Patients Remain Undetectable 96 Weeks After Stopping Bulevirtide

How long a patient with chronic hepatitis D had maintained undetectable HDV RNA before stopping bulevirtide was the only variable a multivariate model retained as a predictor of whether the suppression lasted. Among patients undetectable for at least 96 weeks at the end of treatment, 9 of 10 stayed undetectable across 2 years off therapy; among those undetectable less than 48 weeks, 3 of 32 did.¹

A Severe Disease With an Open-Ended Prescription

Hepatitis D causes the most severe form of chronic viral hepatitis, affecting an estimated 9 million to 19 million people worldwide, and it requires hepatitis B surface antigen (HBsAg) as its envelope protein to propagate. Compared with HBV monoinfection, it carries up to 3-fold higher risk of hepatocellular carcinoma and a mortality rate exceeding 35% over 10 years. Bulevirtide, a first-in-class entry inhibitor that binds the sodium taurocholate cotransporting polypeptide, is approved in the US, the European Economic Area, the UK, Switzerland, Russia, Australia, and Canada.

What nobody knew was whether patients could stop taking the medication, since earlier analyses of this trial reported only on-treatment outcomes. For a daily subcutaneous injection with no defined end point, whether a subset of patients can finish determines both what patients face and what payers commit to.

MYR301 randomized 150 patients 1:1:1, stratified by cirrhosis, to bulevirtide 2 mg daily or 10 mg daily for 144 weeks, or to a 48-week delay followed by 10 mg daily for 96 weeks; the dose approved in the United States delivers 8.5 mg daily. All groups then entered 96 weeks of post-treatment follow-up. Roughly 47% had cirrhosis at baseline, and 54% to 65% were taking a concomitant nucleos(t)ide analogue.

While 92% of patients remained at the end of treatment, 72% completed 48 weeks of follow-up and 57% completed 96 weeks, with withdrawal of consent the most common reason for leaving. Because binary end points used a missing-equals-failure approach, the follow-up response rates run conservative.

What Held and What Did Not

At week 144, virologic response reached 73%, 76%, and 92% in the 2-mg, 10-mg, and delayed-treatment groups, with ALT normalization at 59%, 60%, and 58% and combined response at 57%, 54%, and 56%. Undetectable HDV RNA separated the doses: 29% with 2 mg vs 50% with 10 mg (P = .040), and 52% in the delayed group. Of the 33% with a suboptimal virologic response at week 24, 84% of partial responders and half of nonresponders reached virologic response by the end of treatment.

Stopping undid much of it. The virologic response fell to 33%, 30%, and 32% by week 96 of follow-up; the combined response was 24% in every group, and the undetectable HDV RNA was roughly 20% across the board. HBsAg barely changed throughout, with 4 patients losing it across the entire trial.

Of 65 patients with undetectable HDV RNA at the end of treatment, 64 had follow-up data, and 23 of them (36%) never relapsed. Sorted by how long they had been continuously undetectable before stopping, the rates were 90% at 96 weeks or more, 50% at 48 to fewer than 96 weeks, and 9% below 48 weeks; low baseline HDV RNA and HBsAg predicted it only in univariate analysis. Relapses came early: 93% occurred by follow-up week 24, and none occurred between weeks 48 and 96.

A nationwide Austrian cohort of 61 patients at 10 centers offers an early real-world read on that pattern.² Ten stopped bulevirtide electively after a median of 23 months, 3 after add-on pegylated interferon, and 7 stayed undetectable for a median of 36 months off therapy. The 3 who relapsed had been undetectable for just 3, 4, and 6 months before stopping, and all regained undetectable HDV RNA on retreatment. The investigators identified no consistent stopping rule.

The Cost of Stopping

In MYR301, discontinuation was not benign.¹ ALT rose above 5 times the upper limit of normal in 41% of patients afterward and above 10 times in 10%, most of it by follow-up week 24, asymptomatic, and accompanied by HDV rebound. Patients with cirrhosis were overrepresented, with 43% of that group experiencing a flare above 5 times the upper limit. Twenty patients had hepatic serious adverse events post-treatment, resolving in 17, and 4 were hospitalized.

On-treatment safety was unremarkable by comparison, with no discontinuations for adverse events, no serious adverse events or deaths attributed to the drug, and adherence at or above 93%. Bile salt elevations were dose-dependent but did not produce pruritus.

The trial was open-label, since placebo injections were deemed unethical, and enrolled a predominantly White population. Gilead Sciences sponsored it, conducted the data analysis, and provided writing assistance. The authors framed response-guided stopping as a possibility requiring further study rather than a recommendation, advised at least 6 months of hepatic monitoring after discontinuation, and noted that most patients still benefit from continuing. What the data do support is that the stopping conversation now has a number attached to it.

References

1. Wedemeyer H, Aleman S, Blank A, et al. 144 weeks of bulevirtide monotherapy for chronic hepatitis D: final and post-treatment results from a phase III randomized trial. J Hepatol. 2026;85(3):493-503. doi:10.1016/j.jhep.2026.03.046

2. Schwarz M, Hintersteininger M, Schwarz C, et al. Response-guided bulevirtide ± pegylated interferon alfa-2a: long-term outcomes observed in the nationwide Austrian hepatitis D cohort study. JHEP Rep. 2026;8(6):101835. doi:10.1016/j.jhepr.2026.101835