Author information

¹St Vincent's Hospital Melbourne, Fitrozy, Victoria, Australia.

²Department of Medicine, University of Melbourne, Parkville, Victoria, Australia.

³Victorian Infectious Diseases Reference Laboratory, Department of Infectious Diseases, Royal Melbourne Hospital, University of Melbourne at the Peter Doherty Institute for Infection and Immunity, Melbourne, Victoria, Australia.

⁴Department of Medicine, School of Clinical Medicine, The University of Hong Kong, Hong Kong SAR, China.

⁵Western Health, Melbourne, Australia.

⁶Monash Health, Melbourne, Victoria, Australia.

⁷Liverpool Hospital, Sydney, New South Wales, Australia.

⁸Alfred Health, Melbourne, Victoria, Australia.

⁹Eastern Health, Melbourne, Victoria, Australia.

¹⁰Royal Prince Alfred Hospital, Sydney, New South Wales, Australia.

¹¹Concord Hospital, Sydney, New South Wales, Australia.

¹²Austin Health, Melbourne, Victoria, Australia.

¹³Bankstown-Lindcombe Hospital, Sydney, New South Wales, Australia.

¹⁴St Vincent's Hospital Sydney, Sydney, New South Wales, Australia.

Abstract

Background and aims: Accurate biomarkers to predict outcomes following discontinuation of nucleos(t)ide analogue (NA) therapy are needed. We evaluated serum hepatitis B core-related antigen (HBcrAg) level as a biomarker for predicting outcomes after NA discontinuation.

Methods: Patients with HBeAg-negative chronic hepatitis B (CHB) without cirrhosis were enrolled in a prospective trial evaluating clinical outcomes until 96 weeks after NA discontinuation. End of treatment (EOT) and off-treatment levels of serum HBcrAg, HBsAg, HBV RNA and HBV DNA were used to predict key clinical outcomes including hepatitis flare (ALT ≥5 × ULN and HBV DNA > 2000 IU/mL). The SCALE-B score was calculated for the purposes of model validation.

Results: HBcrAg was tested amongst 65 participants. The median age was 54 years, 54% were male and 83% were Asian. HBcrAg was detectable in 86% patients. HBcrAg level ≥4 log U/mL at EOT was predictive of hepatitis flare [8/10 (80%) vs. 17/55 (31%), p = .001]. The presence of either HBcrAg ≥4 log U/mL or detectable HBV RNA at EOT predicted for both biochemical relapse and hepatitis flare. The SCALE-B model at EOT predicted for virological relapse, biochemical relapse, hepatitis flare and HBsAg loss in this cohort. An increase in the serum HBcrAg level off-treatment was also associated with hepatitis flare. No participant with EOT HBcrAg level ≥4 log U/mL achieved HBsAg loss.

Conclusions: High levels of serum HBcrAg predict for hepatitis flare after stopping NA therapy and low likelihood of HBsAg loss at week 96. People with high levels of serum HBcrAg are not suitable candidates for NA discontinuation.