Skip to main content

ACE-031 (Ramatercept): Human Trials, Safety and Product Identity

By the Peptide Dossier Editorial TeamUpdated How we source

Published

Direct answer: ACE-031, also called ramatercept, is an experimental activin-receptor fusion protein studied for muscle-wasting disease. A small human study found increases in lean mass and thigh-muscle volume, but the subsequent Duchenne muscular dystrophy trial stopped over safety concerns. Those findings do not establish a safe bodybuilding treatment or a proven improvement in everyday function. Its molecular identity also matters: products sold under the same name have failed analytical identity testing. [1–4,8]

Key findings

  • Identity: ramatercept is an ActRIIB–IgG1 Fc fusion protein, not a short peptide or a growth-hormone secretagogue. [1]
  • Human evidence: a single-exposure study measured body-composition changes; the small repeated-treatment Duchenne trial did not establish a statistically significant walking benefit. [2,3]
  • Safety: nosebleeds and telangiectasias were important enough to stop the Duchenne study, even though no serious adverse events were reported in its registry results. [3,4]
  • Newer research: the February 2026 marmoset publication is animal evidence. It does not announce a restarted human program. [9]
  • Product identity: a 2025 investigation found no authentic ACE-031 among its 14 tested internet products. That sample does not establish a market-wide counterfeit rate. [8]

The useful question is not simply whether blocking myostatin can enlarge muscle. It is whether a specific, verified product improves a meaningful outcome in a particular population with an acceptable risk. This guide keeps those questions separate.

What ACE-031 is and how the names relate

The FDA substance-identity record lists ramatercept, ACE-031 and UNII 42HQC6QLEK for the same substance. Its structure combines an extracellular portion of human activin receptor type IIB with the Fc portion of human immunoglobulin G1. Two linked chains form the fusion protein. A substance identifier describes identity; the FDA record explicitly says that having a UNII does not imply regulatory review or approval. [1]

“ActRIIB,” “ACVR2B” and “activin receptor type IIB” refer to the receptor terminology encountered in these papers. They should not be used as automatic synonyms for the finished ACE-031 molecule. A full-length receptor without the Fc component is a different material. That distinction became central to the product-testing study discussed below. [1,8]

RAP-031 is also not an interchangeable spelling. The 2026 paper describes it as a murine analog using a mouse IgG2 Fc component. Likewise, follistatin preparations, receptor-blocking antibodies and other engineered ligand traps should retain their own names and evidence. [9] Our follistatin guide explains why protein isoforms, gene transfer and engineered proteins cannot be collapsed into one injectable-product claim.

Mechanism: a ligand trap rather than a GH secretagogue

ACE-031 acts as a soluble decoy receptor. It binds signaling proteins that would otherwise interact with the activin receptor pathway, including myostatin. The intent is to reduce some of the signals that restrain skeletal-muscle growth. This is a broader approach than selectively targeting myostatin alone. [2,10]

The 2010 original mouse experiment reported growth in muscles rich in both slow and fast fibers after four weeks of treatment. That finding helped establish a biological rationale for development; it was not a trial of strength training, injury rehabilitation or human performance. Its short duration and young animals limit direct translation. [10]

Mechanistic breadth is not automatically an advantage. A pathway with several ligands creates several possible effects, and a convincing muscle-growth mechanism cannot settle clinical safety. The human trial history is therefore essential when interpreting the mechanism. Do not infer that a growth-hormone release test, a myostatin measurement or an animal muscle photograph predicts the benefit of ACE-031 in a person. See the muscle-growth evidence overview for the distinction between hormonal responses, body composition and functional outcomes.

Human evidence: what was actually measured

The table separates published findings from registry records. “Completed” describes a study's recorded status, not whether a treatment succeeded. A registry without posted results should not be treated as a negative trial, a positive trial or proof that no publication exists.

Human study or recordDesign and populationFindings available in checked sourcesMain interpretation limit
Attie et al., 2013; NCT00755638 [2,5]Randomized, blinded, placebo-controlled single ascending-dose study; 48 healthy postmenopausal womenHighest-dose group reported mean day-29 changes of +3.3% total lean mass by DXA and +5.1% thigh-muscle volume by MRI, both p=.03Small, short, single-exposure study; the abstract's percentages are not presented here as placebo-corrected effects or proof of better function
Campbell et al., 2017; NCT01099761 [3,4]Randomized, blinded, placebo-controlled repeated-treatment study; 24 ambulatory boys with Duchenne muscular dystrophySafety concerns stopped the study after the second regimen; walking-distance findings were a non-significant trendEarly termination and small groups prevent a reliable efficacy conclusion or generalization to healthy adults
NCT00952887 [6]Registered repeated-dose study in postmenopausal women; 70 actual participants; completedRetrieved registry record has no posted results tableEnrollment and completion alone do not establish benefit or long-term safety
NCT01239758 [7]Open-label Duchenne extension; 11 actual participants; terminatedRegistry cites preliminary safety data; no posted results table in retrieved recordExtension participants can overlap with the parent study; do not add them as an independent population

