Trevogrumab vs Garetosmab vs Bimagrumab vs Apitegromab: The 2026 Myostatin Inhibitor Comparison

Trevogrumab and garetosmab are two different antibodies from the same Regeneron programme, and they block two different ligands. Trevogrumab (REGN1033) neutralises myostatin, also called GDF-8. Garetosmab (REGN2477) neutralises activin A. Both are fully human IgG4 monoclonal antibodies, both are ligand traps rather than receptor blockers, and both are studied together because myostatin and activin A are the two principal ligands restraining skeletal muscle mass through the same activin type II receptors.

The two most common questions about them have short answers. What is the difference? Trevogrumab binds myostatin and deliberately does not touch GDF11; garetosmab binds activin A, AB and AC and deliberately does not touch activin B, inhibin A, the BMPs, myostatin or GDF11. Why are they searched together? Because Regeneron runs them together in the Phase 2 COURAGE obesity trial, where semaglutide plus trevogrumab plus garetosmab produced the best body-composition result and the worst tolerability in the same arm.

They are also not the only antibodies in this space. Bimagrumab and apitegromab attack the same pathway at different points, and in 2026 all four are in active clinical development. This guide compares them on mechanism, on published trial numbers, and on what each is actually useful for as a research reagent.

Key Takeaways

• Trevogrumab traps myostatin in its mature, latent and pro-forms, with no GDF11 cross-reactivity. Use it when an effect must be attributable to myostatin alone.
• Garetosmab traps activin A, AB and AC, and nothing else in the TGF-beta superfamily. Use it to separate activin signalling from BMP signalling in the same tissue.
• Bimagrumab is not a ligand trap. It blocks the ActRIIA and ActRIIB receptors, so it neutralises myostatin and activin A together. Largest effect, least attribution.
• Apitegromab binds only the pro- and latent forms of myostatin and ignores the mature protein. The narrowest target window of the four.
• In COURAGE, semaglutide alone lost 6.5% of lean body mass at 26 weeks. Adding trevogrumab took that to 3.3%. Adding garetosmab on top took it to 2.0%.
• In BELIEVE, bimagrumab at 30 mg/kg produced a 2.5% increase in lean mass where semaglutide 2.4 mg alone produced a 7.4% loss.
• None of these are peptides. All four are ~145 kDa antibodies grown in mammalian cell culture, and a peptide Certificate of Analysis cannot characterise any of them.

The Comparison, at a Glance

The single most useful distinction in that table is the level of action row. Everything else follows from it.

Ligand Trap vs Receptor Blocker: Why It Decides Your Experiment

Myostatin, activin A, activin B, GDF11 and several BMPs all signal through a small shared set of type I and type II receptors, feeding the same SMAD2 and SMAD3 machinery. That crowding is the central problem in this field.

A receptor blocker like bimagrumab sits on ActRIIA and ActRIIB and stops everything that signals through them. The effect size is the largest available, because you are removing several brakes at once. The cost is that no downstream assay can tell you which ligand mattered.

A ligand trap like trevogrumab, garetosmab or apitegromab binds one ligand in the extracellular space and leaves the receptor free for everything else. The effect is smaller, and it is attributable.

This is why the older soluble ActRIIB-Fc decoy constructs produce impressive phenotypes and unclear papers. They bind myostatin, GDF11, activin A and several BMPs simultaneously. If your study needs to say this ligand did this, a decoy receptor cannot get you there and a selective antibody can.

Trevogrumab (REGN1033)

A fully human IgG4-kappa antibody against myostatin. CAS 1429201-24-0, molecular formula C6374H9884N1696O2018S46, 144,037.80 g/mol.

It binds the mature, latent and pro-forms of myostatin, and it does not cross-react with GDF11 despite roughly 90 percent mature-domain sequence identity between the two. That selectivity is the property worth paying for: GDF11 and myostatin are the pair that most tools in this space cannot separate.

The IgG4 backbone is a deliberate choice. IgG4 has minimal Fc effector function, so a neutralising antibody sequestering a soluble ligand does not also recruit antibody-dependent cytotoxicity against cells displaying it.

See the Trevogrumab research guide and specifications.

Garetosmab (REGN2477)

A fully human IgG4 antibody against activin A. CAS 2097125-54-5, approximately 145.5 kDa.

Its selectivity list is unusually long and unusually clean. It binds activins containing the inhibin beta-A subunit, meaning activin A, activin AB and activin AC. It does not bind or functionally inhibit activin B, inhibin A, BMP-2, BMP-6, BMP-9, BMP-10, GDF8 or GDF11.

