Investigational FSHD Drug Did Little to Change DUX4 Expression

The investigational drug losmapimod did not significantly change expression of the gene behind facioscapulohumeral muscular dystrophy (FSHD), but it was associated with potential improvements in select structural and patient-reported outcomes, the randomized phase IIb ReDUX4 trial showed.

The least squares mean change from baseline in DUX4-driven gene expression from muscle biopsy specimens did not differ significantly between the losmapimod and placebo groups (0.83 vs 0.40; P=0.56), reported John Jiang, PhD, of Fulcrum Therapeutics in Cambridge, Massachusetts, and colleagues in Lancet Neurology.

Though changes in DUX4-driven gene expression were not significant, the researchers noted a smaller increase from baseline in muscle fat infiltration, a secondary outcome, as measured by whole-body musculoskeletal MRI, with a between-group difference of -0.49 percentage points (95% CI -0.86 to -0.12).

The exploratory outcomes of reachable workspace (a motion sensor-based assessment of upper arm function, including the shoulder and proximal arm) and Patient Global Impression of Change scores (a self-assessed rating of change in overall status, on a scale from 1-7, with lower scores indicating more improvement) also improved compared with placebo.

"Although these findings need to be interpreted with caution ... for the first time, an interventional drug has been associated with potential improvements in structural, functional, and participant-reported disease modification outcomes in facioscapulohumeral muscular dystrophy," Jiang and colleagues wrote.

Facioscapulohumeral muscular dystrophy is characterized by weakness and muscle atrophy in the eyes, mouth, shoulders, abdominal muscles, upper arms, and lower legs, with onset early in life, usually before age 20.

The disease is driven by aberrant expression of transcription factor DUX4 in skeletal muscle. Losmapimod was observed to reduce DUX4 in facioscapulohumeral muscular dystrophy myotubes in preclinical studies, Jiang's team said, and a decrease of greater than 50% in DUX4-driven gene expression was hypothesized based on these studies.

However, patients in this study had a wide variation of this gene expression at baseline, potentially contributing to the lack of effect found, the authors noted.

This trial "has challenged the use of changes in DUX4 expression as an endpoint," and this endpoint will not be assessed at all in the authors' follow-up phase III study on losmapimod, wrote Nicol Voermans, MD, PhD, of the Radboud University Medical Center in Nijmegen, Netherlands, and John Vissing, MD, PhD, of the University of Copenhagen in Denmark, in an invited commentary. "Furthermore, the use of MR-assessed change in fat fraction in selected muscles with intermediary disease severity seems useful as an endpoint, and reachable workspace looks to be a sensitive outcome measure."

For this double-blind, placebo-controlled trial, Jiang and team enrolled 80 patients with a confirmed diagnosis of type 1 facioscapulohumeral muscular dystrophy at 17 neurology centers in Canada, France, Spain, and the U.S. from August 2019 through February 2020. Mean age was 45.7, 68% were men, and 88% were white.

All patients had scores of 2-4 on the Ricci clinical severity scale (an assessment of extent and severity of muscle weakness, with higher scores indicating more severe weakness) and at least one skeletal muscle positive for short tau inversion recovery (STIR) on a whole-body muscle MRI, judged to be accessible for needle biopsy. DUX4 activity is more likely to be detected in these STIR-positive muscles. Patients with additional medical conditions that could confound or pose extra risks were excluded.

Participants took either 15-mg losmapimod or placebo orally twice a day for 48 weeks. The study was extended from 24 weeks to 48 weeks because of the COVID-19 pandemic. The primary outcome was change from baseline in DUX4-driven gene expression at week 16 or 36, assessed by quantitative reverse transcription polymerase chain reaction, using a composite measure of selected DUX4-regulated gene transcripts in skeletal muscle.

Losmapimod was well-tolerated, with 29 treatment-emergent adverse events in the losmapimod group (nine drug-related) and 23 in the placebo group (two drug-related). The most common adverse events were injury, poisoning, and procedural complications, followed by gastrointestinal disorders. No treatment discontinuations due to adverse events occurred, and no participants died during the study.

Jiang and colleagues said they were limited by their small sample size and enrollment of patients with type 1 facioscapulohumeral muscular dystrophy only; they are enrolling patients with type 2 disease in an ongoing phase III study. Sensitivity to changes for certain endpoints needs to be further established in the clinic, and missing data from week 48 created a potential bias in the efficacy analysis. Some of the exploratory outcomes did not capture differences between losmapimod and placebo, the authors wrote, so were not sensitive enough to capture differences in disease progression.

Comment