Capricor's Duchenne cell therapy nears an FDA decision on a narrower claim

The FDA is due to act by 22 November 2026 on deramiocel, Capricor Therapeutics' cell therapy for Duchenne muscular dystrophy. Capricor's most recent statements give that date as still in force, and the company reports third-quarter results on 10 November.

The application has had a rough passage. The FDA issued a complete response letter in July 2025. An advisory committee then voted 9 to 3 on 29 July 2026 that the evidence did not establish effectiveness for Duchenne cardiomyopathy. On 24 August Capricor said the agency had accepted a major amendment, adding 24-month open-label data and robustness analyses, and had extended the target date by three months. The company also asked the FDA to consider a refined indication centred on preserving upper limb function.

The trial

HOPE-3 randomised 106 boys and young men aged 10 and over at 20 US sites, 54 to deramiocel and 52 to placebo, with a mean age of about 15. Capricor reports that the therapy slowed decline on the Performance of Upper Limb 2.0 scale by 54% (P = .029).

FDA reviewers saw it differently. According to NeurologyLive, they concluded that neither HOPE-2 nor HOPE-3 met prespecified primary or key secondary endpoints, and judged benefit-risk unfavourable without evidence of clinical benefit. The dispute turns on successive revisions to the statistical analysis plan. The FDA says the last version was finalised the same day the database was locked and unblinded, while Capricor says it was finalised before unblinding. The agency also noted that a revised analysis excluded 23 of the 106 randomised patients.

At the meeting John Teerlink, a professor and cardiologist, spoke of "markedly fragile results that can be changed with different imputation strategies". Linda Marbán, Capricor's chief executive, maintains that HOPE-3 showed a statistically significant benefit on its primary endpoint.

The committee was asked about the heart. The application now leans on the arms.

A weakened cardiac case

Capricor said the committee was not asked to vote on the HOPE-3 primary endpoint or on overall benefit-risk, so the panel never formally ruled on upper limb data. The cardiac case, meanwhile, has softened. In August the company disclosed an issue with the statistical model in its clinical study report, which moved the left ventricular ejection fraction result from 2.4 points (P = .04) to 1.8 points (P = .09). It said the primary endpoint was unaffected.

The company also reported that a July FDA inspection produced a Form 483 with one observation. Hypersensitivity reactions were more frequent on deramiocel, at about 42% against 15% on placebo.

What the extension adds

At the World Muscle Society congress in Hiroshima, which ran from 29 September to 3 October, Capricor presented open-label extension data, including a delayed-start analysis and a comparison with natural history. Of the 106 patients, 82 reached 24 months. Such comparisons can support a randomised result, but they cannot repair doubts about how that result was derived.

Why the decision matters

The FDA's biologics centre accepted the amendment citing significant unmet need in Duchenne, a condition in which regulators have repeatedly had to balance patient pressure against contested data. A decision to approve on the refined claim, against the reasoning of the agency's own reviewers, would be read closely by other rare disease sponsors. So would a second rejection. Either way, the handling of late changes to analysis plans in small paediatric-onset trials looks set to be the lesson that lasts.