Its other research context is bone. Regeneron scientists identified activin A as the driver of heterotopic ossification in fibrodysplasia ossificans progressiva (FOP), where the mutant ACVR1 receptor misreads activin A as an osteogenic signal. In healthy tissue activin A does not drive bone formation; in FOP the mutant receptor converts it into a bone-forming input.

In the Phase 2 FOP study, the 3 mg/kg arm showed a 94 percent reduction in new heterotopic ossification lesions versus placebo (1 lesion vs 19, p=0.0274), and the 10 mg/kg arm a 90 percent reduction (2 vs 19, p=0.0260). The FDA granted Fast Track designation in 2017 and accepted the Biologics License Application for Priority Review with a target action date in August 2026.

See the Garetosmab research guide and specifications.

What COURAGE Actually Showed

COURAGE (NCT06299098) is Regeneron's Phase 2 obesity study combining semaglutide with trevogrumab, with or without garetosmab. Results were presented at the EASD 2025 Annual Meeting.

The trial first quantified the problem it exists to solve: roughly one third of the weight lost on semaglutide alone was lean mass, not fat.

Change in lean body mass from baseline at 26 weeks:

The triplet reached 80.9 percent lean mass preservation and a 27.3 percent increase in fat mass reduction versus semaglutide alone.

And then the part that matters just as much: semaglutide plus trevogrumab was generally well tolerated, but the triplet had a substantially higher rate of discontinuations for tolerability and other adverse events. Adverse events reported in at least 5 percent of any group included muscle spasms, nausea, constipation, fatigue, diarrhoea, headache, vomiting, gastro-oesophageal reflux, upper respiratory tract infection, nasopharyngitis, urinary tract infection, influenza and COVID-19.

The best body composition and the worst discontinuation rate landed in the same arm. That is the single most useful fact about adding activin A blockade on top of myostatin blockade, and it is why the two antibodies are studied as a stack rather than assumed to be additively better.

Bimagrumab (BYM338)

Bimagrumab is the outlier: a human IgG1 antibody that blocks the type II activin receptors themselves rather than any single ligand. Because ActRIIA and ActRIIB are where myostatin and activin A both converge, blocking the receptor neutralises both at once.

The BELIEVE Phase 2b trial (NCT05616013) randomised 507 adults with obesity across nine arms over 72 weeks. Results were presented at ADA 2025 and published in Nature Medicine.

The combination produced roughly 22 percent body weight reduction with 92 percent of the loss coming from fat mass, and up to 17.8 kg total reduction against up to 14.2 kg for semaglutide alone.

Bimagrumab actually adding lean mass where semaglutide removed it is the strongest body-composition signal any of these four has produced. The trade-off is the one described above: receptor-level blockade cannot tell you which ligand did the work.

Apitegromab (SRK-015)

Apitegromab takes the narrowest approach of the four. It binds the pro- and latent forms of myostatin, the inactive precursors, and does not bind mature myostatin at all. Blocking activation rather than the activated protein is what gives it the tightest specificity within the TGF-beta superfamily, and it has been proposed as the reason for its tolerability profile.

It is also the furthest along. In the pivotal Phase 3 SAPPHIRE trial in spinal muscular atrophy it met its primary endpoint on the Hammersmith Functional Motor Scale Expanded at week 52, with 30.4 percent of treated patients improving by more than 3 points versus 12.5 percent on placebo. It is the first muscle-targeted candidate to succeed in a pivotal Phase 3 in SMA. Its Biologics License Application has been accepted with a PDUFA date of 30 September 2026.

In obesity, the EMBRAZE Phase 2 trial paired apitegromab 10 mg/kg with tirzepatide over 24 weeks and reported 1.9 kg less lean mass loss than placebo (P=0.001) at similar total weight loss, a 54.9 percent retention of lean mass relative to placebo.

How to Choose Between Them

The last row is the design COURAGE used, and it is the reason these two are ordered as a pair. Because neither antibody cross-reacts with the other's target, the combination is genuinely additive rather than confounded, which a receptor blocker cannot give you.

None of These Are Peptides, and the COA Proves It

Several research-chemical vendors list trevogrumab and garetosmab under "peptides", sometimes literally titled "trevogrumab peptide vial". That labelling is wrong, and it has practical consequences.