Why the first human study is interesting but insufficient

The Attie study provides a human signal in measured body composition. DXA lean mass and MRI thigh volume answer different measurement questions; neither is interchangeable with a timed walking test or a patient-reported improvement. Its single-dose design also cannot validate repeated administration. [2]

For interpretation, ask three separate questions: Was the measurement changed? Did patients do something clinically important better? Was exposure sufficiently studied to characterize harm? A positive answer to the first does not supply missing answers to the other two. It is especially misleading to turn a result from healthy postmenopausal women into a predicted percentage gain for athletes, younger adults or people with a different disease.

Why the Duchenne result is not proof of functional benefit

The Campbell publication describes a trend toward maintenance of six-minute walking distance with ACE-031 compared with a decline on placebo, but the comparison was not statistically significant. It also describes favorable trends in body composition and bone density. These are hypotheses for further research, not an established treatment effect. [3]

Early stopping makes a simple “worked” or “failed to build muscle” label inadequate. The decisive clinical issue was that the study did not deliver a sufficient benefit-and-safety basis to continue this program as tested. A larger or different future study would need its own evidence; it cannot be supplied retrospectively by animal findings or optimistic summaries.

Safety: what stopped the Duchenne trial

The official parent registry is terminated, with preliminary safety data recorded as the reason. Its posted adverse-event results provide the denominators below. The planned third active regimen enrolled nobody and is omitted; the two enrolled active cohorts each had nine participants, with six receiving placebo. [4]

Reported eventActive cohort 1, n=9Active cohort 2, n=9Placebo, n=6
Epistaxis: nosebleeds150
Telangiectasia: small dilated surface blood vessels050
Serious adverse events000

These are affected-participant counts, not numbers of episodes. People may appear in more than one row, so the rows cannot be added to calculate the number harmed. Tiny cohorts do not provide stable population risk estimates. The registry's adverse-event observation period was 24 weeks. [4]

“No serious adverse events” is a specific reporting category. It is not equivalent to “no important safety problem.” The publication identifies epistaxis and telangiectasias as the potential safety concerns prompting discontinuation. [3] A short reassuring exposure in another population cannot overrule that signal.

The open-label extension was also terminated on preliminary safety data. Its 11 participants should not be counted as 11 independent confirmations of safety or added to the parent's 24 to produce an inflated sample size. [7] These records do not establish a protective cycle, monitoring schedule or human-use protocol. The broader peptide safety guide explains why risk depends on the exact preparation, exposure and population.

What the 2026 marmoset paper adds

Cadena and colleagues published a 14-week study on February 13, 2026, with eight treated common marmosets and four controls. It reported changes in lean-mass trajectory, larger biceps muscle fibers and increased specific force in isolated extensor digitorum longus muscle. [9]

The results section is more qualified than a headline about “mass and strength” suggests. The treatment-by-time interaction for lean mass was significant, while the overall treatment main effect was not. Absolute twitch and tetanic force were not significantly different in the specified comparisons; specific force was higher. An unusual control animal contributed variability. [9]

These are animal and isolated-muscle findings, not demonstrated improvements in human walking, lifting or daily life. The study's small groups and use of different muscles for structural and functional analyses constrain interpretation. Its publication date is not evidence of a new human trial, resumed development or regulatory approval. The paper also discloses former developer-company employment among coauthors. [9]

The product-identity problem: 14 tested internet products

Reichel and colleagues' 2025 analytical investigation examined 14 products sold as ACE-031. None contained the expected receptor–Fc fusion protein. Twelve contained full-length human ACVR2B, with other proteins; the remaining two were identified as other materials. The authors used electrophoresis, immunoblotting, mass spectrometry and an Fc-cleaving protease to investigate identity. [8]

This is an identity finding in a defined sample, not a global market survey. It cannot certify an untested supplier or establish that every current product is identical to those tested. Nor can the study's rat experiments with the misidentified material be used as genuine ACE-031 efficacy or safety evidence. [8]

For a reader assessing a product claim, the editorial implication is direct: a name on a vial does not establish equivalence to the clinical trial material. A purity percentage alone does not answer whether the intended molecule is present. Our interactive COA guide separates identity, content, purity and tests that were not reported; it does not authenticate a particular vendor's product.

