Peptides & growth hormone

Follistatin-344

Updated · 21 sources · Editorial policy

Follistatin-344 is not a drug in the usual sense. The name refers to one of the two forms of the human follistatin precursor protein: the 344-amino-acid version, often shortened to FS344 or FST344 (Shimasaki et al., PNAS 1988; UniProt P19883). Follistatin itself is a natural binding protein that neutralizes activin A, myostatin and GDF-11, three members of the TGF-β family that restrain muscle growth.

Everything that has been tested in people under this name has been a gene transfer, not an injection of the protein. Six patients with Becker muscular dystrophy and six with sporadic inclusion body myositis received the FS344 gene packaged in an adeno-associated virus (AAV1) injected into the quadriceps; three boys with Duchenne muscular dystrophy received a wider-distribution version of the same vector (Mendell et al., Mol Ther 2015; Mendell et al., Mol Ther 2017; NCT02354781 results). All of these were open-label phase 1 studies with no placebo group, and their efficacy claims have been publicly disputed.

No follistatin product is approved by any regulator, there is no published human pharmacokinetic study of injected follistatin protein, and no placebo-controlled trial of follistatin itself has been run in people. Follistatin is explicitly named on the WADA Prohibited List as a myostatin-binding protein, prohibited at all times.

This page summarizes what the published trials, trial registries and regulators report. It is for education only and is not medical advice.

Property Detail
What it is Human follistatin, 344-amino-acid precursor isoform (FS344/FST344); a secreted activin- and myostatin-binding glycoprotein
Drug class Myostatin/activin pathway inhibitor (binding protein)
Approval status Not approved by FDA, EMA or any other authority; no marketed product
Human exposure studied AAV1 or plasmid gene transfer only — no trial has injected the protein itself
Studied gene-transfer doses 3 × 10¹¹ and 6 × 10¹¹ vg/kg per leg (Becker, inclusion body myositis); 2.4 × 10¹² vg/kg total (Duchenne)
Half-life Not established in humans; native follistatin-315 is cleared too fast to work as an injected systemic drug in mice
Key mechanism fact Binds activin A, myostatin and GDF-11; blocking this pathway increased muscle mass in transgenic mice as much as deleting myostatin
US controlled-substance status Not listed in the anabolic steroid schedule (21 CFR 1308.13)
WADA status Prohibited at all times, class S4.3 (myostatin-binding proteins); gene transfer also falls under M3.1

What it is

Follistatin was isolated in 1987 from pig ovarian follicular fluid as a single-chain protein of about 35,000 daltons that suppressed the release of follicle-stimulating hormone (FSH) from cultured rat pituitary cells, with a half-maximal effect at 2.5–6.0 ng/mL. The discoverers named it follistatin to distinguish it from inhibin (Ueno et al., PNAS 1987).

When the human gene was cloned in 1988, two precursor proteins turned up from alternative splicing of the same messenger RNA: one of 344 amino acids and one of 317 (Shimasaki et al., PNAS 1988). Each loses a 29-amino-acid signal peptide when it is secreted, which produces the two mature forms. The protein is built from three repeated follistatin/Kazal domain pairs (UniProt P19883).

Form Precursor Mature protein Behavior
FS344 344 amino acids Follistatin-315 (FS315) Circulates in serum; roughly 10-fold lower affinity for activin than FS288; the form chosen for muscle gene therapy
FS317 317 amino acids Follistatin-288 (FS288) Binds cell-surface heparan sulfate; the form most tied to activin–inhibin signaling in the pituitary and reproductive tissue

The reason gene therapy trials used FS344 rather than FS317 is to limit effects on the pituitary activin–inhibin axis and FSH secretion, which was the original concern about using follistatin as a muscle drug (Al-Zaidy et al., J Neuromuscul Dis 2015).

Follistatin is not a muscle-specific molecule. In healthy people, plasma follistatin rose about 7-fold after three hours of cycling, peaking three hours into recovery, and the arterial–venous measurements showed it was released by the liver rather than the exercising leg (Hansen et al., Endocrinology 2011).