A peptide like BPC-157 is a short chain built by solid-phase chemical synthesis. These four are ~145 kDa four-chain immunoglobulins with interchain disulfide bonds and N-linked glycosylation, grown in Chinese hamster ovary cell culture and purified on Protein A columns. Peptide synthesis cannot produce them at any scale.

That changes the certificate entirely:

Reversed-phase conditions denature an antibody, so an RP-HPLC purity number for one is meaningless. A single ESI-MS charge state is meaningless against a glycan-heterogeneous population. And the property you are actually buying, selectivity, appears on neither.

If a supplier hands you an RP-HPLC trace and a Karl Fischer water content for a monoclonal antibody, that certificate was written for a different kind of molecule. Our guide to verifying purity and COAs covers the peptide side of this in detail.

Frequently Asked Questions

What is the difference between trevogrumab and garetosmab?

They block different ligands in the same pathway. Trevogrumab (REGN1033) binds and neutralises myostatin, also called GDF-8, in its mature, latent and pro-forms, and does not cross-react with GDF11. Garetosmab (REGN2477) binds and neutralises activin A, activin AB and activin AC, and does not bind activin B, inhibin A, BMP-2, BMP-6, BMP-9, BMP-10, GDF8 or GDF11. Both are fully human IgG4 monoclonal antibodies from Regeneron and both are ligand traps rather than receptor blockers.

Why are trevogrumab and garetosmab searched together?

Because Regeneron studies them together. In the Phase 2 COURAGE obesity trial (NCT06299098), semaglutide was combined with trevogrumab with or without garetosmab. Myostatin and activin A are the two principal ligands restraining muscle mass through the same activin type II receptors, so blocking one leaves the other intact. The triplet arm produced the best body composition in the study and also the highest discontinuation rate.

Is trevogrumab better than bimagrumab?

Not better, different. Bimagrumab blocks the ActRIIA and ActRIIB receptors, so it neutralises myostatin and activin A together and produces a larger effect: in BELIEVE it drove a 2.5 percent increase in lean mass where semaglutide alone drove a 7.4 percent loss. Trevogrumab traps only myostatin, so the effect is smaller but can be attributed to a single ligand. Choose receptor blockade for maximum effect and ligand trapping for clean attribution.

What makes apitegromab different from trevogrumab?

Apitegromab binds only the pro- and latent precursor forms of myostatin and ignores the mature, active protein, so it blocks activation rather than the activated ligand. Trevogrumab binds mature, latent and pro-forms alike. Apitegromab has the narrowest target window of the four antibodies here and is the furthest along in development, having met its primary endpoint in the Phase 3 SAPPHIRE trial in spinal muscular atrophy with a PDUFA date of 30 September 2026.

Are trevogrumab and garetosmab peptides?

No, and vendors listing them as peptides are mislabelling them. Both are monoclonal antibodies of roughly 145 kDa, built from two heavy chains and two light chains with interchain disulfide bonds and N-linked glycosylation, produced by recombinant expression in CHO mammalian cell culture. Solid-phase peptide synthesis cannot make a molecule of that size or architecture. The distinction determines how the material is tested, stored and handled.

Are any of these approved medicines?

No. Trevogrumab, garetosmab and bimagrumab are all investigational and not approved in any jurisdiction. Garetosmab's Biologics License Application for fibrodysplasia ossificans progressiva is under FDA Priority Review with a target action date in August 2026, and apitegromab's application for spinal muscular atrophy carries a PDUFA date of 30 September 2026. Material supplied for research is a reagent, unrelated to any clinical supply chain, and is not for human use.

What is the COURAGE trial?

COURAGE (NCT06299098) is Regeneron's Phase 2 study testing whether adding trevogrumab, with or without garetosmab, to semaglutide preserves lean mass during weight loss in adults with obesity. Results presented at EASD 2025 showed lean body mass change at 26 weeks of -6.5% for semaglutide alone, -3.3% adding trevogrumab 200 mg, -3.8% adding trevogrumab 400 mg, and -2.0% for the triplet, with the triplet reaching 80.9% lean mass preservation and 27.3% greater fat mass reduction.

Specifications and Certificates

Both Regeneron antibodies are stocked as research reference standards with batch-matched antibody Certificates of Analysis covering SEC-HPLC monomer purity, SDS-PAGE identity and chain integrity, deglycosylated intact mass, ELISA binding activity and the selectivity counter-screen. Full specifications, pricing per vial and the published certificates are on the Trevogrumab and Garetosmab product pages, and every published certificate is listed at lab results.

For laboratory research purposes only. Not for human consumption or therapeutic use.