Serum HBV RNA levels are associated with risk of hepatitis flare after stopping NA therapy in HBeAg-negative patients.

Abstract

Background and aims: Among HBeAg-negative chronic hepatitis B (CHB) patients on long-term nucelos (t)ide analogue (NA) therapy, treatment discontinuation has been associated with HBsAg loss and durable virological suppression, but also hepatitis flares requiring NA therapy to be restarted. The prediction of clinical outcomes after stopping NA therapy remains challenging. Quantification of circulating HBV RNA is a promising biomarker in CHB patients. We have evaluated the role of serum HBVRNA levels to predict clinical outcomes in a large cohort of patients enrolled in a prospective NASTOP study. Method(s): The Melbourne HBV-STOP study was a prospective multicentre study of NA discontinuation in 110 HBeAg-negative noncirrhotic patients who had achieved long-term virological suppression on treatment. All patients were followed for 96 weeks. In a subset, we performed an exploratory analysis of serum HBVRNA levels for predicting clinical outcomes after stopping NA therapy. HBVRNA levels were measured at baseline and longitudinally using the automated Roche cobas HBVRNA Investigational Assay. Clinical outcomes of interest included rates hepatitis flare (ALT>5xULN), disease remission (HBV DNA<2000 IU/ml and ALT<2xULN) and HBsAg loss, as well as rates of NA re-treatment. Result(s): HBVRNA levels were tested at baseline and off-treatment in 65 patients. At baseline, the median age was 56 yrs, 54% were male, and 75% were Asian. Median HBsAg level was 701 (1.6-19912) IU/ml. Following treatment withdrawal, virological reactivation occurred in all participants, 38% experienced a hepatitis flare (ALT>5xULN), 26% restarted NA therapy and 6% (n = 4) lost HBsAg by 96 weeks. At baseline, serum HBV RNA was detectable in 16/65 (25%). The detection of HBVRNA in serum at baseline was positively associated with risk of hepatitis flare [ALT >5 xULN, 63% vs 31%, OR = 3.8, 95% CI = (1.2, 12.3), p value = 0.03], as well as lower likelihood of sustained disease remission as well as HBsAg loss. HBVRNA levels were undetectable at baseline in all patients who achieved HBsAg loss. Of patients HBVRNA negative at baseline, HBVRNA levels became detectable in 47/49 patients during follow-up. The time to detection of HBVRNAwas shorter for patients stopping TDF than ETV (median 49 vs 90 days, p = 0.02). The detection of HBVRNA over time was associated with rising HBVDNA levels>104 IU/ml. HBVRNA levels>103 cp/ml off-treatment were associated with increased risk of hepatitis flare (ALT>5xULN). Conclusion(s): Serum HBVRNAwas detectable in a minority of HBeAgnegative patients on long-term NA therapy. The detection of any serum HBVRNA prior to stopping NA therapy, and rising HBVRNA offtreatment, were associated with risk of subsequent hepatitis flare, as well as lower likelihood of HBsAg loss. HBVRNA levels show promise as a biomarker for predicting clinical outcomes after stopping NA therapy in patients with HBeAg-negative CHBCopyright © 2022 European Association for the Study of the Liver