FDA Rejects Duchenne Cell Therapy Deramiocel Again, Says Data Integrity Violations Cannot Be Fixed

November 22, 2026:

FDA Rejects Duchenne Cell Therapy Deramiocel Again, Says Data Integrity Violations Cannot Be Fixed
WHITE OAK MD - JULY 20 sign
WHITE OAK, MD – JULY 20: A sign for the Food And Drug Administration is seen outside of the headquarters on July 20, 2020 in White Oak, Maryland.
Sarah Silbiger/Getty Images

The FDA declined to approve Capricor Therapeutics’ deramiocel for Duchenne muscular dystrophy (DMD) a second time on Friday, Nov. 20, two days before the scheduled Nov. 22 action date, saying the Phase 3 HOPE-3 trial behind the application contains “serious data integrity violations” that cannot be remediated after the fact. The decision came from the Office of Therapeutic Products at the agency’s biologics center (CBER). Deramiocel is an investigational cell therapy that is not approved for commercial use in any indication, according to Capricor. The company disputes the FDA’s framing and says it stands behind the integrity of its data.

The FDA also said the two things Capricor added after a July advisory vote, 24 months of open-label follow-up and a narrower upper-limb indication, did not change its conclusion. The reason is how trials produce evidence: a p-value only means something when the analysis was fixed before data were seen and when participants could not identify their arm, and open-label follow-up from the same trial does not restore either safeguard.

Why Did FDA Reject Deramiocel? Unauthorized Analysis Changes and Blinding Violations

The FDA says its Bioresearch Monitoring program inspected Capricor and found serious deficiencies in how HOPE-3 was conducted. The statement names three problems:

  • Unauthorized post hoc changes to the statistical analysis plan.
  • Violations of study blinding procedures.
  • Additional protocol violations.

The agency says these issues, individually and together, make the data unreliable and uninterpretable as substantial evidence of effectiveness. It also says HOPE-3 did not meet prespecified endpoints, which covered upper-limb function and cardiac imaging, and that the integrity problems compounded that concern. A related Regulatory Responsibility Letter dated Nov. 20 addresses the inspection observations, and Capricor can respond to it.

Capricor disputes the framing. Capricor’s release says the letter finds the data insufficient to establish substantial evidence of effectiveness and cites unspecified “other clinical deficiencies,” and CEO Linda Marbán said the company stands behind the integrity of its data. The company says both HOPE trials were randomized, double-blind and placebo-controlled and met their primary endpoints. The release includes no trial data, and Capricor said it is reviewing the letter and will share next steps.

Does One-Observation Inspection Form Match FDA’s Language on Serious Violations?

The inspection finding is the new element in this decision. The FDA’s July clinical-overview slides for the advisory committee raised concerns about analysis changes and possible unblinding but did not mention an inspection. NeurologyLive reports that a July bioresearch inspection produced a Form 483 with a single observation, which Capricor has answered. The contents of that observation are not public in the sources reviewed.

A Form 483 is not the whole record. The FDA says the form is not a final agency determination, does not list every problem, and is weighed alongside the inspector’s report and the company’s written response. So the one-observation form does not by itself rule out a broader basis for the agency’s “serious” language, but the sources reviewed do not include the letter’s text, and the statement does not say whether it will be published. The FDA released Capricor’s first rejection letter in a September 2025 batch of 89 letters.

When that first letter arrived in July 2025, Capricor said it had passed its pre-licensure inspection, a manufacturing-facility review separate from the clinical inspection described above. Capricor’s June 2025 release said that inspection ended with a Form 483 containing several observations, none of which the company said required material changes to its manufacturing process or facility.

Hypersensitivity Reactions in 41.5% of Treated Patients Raised FDA Blinding Questions

HOPE-3 randomized 106 patients with genetically confirmed DMD, 54 to deramiocel and 52 to placebo, with an infusion every three months for 12 months; 84.9% were non-ambulatory. Hypersensitivity reactions occurred in 41.5% of deramiocel recipients (22 of 53) and 15.4% of placebo recipients (8 of 52), and headache in 60.4% versus 3.8%. The FDA called that a distinct side-effect profile that may have partly revealed who received the therapy.

Side effects are a common cause of unblinding, and poor blinding tends to favor the experimental arm and inflate effect sizes. The FDA’s slides add that open-label treatment data could offer further opportunities to infer assignment, and that changes made after data collection add analytic flexibility that can raise false-positive risk.

Primary Endpoint p-Value Ranged From 0.029 to 0.66 Across Five Analysis-Plan Versions

The primary endpoint was the Performance of Upper Limb 2.0 (PUL 2.0) total score. The FDA’s slides show how the result moves with the plan version and with how missing data are handled:

Analysis-plan version Without imputation With imputation

1.1 (prespecified; FDA analysis)

0.66 points, p=0.24

0.92 points, p=0.11

1.7

0.61 points, p=0.64

0.59 points, p=0.66

2.0 (percent change)

2.38%, p=0.26

3.74%, p=0.08

3.0 (percent change)

2.53%, p=0.21

4.16%, p=0.045

3.0 modified in final report

not reported

4.55%, p=0.029

The FDA’s timeline lists version 2.0 on Sept. 26, 2025, version 3.0 on Nov. 24, 2025, unblinding the next day, and further model changes in the clinical study report dated Feb. 3, 2026. The agency called the primary result unstable and dependent on how two placebo patients were handled. Capricor’s national principal investigator, Craig McDonald, told the July advisory committee the trial met its primary endpoint, citing the 4.55% result; the FDA says the study missed it.