How it works in the body

Myostatin signals through the activin type II receptors. Purified myostatin binds ActRIIB, and that binding is blocked by follistatin (Lee and McPherron, PNAS 2001). Follistatin also antagonizes activin A and GDF-11, so it is a broader inhibitor than an antibody aimed at myostatin alone (Yaden et al., JPET 2014).

The practical problem with follistatin as an injectable drug is its pharmacokinetics. In a Lilly program that characterized native follistatin-315, the protein’s intrinsic properties were judged unsuitable for a systemically administered biologic. Fusing it to an antibody Fc fragment and removing its heparan-sulfate binding improved terminal half-life by roughly 100-fold and total exposure by roughly 1,600-fold in mice, and only the engineered version produced a dose-dependent effect in mouse muscle-atrophy models (Datta-Mannan et al., JPET 2013). No equivalent human data exist for follistatin protein in any form.

That is why every human study has delivered the gene instead: a one-time intramuscular injection of a viral vector or plasmid turns muscle fibers into a local factory that secretes the protein for months to years. In macaques, human follistatin stayed above 20 ng per milligram of quadriceps tissue 5 and 15 months after a single AAV1-FS344 injection (Kota et al., Sci Transl Med 2009).

The half-life of 24 hours listed in some compound databases for “follistatin-344” does not correspond to any published human measurement.

Regulatory status

  • No approval anywhere. There is no follistatin product in Drugs@FDA or the openFDA drug label database, and no approved indication (Drugs@FDA).
  • Not a controlled substance in the US. Follistatin is a protein, not an anabolic steroid, and it does not appear in the Schedule III anabolic steroid list (21 CFR 1308.13).
  • Banned in sport at all times. The WADA list names follistatin as an example of a myostatin-binding protein under S4.3, “Agents preventing activin receptor IIB activation.” Substances in S4.3 are non-Specified. Gene transfer of follistatin is separately covered by M3.1, gene and cell doping (WADA 2026 Prohibited List).

What the research shows

Animal work: large effects on muscle mass

Mice carrying a muscle-specific follistatin transgene showed dramatic increases in muscle mass, comparable to mice with the myostatin gene deleted (Lee and McPherron, PNAS 2001). Crossing that transgene into myostatin-null mice pushed muscle mass to about four times wild-type levels, which showed that follistatin blocks more than myostatin alone (Lee, PLoS One 2007).

In cynomolgus macaques, a single AAV1-FS344 injection into one quadriceps produced the results below (Kota et al., Sci Transl Med 2009):

Measure (treated vs untreated) Result
Quadriceps circumference, CMV promoter (n = 3) +15% over baseline at 8 weeks (p = 0.01), stable to week 60
Quadriceps circumference, muscle-specific promoter (n = 3) +10% (p = 0.02), delayed and less durable
Mean muscle fiber diameter at 20 weeks 87.7 µm vs 65.5 µm in controls (p = 0.001)
Twitch/tetanic force, treated vs contralateral leg +12–36% twitch, +12–78% tetanic in 3 evaluable animals
FSH, LH, estradiol, testosterone; liver, kidney, heart No abnormal changes over 15 months

These are animal results. Note also that the macaque paper carries a 2026 erratum (Sci Transl Med 2026).

Becker muscular dystrophy (phase 1/2a, n = 6)

Six ambulatory men with Becker muscular dystrophy received AAV1.CMV.FS344 by direct bilateral quadriceps injection, three at 3 × 10¹¹ vg/kg per leg and three at 6 × 10¹¹ vg/kg per leg. All were given prednisone from one month before injection, tapered off by day 60, as protection against an immune response to the viral capsid (Mendell et al., Mol Ther 2015).

