ACE-031

A recombinant activin receptor type IIB-Fc fusion protein that traps myostatin and related TGF-beta ligands to increase muscle mass; its clinical program went on hold in 2011 after vascular adverse events and the sponsor discontinued development in 2013.

Also referenced as: Ramatercept, ACE031, ActRIIB-IgG1, ActRIIB-IgG1Fc

Also appears in: Other

What is ACE-031?

ACE-031, international nonproprietary name ramatercept, is a recombinant fusion protein: the extracellular ligand-binding domain of the human activin receptor type IIB (ActRIIB, gene ACVR2B) joined to the Fc region of human IgG1. Each chain is 343 amino acids and carries three N-glycosylation sites, and two chains are disulfide-linked into a homodimer; glycosylated and assembled, the molecule is on the order of 100 kDa (Reichel et al., Drug Testing and Analysis, 2025;17(10):1934–46). It is therefore not a peptide — it is a glycosylated antibody-Fc biologic roughly two orders of magnitude larger than the short synthetic peptides it is catalogued beside, and correct production requires mammalian cell culture rather than solid-phase synthesis.

It was developed by Acceleron Pharma, licensed outside North America to Shire AG under an agreement dated September 8, 2010, and tested as a muscle-building agent in healthy volunteers and in boys with Duchenne muscular dystrophy (DMD). It appears in research-market catalogs today because of that muscle-growth rationale, not because any version of it is available as a medicine: the clinical program went on hold in February 2011 and the sponsor discontinued it in 2013. Its use in sport is prohibited: Reichel et al. (2025) place it under chapter S4, “Hormone and metabolic modulators,” sub-chapter S4.3, of the World Anti-Doping Agency Prohibited List, and the same group developed the anti-doping detection methods described below.

How it works

ActRIIB is the type II receptor through which myostatin and several related TGF-beta superfamily ligands suppress muscle growth. A soluble receptor-Fc fusion acts as a ligand trap: it binds those ligands in circulation so they cannot engage receptors on muscle.

• More than myostatin blockade alone. A soluble form of the activin type IIB receptor increased muscle mass by up to 60% in two weeks in wild-type mice, and the effect was attenuated but not eliminated in myostatin-null mice — evidence that at least one ligand besides myostatin normally limits muscle growth through activin type II receptors. That study identified the additional signaling as running through both ActRIIA and ActRIIB but did not establish which specific ligand accounts for the residual effect (Lee et al., Proceedings of the National Academy of Sciences, 2005;102(50):18117–22)
• Hypertrophy across fiber types. In 8-week-old C57BL/6 mice, 28 days of ACE-031 raised mean body weight 16% above controls and soleus, plantaris, gastrocnemius and extensor digitorum longus wet weights by 33%, 44%, 46% and 26%; soleus fiber-type distribution was unchanged, distinguishing it from selective pharmacological myostatin inhibition, which predominantly targets type II fibers (Cadena et al., Journal of Applied Physiology, 2010;109(3):635–42)
• Measurable effect in humans after a single dose. In healthy postmenopausal women given single subcutaneous doses of 0.02–3 mg/kg, the 3 mg/kg group showed a 3.3% increase in total body lean mass by DXA (P = 0.03) and a 5.1% increase in thigh muscle volume by MRI (P = 0.03) at day 29, with mean terminal half-life of 10–15 days, consistent with Fc-mediated recycling (Attie et al., Muscle & Nerve, 2013;47(3):416–23)
• Effects reach beyond muscle. The same Phase 1 study reported statistically significant serum biomarker changes that the authors read as altered bone and fat metabolism, and the DMD trial showed trends toward increased bone mineral density and reduced fat mass — pharmacology consistent with a trap that removes ligands used widely outside skeletal muscle (Campbell et al., Muscle & Nerve, 2017;55(4):458–64)

Research status

Human development is finished. An intervention search on ClinicalTrials.gov returns exactly four registered studies, the most recent of which started in October 2010: two completed Phase 1 studies and two terminated Phase 2 studies. Nothing is recruiting or active.

