Long-term nucleos(t)ide analogue (NUC) therapy suppresses hepatitis B virus (HBV) replication but rarely eradicates the virus, so treatment is usually lifelong. A landmark withdrawal experiment in patients whose intrahepatic covalently closed circular DNA (cccDNA) was undetectable showed that HBV DNA still rebounds in nearly all patients after stopping, establishing that even biopsy-negative cccDNA status cannot identify safe treatment withdrawal. The open problem is to make finite NUC therapy safe and predictable: (1) identify and prospectively validate predictor combinations — intrahepatic cccDNA together with serum hepatitis B core-related antigen (HBcrAg), hepatitis B surface antigen (HBsAg), and HBV RNA — that discriminate patients who will maintain durable off-therapy control (ideally with HBsAg seroclearance) from those who will relapse; (2) determine the optimal threshold and timing for resuming NUCs after virologic rebound, balancing harmful hepatic flares against the possibility of immune-mediated flares that lead to HBsAg loss; and (3) develop and standardize sufficiently sensitive assays for intrahepatic cccDNA and for serum HBV RNA to support these decisions. Substantial 2020–2026 progress — the multiethnic RETRACT-B cohort, the STOP-NUC and Nuc-Stop randomized trials, and guideline updates — has validated end-of-treatment HBsAg level as the dominant patient-selection marker and confirmed agent-specific rebound behavior, but a validated multi-marker prediction rule, a settled re-treatment strategy, and standardized cccDNA/HBV RNA assays are still lacking, so the core problem remains open.

Background

Chronic hepatitis B (CHB), affecting over 250 million people worldwide, is treated with nucleos(t)ide analogues (NUCs) — oral antivirals such as entecavir (ETV), tenofovir disoproxil fumarate (TDF), tenofovir alafenamide, and telbivudine (LdT) — which potently suppress HBV DNA but must typically be taken for life. The viral reservoir that sustains infection is covalently closed circular DNA (cccDNA), a stable episomal form of the HBV genome that persists in hepatocyte nuclei; because NUCs do not degrade cccDNA, cure is rarely achieved. The ideal endpoint, called functional cure, is loss of hepatitis B surface antigen (HBsAg, the viral envelope protein detectable in serum) with sustained viral suppression; HBsAg loss is associated with markedly improved clinical outcomes. Serum markers of residual viral transcriptional activity include HBcrAg (hepatitis B core-related antigen, a composite antigen reflecting core/precore protein production from cccDNA) and HBV RNA (primarily pregenomic RNA released as virion-like particles). Because NUCs block DNA synthesis but not RNA production, these markers can remain detectable even when HBV DNA is fully suppressed. Patients are also stratified by hepatitis B e antigen (HBeAg, a secreted viral antigen whose presence marks a phase of high viral replication) status, which strongly shapes relapse behavior and marker interpretation after withdrawal.

Whether NUCs can ever be safely stopped has long been debated. On one hand, lifelong therapy is costly and adherence-dependent; on the other, withdrawal almost always leads to virologic relapse and sometimes severe hepatitis flares (sharp alanine aminotransferase, ALT, elevations), which occasionally cause hepatic decompensation (acute liver failure with jaundice or ascites). A counterpoint is the observation that some patients achieve HBsAg loss after withdrawal, possibly because the transient return of viral replication re-activates antiviral immunity — a putative 'beneficial flare'. The problem was sharpened by Lai and colleagues (University of Hong Kong; JHEP Reports 2020), who performed the only withdrawal study in patients with biopsy-documented undetectable intrahepatic cccDNA after a median 13.4 years of NUC therapy: 12 of 13 patients who stopped had HBV DNA rebound to 100 IU/mL within 20 weeks and the thirteenth rebounded at week 70, several biochemical flares occurred only after re-treatment was restarted, and patients on TDF or LdT appeared to relapse earlier than those on ETV. This demonstrated that even the most stringent available evidence of viral silence — undetectable cccDNA in a liver biopsy — does not prevent rebound, presumably because minute residual cccDNA below assay sensitivity, biopsy sampling error, or HBsAg production from integrated viral DNA sustain infection. The study therefore posed the questions that define this problem: which combination of markers (intrahepatic cccDNA plus serum HBsAg, HBcrAg, HBV RNA) can predict durable off-therapy control or HBsAg seroclearance; when, after rebound, should NUCs be restarted so as to avoid uncontrolled flares without aborting potentially beneficial immune-mediated ones; and can more sensitive, standardized assays (such as droplet digital PCR, a partition-based PCR method that counts individual DNA molecules, for cccDNA) make these markers clinically reliable. Since 2020 this has become one of the most active questions in hepatology, addressed by the international RETRACT-B observational consortium, the German STOP-NUC randomized trial, the Scandinavian Nuc-Stop randomized trial of re-start strategies, and updated clinical practice guidelines.

