Myostatin inhibitors compared

Myostatin inhibitors are drugs and products that block myostatin, the protein muscle makes to limit its own growth. More than a dozen have been tried since the gene was found in 1997, and the same thing kept happening: nearly every drug that switched off myostatin built muscle, and almost none made anyone stronger.

In September 2026 the first one was approved, for a single disease. This page puts every major attempt on one board: what it blocked, who it was tested in, whether muscle grew, whether strength or daily function improved, and what happened to it.

What is myostatin?

Myostatin, also called GDF-8, is a signalling protein made in muscle. Its job is to hold muscle growth back, which is why it is often called the body's brake on muscle.

It was found in mice. Researchers deleted the gene, and the animals' individual muscles grew two to three times heavier than normal, from both bigger and more numerous fibres (mouse)1. Myostatin works by docking on a receptor on muscle cells called ActRIIB. Follistatin, a natural protein, can grab myostatin before it gets there2.

One detail matters for everything below. Myostatin is not the only brake. A decoy receptor that soaks up myostatin and its relatives still added muscle in mice that had no myostatin at all (mouse)3. Several related proteins press on the same receptor, and drugs that block more of them tend to do more, including more of what nobody wanted.

Myostatin deficiency: the animals and the one child

Nature has run this experiment many times. Myostatin deficiency, meaning little or no working myostatin, happens through at least five different gene changes:

• Belgian Blue cattle carry a small deletion in the gene (cattle)5; Piedmontese cattle carry a different change in the same stretch (cattle)6.
• Whippets with two broken copies are the heavily muscled "bully" type. Dogs with one broken copy look near-normal and race faster on average (dog)7.
• Texel sheep have an intact gene, but a change that lets muscle microRNAs silence it (sheep)8.
• One child in Germany was born with a broken copy from each parent and had unusually large muscles9. The variant is recorded as the cause of "myostatin-related muscle hypertrophy"10. The published report is a single case, not a measured study of strength.

Then a second group asked the obvious question: are mice without myostatin stronger? They were not. The bigger muscles produced no extra maximum force, were weaker for their size, and had fewer mitochondria, the parts of the cell that supply energy (mouse)4. Size is one ingredient of strength, not the whole recipe. Every natural case also grew up without the brake from its first cells. A drug arrives in an adult, or a sick child, whose muscles are already built.

The comparison board

Fourteen attempts, one card each. Read each card on its own: the populations, the measures and the scales differ, so nothing here is ranked.

The antibodies that grab myostatin

The first drugs were antibodies against the myostatin protein itself.

• Stamulumab (MYO-029) was tested in 116 adults with three kinds of muscular dystrophy. It showed only a trend toward larger muscle and no improvement in strength or function; the authors called for more potent drugs11.
• Landogrozumab raised arm-and-leg lean mass by 0.43 kg more than placebo in adults aged 75 and over who had fallen, with a couple of timed tests moving and most measures not12. It missed its main goal after hip replacement13, and in pancreatic cancer the higher-dose arm was stopped because more patients died on the drug than on placebo14.
• Domagrozumab was tested in 120 boys with Duchenne muscular dystrophy. After 49 weeks the stair-climb time differed from placebo by 0.27 seconds (p = 0.94), the muscle-volume gain was not significant, and both studies were terminated for lack of efficacy15.

The decoy receptors: ACE-031 and ACE-083

The next idea was to catch several brakes at once with a decoy of the receptor. [ACE-031](/articles/what-is-ace-031/) (ramatercept) is the best known. One injection in healthy postmenopausal women raised lean mass by 3.3% and thigh muscle volume by 5.1%16. In boys with Duchenne muscular dystrophy the trial was stopped early, and not because of muscle: some boys developed nosebleeds and small dilated blood vessels in the skin and gums17.

The same family of decoys now has approved medicines, for other diseases. Sotatercept is approved in the US for pulmonary arterial hypertension21, and in its pivotal trial nosebleeds and dilated small blood vessels were more common than on placebo20. Luspatercept is approved for certain anaemias22. Neither is approved for muscle.

ACE-083 tried to stay local, injected into one muscle. It grew the biceps by 16.4% in facioscapulohumeral muscular dystrophy18 and a lower-leg muscle by 13.5% in Charcot-Marie-Tooth disease19. One ankle-strength measure moved in the second trial, but function did not improve in any consistent way, and both programmes were closed. A muscle can be made measurably bigger, exactly where the drug went in, and still not be used better.

Follistatin: gene therapy vs the "follistatin 344" peptide

Follistatin is the body's own myostatin binder. Delivered as a gene into the thigh muscles of monkeys, it increased muscle size and strength with no abnormal organ findings (monkey)23.

In people, the only evidence is that same gene therapy, called FS344. In six people with Becker muscular dystrophy, walking distance rose in some and not at all in others, with no placebo group25. In inclusion body myositis, six treated patients walked further over a year than eight untreated patients, a comparison that was neither randomised nor blinded26. These are encouraging signals and the weakest kind of trial evidence.

The follistatin 344 peptide sold online is a different thing. No human trial of injected follistatin has been registered or published. When a lab tested 17 black-market products, only 9 contained follistatin27, and doctors have described 11 bodybuilders who developed a serious fluid build-up behind the retina after injecting it28. The gene therapy and the peptide share a name, not the evidence.