Dose Change in 6-minute walk distance
3 × 10¹¹ vg/kg per leg (12 months) +58 m, +125 m, +9 m
6 × 10¹¹ vg/kg per leg (6 months) −14 m, +108 m, +29 m

Quadriceps strength did not consistently increase in either cohort. Biopsies at 6 months showed less endomysial fibrosis, fewer centrally located nuclei and a more normal fiber-size distribution. No adverse events were attributed to gene transfer, and FSH, LH, estradiol and testosterone stayed normal throughout.

Sporadic inclusion body myositis (n = 6)

Six patients received rAAV1.CMV.huFS344 at 6 × 10¹¹ vg/kg per leg, again with prednisone and, this time, a prescribed exercise program (a stationary bicycle for 15 minutes plus leg lifts with a 5-pound ankle weight, three times a week). Annualized to one year, the treated patients gained +56.0 m on the 6-minute walk test while eight untreated patients from the same clinic, matched for age, sex and baseline walking distance, lost −25.8 m (p = 0.01). Four of six treated patients improved by 58–153 m; two improved by 5–23 m. No treatment-related adverse events were reported (Mendell et al., Mol Ther 2017).

This result is contested. A detailed critique in the same journal argued that the trial’s design cannot support an efficacy claim: it was open-label with no randomized comparator, patients received four potentially active interventions at once (gene transfer, high-dose prednisone, exercise and participation effects), the registered primary endpoint was safety rather than walking distance, the “annualized” walk-test measure was created after the fact, and adverse-event percentages in the paper were calculated per event rather than per patient — a fall rate reported as 6% was 2 of 6 patients, or 33% (Greenberg, Mol Ther 2017).

Duchenne muscular dystrophy (n = 3)

A separate phase 1/2 study injected a total of 2.4 × 10¹² vg/kg into gluteal, quadriceps and tibialis anterior muscles of three boys with Duchenne muscular dystrophy. The registry results report no dose-limiting toxicities, one serious adverse event (a head injury from a fall), and on the efficacy side 0 of 3 participants with an increase in 6-minute walk distance and 1 of 3 with an improved North Star Ambulatory Assessment at two years (NCT02354781).

Follistatin-based drug candidates

The most rigorously tested molecule in this family is not follistatin-344 but ACE-083, an engineered follistatin-based protein designed to stay in the muscle it is injected into. Its record is informative because it was placebo-controlled:

Study Design Muscle volume Function
Healthy postmenopausal women, phase 1 (Glasser et al., Muscle Nerve 2018) Randomized, double-blind, placebo-controlled; 58 women, 50–200 mg +14.5% rectus femoris, +8.9% tibialis anterior No significant change in mean strength
Facioscapulohumeral dystrophy, phase 2 (Statland et al., Muscle Nerve 2022) Randomized, double-blind, placebo-controlled part; 55 evaluable +16.4% biceps (90% CI 9.8–23.0, p under 0.0001); +9.5% tibialis anterior (90% CI 3.2–15.9, p = 0.01) No consistent improvement in functional or patient-reported outcomes

In other words, when a follistatin-based agent was tested against placebo, it reliably made the injected muscle bigger and did not reliably make patients stronger or more functional. The most common adverse events were mild-to-moderate injection-site reactions.

What has not been studied

  • No trial registered on ClinicalTrials.gov has administered follistatin protein to humans by injection; the human interventions on record are AAV or plasmid gene transfer (ClinicalTrials.gov search).
  • Two plasmid gene-therapy studies in healthy volunteers (43 and 14 participants) are listed as completed but have no posted results or publications (NCT06411366; NCT07285629), and an AAV9-follistatin study in age-related muscle loss is recruiting (NCT07443826).
  • There is no human study of follistatin in healthy athletes, no dose-response data, and no measurement of body composition under placebo control.

Side effects and risks

The honest summary

For products labeled “follistatin-344,” the risk profile is unknown. Nothing has been injected into a human in a registered trial, so there is no adverse-event table to report, no exposure data, and no way to anticipate what a given amount of protein does.

