Follistatin-344 Is a Gene Therapy Payload, Not a Peptide — and Myostatin Inhibition Adds Mass Without Function

Follistatin-344 is the 344-amino-acid precursor of follistatin, a human glycoprotein of about 38,000 daltons that blocks myostatin, and all five human studies of it registered on ClinicalTrials.gov deliver its gene rather than the protein. A glycoprotein is a protein with sugar chains attached, and a dalton (Da) is the unit of molecular mass. Myostatin, also called GDF-8, is the signal that limits how large skeletal muscle grows, which is why anything that blocks it is sold for muscle growth.

Drugs that block the same pathway have added lean mass, the body's weight minus its fat, without improving how far patients walked, in randomised trials of 251 and 217 people. The one myostatin inhibitor approved on a gain in function, apitegromab (Isembyld), is a different molecule, approved by the FDA on 11 September 2026 for spinal muscular atrophy (SMA). The human evidence for follistatin itself is sparse and comes entirely from gene delivery. What vials sold under the name contain is a separate question with its own record, and the certificates on file are listed on the Follistatin lab tests page. Peptigrity files the compound with the other tissue repair & healing peptides on its category page.

The short version

• Follistatin-344 is the 344-amino-acid precursor of a human protein that cells trim to a 315-amino-acid form as they release it into the blood; it is not a short synthetic peptide.

• No study has injected follistatin protein into a person; the human evidence comes from gene therapy trials of 3 to 15 patients without a control group, which reported mixed walking results, and from plasmid studies that have posted no results (as of September 2026).

• Drugs that block the same pathway added lean mass without significantly improving walking or physical performance in trials of 251 and 217 patients; apitegromab, the one approved on a functional gain, is a different molecule, and the FDA reports an increased risk of fractures with it.

• Follistatin-344 vials list at a median $103.00 per milligram before shipping, taxes and customs (verified September 2026), and no certificate on file shows a protein-scale mass or a sugar analysis that would settle which follistatin form a vial holds; one measured 60% of the labelled amount.

• As of September 2026 follistatin has no route into a US compounding pharmacy, and the World Anti-Doping Agency prohibits it by name at all times.

What is Follistatin-344, and is it a peptide?

Follistatin-344 is a glycoprotein precursor, not a peptide: human follistatin is made as a 344-amino-acid chain of 38,007 Da whose first 29 amino acids form a signal peptide, a leader sequence the cell cuts off as it secretes the protein. What leaves the cell is FS-315, a 315-amino-acid chain, and a shorter splice form, FS-288, comes from the same gene. An intact 344-amino-acid chain is the unprocessed precursor, not the protein that circulates in blood.

Those figures come from the UniProt record FST - Follistatin - Homo sapiens (Human), entry P19883, which also lists two sites where sugar chains attach and 18 disulfide bonds, chemical bridges between pairs of cysteine amino acids that hold the folded protein in shape. The amino acids in a chain are also called residues, and an isoform is a variant of a protein made from the same gene. The gene therapy team that tested follistatin in patients described FS-315 in a 2015 review in the Journal of Neuromuscular Diseases as the serum form, with a 10-fold lower affinity for activin than FS-288.

The biology behind the interest is real. Follistatin binds activins, myostatin, GDF-11 and some bone morphogenetic proteins outside the cell and blocks the site these signals use to reach their receptors, according to the same record; activins, GDF-11 and bone morphogenetic proteins are signalling proteins of the same family as myostatin. Neutralising myostatin is the mechanism every muscle claim in this category rests on.

What is not real is the implied product. The number 344 in the product name is the precursor count, and the first 29 of those residues are removed by the cell on the way out. Nothing with 344 residues intact circulates in a human being. That is a fact about the molecule, available from a public sequence database, and it is the first thing a buyer should check.

Peptigrity keeps a Follistatin page for the protein family alongside the Follistatin-344 page, and four of the seven certificates on the family's lab-test page name the 344 form.

Can a 344-residue glycoprotein be made by peptide synthesis?

A 344-residue glycoprotein cannot realistically be made by solid-phase peptide synthesis, the stepwise chemistry behind almost every research peptide. Peptigrity's synthesis guide puts the routine limit at about 50 amino acids, and at an excellent 99.5% efficiency per coupling only about 18% of 344-residue chains would come out full length. Standard solid-phase synthesis also attaches no sugar chains, and follistatin's 18 disulfide bonds must pair correctly. A genuine follistatin protein is made in cells.

The 18% figure is our arithmetic: it applies the per-step efficiency given in How Are Peptides Made to 344 couplings, and at 99% per step the share falls to about 3%. Each coupling adds one amino acid, every step is imperfect, and the errors compound as the chain grows.

