The idea, and why it keeps being tested
Myostatin is a protein of the TGF-beta family that tells skeletal muscle to stop growing. Animals born without it carry far more muscle, and that observation has made blocking myostatin one of the most tested ideas in muscle medicine — first for muscle-wasting diseases, and now for a new reason: keeping muscle while people lose weight on GLP-1 drugs.
The pathway has three places to intervene:
- The precursor. Myostatin is made inactive and must be cut to signal. Block the precursor and it never switches on.
- The signal. Bind active myostatin before it reaches a receptor.
- The receptor. Block the activin type II receptors it signals through — which also carry other signals.
What is actually in trials: proteins, not peptides
Every myostatin-pathway candidate with a human result in this library is a large protein. None is a peptide.
| Compound | What it is | Where it acts | Human record |
|---|---|---|---|
| Apitegromab | antibody | the precursor | 188-person phase 3 in SMA; 102-person phase 2 with tirzepatide |
| Trevogrumab | antibody | mature myostatin | 1,005-person phase 2 with semaglutide, no results posted |
| Bimagrumab | antibody | the receptors | 507-person phase 2 with semaglutide; a failed phase 2b in inclusion body myositis |
| ACE-031 | receptor-fusion protein | a decoy receptor | stopped in Duchenne for safety |
| Follistatin | the body's own binding protein | the signal | no completed trial of the protein |
The search phrase "myostatin inhibitor peptides" therefore describes a category that, on the evidence, does not exist in the clinic. What exists is a set of antibodies, one fusion protein, and a natural binding protein sold in vials.
What the trials found
Apitegromab — the split result. Its phase 3 in spinal muscular atrophy, SAPPHIRE, improved motor function by 1.8 points on the pooled doses (p=0.019) and 1.4 points on the 20 mg/kg dose alone, a comparison whose confidence interval crossed zero (p=0.11). With tirzepatide, in EMBRAZE, the apitegromab group lost 1.9 kg less lean mass at 24 weeks. The detail is on the apitegromab entry, which sets out the split phase 3.
Bimagrumab — the dose matters. In BELIEVE, bimagrumab 30 mg/kg with semaglutide produced −17.79 kg at 48 weeks against −14.24 kg on semaglutide alone. At 10 mg/kg the combination produced −14.31 kg — no different. Its first large test, in inclusion body myositis, missed on walking distance at every dose.
Trevogrumab — waiting. COURAGE, a 1,005-person phase 2 alongside semaglutide, passed its primary completion date of 2026-06-16 with no results posted when checked. Its earlier 253-person sarcopenia trial is marked completed and has no published result under either of its names.
ACE-031 — the warning. A fusion of the activin receptor type IIB with an antibody fragment, it acted as a decoy that soaked up myostatin and related signals. In boys with Duchenne muscular dystrophy it showed trends toward more lean mass and bone density and a maintained walking distance — none statistically significant — and the study was stopped after the second dosing regimen because of nosebleeds and telangiectasias (small dilated blood vessels) (Muscle & Nerve, 2017). ClinicalTrials.gov records two ACE-031 studies as terminated "based on preliminary safety data." The lesson for the class: the receptors carry more than myostatin, and a drug that blocks them broadly blocks those other signals too. Apitegromab's precursor-only design is a direct answer to that problem.
What is sold: follistatin-344
The product most often sold as a myostatin peptide is follistatin, usually labelled follistatin-344. The underlying biology is real — follistatin binds and neutralises myostatin in the body. The product evidence is not:
- follistatin is a glycoprotein of 288 or 315 amino acids, not a short peptide; "344" refers to a precursor form;
- its only registered administration study delivers a gene for it through a viral vector, not the protein;
- no completed trial has injected follistatin protein into people.
The full record is on the follistatin entry, which separates the protein from the gene therapy.
What the record supports
- Blocking myostatin changes body composition in people — more lean mass, less fat — in several trials.
- Turning that into function has been harder: one phase 3 split by dose, one phase 2b failed.
- Every candidate with a human result is a large protein. No peptide myostatin inhibitor has a human trial in the records checked.
- Nothing in the class is approved in the United States as of 2026-09-21.
The performance side of the question — what athletes are told these compounds do — is MuscleLedger's subject, and the lean-mass question in GLP-1 treatment is set out by GLP1Ledger. In this library the neighbouring entries are growth-factor analogue IGF-1 LR3 and the index in what peptides are used for.
Last checked
2026-09-21. ACE-031 records from the ClinicalTrials.gov v2 API and its Duchenne trial abstract (PMID 27462804) on PubMed; the apitegromab, bimagrumab, trevogrumab and follistatin figures from those entries' own sources, each checked on the dates given there.