A Gene Transfer Therapy Study to Evaluate the Safety and Efficacy of Delandistrogene Moxeparvovec (SRP-9001) in Participants With Duchenne Muscular Dystrophy (DMD)

Rationale

Duchenne muscular dystrophy is caused by loss-of-function variants in the DMD gene, so muscle fibers lack dystrophin and degenerate. The DMD gene is too large for an AAV vector, so delandistrogene moxeparvovec (SRP-9001) delivers a shortened micro-dystrophin transgene that keeps the key functional domains of dystrophin. The transgene is packaged in an AAVrh74 capsid, chosen for efficient muscle transduction and low pre-existing immunity, and is driven by the muscle-specific MHCK7 promoter with a cardiac enhancer. One intravenous dose was expected to express micro-dystrophin in skeletal and heart muscle and slow the loss of motor function.

Results

The dose was 1.33 x 10^14 vector genomes per kg. At week 52 (Part 1, data cut 13 September 2023), the NSAA total score rose by a least squares mean of 2.57 points with delandistrogene moxeparvovec (n=63) and 1.92 with placebo (n=61 analysed); the difference was 0.65 (95% CI -0.45 to 1.74; P=0.2441), so the primary endpoint was not met. Time to rise from the floor (-0.64 s) and 10-meter walk/run (-0.42 s) favored treatment, but the authors state no statistical significance can be claimed. Week-12 micro-dystrophin by western blot was 34.29% of normal in 17 treated patients versus 0.00% in 14 on placebo. Serious adverse events occurred in 14 of 63 treated and 5 of 62 on placebo; 7 treated patients (11.1%) had 10 treatment-related serious events, mainly acute liver injury, and all resolved. There were no deaths. At 2 years, treated patients (n=63) differed from a propensity-weighted external control (n=143) on NSAA change (+2.63 vs -0.25 points, nominal P=0.0001).

