ACE-031

Healthy postmenopausal women, single ascending dose: 48 participants

The first published human study was a double-blind, placebo-controlled single-dose trial in 48 healthy postmenopausal women randomised 3:1 to one subcutaneous dose of ACE-031 from 0.02 to 3 mg/kg or placebo (ClinicalTrials.gov NCT00755638). The endpoints were safety, pharmacokinetics and pharmacodynamics. ACE-031 was reported as generally well tolerated, with injection site erythema among the adverse events. Exposure 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 was 3.3% higher by dual x-ray absorptiometry (p = 0.03) and thigh muscle volume 5.1% higher by magnetic resonance imaging (p = 0.03), both statistically significant. Serum biomarker changes were read by the authors as signs of effects on bone and fat metabolism (PMID: 23169607).

Healthy postmenopausal women, multiple ascending dose: 70 participants, unpublished

A second phase 1 study randomised 70 healthy postmenopausal women to multiple ascending doses of ACE-031 or placebo between August 2009 and February 2011 (ClinicalTrials.gov NCT00952887). The registry lists it as completed with no results posted, and no publication of it could be retrieved. Whatever that study found is not in the public record.

Duchenne muscular dystrophy, phase 2: 24 boys, stopped early

The main efficacy test was a randomised, double-blind, placebo-controlled ascending-dose trial in ambulatory boys with Duchenne muscular dystrophy, with ACE-031 given subcutaneously every two to four weeks (ClinicalTrials.gov NCT01099761). The registry records 24 participants across three groups: 9 on 0.5 mg/kg every four weeks, 9 on 1.0 mg/kg every two weeks and 6 on placebo. A planned 2.5 mg/kg group has 0 participants at risk in the posted results, so the trial stopped before that cohort was dosed.

The primary objective was safety, not efficacy. ACE-031 was not associated with serious or severe adverse events. The trial was stopped after the second dosing regimen because of potential safety concerns, specifically epistaxis and telangiectasias. On the efficacy side the report describes trends rather than results: a trend toward maintenance of 6-minute walk distance in the ACE-031 groups against a decline in the placebo group, which was not statistically significant, plus trends toward increased lean body mass and bone mineral density and reduced fat mass. The authors concluded that non-muscle-related adverse events contributed to the decision to discontinue the study, while describing myostatin inhibition as a promising approach for the disease (PMID: 27462804).

The posted registry results give the counts behind that decision, over 24 weeks (12 weeks of treatment and 12 weeks of follow-up). Epistaxis was recorded in 1 of 9 on the lower dose, 5 of 9 on the higher dose and 0 of 6 on placebo. Telangiectasia was recorded in 0 of 9, 5 of 9 and 0 of 6. Injection site erythema was recorded in 3 of 9, 6 of 9 and 3 of 6, and headache in 1 of 9, 3 of 9 and 1 of 6 (ClinicalTrials.gov NCT01099761, posted results).

Duchenne open-label extension: 11 boys, also stopped

An open-label extension enrolled 11 boys from the core study to follow safety, tolerability and pharmacodynamics (ClinicalTrials.gov NCT01239758). It was terminated with the reason recorded as "This study was terminated based on preliminary safety data", and no results were posted. The core trial carries the matching note "Based on preliminary safety data." Those registry entries, and the published paper, are the whole public account of why ACE-031 development in Duchenne muscular dystrophy ended.

Common marmosets: a 2026 primate study

Fifteen years after the human trials stopped, the original research group published a non-human primate study. Common marmosets were randomised to ACE-031 or vehicle for 14 weeks, with weekly body composition measurement and terminal muscle morphometry and contractile testing. There was a significant main effect of time and a time by treatment interaction for lean body mass, with the ACE-031 animals higher at euthanasia than at baseline and no such change in vehicle controls. Biceps brachii showed significant increases in cross-sectional area of both type I and type II fibres, and extensor digitorum longus muscle tested outside the body showed increases in both absolute and specific force production (PMID: 41686840). Specific force is force corrected for muscle size, so this primate result differs from the earlier mdx mouse finding where normalised force did not improve.

Mice

The mouse work predates the human trials and is consistent. In normal C57BL/6 mice, 28 days of ACE-031 raised body weight 16% and the wet weights of four hind limb muscles by 26% to 46%, with fibre cross-sectional area up in both fibre types and no shift in soleus fibre-type distribution (PMID: 20466801). In castrated and sham-operated mice, an ActRIIB-mFc version raised lean mass 23% and 26%, cut adiposity 30% and 36%, prevented fatty liver in the castrated animals and prevented bone loss (PMID: 20573726). In dystrophic mdx mice, a murine-Fc version raised body mass 41% and forward pulling tension 42.5% without changing force normalised to body mass (PMID: 21462203). In streptozotocin-diabetic mice the same class of molecule increased muscle mass but worsened hyperglycaemia and raised corticosterone (PMID: 25561902). In chemotherapy-induced cachexia, metabolomic profiling found the Fc fusion partially rescued the glucose and lipid disturbances caused by Folfiri (PMID: 31438622).

