Larsucosterol flops on clinical benefit in phase 2b alcohol-associated hepatitis trial
Key takeaways:
- The difference in 90-day mortality or LT rate between the larsucosterol groups and placebo group was not statistically significant.
- However, the results provide clinical equipoise for a phase 3 trial.
The AHFIRM trial did not demonstrate a beneficial effect of larsucosterol on 90-day mortality or the need for liver transplant among patients with severe alcohol-associated hepatitis, or AH, according to a study published in NEJM Evidence.
“Many marketed drugs and investigational agents have been evaluated for AH over decades, but none have proven consistently effective or received regulatory approval,” Mitchell Shiffman, MD, lead study author and director of the Liver Institute of Virginia, said in a related Durect Corp. press release. “Currently, a third of severe AH patients won’t survive beyond 3 months. This patient community urgently needs a therapy that can improve standard of care and save lives.”
As Healio previously reported, a phase 2a, multicenter, open-label, dose-escalation study demonstrated that larsucosterol (Durect) was safe and well-tolerated at 30 mg, 90 mg and 150 mg among patients with AH.
To further evaluate its safety and efficacy, Shiffman and colleagues conducted the randomized, double-blind, placebo-controlled phase 2b AHFIRM trial that enrolled 307 patients with severe AH from 46 centers in the U.S. and 16 centers in Europe, the U.K. and Australia from January 2021 to September 2023.
The researchers randomly assigned patients 1:1:1 to receive an IV infusion of 30 mg of larsucosterol (n = 102; 68% men; 81.4% white; median age, 44 years), 90 mg of larsucosterol (n = 102; 47% men; 81.4% white; median age, 43 years) or placebo (n = 103; 50% men; 83.5% white; median age, 47 years) and followed them for 90 days.
The researchers administered a second dose to patients who remained hospitalized after 72 hours.
The primary endpoint of the study was 90-day mortality or LT rate; the key secondary endpoint was 90-day mortality.
Overall, 301 patients received at least one dose of treatment.
The researchers reported that there were numerically more deaths at 90 days but fewer LTs in the placebo group (n = 25; n = 5) than in the 30 mg (n = 15; n = 6) and 90 mg (n = 17; n = 9) larsucosterol groups. However, the researchers noted these differences were not statistically significant, with win probability differences of 0.078 ± 0.061 for 30 mg against placebo and 0.039 ± 0.062 for 90 mg against placebo.
Among patients enrolled from centers in the U.S. (n = 232), there were eight deaths and five LTs among those who received 30 mg larsucosterol (n = 73), 10 deaths and eight LTs among those who received 90 mg larsucosterol (n = 77) and 21 deaths and four LTs among those who received placebo (n = 77), according to the researchers.
To account for regional differences in hospitalization-to-treatment time, the researchers conducted a post-hoc analysis of the patients treated within less than 10 days of hospitalization, observing 90-day mortality rates of seven out of 74 among those treated with 30 mg larsucosterol, 13 out of 77 among those treated with 90 mg larsucosterol and 20 out of 79 among those treated with placebo.
The researchers noted that most treatment-emergent adverse events could be attributed to complications of alcohol-associated hepatitis or underlying alcohol-associated liver disease.
“We look forward to further demonstrating larsucosterol’s potential in our upcoming phase 3 trial,” James E. Brown, DVM, study author and Durect’s president and CEO, said in the release. “If successful, the FDA has agreed that a single phase 3 trial could be sufficient to support a new drug application. We plan to initiate the trial in 2025, pending funding.”
Reference:
- Durect Corporation announces publication of larsucosterol phase 2b results in NEJM Evidence. https://www.durect.com/2025/01/durect-corporation-announces-publication-of-larsucosterol-phase-2b-results-in-nejm-evidence/. Published Jan. 28, 2025. Accessed Feb. 10, 2025.
Collapse
Read more about