• In 125 ambulatory boys aged 4 to under 8 years with Duchenne muscular dystrophy on stable corticosteroids, one intravenous dose of delandistrogene moxeparvovec (1.33 x 10^14 vector genomes per kg) did not significantly improve the North Star Ambulatory Assessment (NSAA) total score at week 52 versus placebo in the phase 3 EMBARK trial: least squares mean change 2.57 points (n=63) versus 1.92 points (n=61 analysed), difference 0.65 (95% CI -0.45 to 1.74; P=0.2441). (Mendell 2024)
• In the phase 3 EMBARK trial in boys aged 4 to under 8 years with Duchenne muscular dystrophy, the two key secondary timed function tests favored delandistrogene moxeparvovec over placebo at week 52, although the authors state no statistical significance can be claimed after the primary endpoint was not met: time to rise from the floor changed by -0.27 s versus +0.37 s (difference -0.64 s; 95% CI -1.06 to -0.23) and the 10-meter walk/run by -0.34 s versus +0.08 s (difference -0.42 s; 95% CI -0.71 to -0.13), n=63 versus 61. (Mendell 2024)
• At week 52 of the phase 3 EMBARK trial in boys aged 4 to under 8 years with Duchenne muscular dystrophy, delandistrogene moxeparvovec versus placebo gave these other secondary differences (95% CI): stride velocity 95th centile +0.10 m/s (0.00 to 0.19; n=57 vs 61), 100-meter walk/run -3.29 s (-8.28 to 1.70; n=59 vs 57), time to ascend 4 steps -0.36 s (-0.71 to -0.01; n=62 vs 60), PROMIS Mobility 0.05 (-0.08 to 0.19), PROMIS Upper Extremity -0.04 (-0.24 to 0.17) and NSAA skills gained or improved 0.19 (-0.67 to 1.06); none is claimed as statistically significant. (Mendell 2024)
• In a prespecified sensitivity analysis of the phase 3 EMBARK trial (delandistrogene moxeparvovec versus placebo, boys aged 4 to under 8 years with Duchenne muscular dystrophy, week 52), a global statistical test across six functional endpoints (NSAA, time to rise, 10-meter walk/run, stride velocity 95th centile, 100-meter walk/run, time to ascend 4 steps) gave P=0.0044, not controlled for multiplicity; in a post hoc analysis 3.2% of treated patients versus 16.4% of placebo patients progressed to a time to rise above 5 s (odds ratio 0.091; 95% CI 0.01 to 0.61). (Mendell 2024)
• In the phase 3 EMBARK trial, muscle biopsies in boys aged 4 to under 8 years with Duchenne muscular dystrophy showed micro-dystrophin expression after one intravenous dose of delandistrogene moxeparvovec (1.33 x 10^14 vector genomes per kg): western blot adjusted for muscle content was 34.29% of normal (SD 41.04; n=17) at week 12 versus 0.00% in placebo (n=14), and 45.68% (SD 39.75; n=16) at week 64, with 38.60% dystrophin-positive fibers at week 64. (Mendell 2024; Mendell 2026)
• At 2 years, 63 boys aged 4 to under 8 years with Duchenne muscular dystrophy treated with delandistrogene moxeparvovec in Part 1 of the phase 3 EMBARK trial changed from baseline by +2.63 NSAA points versus -0.25 in a propensity-weighted external control (n=143; difference 2.88; 95% CI 1.43 to 4.33), by +0.65 s versus +2.71 s on time to rise (difference -2.06 s; 95% CI -3.43 to -0.70) and by -0.04 s versus +1.32 s on the 10-meter walk/run (difference -1.36 s; 95% CI -2.24 to -0.47); P values are nominal, and the external control came from the FOR-DMD, PRO-DMD-01 and CINRG DNHS studies. (Mendell 2026)
• Through week 52 of the phase 3 EMBARK trial, serious adverse events occurred in 14 of 63 boys aged 4 to under 8 years with Duchenne muscular dystrophy given delandistrogene moxeparvovec (21 events) and in 5 of 62 given placebo; 7 treated patients (11.1%) had 10 treatment-related serious events: acute liver injury (including three transient liver enzyme elevations, one hepatotoxicity and one liver injury), myocarditis, rhabdomyolysis, nausea, vomiting and pyrexia. All resolved. No serious event was treatment-related in the placebo group. (Mendell 2024)
• In the phase 3 EMBARK trial, transient liver enzyme elevations (glutamate dehydrogenase, gamma-glutamyl transferase, alanine or aspartate transaminase) were treatment-related adverse events in 41.3% of boys aged 4 to under 8 years with Duchenne muscular dystrophy given delandistrogene moxeparvovec (1.33 x 10^14 vector genomes per kg), starting within 90 days of infusion (median 42 days); they resolved spontaneously or after increased corticosteroids, and none progressed to liver failure. (Mendell 2024)
• Through about 104 weeks of the phase 3 EMBARK trial, no deaths, no study discontinuations due to adverse events, no thrombotic microangiopathy and no clinically significant complement-mediated adverse events occurred in boys aged 4 to under 8 years with Duchenne muscular dystrophy given delandistrogene moxeparvovec. Between weeks 52 and 104, 352 new adverse events were recorded in 63 treated patients, 34 treatment-related in 15 patients (23.81%), most often troponin-I increase (4 patients) and proteinuria and headache (2 each); one patient had two serious treatment-related rhabdomyolysis events that resolved. (Mendell 2024; Mendell 2026)
• In a post hoc analysis of the phase 3 EMBARK trial in boys aged 4 to under 8 years with Duchenne muscular dystrophy, caregiver-reported global impression at week 52 favored delandistrogene moxeparvovec over placebo: the multi-domain responder index treatment difference was 1.7 (95% CI 0.90 to 2.5) for caregiver-rated change (CaGI-C) and 1.1 (95% CI 0.30 to 1.9) for severity (CaGI-S); results are exploratory and nominal. (McDonald 2025)

Design implications

Expression of micro-dystrophin at about one third of normal did not produce a significant NSAA gain at one year in boys aged 4 to 7 who were still gaining skills on corticosteroids. Trials of dystrophin-restoring therapy at this age need endpoints and follow-up that detect early divergence: timed function tests and wearable stride velocity moved in favor of treatment where NSAA did not. A 1-year placebo arm with crossover leaves long-term claims resting on an external control, so a controlled durability readout needs a longer parallel control or prospectively matched natural history. Systemic AAVrh74 at 1.33 x 10^14 vg/kg caused transient liver enzyme elevations in 41% of treated boys within 90 days despite added corticosteroids, so liver monitoring is required. No complement-mediated events or thrombotic microangiopathy occurred, consistent with a non-human-primate capsid and screening out patients with elevated anti-AAVrh74 antibodies.

Limitations

One randomized comparison in 125 treated boys on stable corticosteroids, placebo-controlled for only 52 weeks. The primary endpoint was not met, so secondary results are descriptive and not adjusted for multiplicity. Week-104 efficacy compares 63 treated patients with a propensity-weighted external control whose corticosteroid dosing is unknown. Micro-dystrophin was measured in a biopsy subset (n=31 at week 12). Sponsor-run trial; authors include sponsor employees.

References

1. Truong et al. (2015). Development of an intein-mediated split–Cas9 system for gene therapy. Nucleic Acids Res - Truong 2015 Split-Cas9 Intein
2. Lagos-Quintana et al. (2002). Identification of tissue-specific microRNAs from mouse. Curr Biol - Lagos-Quintana 2002 Tissue miRNAs