A cell does three things the research-peptide production line does not do at this scale: it builds the chain from DNA, folds it so the right cysteines pair, and attaches sugars at the two sites UniProt records. A genuine follistatin product is therefore a biologic, made by recombinant expression (protein produced by engineered cells from inserted DNA), then folded and purified, and biologics carry a different quality regime from peptide synthesis. Whether that is what any vendor is selling is not something a buyer can determine from a certificate of analysis, the lab report for one batch.

Follistatin is not the only muscle-growth factor sold in vials whose identity is hard to pin down, and the analysis of PEG-MGF found no defined molecular weight for that compound at all.

The 344-residue count, the signal peptide at 1–29, the 38 kDa mass, the sugar sites and the disulfide bonds are all directly verifiable from UniProt; a kilodalton (kDa) is 1,000 daltons. The inference about synthesis follows from them, and it is the question a buyer should be asking rather than a claim about any specific product.

Has follistatin ever been given to humans as a protein?

No registered human study has given follistatin as a protein, as of September 2026. All five follistatin studies on ClinicalTrials.gov deliver the gene, carried by a modified virus or a plasmid, and the body makes the protein itself. Published results cover 12 patients in two uncontrolled reports: 4 of 6 with Becker muscular dystrophy walked 29 to 125 metres farther in six minutes, and 6 with inclusion body myositis gained a median 56 metres a year. A 3-boy Duchenne trial posted results in which none walked farther.

An adeno-associated virus (AAV) vector is a modified virus that does not cause disease and is used to carry a gene into cells; a plasmid is a ring of DNA injected without a virus. Becker muscular dystrophy (BMD) is a variant of the dystrophin deficiency behind Duchenne muscular dystrophy (DMD), and sporadic inclusion body myositis (sIBM) is an inflammatory muscle disease most common after age 50. Open-label means everyone involved knew who received the treatment.

The primate study behind the human work, Follistatin gene delivery enhances muscle growth and strength in nonhuman primates (Kota et al., 2009), injected an AAV carrying the FS344 gene into the thighs of cynomolgus macaques, a monkey species used in drug testing, and recorded pronounced, durable increases in muscle size and strength without abnormal changes in key organs.

The first human trial, run at Nationwide Children's Hospital and registered as NCT01519349, injected the same construct, rAAV1.CMV.huFollistatin344, into the thigh (quadriceps) muscles of 15 patients from 2012, and lists no results on the registry. Its Becker results were published as A phase 1/2a follistatin gene therapy trial for becker muscular dystrophy (Mendell et al., 2015): 4 of 6 patients walked 29 to 125 metres farther in six minutes, 2 did not improve, no adverse effects were encountered, and muscle biopsies showed less fibrosis (scarring) and larger fibres.

The inclusion body myositis results appeared as Follistatin Gene Therapy for Sporadic Inclusion Body Myositis Improves Functional Outcomes (Mendell et al., 2017): 6 patients injected in both thighs changed their walking distance by a median of +56.0 metres a year, against −25.8 metres a year in 8 untreated clinic patients matched for age, sex and baseline measures (p = 0.01), and the protocol included an exercise regimen.

That paper was challenged in the same journal. Unfounded Claims of Improved Functional Outcomes Attributed to Follistatin Gene Therapy in Inclusion Body Myositis (Greenberg, 2017) argues that the design could not support an efficacy claim: the trial had no control group, the treated patients also received about 60 days of high-dose prednisone, a monitored exercise programme and the placebo effect of an open-label study, and the paper reported a different primary outcome from the one registered. Mendell replied in the same issue.

The one follistatin trial with results posted on the registry points the other way. Clinical Intramuscular Gene Transfer of rAAV1.CMV.huFollistatin344 Trial to Patients With Duchenne Muscular Dystrophy (NCT02354781) gave 3 boys a total dose of 2.4 trillion vector genomes per kilogram between 2015 and 2017: none walked farther on the six-minute test after two years, 1 of 3 improved on the North Star Ambulatory Assessment, a motor-function scale for Duchenne, and no dose-limiting toxicity, a side effect severe enough to halt dosing, was recorded.

The other three registered studies are commercial and unreported. Minicircle's NCT06411366 injected a follistatin plasmid once into 43 healthy adults in Roatán, Honduras, from August 2022 to August 2023, with no control group; it was registered in May 2024, after it had ended, and describes a plasmid "intended to express and secrete bioidentical human follistatin into serum circulation." Its klotho-and-follistatin successor, NCT07285629, finished its main data collection in April 2026 with 14 participants, and NCT07443826, from Unlimited Biotechnology, is recruiting 12 adults with age-related muscle decline, with a start date of June 2026, for AAV-follistatin combined with a plasmid carrying VEGF, a gene for blood-vessel growth. None of the three has posted results or a peer-reviewed report as of September 2026.