Approval and sport status as checked November 20, 2026

The checked record does not establish an approved ACE-031 treatment. The clinical reports and later research describe a stopped development program, and the FDA identity listing is not a product approval. An identifier, a research label and an approved medicine are different categories. [1,3,8]

FDA explains that approval examines product-specific evidence, including reliable manufacture of the expected identity, strength, quality and purity. A laboratory finding or ingredient name does not substitute for that finished-product review. This general regulatory explanation does not adjudicate every country's law or each seller's legal status. [11]

For sport governed by the World Anti-Doping Code, the 2026 Prohibited List, effective January 1, 2026, explicitly includes ACE-031 among decoy activin receptors in section S4.3. This class is prohibited at all times, in and out of competition. [12] Prohibition is a sport-rule classification, not proof that a substance improves performance or has regulatory approval. Athletes should check their governing body's applicable requirements rather than infer permission from a “research use” label.

Reader questions

Is ACE-031 the same thing as ramatercept?

Yes, those names identify the same experimental fusion protein in the FDA substance record. A product using either name still needs its identity established; naming alone does not demonstrate equivalence to the research material. [1]

Did ACE-031 increase muscle in humans?

The single-exposure study reported lean-mass and thigh-volume increases. That is a narrower finding than proven functional improvement, sustained benefit or safe repeated use. The Duchenne trial's walking-distance trend was not statistically significant. [2,3]

Does the 2026 paper show that the human safety concerns were resolved?

No. It reports an animal experiment and does not replace a human safety evaluation. New publication does not erase the stopped clinical trial. A claim of resolved risk would need direct, appropriately designed human evidence.

Can a bodybuilding dose or cycle be taken from the trials?

The checked studies do not validate a bodybuilding regimen. Trial exposure in a controlled research population cannot be converted into a safe dose, schedule or supplier recommendation for a different person and product. Reconstitution instructions would not resolve the underlying evidence or identity gaps.

What evidence would change this assessment?

Useful new evidence would identify the exact product and population, prospectively assess vascular safety, distinguish body composition from meaningful function and report denominators and follow-up. A newly registered study alone would establish a research plan; completed, reported findings would be needed to assess its results. A product-identity claim would require evidence for that actual preparation rather than borrowing a trial compound's name.

Editorial methods and references

Prepared by the Peptide Dossier Editorial Team with AI assistance under the editorial policy. Sources checked November 20, 2026. This is a targeted editorial synthesis, not a systematic review or clinician-reviewed treatment recommendation. We read the original human-study abstracts, the official trial records and posted parent-trial results, and the full text of the 2025 analytical and 2026 animal papers. Unavailable full human-study text was not represented as reviewed. There is no claim of personal treatment experience. Study dates, publication dates and this page's review date are distinct.

  1. FDA substance identity: ramatercept, UNII 42HQC6QLEK. Includes identity synonyms and the identifier-versus-approval qualification.
  2. Attie KM et al. A single ascending-dose study of muscle regulator ACE-031 in healthy volunteers. Muscle & Nerve. 2013;47:416–423. PMID 23169607; original publisher abstract.
  3. Campbell C et al. Myostatin inhibitor ACE-031 treatment of ambulatory boys with Duchenne muscular dystrophy. Muscle & Nerve. 2017;55:458–464. PMID 27462804; original publisher abstract; first online in 2016.
  4. ClinicalTrials.gov: NCT01099761, Duchenne randomized trial and posted results. Completion June 2011; record last updated October 13, 2022.
  5. ClinicalTrials.gov: NCT00755638, single-dose healthy-volunteer study. Completed; 48 actual participants.
  6. ClinicalTrials.gov: NCT00952887, repeated-dose healthy-volunteer study. Completed; 70 actual participants; retrieved record has no posted results table.
  7. ClinicalTrials.gov: NCT01239758, Duchenne open-label extension. Terminated; 11 actual participants.
  8. Reichel C et al. Gel Electrophoretic Detection of Black Market ACE-031. Drug Testing and Analysis. 2025;17:1934–1946. DOI 10.1002/dta.3898; original analytical study.
  9. Cadena SM et al. ACE-031, a soluble activin type IIB receptor, increases muscle mass and strength in the common marmoset. PLoS One. February 13, 2026;21:e0342666. Original animal study; methods and results distinctions retained.
  10. Cadena SM et al. Administration of a soluble activin type IIB receptor promotes skeletal muscle growth independent of fiber type. Journal of Applied Physiology. 2010;109:635–642. DOI 10.1152/japplphysiol.00866.2009; original mouse-study abstract.
  11. FDA: Concerns About Unapproved Drugs. General product-specific review context, not an ACE-031 approval determination.
  12. WADA: 2026 Prohibited List. Effective January 1, 2026; section S4.3.