Off-Therapy Response After Nucleos(t)ide Analogue Withdrawal in Patients With Chronic Hepatitis B: An International, Multicenter, Multiethnic Cohort (RETRACT-B Study)

G. Hirode, H. S. J. Choi, C. H. Chen, T. H. Su, W. K. Seto, S. Van Hees, M. Papatheodoridi, S. Lens, G. Wong, S. M. Brakenhoff, R. N. Chien, J. Feld, M. J. Sonneveld, H. L. Y. Chan, X. Forns, G. V. Papatheodoridis, T. Vanwolleghem, M. F. Yuen, Y. C. Hsu, J. H. KaoM. Cornberg, B. E. Hansen, W. J. Jeng, H. L. A. Janssen*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

76 Citations (Scopus)
93 Downloads (Pure)

Abstract

Background & Aims: Functional cure, defined based on hepatitis B surface antigen (HBsAg) loss, is rare during nucleos(t)ide analogue (NA) therapy and guidelines on finite NA therapy have not been well established. We aim to analyze off-therapy outcomes after NA cessation in a large, international, multicenter, multiethnic cohort of patients with chronic hepatitis B (CHB). Methods: This cohort study included patients with virally suppressed CHB who were hepatitis B e antigen (HBeAg)–negative and stopped NA therapy. Primary outcome was HBsAg loss after NA cessation, and secondary outcomes included virologic, biochemical, and clinical relapse, alanine aminotransferase flare, retreatment, and liver-related events after NA cessation. Results: Among 1552 patients with CHB, cumulative probability of HBsAg loss was 3.2% at 12 months and 13.0% at 48 months of follow-up. HBsAg loss was higher among Whites (vs Asians: subdistribution hazard ratio, 6.8; 95% confidence interval, 2.7–16.8; P < .001) and among patients with HBsAg levels <100 IU/mL at end of therapy (vs ≥100 IU/mL: subdistribution hazard ratio, 22.5; 95% confidence interval, 13.1–38.7; P < .001). At 48 months of follow-up, Whites with HBsAg levels <1000 IU/mL and Asians with HBsAg levels <100 IU/mL at end of therapy had a high predicted probability of HBsAg loss (>30%). Incidence rate of hepatic decompensation and hepatocellular carcinoma was 0.48 per 1000 person-years and 0.29 per 1000 person-years, respectively. Death occurred in 7/19 decompensated patients and 2/14 patients with hepatocellular carcinoma. Conclusions: The best candidates for NA withdrawal are virally suppressed, HBeAg- negative, noncirrhotic patients with CHB with low HBsAg levels, particularly Whites with <1000 IU/mL and Asians with <100 IU/mL. However, strict surveillance is recommended to prevent deterioration.

Original languageEnglish
Pages (from-to)757-771.e4
JournalGastroenterology
Volume162
Issue number3
DOIs
Publication statusPublished - Mar 2022

Bibliographical note

on Behalf of theRETRACT-B Study Group

Fingerprint

Dive into the research topics of 'Off-Therapy Response After Nucleos(t)ide Analogue Withdrawal in Patients With Chronic Hepatitis B: An International, Multicenter, Multiethnic Cohort (RETRACT-B Study)'. Together they form a unique fingerprint.

Cite this