History

Follistatin is an endogenous glycoprotein that binds and blocks myostatin (GDF-8) and activin A. The "344" refers to the 344-amino-acid splice isoform (FS344), chosen for systemic gene-transfer work over the FS288 and FS315 variants. The only meaningful human testing came from the Nationwide Children's Hospital / Mendell group, which delivered the human FS344 gene by AAV (serotype 1, construct rAAV1.CMV.huFollistatin344) via intramuscular injection in two small open-label trials — in Becker muscular dystrophy and sporadic inclusion body myositis (NCT01519349; Mendell et al., Molecular Therapy 2015) and in Duchenne muscular dystrophy (NCT02354781, n=3). The reported functional improvements, especially in the inclusion-body-myositis cohort, were credibly disputed in a peer-reviewed rebuttal (Greenberg, Molecular Therapy 2017).

Follistatin 344 (FS344) is a follistatin protein that blocks myostatin and activin — the body’s natural “brakes” on muscle growth. It is biologically real and active, but clinically unproven: its only legitimate human testing was small, early-phase gene therapy for muscular dystrophy, the headline benefits were challenged in the medical literature, and there is no human evidence for the muscle-building or fat-loss uses it is sold for. This page is educational and harm-reduction oriented. It contains no doses, cycles, or sourcing.

What it is

Follistatin is an endogenous secreted glycoprotein that acts as a high-affinity binding antagonist of several TGF-β superfamily ligands — most importantly myostatin (GDF-8) and activin A (and to some degree GDF-11). By binding these ligands, it stops them from engaging the activin type II receptors (ActRIIA/ActRIIB), which blocks the downstream SMAD2/3 signaling that normally restrains muscle growth. The net effect is removing a “brake” on skeletal-muscle hypertrophy rather than directly stimulating anabolism.

The “344” denotes a specific isoform — the 344-amino-acid splice variant (FS344), as opposed to the shorter FS288 and longer FS315 isoforms. FS344 is a precursor form; differing heparin/cell-surface binding affinities among the isoforms (FS288 binds cell surfaces most avidly) influenced the choice of FS344 for systemic gene-therapy work.

An important caveat: because follistatin antagonizes activin and related ligands broadly — not only myostatin — it is a relatively non-specific pathway inhibitor. That matters for interpreting both how well it works and how safe it is.

It is also worth being clear about modality. The legitimate human work used follistatin as an AAV gene therapy — DNA delivering the FS344 gene by intramuscular injection — not as an injectable peptide. The “Follistatin 344 peptide” sold by research-chemical and gray-market vendors is research-grade material with no clinical development behind it as an injectable protein, and is not the same thing as the trial gene therapy.

The claims

Gray-market sellers market injectable “Follistatin 344” for muscle growth, fat loss, faster recovery, and even hair regrowth. The dramatic imagery often invoked — large lean-mass gains, “double-muscled” physiques — comes from animal and preclinical work on myostatin/follistatin manipulation in rodents and non-human primates, not from humans. None of these vendor claims are supported by controlled human data.

What the evidence actually shows

Two small open-label gene-therapy trials are the only meaningful human data, both using the construct rAAV1.CMV.huFollistatin344 (reported in the muscular-dystrophy/myositis paper as AAV1.CMV.FS344 — the same agent).

Becker muscular dystrophy + sporadic inclusion body myositis (NCT01519349, Phase 1). The Becker cohort — 6 ambulatory male patients — was reported in Mendell JR et al., Molecular Therapy 2015;23(1):192–201 (PMID 25322757). Six-minute-walk-test changes were heterogeneous: roughly +58 m and +125 m at lower dose with one patient unchanged, and +108 m and +29 m at higher dose with one patient not improved. Post-treatment biopsies described reduced fibrosis and muscle hypertrophy, especially at the higher dose, and the authors reported no adverse effects.

