Clinical Utility of SCALE-B to Predict Hepatitis B Virus Relapse, Hepatitis B Surface Antigen Loss After Antiviral Cessation in Asian Patients After 2-Year Follow-up

BackgroundDiscontinuation of antiviral therapy in chronic hepatitis B (CHB) patients leads to a higher hepatitis B surface antigen (HBsAg) loss; yet, clinical relapse (CR) may occur. SCALE-B score was developed to predict off-treatment CR; however, validation of SCALE-B beyond a 48-week follow-up is rare. We studied whether SCALE-B and hepatitis B virus ribonucleic acid (HBV RNA) could predict outcomes in CHB patients after a 2-year follow-up.MethodsA total of 92 Thai CHB patients who stopped antiviral treatment were followed up; baseline characteristics, quantitative hepatitis B surface antigen (qHBsAg), hepatitis B core-related antigen (HBcrAg), and HBV RNA were collected at the time of discontinuation, and SCALE-B scores were calculated. Patients were followed up every 12 weeks for 48 weeks, and then, the intervals were upon primary doctors. Follow-up data regarding virological relapse (VR), CR, and HBsAg loss were obtained.ResultsThe median follow-up duration was 142 weeks; the cumulative incidences of VR, CR, and HBsAg loss were 65.2, 33.7, and 7.6%, respectively. After 48 weeks, VR and CR plateaued, but HBsAg loss increased from 2.2 to 7.6%. According to the SCALE-B strata, VR, CR, and HBsAg loss were significantly different. The highest stratum (≥ 320) was associated with higher VR, CR, and lesser HBsAg loss when compared to the lowest stratum, with adjusted hazard ratios of 5.0 (95% CIs: 1.8–14.4), 10.44 (95% CIs: 1.4–79.1), and 0.04 (95% CIs: 0.004–0.43), respectively.ConclusionAt a median follow-up of 2.5 years after discontinuing therapy, HBsAg loss in Thai patients was found to increase over time. SCALE-B is a valuable tool for predicting CR, VR, and HBsAg loss; HBV RNA is not significantly associated with long-term outcomes.Clinical Trial Registration[www.ClinicalTrials.gov], identifier [TCTR20180316007].

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.