Problem Statement

For patients with chronic hepatitis B on long-term nucleos(t)ide analogue (NUC) therapy, define a clinically deployable strategy for finite NUC therapy, substantiated with the available evidence and including a design for its prospective validation, jointly comprising: (1) a predictor set — the combination of markers selected from intrahepatic cccDNA, serum HBsAg, HBcrAg, HBV RNA, and relevant clinical covariates — that best discriminates patients who can discontinue NUCs with durable off-therapy control from those who will relapse, where durable off-therapy control is the sustained off-therapy response (HBV DNA below 2,000 IU/mL with normal ALT and no re-treatment, maintained through the end of the off-treatment follow-up), with HBsAg seroclearance as the ideal sub-outcome, and where quantified discrimination and calibration are to be estimated from current evidence for the HBeAg-negative and HBeAg-positive populations where the predictor is to be used, together with the performance target that its prospective validation must confirm; (2) a monitoring and re-treatment rule specifying when NUCs should be resumed after virologic rebound — the ALT and HBV DNA thresholds and their timing — that optimally balances the risk of uncontrolled hepatic flare and decompensation against the possibility of immune-mediated flares promoting HBsAg loss; and (3) the assay foundation making these decisions reliable: the sensitive and standardized assays required for intrahepatic cccDNA and for serum HBV RNA, specified by the limits of detection, reproducibility, and cross-laboratory performance needed for clinical decision-making. These three components serve one objective — making treatment withdrawal safe and predictable.

