ACE-031: The Myostatin Blocker That Was Stopped, and Why

At a Glance

ACE-031 is the most instructive failure in this part of the library. It is the compound that proved the myostatin pathway can be moved in humans, with hard endpoints: a single subcutaneous dose produced a 3.3% increase in total lean mass by DXA and a 5.1% increase in thigh muscle volume by MRI within 29 days (Attie 2013). Nothing in the grey-market myostatin space comes close to that quality of evidence.

It is also the compound that showed why blocking this pathway broadly is dangerous. The phase 2 trial in ambulatory boys with Duchenne muscular dystrophy was stopped after the second dosing regimen because of epistaxis and telangiectasias (Campbell 2017). Those are not nuisance findings in a paediatric population, and they were not dose-finding noise: they follow from what the molecule is, a decoy receptor that cannot distinguish myostatin from vascular morphogenetic proteins.

The honest verdict is that ACE-031 belongs in this library as history and as a warning, not as an option. On top of the safety issue there is a manufacturing reality: a 130 kDa glycosylated Fc-fusion protein requires mammalian cell culture and cannot be produced by the solid-phase peptide synthesis that makes everything else on these shelves. Anything sold in a vial labelled ACE-031 should be treated as unidentified material until proven otherwise.

How it works

Myostatin restrains skeletal muscle mass by signalling through activin type II receptors, principally ActRIIB. The mechanistic foundation was established in transgenic mice: the myostatin C-terminal dimer binds ActRIIB, and mice expressing a dominant-negative ActRIIB, the myostatin propeptide, or follistatin in skeletal muscle all developed dramatic increases in muscle mass comparable to myostatin knockouts (Lee 2001).

ACE-031 took the direct route to the same effect: take the ligand-binding domain of ActRIIB, fuse it to an antibody Fc region for a long half-life, and let it circulate as a sponge. Myostatin that binds the decoy never reaches the real receptor.

The problem is written into the design. ActRIIB is not a myostatin-specific receptor. It binds activin A and activin B, GDF-11, and bone morphogenetic proteins including BMP-9 and BMP-10, which have established roles in vascular homeostasis and endothelial quiescence. A soluble decoy sequesters all of them in proportion to its affinity. That is the accepted explanation for the vascular adverse events, and it is a class effect of broad ActRIIB inhibition rather than something a lower dose avoids.

Half-life matters too. Fc fusion gave ACE-031 an elimination half-life of 10 to 15 days (Attie 2013), which is why dosing was every two to four weeks, and why an adverse effect, once it appears, persists for weeks.

What the evidence shows

Human trials

Phase 1, single ascending dose in healthy volunteers. 48 healthy postmenopausal women were randomised 3:1 to a single subcutaneous dose of ACE-031 at 0.02 to 3 mg/kg or placebo. The compound was generally well tolerated, with injection-site erythema among the adverse events. Mean AUC and Cmax rose linearly with dose and the mean half-life was 10 to 15 days. At day 29 in the 3 mg/kg group, total body lean mass rose by 3.3% (p = 0.03, DXA) and thigh muscle volume by 5.1% (p = 0.03, MRI), with serum biomarker changes suggesting improvements in bone and fat metabolism (Attie 2013).

Read that again in the context of this whole batch. One subcutaneous dose, 29 days, imaging-confirmed muscle volume increase. This is what a real effect on human muscle looks like when it is measured properly, and it came from a biologic developed by a pharmaceutical company, not from a peptide protocol.

Phase 2 in Duchenne muscular dystrophy, halted. ACE-031 was given subcutaneously every two to four weeks to ambulatory boys with Duchenne muscular dystrophy in a randomised, double-blind, placebo-controlled, ascending-dose trial whose primary objective was safety. ACE-031 was not associated with serious or severe adverse events, but the study was stopped after the second dosing regimen because of potential safety concerns, specifically epistaxis and telangiectasias. There were non-significant trends toward maintenance of six-minute walk distance against decline on placebo, increased lean body mass and bone mineral density, and reduced fat mass. The authors concluded that non-muscle-related adverse events drove the decision to discontinue, while maintaining that myostatin inhibition remains a promising approach (Campbell 2017).

What came after

The field moved toward more selective targeting. Bimagrumab, a monoclonal antibody against activin type II receptors, increased thigh muscle volume in men with casting-induced atrophy (Rooks 2017), and its effects on pituitary neurohormonal axes have been characterised separately (Garito 2018). It is now being investigated in obesity, where preserving lean mass during substantial weight loss is the central problem (Kaur 2024). Ligand-selective antibodies such as apitegromab and trevogrumab followed the same logic of narrowing the target.

The sobering counterpoint: a 2025 review of myostatin inhibition concludes that despite extensive research and promising preclinical results, clinical trials in muscular dystrophy have failed to yield substantial improvements in muscle function or fitness, citing drug specificity, species differences in serum myostatin concentrations, and the need for neural input for functional gain (Wetzlich 2025). More muscle is not automatically more function.

What does not exist

No completed efficacy trial of ACE-031. No long-term safety data. No published human data of any kind on material sold outside the trial system under this name.

Dosing

For informational purposes only — this summarises doses used in research and practice; it is not a prescription or individual medical advice.

ACE-031 has no legitimate supply and no approved use. The doses below are what was given in the two trials, included because they are the factual record and because they expose how far the circulating numbers are from anything a vial could deliver.

Why those numbers cannot be taken at face value outside a trial. A 3 mg/kg dose in an 80 kg adult is 240 mg of a 130 kDa fusion protein. That is a quantity and a molecule type produced in bioreactors with purification and glycosylation control, supplied in milligram-per-millilitre formulations with stabilisers. Vials offered as ACE-031 are typically labelled in single-digit milligrams, one to two orders of magnitude below the trial dose, before any question about whether the contents are the molecule at all.