What the gene-transfer trials observed

Across the published AAV1-FS344 reports (6 patients with Becker dystrophy, 6 with inclusion body myositis, 3 with Duchenne dystrophy), investigators reported no adverse events attributed to gene transfer, normal liver, kidney and bone-marrow panels, and no change in FSH, LH, estradiol or testosterone (Mendell et al., Mol Ther 2015; Mendell et al., Mol Ther 2017). These are small, short, uncontrolled groups of patients with muscle disease, and the published safety reporting has itself been criticized as presented in a misleading way (Greenberg, Mol Ther 2017). “No signal in 15 patients” is not the same as “safe.”

Theoretical concerns that drove the drug design

  • Pituitary and reproductive effects. Follistatin was discovered as an inhibitor of FSH release (Ueno et al., PNAS 1987). The FS344/FS315 form was selected specifically to reduce the chance of interfering with the activin–inhibin axis (Al-Zaidy et al., J Neuromuscul Dis 2015); the trials monitored these hormones for that reason.
  • Breadth of inhibition. Because follistatin also binds activin A and GDF-11, which act in many tissues besides muscle, broad systemic exposure is not pharmacologically equivalent to blocking myostatin alone (Yaden et al., JPET 2014).
  • Immune response. T-cell responses to the AAV1 capsid and to follistatin were measurable in some gene-therapy patients, although anti-follistatin antibody titers stayed low (Mendell et al., Mol Ther 2015).

Permanence

Gene transfer is not reversible. Macaque muscle was still producing human follistatin 15 months after one injection (Kota et al., Sci Transl Med 2009). Any unwanted effect of a transferred gene cannot be stopped by discontinuing a dose.

Quality and product identity

“Follistatin-344” as a product name is a category error worth understanding: 344 describes the length of a precursor protein coded by a gene, and the number comes from gene-therapy constructs. Because no follistatin product is approved, there is no pharmacopeial monograph, no regulated potency or purity standard, and no label whose claims can be checked against a regulator’s review. Recombinant follistatin is also a glycoprotein, and the Lilly work showed that how it is glycosylated during manufacture materially changed how fast it was cleared in mice and monkeys (Datta-Mannan et al., Drug Metab Dispos 2015). Identity, potency and clearance of unapproved follistatin preparations are therefore unverifiable.

Key takeaways

  • The number refers to a gene, not a dose. FS344 is the 344-amino-acid follistatin precursor; it matures into circulating follistatin-315. The alternative 317-amino-acid precursor gives follistatin-288, which is more tied to pituitary FSH regulation.
  • Follistatin is a broad TGF-β blocker. It neutralizes activin A and GDF-11 as well as myostatin, which is why follistatin transgenic mice outgrow myostatin-knockout mice.
  • All human data come from gene transfer, not injections. Fifteen patients across published reports in Becker dystrophy, inclusion body myositis and Duchenne dystrophy received AAV1-FS344 in open-label phase 1 studies. There is no registered trial of injected follistatin protein in humans.
  • The efficacy claims are disputed and the Duchenne data were negative. The inclusion body myositis result was challenged in print for lacking a randomized comparator and for post-hoc endpoints; the Duchenne registry reports 0 of 3 participants improving on the walk test.
  • Placebo-controlled follistatin biology grew muscle without improving function. ACE-083 increased injected-muscle volume by about 9–16% versus placebo yet produced no consistent gain in strength or patient-reported outcomes.
  • No approval, unknown pharmacokinetics, banned in sport. No regulator has approved a follistatin product, no human half-life has been published, and WADA prohibits follistatin at all times under S4.3, with gene transfer covered by M3.1.