References

  1. Lai C-L, Wong DK-H, Wong GT-Y, Seto W-K, Fung J, Yuen M-F. Rebound of HBV DNA after cessation of nucleos/tide analogues in chronic hepatitis B patients with undetectable covalently closed circular DNA. JHEP Reports 2020;2(3):100112. doi:10.1016/j.jhepr.2020.100112, https://doi.org/10.1016/j.jhepr.2020.100112 (PMC7242874)
  2. Hirode G, Choi HSJ, Chen C-H, Su T-H, Seto WK, Van Hees S, Papatheodoridi M, Lens S, Wong G, Brakenhoff SM, Chien R-N, Feld J, Sonneveld MJ, Chan HLY, Forns X, Papatheodoridis GV, Vanwolleghem T, Yuen MF, Hsu Y-C, Kao JH, Cornberg M, Hansen BE, Jeng W-J, Janssen HLA; RETRACT-B Study Group. Off-Therapy Response After Nucleos(t)ide Analogue Withdrawal in Patients With Chronic Hepatitis B: An International, Multicenter, Multiethnic Cohort (RETRACT-B Study). Gastroenterology 2022;162(3):757-771. doi:10.1053/j.gastro.2021.11.002, https://doi.org/10.1053/j.gastro.2021.11.002
  3. van Bömmel F, Stein K, Heyne R, Petersen J, Buggisch P, Berg C, Zeuzem S, Stallmach A, Sprinzl M, Schott E, Pathil-Warth A, von Arnim U, Keitel V, Lohmeyer J, Simon K-G, Trautwein C, Trein A, Hüppe D, Cornberg M, Lammert F, Ingiliz P, Zachoval R, Hinrichsen H, Zipprich A, Klinker H, Schulze zur Wiesch J, Schmiedeknecht A, Brosteanu O, Berg T. A multicenter randomized-controlled trial of nucleos(t)ide analogue cessation in HBeAg-negative chronic hepatitis B. Journal of Hepatology 2023;78(5):926-936. doi:10.1016/j.jhep.2022.12.018, https://doi.org/10.1016/j.jhep.2022.12.018
  4. Johannessen A, Reikvam DH, Aleman S, Berhe N, Weis N, Desalegn H, Stenstad T, Heggelund L, Samuelsen E, Karlsen LN, Lindahl K, Pettersen FO, Iversen J, Kleppa E, Bollerup S, Winckelmann AA, Brugger-Synnes P, Simonsen HE, Svendsen J, Kran AB, Holmberg M, Olsen IC, Rueegg CS, Dalgard O. Clinical trial: An open-label, randomised trial of different re-start strategies after treatment withdrawal in HBeAg negative chronic hepatitis B. Alimentary Pharmacology & Therapeutics 2024;60(4):434-445. doi:10.1111/apt.18147, https://doi.org/10.1111/apt.18147
  5. Dongelmans EJ, Hirode G, Hansen BE, Chen C-H, Su T-H, Seto WK, Furquim d'Almeida A, van Hees S, Papatheodoridi M, Lens S, Wong GLH, Brakenhoff SM, Chien R-N, Feld JJ, Chan HLY, Forns X, Papatheodoridis GV, Vanwolleghem T, Yuen MF, Hsu Y-C, Kao J-H, Cornberg M, Sonneveld MJ, Jeng W-J, Janssen HLA; RETRACT-B study group. Predictors of hepatic flares after nucleos(t)ide analogue cessation - Results of a global cohort study (RETRACT-B study). Journal of Hepatology 2025;82(3):446-455. doi:10.1016/j.jhep.2024.08.015, https://doi.org/10.1016/j.jhep.2024.08.015
  6. Dongelmans E, Hirode G, van Bömmel F, Jeng W-J, Chien R-N, Chen C-H, Kilany M, Su T-H, Kao J-H, Seto WK, Liang LY, Reikvam DH, Holmberg M, Furquim d'Almeida A, Papatheodoridi M, Maasoumy B, Hansen B, Post R, Pocurull A, Terrault N, Ghany M, Johannessen A, Lok A, Vanwolleghem T, Lens S, Cornberg M, Yuen MF, Wong G, Sonneveld M, Papatheodoridis G, Berg T, Hsu Y-C, Feld J, Janssen H; RETRACT-B study group. Long-term outcomes after nucleos(t)ide analogue cessation in chronic hepatitis B - follow-up from the RETRACT-B cohort. Journal of Hepatology 2026;85(1):61-70. doi:10.1016/j.jhep.2026.03.030, https://doi.org/10.1016/j.jhep.2026.03.030
  7. Hirode G, Hansen BE, Chen C-H, Su T-H, Wong GLH, Seto WK, d'Almeida AF, Papatheodoridi M, Brakenhoff SM, Lens S, Choi HSJ, Chien R-N, Feld JJ, Forns X, Sonneveld MJ, Papatheodoridis GV, Vanwolleghem T, Yuen MF, Chan HLY, Kao J-H, Hsu Y-C, Cornberg M, Jeng W-J, Janssen HLA; RETRACT-B study group. Limited Sustained Remission After Nucleos(t)ide Analog Withdrawal: Results From a Large, Global, Multiethnic Cohort of Patients With Chronic Hepatitis B (RETRACT-B Study). American Journal of Gastroenterology 2024;119(9):1849-1856. doi:10.14309/ajg.0000000000002759, https://doi.org/10.14309/ajg.0000000000002759