Titration. Not applicable. There is no legitimate protocol to titrate, and the adverse events that stopped development appeared at trial doses in a monitored setting.

For the compounds in this space that can at least be handled coherently, see the peptide injection guide and peptides for muscle growth.

What people report: what works and what doesn’t

Sources: the aggregated community data set (Peptide Protocol Wiki, 15 reports), a second compilation (Peptide DB), and the published trials. This is the thinnest report base in the batch, and the aggregated data describe the discussion as mostly retrospective and educational rather than reflecting active use. Patterns, not proof.

Reported to work

• Nothing that can be attributed to ACE-031 with any confidence. The report base is too small and product authenticity is the dominant confounder.

Mixed or unclear

• Whether any available material is actually ACE-031. The aggregated community data flag this explicitly: a large complex fusion protein cannot be produced by standard peptide synthesis, requires mammalian cell culture, and claims of availability from research suppliers warrant strong scepticism.
• Historical interest versus use. Most discussion is about why the programme stopped and what the selective successors do, rather than about results.

Reported not to work, or disappointing

• The expectation of safe myostatin blockade. Community discussion is described as often underestimating the vascular safety signal that ended development, which is the single most important thing to understand about this molecule.
• Side effects: the serious ones come from the trials rather than from reports. Epistaxis, telangiectasias and gingival bleeding appeared in clinical use at trial doses. The absence of community reports of these events most likely reflects product authenticity and dose, not a benign profile.

What a properly designed trial would need to show: for any ActRIIB-directed agent, that muscle gain translates into function, and that the vascular ligands are spared. That is precisely the design brief the selective successors were built against, and the 2025 review suggests the function half of it remains unsolved (Wetzlich 2025).

Safety and side effects

The documented signal. Epistaxis, telangiectasias and gingival bleeding in the Duchenne trial, serious enough to stop the study after the second dosing regimen even though no serious or severe adverse events were formally recorded (Campbell 2017). In the healthy-volunteer study the main reported adverse event was injection-site erythema (Attie 2013), a reminder that a single dose in adults looked benign and the problem emerged with repeated dosing in children.

Why it is mechanistic rather than incidental. Broad ActRIIB inhibition traps BMP-9 and BMP-10, ligands required for vascular integrity, alongside myostatin. This is why the field responded by narrowing the target rather than lowering the dose.

What a 10 to 15 day half-life means for risk. An adverse effect from a biologic with that half-life cannot be reversed by stopping. Exposure persists for weeks after the last injection.

Immunogenicity. Fc-fusion biologics can generate anti-drug antibodies affecting exposure and safety. No immunogenicity data exist for non-pharmaceutical material.

Product identity as a safety issue in its own right. For this molecule the usual grey-market concerns are sharpened by the manufacturing requirement. Anti-doping laboratories have documented unidentified fusion proteins in confiscated black-market products, including an 18 kDa construct consisting of bacterial thioredoxin fused to an unassigned peptide, with thioredoxin serving as a solubility tag in recombinant protein production (Walpurgis 2014). That is exactly the signature of bacterially expressed material being sold as something it is not. See peptide sourcing.

Legal and regulatory status

United States. Never approved; development discontinued in 2013. ACE-031 is an investigational biologic, which means that outside an authorised clinical trial there is no lawful route to it. A biologic of this type would require a biologics licence application, and it is not eligible for pharmacy compounding under section 503A: it does not appear in category 1 or 2 of FDA’s 503A bulk drug substance list (updated 14 May 2026), and bulk substances may be used in 503A compounding only when they meet a statutory condition. See are peptides legal.

European Union and Germany. No authorisation. Germany’s Anti-Doping Act (AntiDopG) annex includes a category for substances that modify myostatin function, listing follistatin and its derivatives and stamulumab, an anti-myostatin antibody. A soluble ActRIIB decoy sits squarely in that functional category, and § 2(3) makes it an offence to acquire, possess or import more than a “non-small quantity” of a listed substance for the purpose of doping in sport; the quantity regulation (DmMV, in force 16 March 2023) sets 450 mg for that group.

Sport. Prohibited under the WADA list in the hormone and metabolic modulators section, which covers agents preventing activin receptor IIB activation, including decoy activin receptors. See peptides and anti-doping rules.

How ACE-031 fits, and what it teaches

Three lessons come out of this story, and they apply well beyond ACE-031.

First, the pathway works in humans. Anyone sceptical that myostatin inhibition can add measurable muscle should read the phase 1 data. A single dose moved MRI-measured muscle volume in a month.

Second, specificity is the whole problem. The decoy approach succeeded at adding muscle and failed at leaving the vasculature alone. The compounds that followed were designed around exactly that constraint.

Third, more muscle is not automatically more function, which is the conclusion of the most recent review of the field (Wetzlich 2025). For anyone whose actual concern is age-related muscle loss, sarcopenia prevention and the muscle-preservation work with bimagrumab and semaglutide are where the useful evidence lives.

Related Articles

• Follistatin: Myostatin Blockade and the Injectable Gap — the other myostatin-pathway agent sold outside medicine.
• Bimagrumab + Semaglutide: Preserving Muscle While Losing Fat — the selective successor, with trial data.
• Peptides for Muscle Growth: What the Evidence Actually Supports — the comparative hierarchy.
• Sarcopenia Prevention: Preserving Muscle Mass After 40 — the evidence-based approach.
• Peptide Sourcing: Pharmaceutical-Grade vs. Grey Market — why a 130 kDa fusion protein in a research vial is a red flag.