Rats, as a tool rather than a treatment

One anatomy group at Dalian Medical University has used ACE-031 injections into rat suboccipital muscles in two studies, purely to create a model of local muscle hyperplasia (PMID: 36054096, PMID: 37919345). The pressure and fluid findings are about neck anatomy, not about ACE-031; the hyperplasia model is the only part that bears on it.

What is sold as ACE-031 is usually not ACE-031

A 2025 anti-doping laboratory study bought and analysed 14 black-market products labelled as ACE-031. Only 12 contained a protein that reacted to an ACVR2B antibody, and mass spectrometry and immunoblotting showed those 12 contained full-length human activin receptor IIB rather than the Fc fusion protein. Treatment with IdeS protease, which cleaves IgG, failed to cut the products, confirming the absence of an Fc fusion. Each product also contained many proteins besides the main one, which ran at about 58.4 kDa. The study notes that ACE-031 has not been pharmaceutically approved and is sold as a research chemical, and that in rats given 10 mg/kg the material was detectable in serum for up to 48 hours (PMID: 40312924).

What the evidence does not contain

There is no completed phase 3 trial of ACE-031, no approval anywhere, and no published long-term human data. The only published human efficacy signals are a 3.3% lean mass change at day 29 after one dose in 48 healthy postmenopausal women, and non-significant trends in 18 treated boys with Duchenne muscular dystrophy. The highest phase reached is phase 2, which is also how ChEMBL records ramatercept (ChEMBL CHEMBL2108674). No study has tested ACE-031 in healthy men, in athletes, or in older adults with sarcopenia.

Human safety data for ACE-031 comes from 48 healthy postmenopausal women given one dose and 18 boys with Duchenne muscular dystrophy given repeat doses. That is the entire published human record.

The findings that stopped the programme. The phase 2 Duchenne trial was halted after the second dosing regimen because of potential safety concerns, named in the publication as epistaxis and telangiectasias (PMID: 27462804). The registry's posted results give the counts over 24 weeks: epistaxis in 1 of 9 boys on 0.5 mg/kg every four weeks, 5 of 9 on 1.0 mg/kg every two weeks and 0 of 6 on placebo; telangiectasia in 0 of 9, 5 of 9 and 0 of 6 (ClinicalTrials.gov NCT01099761). Both trial records carry termination notes referring to preliminary safety data (ClinicalTrials.gov NCT01099761, NCT01239758).

No serious or severe adverse events were reported. The Duchenne publication states that ACE-031 was not associated with serious or severe adverse events, and that non-muscle-related adverse events contributed to the decision to discontinue (PMID: 27462804). The problem was not acute danger in those 18 boys; it was a vascular signal in a drug intended for long-term use in children.

Injection site reactions. Injection site erythema appears in both human studies: among the adverse events in the single-dose study (PMID: 23169607), and in 3 of 9, 6 of 9 and 3 of 6 participants across the two dose groups and placebo in the Duchenne trial (ClinicalTrials.gov NCT01099761).

Headache. Recorded in 1 of 9, 3 of 9 and 1 of 6 in the same posted results (ClinicalTrials.gov NCT01099761).

An animal glucose signal. In streptozotocin-diabetic mice, a soluble myostatin and activin receptor Fc fusion increased muscle mass but reproducibly worsened hyperglycaemia within about a week and raised serum corticosterone. The authors concluded such inhibitors "may be counterproductive" in type 1 diabetes (PMID: 25561902). This is mouse data and has never been tested in people with diabetes.

Mass without strength, in one model. In mdx mice, the construct raised body mass and absolute pulling force but left force normalised to body mass unchanged (PMID: 21462203). A later marmoset study did report increased specific force (PMID: 41686840). The two results are in different species and different muscles.

Product identity. Material sold under this name has been shown by an anti-doping laboratory to contain something else: of 14 black-market products labelled ACE-031, 2 contained no ACVR2B-reactive protein at all and the other 12 contained full-length human activin receptor IIB rather than the Fc fusion, alongside many additional proteins (PMID: 40312924). Anyone describing effects or side effects from a purchased vial is probably not describing ACE-031.

Not characterised at all. No published human data exist on immunogenicity over repeated dosing, on effects after more than 12 weeks of treatment, on bone or vascular outcomes beyond the short trials, or in any adult population other than postmenopausal women.