Isembyld Is the First Myostatin Drug Approved. Read the Label

On September 11, the FDA approved Isembyld (apitegromab-mstn), a monoclonal antibody from Scholar Rock that blocks the activation of myostatin, the protein that limits muscle growth. It is the first myostatin targeted antibody to win approval. The indication is spinal muscular atrophy (SMA) in adults and in children 2 and older who are already taking a treatment aimed at the SMN2 gene. The dose is 10 mg per kilogram as an intravenous infusion every four weeks, each lasting 60 to 120 minutes. Anyone watching the muscle preservation race in obesity, which we covered in our piece on myostatin blockers and GLP-1 drugs, will want the actual label numbers.

What the pivotal trial found

The label describes Study 1, known as SAPPHIRE (NCT05156320). It enrolled 188 patients aged 2 to 21 with 5q SMA, all non ambulatory and all already on nusinersen or risdiplam. They were randomized one to one to one to apitegromab 20 mg per kilogram, 10 mg per kilogram or placebo, every four weeks for about a year. The main efficacy population was the 156 patients aged 2 to 12: 53 on the recommended dose and 50 on placebo.

The primary endpoint was the change in the Hammersmith Functional Motor Scale Expanded (HFMSE), a 33 item test of activities such as sitting unassisted, with a maximum of 66. At one year, the apitegromab group improved by 1.0 point and the placebo group declined by 1.2 points, a difference of 2.2 points (95% CI 0.49 to 3.95). The FDA's table labels the p value of 0.0121 as nominal. A larger share of treated children gained 3 points or more: 34.2% against 13.5% (odds ratio 3.8, 95% CI 1.33 to 10.90). The 20 mg per kilogram dose did not add efficacy, which is why the label recommends 10. Results in older patients, who received the higher dose, showed a consistent trend, according to the label.

Two points on a 66 point scale is small in absolute terms, and the confidence interval starts just under half a point. In SMA, though, the comparison is stark: children on background therapy without the antibody drifted downward. Stopping a decline, or nudging it up, is how the field defines progress after the first generation of drugs kept babies alive but left motor function uneven.

The warning on the label

Fractures carry the label's one warning. In Study 1, 5 of 53 patients (9%) on the recommended dose and 3 of 53 (6%) on the higher dose had fractures, against 1 patient (2%) on placebo, and femur fractures occurred in three treated patients. In an ongoing extension study at 20 mg per kilogram, fractures occurred in 22 patients (9%), including 5 serious femur fractures. No one stopped treatment for a fracture, most affected patients had low bone density, earlier fractures or contractures, and some fractures had no clear cause. Rat studies showed effects on the femoral head at exposures below those in people. A Leerink Partners analyst quoted by BioPharma Dive called the warning unexpected.

Other common adverse reactions, at 20% or more and more often than on placebo, included upper respiratory infections (66% against 54%), vomiting (30% against 16%) and gastroenteritis (23% against 8%). Creatine phosphokinase rose in 28% of treated patients against 12% on placebo, mostly without symptoms. Pneumonia was serious in three treated patients and none on placebo.

Price, access and the Roche contrast

Scholar Rock set a wholesale acquisition cost of $11,659 per vial, according to Reuters. Executives told investors a typical annual cost would be about $310,000 for a patient of 35 to 45 kilograms, who would use roughly three vials every four weeks. Reuters calls that a net figure and BioPharma Dive calls it a list price, so the number comes with a caveat. The path was bumpy: the FDA rejected the drug in September 2025 over a manufacturing site, and Scholar Rock later swapped facilities. Its European application was withdrawn on August 13, 2026, according to the European Medicines Agency.

The contrast with Roche's antibody is sharp. Emugrobart (GYM329), a related antibody designed to clear latent myostatin, was discontinued by Roche in March 2026 for SMA and for facioscapulohumeral muscular dystrophy after it did not consistently deliver the hoped for gains in muscle growth and function. On September 28, Chugai announced that development in obesity would also stop. An interim analysis of the Phase II GYMINDA study, which paired emugrobart with GLP-1/GIP drugs, concluded that meeting the prespecified goal of clinically meaningful weight loss was unlikely. Chugai says the drug was well tolerated and that it is preparing to restart SMA development, while also exploring out licensing.

The tidy story, that blocking myostatin builds muscle, has taken a beating, and the label shows why. One antibody earned a modest and statistically nominal motor gain with a bone warning, and a cousin was dropped in three settings. I would treat Isembyld as proof that a myostatin pathway drug can move a clinical scale in a disease where muscle is the bottleneck, not as proof that the class will keep muscle on people who lose weight. The obesity trials we reviewed measured lean mass in kilograms. The SMA approval rests on function. Only a trial that measures what people can do will settle the obesity question.