Set side by side, the follistatin record is small, uncontrolled and entirely gene delivery:

Does myostatin inhibition actually build usable muscle?

Myostatin inhibition has built measurable muscle far more reliably than usable function. In RESILIENT (n = 251), the antibody bimagrumab raised lean body mass to 102.8% of baseline at its top dose against 97.2% on placebo, yet no dose improved the six-minute walk significantly; in a 217-person trial in sarcopenia, age-related muscle loss, lean mass rose (p < 0.001) while walking and physical-performance scores did not. The one functional success, apitegromab, came in children with spinal muscular atrophy already on gene-targeted treatment.

These pharmaceutical trials are the strongest evidence on the pathway, because they were placebo-controlled, they posted their results, and no one running them was selling the compound to consumers. The drugs in them are antibodies, lab-made immune proteins each built to bind one target in the myostatin pathway.

Which trials found mass without function?

Five placebo-controlled trials of myostatin-pathway drugs found more muscle, or a trend toward it, without a significant gain in walking, strength or physical performance: bimagrumab in inclusion body myositis (n = 251) and in sarcopenia (n = 217), landogrozumab in older adults who had fallen (n = 201), stamulumab in adult muscular dystrophies (n = 116), and apitegromab alongside tirzepatide in obesity (n = 102). A sixth, domagrozumab in Duchenne muscular dystrophy, was stopped for lack of efficacy.

The largest, Efficacy and Safety of Bimagrumab/BYM338 at 52 Weeks on Physical Function, Muscle Strength, Mobility in sIBM Patients (RESILIENT, NCT01925209), was a phase 2b/3 trial of Novartis's antibody bimagrumab in 251 patients with inclusion body myositis. Its primary endpoint was the change in six-minute walking distance at week 52, shown below as least-squares means, the trial's model-adjusted averages, beside the lean body mass measured at the same visit:

All three walking comparisons were non-significant, while the top dose held lean body mass at 102.8% of baseline (95% confidence interval, the range the true value most likely falls in, 101.4–104.2) against 97.2% on placebo (95.9–98.5). Mass up; walking unchanged.

InvestiGAIT (NCT02333331), a phase 2 dose-ranging trial of bimagrumab in 217 older adults with sarcopenia, the age-related loss of muscle mass and function, is the cleanest demonstration of the split, because both outcome families sit inside the same trial. SPPB is the Short Physical Performance Battery, a 0–12 score of balance, walking speed and rising from a chair; the lean-mass figures are the estimates posted on the registry against placebo, where 1.00 means no difference.

Every functional endpoint was non-significant; lean body mass was significant at p < 0.001 in two of three arms.

The other programmes fit the same pattern. Eli Lilly's landogrozumab raised appendicular lean mass, the muscle of the arms and legs, by 0.303 kg against a 0.123 kg loss on placebo in 201 older adults who had fallen (p < 0.001), while stair-climbing time, repeated chair stands and usual gait speed did not differ significantly (p = 0.073, 0.191 and 0.478), according to the 24-week results posted for NCT01604408. Wyeth's stamulumab (MYO-029) produced no improvement in exploratory measures of strength or function in 116 adults with muscular dystrophy, with a trend toward larger muscles in a few, in a trial published in Annals of Neurology in 2008 that was not powered to test efficacy. Pfizer's domagrozumab trial in 121 boys with Duchenne muscular dystrophy (NCT02310763) was terminated on 30 August 2018 for "Lack of Efficacy."

The newest trials use the pathway to protect muscle during weight loss on drugs such as tirzepatide. In EMBRAZE, 102 adults with overweight or obesity took tirzepatide with apitegromab or with placebo; apitegromab reduced lean-mass loss by 1.9 kg at 24 weeks (P = 0.001), yet exploratory tests of physical function, handheld force testing and a chair sit-to-stand test, showed no notable differences, according to the trial report published in Nature Medicine in June 2026.

That is not a failure of dosing or a single underpowered study. It is a consistent finding across programmes, and one plausible reading is that blocking myostatin adds tissue without adding the neural drive, tendon adaptation and motor control that turn tissue into performance.

Which trial found a gain in function?

Apitegromab's SAPPHIRE trial is the only phase 3 trial we found in which blocking myostatin improved function, and the FDA approved the drug as Isembyld on 11 September 2026 for spinal muscular atrophy in patients aged 2 and older already receiving an SMN2-targeted treatment. Among 156 children aged 2 to 12, those on 10 mg/kg improved on a motor-function scale while those on placebo declined, and 34.2% against 13.5% reached a clinically meaningful improvement.

