Letter: Safety after cessation of nucleos(t)ide analogue therapy in patients with chronic hepatitis B infection

We read with great concern of the study by Hsu et al¹ on severe hepatitis flare with decompensation after cessation of nucleos(t)ide analogue (NUC) based on a large database of patients with chronic hepatitis B (CHB) infection in Taiwan. The study showed 4-year cumulative incidence of off-therapy hepatic decompensation and mortality/liver transplantation 1.8% and 0.7%, respectively; the 4-year incidence of hepatic decompensation decreased to 1.3% in patients without evidence of cirrhosis and 1.1% in those adhering to a standardised stopping rule. In addition to the inherent problems of an administrative database, several points deserve clarification and further discussion.

First, HBeAg-positive and negative CHB are different disease entities with different risks of off-therapy hepatic decompensation. In our off-NUC cohort including 295 HBeAg-positive and 1234 HBeAg-negative patients, the cumulative decompensation incidence was higher in HBeAg-positive than in HBeAg-negative patients: 1-year 2.4% vs. 0.97%, 2-year 2.4% vs. 1.2%.^{2, 3} Among patients without cirrhosis in our off-NUC cohort, the 2-year cumulative incidence of decompensation was 0.8% and 0.1%, respectively, but 8-fold higher in HBeAg-positive patients. Given such differences, HBeAg-positive and HBeAg-negative patients need to be addressed separately. Second, off-therapy flares in patients stopping tenofovir disoproxil fumarate (TDF) occur more frequently and are more severe than off-entecavir flares, and TDF use and cirrhosis are independent predictors of severe flare and hepatic decompensation.^{2, 4} This study did not stratify the decompensation risk by NUC types and did not include the NUC types into the regression analysis. Third, most patients in this study lacked the information of consolidation duration, and the proportion of self-discontinuation of NUC or loss-to-follow-up, all of which are crucial factors associated with off-therapy severe flare.⁵ The 5-year cumulative incidence of on-treatment loss-to-follow-up may be up to 11%.^{6, 7}

Patients who stop NUC by themselves or who are lost to follow-up may have similar risk of hepatitis flare but higher risk of decompensation or liver-related mortality because they would be not properly monitored and treated in time.⁸ Of note, the rates of decompensation in off-NUC studies are not higher than those from long-term NUC treatment, even in patients with cirrhosis.⁹

Safety after stopping NUC is of paramount importance. However, this database study may have provided inflated data of off-NUC hepatic decompensation as compared with those from hospital-based studies. This may highlight the importance of proper off-NUC monitoring and education of patients.

Yen-Chun Liu: Formal analysis (lead); writing – original draft (lead). Wen-Juei Jeng: Conceptualization (lead); supervision (equal). Rong-Nan Chien: Supervision (equal); writing – review and editing (lead).

The authors have no financial and personal relationships with other people or organisations that could inappropriately influence (bias) their work.

This article is linked to Hsu et al papers. To view these articles, visit https://doi.org/10.1111/apt.17614 and https://doi.org/10.1111/apt.17681