The weight-loss turn: bimagrumab, trevogrumab, taldefgrobep

GLP-1 drugs such as semaglutide and tirzepatide cause weight loss that includes some lean mass, so the question changed from building muscle to keeping it.

One word needs unpacking first. These trials measure lean mass with a DXA scan, which splits the body into fat, bone and everything else. "Everything else" includes muscle, but also water, organs, skin and blood. Keeping lean mass is not the same as keeping working muscle, and the authors of the largest obesity trial asked future studies to measure muscle by MRI alongside physical function33.

• [Bimagrumab](/articles/what-is-bimagrumab/) blocks the receptor, so it shuts out myostatin and its relatives together. In inclusion body myositis it did not improve walking at any dose29. In older adults thigh muscle grew, and walking improved only in a subgroup of slow walkers picked out afterwards30. In lung disease it "safely increased skeletal muscle mass but did not improve functional capacity"31. In obesity with type 2 diabetes it cut fat mass by 20.5% while adding lean mass32, and with semaglutide 92% of the weight lost was fat, but grip strength and patient-reported outcomes did not significantly improve and more people stopped for side effects33.
• Trevogrumab (anti-myostatin), given with semaglutide, prevented about half of the lean-mass loss. Adding garetosmab (anti-activin A) kept even more, but 28.3% of that group stopped for side effects and two people died; the company has not linked the deaths to the drug. No strength results have been released35. Garetosmab itself was approved by the FDA on 19 Aug 2026 for fibrodysplasia ossificans progressiva, a bone disease36.
• Taldefgrobep alfa failed a futility test in Duchenne37 and missed its main goal in spinal muscular atrophy, where 35% of patients had no measurable myostatin at the start. A fat-loss signal moved it into obesity trials38.

Apitegromab: the first approval, and what it covers

Apitegromab (ISEMBYLD) does not grab active myostatin. It stops the inactive form being switched on, which makes it more selective. On 11 Sep 2026 the FDA approved it for spinal muscular atrophy in people aged 2 and over, only alongside SMN2-targeted treatment41.

In the phase 3 trial, the motor-function score improved 1.8 points more than placebo with the doses pooled (p = 0.019); the higher dose alone did not clearly separate (p = 0.11)39. The FDA's first answer, in 2025, was a complete response letter that cited only an outside plant that filled the vials, not the drug's efficacy or safety data42. The label warns of fractures: 9% on the drug vs 2% on placebo41. In obesity, with tirzepatide, it cut lean-mass loss by 1.9 kg, and physical function did not differ40.

The approval narrows the claim instead of widening it: one disease, one motor score, alongside other treatment.

Why more muscle did not mean more strength

The research points to four reasons that fit together:

1. What gets measured. Most trials count lean mass or muscle volume, not what the muscle can do.
2. Quality. In mice, muscle grown without myostatin made less force for its size and had fewer mitochondria (mouse)4.
3. Timing. The natural mutations shape a body from its first cells; a drug arrives in a body already built, often one a disease is damaging.
4. Safety. The pathway also runs through blood vessels and bone, so trials were stopped before function had a chance to show.

Does blocking myostatin make the heart grow?

The heart is a muscle too, and this is the question people ask most. The human data so far are reassuring on this narrow point:

• In 68 healthy adults aged 60 to 86 given bimagrumab for six months, cardiac MRI showed no clinically relevant change in heart-muscle mass or pumping34.
• In the bimagrumab myositis trial, heart tracings and ultrasound showed no signal29.
• Pigs engineered to make extra follistatin showed no heart enlargement (pig)24, and monkeys given follistatin gene therapy showed no abnormal organs (monkey)23.

That is the absence of a signal, not a clean pass: it does not cover every drug, every person or every length of use. But the side effects that actually ended trials were in blood vessels, bones and frail patients, not in an enlarged heart.

YK-11 and "natural myostatin blockers"

YK-11 is a steroid-like lab compound that acts on the androgen receptor. Its "myostatin inhibitor" reputation comes from one experiment in mouse muscle cells, where it raised follistatin (mouse cells)43. It has no human trial of any kind and is not approved anywhere.

Supplements sold as natural myostatin blockers (epicatechin, fortetropin, creatine) have small studies behind them. The clearest test put 24 young men's legs in a brace for two weeks: fortetropin stopped their blood myostatin from rising, and their muscle size and strength fell just the same as on placebo44. A number in the blood moved; the muscle did not.

What the studies found

Latest research (September 2026)

• 11 Sep 2026: apitegromab approved by the FDA for spinal muscular atrophy alongside SMN2-targeted therapy, the first myostatin-pathway drug approved for a muscle condition41.
• 19 Aug 2026: garetosmab (anti-activin A) approved by the FDA for fibrodysplasia ossificans progressiva, a bone disease36.
• Still open: the obesity trials are short, measured in months. The longer maintenance results of the trevogrumab trial are not public, physical function was only an exploratory measure in the apitegromab obesity study, and no trial has followed healthy people on these drugs for years.

We update this section as those results report.

The short version

Blocking myostatin reliably changes the scan. It does not reliably change what the muscle can do. The side effects that mattered were outside the muscle, in blood vessels, bones and frail patients. And a gene missing from birth is not the same as a drug given to an adult. For the individual compounds, see what is follistatin, what is ACE-031 and what is bimagrumab. Educational research overview only, not medical advice. Nothing here is a dose or a recommendation.