On heart function, the prespecified analysis showed a Month 12 ejection-fraction difference of -0.04% (p=0.97), while Capricor’s ranked-change analysis gave p=0.041. The FDA said ranked change lacks clinical interpretation. In FDA slide figures, raw mean ejection fraction at Month 12 was 54.2% on deramiocel and 58.3% on placebo, but baseline values were 55.3% and 59.3%, so the gap was nearly the same before treatment. The raw gap alone does not show that heart function worsened on deramiocel.

Open-Label Extension and Upper-Limb Indication Failed to Change FDA’s Conclusion

Capricor’s first application, a rolling submission announced complete on Jan. 2, 2025 and based on the earlier HOPE-2 study, drew a complete response letter in July 2025 for lack of substantial evidence of effectiveness and outstanding manufacturing items. After Capricor submitted HOPE-3 data, the company announced on March 10, 2026 that the agency had lifted that letter and resumed review for DMD cardiomyopathy. On July 29, the advisory committee then voted 9 to 3 that the evidence did not show effectiveness in cardiomyopathy. The vote was non-binding and covered only cardiomyopathy; Capricor said it was encouraged by the committee’s discussion of the skeletal muscle effect and noted there was no vote on benefit-risk.

Capricor then filed a major amendment with 24-month open-label extension findings from HOPE-3-OLE and switched its requested indication from cardiomyopathy to upper-limb impairment in DMD. NeurologyLive says the proposal covered patients older than 9 and that the action date moved from Aug. 22 to Nov. 22. The FDA says it reviewed the extension data and the new indication and found the standard still unmet, and its statement gives no extension results.

Open-label follow-up has no blinded control, which limits its weight. That applies to a 76% slowing figure Capricor showed at the 2026 World Muscle Society Congress, which NeurologyLive describes as an open-label comparison of patients who switched from placebo to deramiocel.

Capricor Offers No Resubmission Timeline; Call Is Scheduled for Monday at 8:30 a.m. ET

The release gives no timeline for a resubmission or appeal, though its forward-looking statements refer to the timing of any resubmission. Capricor will hold a call and webcast on Monday, Nov. 23, at 8:30 a.m. ET.

The company reported about $198.5 million in cash and securities at Sept. 30 and says that funds planned operations for at least 12 months while it reviews costs to extend its runway. RTTNews reported that Capricor shares fell 78.52% to $1.63 in after-hours trading from a $7.59 close. NeurologyLive reports Capricor has begun regulatory engagement in Europe and Japan. Capricor is also in litigation with distributor Nippon Shinyaku and its U.S. unit NS Pharma, having sued in May to rescind the U.S. distribution agreement; its allegations in that case have not been adjudicated.

Several things remain unknown: the full text of the letter, what the other “clinical deficiencies” are, what the inspection form said, and whether the FDA sees any route to approval short of a new trial. The sources reviewed also do not say what happens to patients now in open-label extension studies.

Deramiocel Remains Investigational: What Duchenne Families Can Do Now

For families, the practical point is that no approved deramiocel exists and the sources reviewed describe no way to obtain it. Approved Duchenne options include corticosteroids, exon-skipping drugs for specific mutations, the gene therapy Elevidys, vamorolone and givinostat, each with its own eligibility limits. A neuromuscular care team can say which options and clinical trials fit a particular patient, and Monday’s call may clarify what Capricor plans next.

This article is for information only and is not medical advice.


Frequently Asked Questions

Is deramiocel approved by the FDA?

No. Deramiocel is an investigational cell therapy, and the FDA’s second rejection means it remains unapproved in the United States. Capricor says it is not approved for commercial use in any indication.

What does it mean that the violations “cannot be remediated post-hoc”?

The FDA’s statement says the data cannot be remediated post-hoc. In plain terms, the agency says the problems lie in how HOPE-3 was run, so new analysis or added data from the same trial cannot repair them. Analysis choices made after seeing results raise false-positive risk, and visible side-effect differences can reveal treatment assignment, which is why both are treated as threats to a result’s validity.

Can Capricor try again?

Capricor has announced no resubmission plan or appeal, and the FDA says the company can respond to the related letter. After the first rejection in July 2025, Capricor said in March 2026 that the FDA had lifted the first rejection once HOPE-3 data arrived, but this letter says the HOPE-3 data problems cannot be fixed after the fact.

What treatments are approved for Duchenne muscular dystrophy right now?

Approved options include corticosteroids, exon-skipping drugs for subsets of patients, the gene therapy Elevidys, vamorolone and givinostat, each with its own eligibility limits. Families should confirm current labels with their care team.

Source link