FDA experts weigh a Duchenne cell therapy as agency staff flag a post-trial statistical change

Capricor Therapeutics' cell therapy for Duchenne muscular dystrophy, a rare, progressive muscle-wasting disease that mostly affects boys and is often fatal by early adulthood, goes before an FDA advisory committee on Wednesday. The committee's job is narrower than a verdict on the drug. It is to weigh a single technical question: whether the company can be credited with a positive Phase 3 trial it edited the playbook for after seeing the results.

In its December 2025 topline release, Capricor said its HOPE-3 trial preserved upper-arm function in patients who had already lost the ability to walk and staved off the cardiac decline that usually becomes fatal in this group, with the result reaching statistical significance on both the upper-limb primary endpoint and a cardiac secondary endpoint. The FDA's Monday briefing, summarized in STAT's live coverage, concluded the opposite: that deramiocel did not separate from placebo on the upper-limb or cardiac measures.

Both cannot be true in the way each side is framing it. The disagreement is not really about the drug. It is about a document the trial leaves behind: the statistical analysis plan, the file that says in advance which patients count, which endpoints count, and how missing data is handled. The FDA flagged — in its briefing document for BLA 125842 — that Capricor made "numerous" changes to that plan after the trial was completed, calling the late edits a further muddying of the outcome. A post-hoc edit of a statistical analysis plan is, in effect, the sponsor choosing the analytical lens after seeing the picture, and the FDA's contrary read on the endpoints tracks that critique rather than the underlying measurements.

Capricor has argued, in its release and earlier STAT coverage of the topline, that the changes were minor and pre-specified before the database was unblinded. Both positions are on the public record; the committee's job on Wednesday is to weigh which version of the trial the agency should credit.

The patient population Capricor is targeting gives the stakes a concrete shape. The application is for teenagers and young men who have already lost the ability to walk, the subgroup in which heart failure has become the leading cause of death. That is why the cardiac endpoint is itself a survival measure, and why a one-point difference on a clinical score, on either arm, has become the load-bearing question of the program. The same population has very few approved treatment options, which is why the committee's read on the statistical analysis plan dispute will be the live signal on Wednesday, not a verdict on approval.

A favorable vote does not equal approval. The committee, the FDA's Cellular, Tissue, and Gene Therapies Advisory Committee, is a panel of outside experts whose vote is non-binding; the FDA's later action on Capricor's application is a separate decision, and even a positive or split vote would leave the agency free to require an additional trial. A negative vote, or a vote calling for more data, makes the path materially harder in a population with very limited options. The GlobeNewswire mirror of the company's release is also on the public record. What changes on Wednesday is whether outside experts ratify the agency's lens or push back on it. For teenagers and young men in Capricor's target population, the time cost of any additional trial would fall on a window where heart failure is the leading cause of death and approved options are scarce.