Mirum Pharmaceuticals said Nov. 10 that brelovitug met the primary endpoint in AZURE-4, the second of two Phase 3 studies that will support its U.S. application for chronic hepatitis delta. At 24 weeks, 59% of patients on 300 milligrams once weekly and 44% on 900 milligrams every four weeks reached the combined endpoint of virologic response and normalization of alanine aminotransferase, compared with 0% in the delayed-treatment arm. The company said it plans to submit a biologics license application in the first half of 2027 and launch in the fourth quarter of 2027, if approved.

The result is a second consecutive pivotal win for a drug Mirum acquired with Bluejay Therapeutics in January, and it gives the company a complete U.S. registration package for a disease that had no approved therapy until six months ago. It does not answer the question that will shape the drug's commercial fate: how a weekly or monthly antibody compares, in practice, with the daily injection that reached the market first.

A disease that borrows what it needs

Hepatitis delta is the most severe form of viral hepatitis. The virus cannot replicate on its own. It borrows the surface antigen of hepatitis B to assemble and infect new cells. People can acquire both viruses at once or acquire delta later as a superinfection, which accelerates cirrhosis and liver cancer faster than hepatitis B alone. Mirum estimates that about 230,000 people in the United States and Europe have the infection and that more than half will die of liver-related causes within 10 years of diagnosis. Testing is uneven, and many clinicians see few cases, so the pool of diagnosed patients is smaller than the estimate.

For decades, the only option was off-label pegylated interferon, which many patients could not tolerate. Gilead's Hepcludex changed that in May 2026, when the FDA granted accelerated approval to bulevirtide, an 8.5 milligram once-daily subcutaneous injection for adults without cirrhosis or with compensated cirrhosis. The approval rested on a Phase 3 trial in which 48% of treated patients reached a combined response at week 48, compared with 2% in a delayed-treatment arm. Continued approval depends on confirmatory evidence, and the label carries a boxed warning for severe hepatitis B and delta flares after treatment stops.

That approval changed the question for the field. The question is no longer whether hepatitis delta can be treated. It is which treatment, for how long, and whether the virus can be pushed into a remission that lasts after the drug stops.

A different way to attack the virus

Brelovitug is a fully human IgG1 monoclonal antibody that binds hepatitis B surface antigen on both the hepatitis B and hepatitis D virions. The design aims to neutralize and clear the virions and to deplete the subviral particles that carry HBsAg, which the delta virus needs to assemble. The antibody is pan-genotypic, and Mirum is developing it as a single agent. That distinguishes it from Vir Biotechnology's approach, which pairs an antibody with an siRNA in a monthly two-drug regimen. Vir reported 44% combined response at week 48 in its pivotal ECLIPSE-1 trial last week, with 79% of treated patients reaching undetectable HDV RNA. The earlier report described how the two programs are chasing the same virus from different angles.

Brelovitug carries FDA Breakthrough Therapy and Orphan Drug designations, plus PRIME and Orphan designations from the European Medicines Agency. Those designations do not guarantee approval, but they signal that regulators see the unmet need as serious.

The trial, in numbers

AZURE-4 enrolled 113 treatment-naive patients and randomized them 2:1:1 to 300 milligrams once weekly by self-administered subcutaneous injection, 900 milligrams every four weeks, or delayed treatment beginning at week 12. The population had substantial disease: 53% had cirrhosis, about half had baseline HDV RNA above 6 log10 IU/mL, and the mean baseline ALT was 122 U/L. That matters because earlier-stage trials often enroll healthier patients, and results are harder to generalize to the people most at risk of liver failure.

At week 24, the combined endpoint was 59% in the 300 milligram arm and 44% in the 900 milligram arm, both against 0% in the delayed arm, with p values of 0.0001 or lower. Virologic response was nearly universal: 95% and 96%. ALT normalization was 60% and 44%. The stricter virologic measures were lower: 31% and 22% of patients had HDV RNA below the limit of quantification, and 14% and 11% had target-not-detected results.

The pattern is consistent with AZURE-1, the first Phase 3 study, which reported a 56% combined response for the 300 milligram arm and 45% for the 900 milligram arm at week 24. The two studies used the same primary endpoint and similar designs, and neither delayed-treatment arm produced a single combined responder, Mirum said. Nancy Shulman, Mirum's executive vice president of clinical development, said the results showed "consistent efficacy in patients with substantial disease burden."