• Phase 1: NCT00755638 randomized 48 healthy postmenopausal volunteers to a single dose of ACE-031 or placebo (3:1), ran from September 2008 to July 2009, and matches the published single ascending-dose report in design, size and population; the registry record itself posts no results and cites no publication, so that pairing is an inference from the study description rather than a registry-stated link. The publication reports the compound was generally well tolerated, with injection-site erythema among the adverse events and dose-linear exposure (Attie et al., 2013). A second, larger Phase 1 in healthy postmenopausal women (NCT00952887, 70 participants) completed in February 2011 with no results posted and no publication
• Phase 2 in DMD, stopped: NCT01099761 enrolled 24 ambulatory boys in a randomized, double-blind, placebo-controlled ascending-dose design with a safety primary objective — 18 assigned to ACE-031 across two dosed regimens, 6 to placebo, with a planned third regimen never dosed — and the open-label extension NCT01239758 enrolled 11. Both were terminated in 2011 “based on preliminary safety data.” The published report states that ACE-031 was not associated with serious or severe adverse events but that the study was stopped after the second dosing regimen over potential safety concerns of epistaxis and telangiectasias; a non-significant trend toward maintained 6-minute walk distance versus decline on placebo was seen, along with trends for increased lean mass and bone mineral density and reduced fat mass (Campbell et al., 2017). The results posted on the registry record classify both events as vascular disorders: epistaxis in 1 of 9 and 5 of 9 participants in the two dosed ACE-031 groups versus 0 of 6 on placebo, and telangiectasia in 5 of 9 in the higher-exposure group versus 0 of 6 on placebo
• Regulatory and corporate record: the FDA placed ACE-031 on clinical hold on February 8, 2011. In April 2013 Acceleron and Shire determined not to further advance development, Shire terminated the collaboration effective June 30, 2013, all rights reverted to Acceleron, and the company stated in its FY2013 Form 10-K that it had “no plans to continue the development of ACE-031,” having concluded from further animal toxicology that the risk-benefit profile was not appropriate for the intended patient population. No ACE-031 study has been registered in the fifteen years since; the registry now lists the sponsor as Acceleron Pharma, a wholly owned subsidiary of Merck & Co., with no new registration under that ownership
• Preclinical publishing continued after the program closed: a common marmoset study reporting increased muscle mass and strength appeared only in 2026, fifteen years after human dosing stopped (Cadena et al., PLOS One, 2026;21(2):e0342666). The broader receptor-Fc class did not end with ACE-031 — luspatercept, a differently engineered ACVR2B-Fc fusion protein approved for anemia, and sotatercept, an ActRIIA-Fc fusion protein approved for pulmonary arterial hypertension, both came out of the same company’s pipeline — but no approved product contains ACE-031 itself
• Product-identity evidence: 14 products sold under the ACE-031 name were analyzed by SDS-PAGE, Western blotting, mass spectrometry and IdeS protease digestion. Twelve contained an ACVR2B-immunoreactive protein, and in all twelve that protein was full-length human activin receptor IIB rather than the Fc-fusion protein — IdeS, which cleaves IgG below the hinge, could not cut them, and the observed mass matched non-glycosylated ACVR2B, which the authors read as bacterial rather than mammalian expression. Those twelve also contained many additional proteins. Of the remaining two, one held proteins with a different mass profile and one contained no protein at all, with a growth hormone secretagogue identified instead (Reichel et al., 2025)

Common dosage forms

Key considerations

• Status basis. ACE-031 is classified Research Market because no regulator in any jurisdiction has approved it and no sponsor has developed it since the February 2011 clinical hold and the April 2013 decision to stop, with no study registered in the fifteen years since.
• Development was discontinued, not paused. A clinical hold in February 2011, two terminated Phase 2 studies, and a sponsor statement of no plans to continue are the whole regulatory record. Material sold under the name is not a withdrawn drug product being resold; it is manufactured independently of the original program
• The signal that stopped it was vascular, not muscular. Epistaxis and telangiectasias — not muscle toxicity — ended the DMD program, and the posted trial results file both under vascular disorders. ActRIIB traps bind a family of ligands with roles well outside skeletal muscle, but the published report attributes the decision to discontinue to these non-muscle adverse events without establishing a mechanism; proposals that off-target binding of vascular BMP ligands explains the bleeding remain hypotheses, not trial findings. The DMD cohort was 24 boys with the extension at 11, so safety characterization is thin by any standard
• A ~100 kDa recombinant Fc-fusion protein is unlikely to be what research-market vials labeled “ACE-031” contain. The one published analysis of market products found that 12 of 14 contained full-length activin receptor IIB without an Fc fusion — a different molecule with different pharmacokinetics — alongside substantial additional protein, one contained unrelated proteins, and one contained no protein at all. Unlike a short synthetic peptide, correct production requires mammalian expression, Fc-dimer assembly, and glycosylation, and vial contents cannot be judged from appearance or from a purity figure alone. This is a statement about what that published analysis measured, not about any particular seller
• Evidence gaps that matter: every human efficacy observation is a surrogate — DXA lean mass, MRI thigh volume, serum biomarkers, a non-significant 6-minute-walk trend. No trial measured strength or function against a prespecified efficacy endpoint, the larger of the two Phase 1 studies was never published, and pediatric exposure ended after the second dosing regimen
• Naming confusions: ACE-031 and ramatercept are the same molecule, and “ActRIIB-Fc” or “ACVR2B-Fc” names a class rather than this construct. Luspatercept, developed as ACE-536, is a different engineered ActRIIB-Fc protein approved for anemia; sotatercept is built on ActRIIA and is approved for pulmonary arterial hypertension; both have ligand profiles and indications distinct from ACE-031. ACE-031 is also commonly grouped with follistatin 344 and myostatin-targeting antibodies under “myostatin inhibitors,” though the mechanism, molecular class, and evidence base differ in each case