α treatment (P ≤.05), but not the type IFN-α given, treatment strategy, treatment duration, nor baseline alanine aminotransferase or HBV DNA. Men reached borderline significance (P =.056).

Of the 267 patients, 147 (55.1%) were re-treated: 11.6% (17/147) with IFN-α, 60.5% (89/147) with nucleo(s)tide analogues (NAs), and 27.9% (41/147) with a combination of IFN-α and NAs. In a subanalysis of the 147 patients pretreatment after the first course of IFN-α, the incidence rates of HBeAg and HBsAg loss were 10.4 (95% CI, 8.7-12.6) and 1.0 (95% CI, 0.7-1.6) per 100 PY, respectively. Seroclearance rate was higher in patients who were re-treated with IFN-α vs NAs (2.9 [95% CI, 1.4-6.1] vs 0.5 [95% CI, 0.2-1.2] per 100 PY; P =.008).

HBeAg loss was significantly associated with a decreased incidence rate of adverse clinical outcomes. Compared with those who did not achieve HBeAg loss, incidence rates of clinical events were lower in patients who did achieve HBeAg loss (0.4 vs 3.1 per 100 PY; P <.001). While the incidence event rate was lower in the 45 patients who achieved HBsAg loss vs the 222 patients who did not (0.8 vs 1.2 per 100 PY) when researchers included all follow-up time, this difference was not statistically significant (P =.413). However, when follow-up was censored at 10 years from baseline, both HBeAg and HBsAg loss became strongly associated with improved patient outcomes (HBeAg loss: 0.1 vs 2.7 per 100 PY; P =.001; HBsAg loss: 0.1 vs 1.0 per 100 PY; P =.052). Early response to IFN-α therapy was associated with a significantly higher incidence rate of HBsAg loss (6.7 vs 1.7 per 100 PY; P <.001) and lower incidence rate of clinical events (0.3 vs 1.4 per 100 PY; P =.020).

Inherent limitations include infrequent measurements for HBeAg and HBsAg as well as a lack of confirmatory testing. Time-related biased could be present when incidence rates of clinical events were compared, as follow-up was calculated from baseline to response or end of follow-up for both HBeAg/HBsAg loss and persistent groups.

Incidence rates of HBsAg loss did not differ significantly between those who received short-term (≤24 weeks) vs long-term (>24 weeks) therapy. “The lack of superiority suggests that short-term IFN-α treatment could also be effective in inducing favorable response in the long term,” researchers noted. “However, as these are univariable effects and studies have reported on the superiority of long-term over short-term treatment at 6 months post-treatment, further investigation would be needed to adjust for any confounders and reveal its true effects.”

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


Choi HSJ, van Campenhout MJH, van Vuuren AJ, et al. Ultra long-term follow-up of interferon-alpha treatment for HBeAg-positive chronic hepatitis B virus infection. Published online September 2, 2020. Clin Gastroenterol Hepatol. doi: 10.1016/j.cgh.2020.09.004