A first-in-human clinical trial of an experimental epigenetic therapy has produced an outcome long considered elusive in the progressive genetic muscle degeneration disease facioscapulohumeral muscular dystrophy (FSHD): measurable gains in lean muscle volume.
Epicrispr Biotechnologies announced updated interim results from its ongoing Phase I/II study of EPI-321, reporting that all three FSHD patients who reached the six-month evaluation point experienced increases in lean muscle mass following a single intravenous (IV) infusion of the therapy. The findings represent what the company says is the first clinical evidence that a treatment may be capable of increasing muscle volume in FSHD, which affects an estimated 870,000 people worldwide and is one of the most common forms of muscular dystrophy.
Although the study remains in its early stages with only nine total patients across two dosing cohorts, the promising findings were supported by imaging data, functional measurements, and biomarker changes that together suggest the therapy may be altering the biological drivers of the disease.
FSHD lacks disease-modifying treatments to intervene in the characteristic loss of skeletal muscle function, starting in the face, shoulders, and upper arms and spreading throughout the body. “We were pretty blown away,” Epicrispr CEO Amber Salzman told Inside Precision Medicine. “Every single patient gained lean muscle volume. No one has seen that.”
A silencing GEM
FSHD is caused by hypomethylation of the D4Z4 region, which is a polymorphic variable number tandem repeat (VNTR) array made up of 3.3 kilobase units, with each unit encoding the DUX4 gene. This hypomethylation leads to abnormal activation of the DUX4 gene, triggering muscle cell death and tissue degeneration, resulting in progressive muscle weakness, asymmetry, and fat infiltration.
Unlike conventional gene-editing therapies that permanently alter DNA sequences, Epicrispr’s therapy uses their Gene Expression Modulation System (GEMS). Epicrispr’s EPI-321 is an adeno-associated virus (AAV)-delivered epigenetic gene therapy that restores D4Z4 methylation and suppresses DUX4 expression, and preclinical studies show improved muscle function and reduced muscle cell death. “We don’t cut DNA,” Salzman said. “We’re not using Cas9 in the traditional editing sense. We use a dead Cas protein and an epigenetic mechanism to silence the gene.”
According to Epicrispr, preclinical studies demonstrated that the epigenetic modifications persist through many rounds of cell division, suggesting long-term suppression may be possible after a single treatment.
In addition to the therapeutic innovation, Epicrispr has also made headway into the long-standing challenge of identifying reliable molecular biomarkers for FSHD to evaluate activity downstream of DUX4. Traditional muscle biopsies provide information from only a small tissue sample and can be difficult to interpret when substantial fat replacement has already occurred.
Instead, Epicrispr collaborated with researchers at the University of Colorado investigating circulating cell-free DNA signatures associated with DUX4-driven disease activity. The biomarker, known as CBT17, emerged from studies comparing blood samples from approximately 50 healthy individuals and 50 FSHD patients. The biomarker may provide a whole-body readout of disease activity, offering evidence that DUX4 suppression is occurring beyond individual muscles examined through biopsy.
As of the May 12, 2026 data cutoff, EPI-321 has demonstrated a manageable safety profile, with nine patients treated across two dose cohorts. One IV infusion was administered to six patients in the first cohort at a target dose of 2×10¹³ vg/kg and to three patients in the second cohort at a target dose of 4×10¹³ vg/kg.
The company reported a “manageable” safety profile, with no severe dose-limiting toxicities disclosed to date. Salzman was careful not to overstate the findings. “I never want to be hubristic when it comes to AAV,” she said. “You’re giving people a lot of virus.”
She described the treatment experience as somewhat analogous to vaccine-related immune responses, noting that prophylactic immunosuppression is used as part of the protocol. “It’s manageable,” she said. “I’m not going to say it’s favorable. It’s manageable.” No serious grade 3 or grade 4 treatment-related safety concerns have been reported so far.
Baseline levels of CBT17 in treated patients fell substantially after therapy. “When we looked at baseline and three months, our patients’ levels came down so that it was just about in the healthy range,” she said.