The single most important honesty point: these are not controlled-trial results. A published methodological rebuttal — Greenberg SA, Molecular Therapy 2017 (PMID 28927986) — argued that the uncontrolled, open-label design (with concurrent prednisone and exercise and no comparator), a changed/post-hoc “annualized” walk-test outcome versus the registered safety primary endpoint, and a misstated adverse-event (falls) frequency mean the data cannot support efficacy claims, particularly for the inclusion-body-myositis portion.

Duchenne muscular dystrophy (NCT02354781, Phase 1/2). This trial enrolled 3 patients, with safety as the primary endpoint (dose-limiting toxicity through 2 years). One serious adverse event was a fall-related head injury. No peer-reviewed efficacy readout was located; treat it as inconclusive.

Weight loss and metabolic use: no human evidence at all. There are no completed human trials of follistatin (any isoform) for obesity, fat loss, or metabolic disease. Effects on body composition in animals are an extrapolation, not a human finding.

So the picture is a genuine biological mechanism with a tiny, disputed early-phase human signal in muscle disease — and nothing supporting the fitness uses it is sold for.

Legal and regulatory status

Follistatin is not approved by any regulator (FDA or EMA) for any indication — not for muscle, weight loss, or metabolic disease. There is no approved follistatin drug. The legitimate human work was an investigational gene-transfer biologic that reached only early-phase testing.

In sport, follistatin is prohibited by WADA at all times under the 2026 Prohibited List, category S4 (Hormone and Metabolic Modulators), subsection S4.3 — “Agents preventing activin receptor IIB activation.” This subsection explicitly covers myostatin-binding proteins (e.g., follistatin, myostatin propeptide), decoy/soluble activin receptors and anti-ActRIIB antibodies (e.g., ACE-031, bimagrumab), and myostatin-neutralizing antibodies (e.g., apitegromab, domagrozumab). So follistatin sits in the same activin-receptor-IIB category as ACE-031. (The subsection numbering has shifted across editions — it was S4.4 in some earlier lists — but the heading wording and follistatin’s inclusion are stable.) For contrast, IGF-1 and mechano growth factors sit under a different category, S2 (peptide hormones, growth factors, and mimetics).

Safety

Human safety data are extremely thin — derived from roughly a dozen patients total across two tiny, uncontrolled gene-therapy studies. The “no adverse effects” in the Becker paper reflects a handful of subjects and limited follow-up, not a generalizable safety profile.

Mechanistic concerns are unresolved in humans. Because follistatin/activin signaling is pleiotropic, broad antagonism raises plausible concerns around reproductive and endocrine effects (activin biology), and the myostatin/activin–ActRIIB axis has cardiovascular roles. The most-cited real-world signal for this broader pathway comes from a related drug, not follistatin itself: the ACE-031 program (a soluble/decoy ActRIIB receptor) was halted by its data safety monitoring board and discontinued (Acceleron, around 2013) after non-muscle adverse effects — notably epistaxis (nosebleeds) and telangiectasias (cutaneous vascular lesions) — attributed to off-target inhibition of BMP9/BMP10 vascular signaling. This is well-established class context for the activin-receptor pathway, not a follistatin-specific finding, but it illustrates why broad pathway inhibition warrants caution.

Gray-market injectable “Follistatin 344” peptide carries the usual research-chemical risks — no GMP manufacturing, unknown purity and identity, endotoxin and sterility risk — layered on top of an unproven mechanism. There is no established safe use and no dosing guidance exists or should be inferred. None of this is medical advice.

Bottom line

Follistatin 344 is a myostatin/activin-pathway antagonist that is scientifically real and biologically active, but clinically unproven. Its only legitimate human testing was small early-phase AAV gene therapy in muscular dystrophy, the headline functional claims — especially in inclusion body myositis — were credibly disputed in the peer-reviewed literature, and there is no human evidence for weight loss or metabolic use. It is not approved for anything and is WADA-prohibited at all times under S4.3. The marketing runs far ahead of the evidence.

Evidence grade: Preliminary human.