Follistatin / myostatin inhibitors

The brake and what happens when it is missing

The body has a brake on muscle growth, and it is called myostatin. This is a signaling substance that muscle produces itself to limit its own growth. Not out of malice; muscles cost energy. Follistatin is the obvious countermeasure, an endogenous protein that intercepts myostatin. More follistatin means less active myostatin means less brake.

If the brake fails, you see it immediately. There are cattle breeds with a defect in the myostatin gene that look absurdly muscular, and the same in dogs. In 2004 the New England Journal of Medicine published the case of a German boy who was strikingly muscular from birth. The cause was a mutation at a splice site of the myostatin gene, in him on both gene copies, in his mother on one; she was unusually muscular, but without complaints.

Three attempts, three ways to fail

The pharmaceutical industry has worked on this path for more than 20 years, with proper endpoints and a great deal of money. Until September 2026 there was not a single approved myostatin inhibitor. The three best-known attempts failed in three different ways: because of safety findings, because of lack of efficacy, and because additional muscle mass did not translate into function.

The first attempt, ACE-031, was a decoy receptor that captures myostatin before it reaches the muscle. It was stopped in 2011 because of nosebleeds, gum bleeding and dilated small blood vessels in the skin. The reason lies in the biology of the docking point: a whole family of signaling substances depends on this receptor, and some of them regulate blood vessel growth. A switch that has only one label in the textbook has several in the body.

More mass is not more function

The third attempt is the most instructive. Bimagrumab, an antibody against the same receptor, was tested in the RESILIENT study in inclusion body myositis. The primary endpoint was missed: the 6-minute walking distance declined progressively from week 24 to week 104 in all groups. In all of them.

What is remarkable about this is that bimagrumab demonstrably builds muscle mass. That is undisputed and the reason why it is currently being rediscovered in combination with weight-loss injections. It makes people more muscular, and they still could not walk better. Muscle is simply not just volume, but volume times neural control times tendon quality times coordination. In a disease that destroys the muscle fibers themselves, more mass made of broken fibers is simply more broken mass.

What is sold today

In Honduras there is a special economic zone called Próspera with its own regulatory framework, and providers there offer follistatin gene therapy. According to the investigation by MIT Technology Review, the provider that charges 25,000 dollars for it has not published any robust clinical study data. A gene therapy researcher quoted says that much of the evidence for follistatin comes from rodents.

The difference from a peptide is decisive: you stop a peptide, and after days to weeks it is gone. A gene therapy is meant to stay; that is its whole purpose. If something goes wrong, and ACE-031 showed that this signaling pathway is linked to blood vessels, there is no off switch. With plasmid-based methods this sounds shorter-lived, with virus-based ones more permanent; neither has been tested.

What the studies show

Case report, New England Journal of Medicine 2004

A German boy, examined at 4.5 years, strikingly muscular from birth. The cross-section of the thigh muscle was 7.2 standard deviations above the mean of children of the same age; he held 2 dumbbells of 3 kilograms each horizontally with outstretched arms. Motor and mental development were normal, and cardiac examinations up to 4.5 years were unremarkable.

ACE-031 in Duchenne muscular dystrophy, stopped 2011

A decoy receptor, tested in boys with Duchenne muscular dystrophy, stopped in April 2011 because of nosebleeds, gum bleeding and telangiectasias. Everything resolved completely after discontinuation, and considered individually the companies did not call it a serious safety problem, but the authorities first wanted the cause understood. In 2013 the program was discontinued.

Domagrozumab, phase 2 in Duchenne, result 2018

An antibody directly against myostatin, tested in 121 boys aged 6 to 15. The primary endpoint was the time to climb 4 stairs after 1 year versus placebo, that is, an endpoint that measures what matters to families. Result in 2018: missed. The manufacturer wrote that the secondary endpoints also did not support a treatment effect, and that the studies had not been ended for safety reasons.

Bimagrumab, RESILIENT study, phase 2b in inclusion body myositis

An antibody against the receptor, 4 groups with 3 dose levels and placebo, as an infusion every 4 weeks, with extension to week 104. The 6-minute walking distance declined progressively from week 24 to week 104 in all groups; the primary endpoint was missed. Tolerability was good: side effects in 91 percent under bimagrumab versus 89 percent under placebo, serious events comparable.

Apitegromab, SAPPHIRE study, phase 3 in spinal muscular atrophy

An antibody that prevents the activation of myostatin, tested in 188 children and young adults aged 2 to 21 who cannot walk and were already receiving standard therapy, as an infusion every 4 weeks versus placebo. After 12 months, the motor scale in the 2- to 12-year-olds was on average 1.8 points above placebo, statistically significant; the higher dose alone missed significance. Side effects occurred about as often as under placebo, and nobody discontinued because of them. The FDA approved apitegromab on September 11, 2026; the manufacturer withdrew the European application on August 13, 2026, because a manufacturing site did not meet the requirements of Good Manufacturing Practice.

Follistatin gene therapy, phase 1/2a in Becker muscular dystrophy (2015)

Six patients received a virus carrying the follistatin gene directly into both thigh muscles, without a control group. The 6-minute walking distance improved by 29 to 125 meters in 4 of 6, and not at all in 2; no side effects were observed. This is a small, uncontrolled single-center study in a severe disease, not evidence for the product offered today.

What is missing in healthy people

All three failed programs were in sick people, in Duchenne muscular dystrophy and in inclusion body myositis. That it might look different in healthy people is a fair objection, but nobody has tested it. There is no study that gave a healthy person a myostatin inhibitor and then measured strength over a relevant period against placebo. The assumption that it works in healthy people is exactly that: an assumption.

Status, approval and legal

In the US, the antibody apitegromab (Isembyld) has been approved since September 11, 2026, for people aged 2 and over with spinal muscular atrophy who are already receiving an SMN therapy. In the EU, the manufacturer withdrew the application in August 2026 because of deficiencies at a manufacturing site, not because of the study data. No myostatin inhibitor is approved for muscle building in healthy people. The injectable peptide sold as follistatin is a gray-market product. The gene therapy is offered in a special economic zone whose regulatory framework was created specifically so as not to have to ask the usual questions.

Safety

The documented safety signal comes from the ACE-031 program: nosebleeds, gum bleeding and telangiectasias, regressing after discontinuation, with unexplained cause. The reason is structural, because a family of signaling substances depends on the blocked receptor, and some of them regulate blood vessel growth. With bimagrumab tolerability was good; with domagrozumab the studies were not ended for safety reasons. The dosage form remains decisive: a peptide can be stopped, a gene therapy cannot.

Sources

• Schuelke et al., N Engl J Med 2004 (case report on a myostatin mutation in a child)
• Campbell et al., Muscle Nerve 2017 (ACE-031 in Duchenne muscular dystrophy, stopped 2011 and discontinued 2013)
• Wagner et al., Neuromuscul Disord 2020 (domagrozumab, phase 2 in Duchenne muscular dystrophy, result 2018)
• Hanna et al., Lancet Neurol 2019 (bimagrumab, RESILIENT study in inclusion body myositis)
• MIT Technology Review, investigation into follistatin gene therapy in the Próspera special economic zone
• Crawford et al., Lancet Neurol 2025 (apitegromab, SAPPHIRE study, phase 3 in spinal muscular atrophy)
• Mendell et al., Mol Ther 2015 (follistatin gene therapy, phase 1/2a in Becker muscular dystrophy)
• EMA, withdrawal letter for the marketing authorization application Isembyld (apitegromab), August 13, 2026
• FDA, approval of Isembyld (apitegromab), September 11, 2026