Functional measures match MRI
The headline result comes from whole-body MRI analyses conducted six months after treatment. Among the first three evaluable patients, lean muscle volume increased in every individual compared with baseline measurements. On average, patients gained approximately 370 milliliters of lean muscle tissue, equivalent to roughly 0.8 pounds of muscle mass. Individual gains ranged from about 0.5 to 1.3 pounds. For a disease characterized by chronic muscle loss, even stabilizing muscle mass would be considered a meaningful achievement. Actual gains are potentially more significant.
The MRI analyses were performed in collaboration with Springbok Analytics, a company specializing in AI-powered muscle imaging. In addition to being widely used in professional sports to monitor injury recovery and rehabilitation, as well as a growing number of clinical settings, FSHD is one of Springbok’s primary clinical focus areas, along with disorders such as Charcot-Marie-Tooth disease and Duchenne muscular dystrophy. Using a specialized MRI protocol that runs on standard scanners, the company can quantify changes across as many as 140 individual muscles throughout the body, measuring muscle volume, muscle composition, and fat infiltration, producing objective data on muscle health.
According to Salzman, Springbok compared patients’ baseline scans against machine-learning models trained on longitudinal imaging datasets from more than 100 FSHD patients. “They can predict what it’s going to look like in six months—how much muscle patients will lose and where,” she said. “Then we sent them the six-month scans, and every patient gained lean muscle volume.”
The gains were not evenly distributed throughout the body. Patients appeared to benefit most in muscles that still retained substantial healthy tissue before treatment. “If you have too much fat in that muscle, there’s not that much we can do,” Salzman explained. “But if you have some residual muscle, it can not only stop getting worse because we’re cutting off the poison, but you can also regenerate muscle.”
One participant, a post-menopausal woman, gained approximately 1.3 pounds of lean muscle mass according to the MRI analysis. The increase was particularly notable because age-related muscle loss would normally be expected in that population. Earlier data released by Epicrispr showed favorable trends across several functional and strength assessments at the three-month mark. These measures included tests commonly used in neuromuscular disease studies, such as the Timed Up and Go test, 10-meter walk/run assessments, and quantitative muscle testing.
Because many functional endpoints depend on patient effort, interpreting changes in small open-label studies can be challenging. Investigators and investors alike often question whether participants simply perform better because they know they received treatment. Salzman acknowledged those concerns. “You could say maybe they had a good day, and that’s why the functional measures were better,” she said.
MRI measurements, however, are not subject to effort-dependent variability. “You can’t fake MRI,” she added. Importantly, the regions showing muscle growth on imaging appeared to correspond with areas where patients demonstrated improved performance.
One participant who gained substantial upper-body muscle volume showed corresponding gains on upper-body strength assessments, while lower-body improvements were more limited in areas where muscle loss had already become severe. “The whole story is holding together,” Salzman said. “Their functional measures are getting better; their lean muscle is increasing.”
Taken together, the MRI, biomarker, and functional data provide multiple independent signals supporting biological activity. “You can’t fake blood; you can’t fake MRIs,” Salzman said. “Those are totally supportive.”
Epitome of epigenetic editing?
Beyond FSHD, researchers are closely watching EPI-321 because it represents one of the first clinical tests of an epigenetic CRISPR-based therapeutic strategy. “This is an important milestone not just for FSHD but for epigenetic medicine,” Salzman said. The approach could potentially be adapted to other diseases caused by harmful gene activation or insufficient gene silencing.
Still, significant hurdles remain before EPI-321 can be considered a proven therapy. The current dataset includes only three patients with six months of follow-up. Epicrispr plans to present additional data from six patients at the World Muscle Society Annual Congress in September 2026.
The company expects the initial Phase I/II study to complete enrollment and generate a full readout in mid-2027. Yet, larger studies will be needed to determine whether muscle gains persist, whether they translate into meaningful long-term functional benefits, and whether safety remains acceptable over time.
Regulators will also require evidence linking increases in lean muscle volume to tangible improvements in patient function and quality of life. To that end, Epicrispr is preparing discussions with the U.S. Food and Drug Administration regarding potential future endpoints and whether MRI-based muscle measurements could eventually serve as surrogate markers of clinical benefit.
Whether those early signals hold up in larger populations remains to be seen. But for a disease in which muscle loss has long been viewed as inevitable, the possibility of reversing that trend, even modestly, marks a notable moment for the FSHD field.
The post Pump You Up: Epigenetic Editor Drives Muscle Growth in FSHD Patients appeared first on Inside Precision Medicine.