Sources

  1. Ueno N, Ling N, Ying SY, Esch F, Shimasaki S, Guillemin R. Isolation and partial characterization of follistatin: a single-chain Mr 35,000 monomeric protein that inhibits the release of follicle-stimulating hormone. Proc Natl Acad Sci U S A. 1987;84(23):8282–8286. PubMed.
  2. Shimasaki S, Koga M, Esch F, et al. Primary structure of the human follistatin precursor and its genomic organization. Proc Natl Acad Sci U S A. 1988;85(12):4218–4222. PubMed.
  3. UniProt Consortium. Follistatin (FST), human — entry P19883. UniProt.
  4. Lee SJ, McPherron AC. Regulation of myostatin activity and muscle growth. Proc Natl Acad Sci U S A. 2001;98(16):9306–9311. PubMed.
  5. Lee SJ. Quadrupling muscle mass in mice by targeting TGF-beta signaling pathways. PLoS One. 2007;2(8):e789. PubMed.
  6. Kota J, Handy CR, Haidet AM, et al. Follistatin gene delivery enhances muscle growth and strength in nonhuman primates. Sci Transl Med. 2009;1(6):6ra15. PubMed. Erratum: Sci Transl Med. 2026;18(861):eaek4223. PubMed.
  7. Mendell JR, Sahenk Z, Malik V, et al. A phase 1/2a follistatin gene therapy trial for Becker muscular dystrophy. Mol Ther. 2015;23(1):192–201. PubMed.
  8. Al-Zaidy SA, Sahenk Z, Rodino-Klapac LR, Kaspar B, Mendell JR. Follistatin gene therapy improves ambulation in Becker muscular dystrophy. J Neuromuscul Dis. 2015;2(3):185–192. PubMed.
  9. Mendell JR, Sahenk Z, Al-Zaidy S, et al. Follistatin gene therapy for sporadic inclusion body myositis improves functional outcomes. Mol Ther. 2017;25(4):870–879. PubMed.
  10. Greenberg SA. Unfounded claims of improved functional outcomes attributed to follistatin gene therapy in inclusion body myositis. Mol Ther. 2017;25(10):2235–2237. PubMed.
  11. Nationwide Children’s Hospital. Phase I/II clinical intramuscular gene transfer of rAAV1.CMV.huFollistatin344 trial to patients with Duchenne muscular dystrophy — study results. NCT02354781. ClinicalTrials.gov.
  12. Nationwide Children’s Hospital. Follistatin gene transfer to patients with Becker muscular dystrophy and sporadic inclusion body myositis. NCT01519349. ClinicalTrials.gov.
  13. Datta-Mannan A, Yaden B, Krishnan V, Jones BE, Croy JE. An engineered human follistatin variant: insights into the pharmacokinetic and pharmocodynamic relationships of a novel molecule with broad therapeutic potential. J Pharmacol Exp Ther. 2013;344(3):616–623. PubMed.
  14. Yaden BC, Croy JE, Wang Y, et al. Follistatin: a novel therapeutic for the improvement of muscle regeneration. J Pharmacol Exp Ther. 2014;349(2):355–371. PubMed.
  15. Datta-Mannan A, Huang L, Pereira J, Yaden B, Korytko A, Croy JE. Insights into the impact of heterogeneous glycosylation on the pharmacokinetic behavior of follistatin-Fc-based biotherapeutics. Drug Metab Dispos. 2015;43(12):1882–1890. PubMed.
  16. Hansen J, Brandt C, Nielsen AR, et al. Exercise induces a marked increase in plasma follistatin: evidence that follistatin is a contraction-induced hepatokine. Endocrinology. 2011;152(1):164–171. PubMed.
  17. Glasser CE, Gartner MR, Wilson D, Miller B, Sherman ML, Attie KM. Locally acting ACE-083 increases muscle volume in healthy volunteers. Muscle Nerve. 2018;57(6):921–926. PubMed.
  18. Statland JM, Campbell C, Desai U, et al. Randomized phase 2 study of ACE-083, a muscle-promoting agent, in facioscapulohumeral muscular dystrophy. Muscle Nerve. 2022;66(1):50–62. PubMed.
  19. World Anti-Doping Agency. Prohibited List 2026, class S4.3 and M3.1. WADA 2026 Prohibited List.
  20. Drug Enforcement Administration. Schedule III anabolic steroids. 21 CFR 1308.13. eCFR.
  21. Food and Drug Administration. Drugs@FDA database (no approved follistatin product). FDA.