  8. Fan R, Peng J, Xie Q, Tan D, Xu M, Niu J, Wang H, Ren H, Chen X, Wang M, Sheng J, Tang H, Bai X, Wu Y, Zhou B, Sun J, Hou J; Chronic Hepatitis B Study Consortium. Combining Hepatitis B Virus RNA and Hepatitis B Core-Related Antigen: Guidance for Safely Stopping Nucleos(t)ide Analogues in Hepatitis B e Antigen-Positive Patients With Chronic Hepatitis B. Journal of Infectious Diseases 2020;222(4):611-618. doi:10.1093/infdis/jiaa136, https://doi.org/10.1093/infdis/jiaa136
  9. Tseng T-N, Jeng W-J, Hu T-H, Wang J-H, Hung C-H, Lu S-N, Chen C-H. Combined baseline HBcrAg and end-of-treatment HBsAg predict HBV relapse after entecavir or tenofovir cessation. Journal of Antimicrobial Chemotherapy 2023;78(2):436-439. doi:10.1093/jac/dkac409, https://doi.org/10.1093/jac/dkac409
  10. Ohlendorf V, Wübbolding M, Höner zu Siederdissen C, Bremer B, Deterding K, Wedemeyer H, Cornberg M, Maasoumy B. Limited Value of HBV-RNA for Relapse Prediction After Nucleos(t)ide Analogue Withdrawal in HBeAg-negative Hepatitis B Patients. Journal of Viral Hepatitis 2025;32(4):e14026. doi:10.1111/jvh.14026, https://doi.org/10.1111/jvh.14026
  11. Hayashi S, Isogawa M, Kawashima K, Ito K, Chuaypen N, Morine Y, Shimada M, Higashi-Kuwata N, Watanabe T, Tangkijvanich P, Mitsuya H, Tanaka Y. Droplet digital PCR assay provides intrahepatic HBV cccDNA quantification tool for clinical application. Scientific Reports 2022;12(1):2133. doi:10.1038/s41598-022-05882-9, https://doi.org/10.1038/s41598-022-05882-9
  12. Holmberg M, Niehrs A, Dalgard O, Berhe N, Desalegn H, Aleman S, Weis N, Stenstad T, Heggelund L, Samuelsen E, Karlsen LN, Lindahl K, Kleppa E, Winckelmann AA, Brugger-Synnes P, Simonsen HE, Svendsen J, Björkström NK, Reikvam DH, Johannessen A. Hepatic flares, their immune signatures, and ALT variability after nucleos(t)ide analogue cessation in HBeAg-negative hepatitis B. JHEP Reports 2026;8(7):101875. doi:10.1016/j.jhepr.2026.101875, https://doi.org/10.1016/j.jhepr.2026.101875
  13. Chen Y-C, Hsu C-W, Chien R-N. Higher HBeAg-reversion virological relapse and lower sustained remission after treatment cessation in tenofovir-treated HBeAg-positive patients. Journal of Medical Virology 2023;95(11):e29213. doi:10.1002/jmv.29213, https://doi.org/10.1002/jmv.29213
  14. Sattayalertyanyong O, Bandidniyamanon W, Limsrivilai J, Chotiyaputta W, Nimanong S, Tanwandee T. Long-term HBsAg clearance, relapse, and safety following tenofovir or entecavir discontinuation in non-cirrhotic, HBeAg-negative chronic hepatitis. Scientific Reports 2026;16(1):10404. doi:10.1038/s41598-026-40972-y, https://doi.org/10.1038/s41598-026-40972-y
  15. European Association for the Study of the Liver (Cornberg M, Sandmann L, Jaroszewicz J, Kennedy P, Lampertico P, Lemoine M, Lens S, Testoni B, Wong GLH, Russo FP). EASL Clinical Practice Guidelines on the management of hepatitis B virus infection. Journal of Hepatology 2025;83(2):502-583. doi:10.1016/j.jhep.2025.03.018, https://doi.org/10.1016/j.jhep.2025.03.018
  16. Maung ST, Chaiteerakij R. Scoping Review on Strategies for Safe Nucleot(s)ide Analogue Discontinuation and Optimising Functional Cure in Chronic Hepatitis B. Journal of Viral Hepatitis 2025;32(7):e70040. doi:10.1111/jvh.70040, https://doi.org/10.1111/jvh.70040
  17. Lai JC, Colombatto P, Wong GL, Brunetto MR. Why and when could nucleos(t)ide analogues treatment be withdrawn? Digestive and Liver Disease 2025;57(2):558-563. doi:10.1016/j.dld.2024.10.012, https://doi.org/10.1016/j.dld.2024.10.012