The combined endpoint is a surrogate, not a hard clinical outcome. It counts a meaningful drop in HDV RNA together with a return of ALT to the normal range, which reflects both viral suppression and reduced liver injury. Regulators accepted the same kind of endpoint for bulevirtide's accelerated approval. Mirum's application will test whether the approach holds for a second drug. The open question is whether the early responses translate into fewer cases of cirrhosis, liver cancer, and liver-related death, which is the benefit patients ultimately need. Answering that takes years and a confirmatory program.

The safety picture

Brelovitug was well tolerated in AZURE-4, according to the company. The most common adverse event was flu-like illness, reported in 17.2% of the 300 milligram arm and 18.5% of the 900 milligram arm, with most cases following the first dose. Injection-site reactions were reported in 1.7% and 7.4% of patients. Mirum said there were no new safety signals across the program.

Those numbers are reassuring for an antibody that has to be given repeatedly, but they are not the whole safety story. A 24-week trial with about 85 treated patients is not designed to detect rare events, and the follow-up is short for a chronic infection measured in years. The longer-term data Mirum presented at The Liver Meeting offer some comfort: in 15 patients with cirrhosis who received brelovitug for up to 108 weeks in a Phase 2 study, all 15 had a virologic response and all 15 had ALT normalization. Fifteen patients is a small group, and the extension was open-label, but the direction is consistent with the controlled results.

What the comparison can and cannot show

The obvious question is how brelovitug compares with bulevirtide. The two have not been tested head to head. Their trials differ in size, duration, endpoint timing, and patient mix. Bulevirtide's 48% combined response came at week 48 in the MYR301 trial. Mirum's 59% and 44% came at week 24 in AZURE-4. A higher response at an earlier time point is encouraging, but it is not proof that the drug is more effective, and the field has been burned by cross-trial comparisons before. The more meaningful comparison will come from long-term durability, real-world use, and the dosing burden.

Dosing is where brelovitug may have an edge. Bulevirtide is a daily injection. Brelovitug's 300 milligram dose is self-administered once a week, and the 900 milligram dose is given every four weeks. For patients who face years of treatment, fewer injections can matter for adherence, though the tradeoff is that the higher-dose, less-frequent arm produced a lower response rate than the weekly arm in AZURE-4.

The competitive field is also filling in. Vir's combination is in a pivotal trial and reported functional cure data last week. Other programs are in earlier stages. The first approved drug has a head start, but the pool of patients is large enough, and the treatment options thin enough, that a second or third option can find a market if the data hold up.

The regulatory clock

Mirum plans to submit its BLA in the first half of 2027, with a potential launch in the fourth quarter of 2027. The application will rest on AZURE-1 and AZURE-4, plus the long-term Phase 2 data. The FDA will review the manufacturing process, the dosing regimen, and the safety database. Mirum has not said publicly which dose it will propose for approval. The 300 milligram weekly arm had the stronger response, while the 900 milligram every-four-weeks arm may be easier to administer in some settings.

Analysts have been positive. H.C. Wainwright raised its price target to $195 from $140 after the data, according to TipRanks. The raise suggests that the second win reduced the perceived risk in the filing rather than surprising the market, which had already seen AZURE-1 succeed in September. The commercial question is now less about whether the drug works and more about how quickly it can reach patients and how payers will treat a new entrant in a small, expensive category.

What to watch

The full AZURE-4 results will be presented at an upcoming medical congress. The BLA submission will show which dose Mirum chooses and how it frames the risk-benefit case. The FDA's review will show whether the agency wants a confirmatory trial, as it did for bulevirtide, or a broader label. And the first year of sales will show whether once-weekly dosing and the antibody mechanism translate into uptake against an established daily injection.

For patients, the near-term change is not a new option yet. It is the prospect of one. The drug still has to clear the FDA, and the company still has to show that the responses deepen and hold. The 24-week data are a strong start, not a finish line.

Primary sources

  1. Mirum Pharmaceuticals' Nov. 10 press release for the AZURE-4 primary endpoint, the trial design, the efficacy and safety data, the long-term Phase 2 results, and the BLA timing.
  2. Mirum's Form 8-K filed Nov. 10 for the same disclosure.
  3. The FDA's May 22, 2026 announcement of the Hepcludex accelerated approval for the bulevirtide comparison.
  4. TipRanks for the H.C. Wainwright price target change.
  5. Mavengity's Nov. 9 report on Vir Biotechnology's ECLIPSE-1 data for the competitive context.