The agency's announcement, FDA Approves First Therapy to Target Muscle Loss in Spinal Muscular Atrophy, calls Isembyld the first therapy approved for SMA that directly targets muscle loss and describes the 52-week trial: 188 participants aged 2 to 21 who could not walk independently, all already on an approved SMN2-targeted drug, given 10 or 20 mg/kg of apitegromab or placebo by intravenous infusion every four weeks. SMA is an inherited disease in which the motor nerve cells that drive muscle die for lack of a protein called SMN; SMN2-targeted drugs such as nusinersen and risdiplam help the body make enough of it.

The full trial report in The Lancet Neurology (2025) put the combined-dose gain at 1.8 points on the Hammersmith Functional Motor Scale Expanded, a standard motor-ability scale for SMA (p = 0.019), and found that the 20 mg/kg dose alone did not differ significantly from placebo (p = 0.11).

What the approval changes for follistatin is narrow. Apitegromab is an antibody that stops myostatin being activated, given by vein, and its gain came in children whose motor nerve cells were already being treated; nothing in the trial speaks to healthy adults, to follistatin or to a research vial. The same FDA announcement reports an increased risk of fractures, including serious fractures, and says the drug may cause fetal harm and may affect reproductive function.

How does follistatin compare with the myostatin drugs that reached trials?

Follistatin is the only product in this comparison never tested in a controlled human trial: no study has injected the protein, and its gene therapy trials enrolled 3 to 15 patients without a control group. Bimagrumab and apitegromab went through placebo-controlled trials of 102 to 251 patients, and ACE-031, a decoy receptor sold in research vials like follistatin, was stopped in a 24-boy trial on safety grounds. Only apitegromab has reached approval.

The medians are per milligram across 1 mg vials on each compound's price page, read on 27 September 2026, and exclude shipping, taxes and customs; the Follistatin-344 price comparison lists 5 shops in stock at $69.99 to $165.00 per milligram (verified September 2026).

A decoy receptor is a free-floating copy of a receptor that soaks up myostatin and activins before they reach muscle. The decoy sold for the same pathway on the same research shelves has its own guide, ACE-031, with the one ACE-031 certificate on file, and it is the natural next comparison for anyone weighing follistatin.

Which follistatin claims survive the evidence?

One of nine common follistatin claims survives the evidence: follistatin binds and neutralises myostatin and the activins, which is established biology. The other eight fail on identity, manufacturing, the trial record, function or regulation, including the claim that it is not banned in sport, which is false because the World Anti-Doping Agency names follistatin in section S4.3 of its list. Walking gains after follistatin gene therapy remain unconfirmed outside open-label studies of 3 to 6 patients.

Read the verdict column downward and the structure of the problem is visible: the only surviving row is the one about what follistatin does inside a body that made it itself.

What dose of follistatin has been tested in humans?

No dose of follistatin protein has been tested in humans, so follistatin has no established human dose. The only human dosing figures are gene therapy doses, counted in vector genomes per kilogram, the number of copies of the gene-carrying virus per kilogram of body weight: 300 billion to 600 billion per leg, injected once into the thigh muscles in the published trials, and 2.4 trillion in total in the Duchenne trial. No figure converts vector genomes into milligrams of protein.

Peptigrity publishes no follistatin dose. The Follistatin Concentration Calculator converts a labelled vial's milligrams and the water added into syringe units; it cannot say which follistatin form, or how much active protein, a milligram represents, and no published human study supplies a milligram figure to enter.

What side effects have been recorded?

No side-effect data exist for injected follistatin protein, because no study has given it. In the gene therapy trials, the Becker study reported no adverse effects in 6 patients, and the Duchenne study recorded no dose-limiting toxicity in 3 boys, pain in all three and one serious event, a head injury from a fall. The pathway's documented risks come from other molecules: nosebleeds and dilated skin blood vessels with ACE-031, and fractures with apitegromab.

In ACE-031's placebo-controlled trial in boys with Duchenne muscular dystrophy (NCT01099761), nosebleeds occurred in 5 of 9 boys on 1.0 mg/kg and 1 of 9 on 0.5 mg/kg, and telangiectasia, small widened blood vessels near the surface of the skin or gums, in 5 of 9 on the higher dose, against none of 6 on placebo; the registry gives the reason for stopping as "Based on preliminary safety data." Stamulumab caused skin hypersensitivity reactions at its two higher doses, 10 and 30 mg/kg, in the 2008 trial. For apitegromab, the FDA reports an increased risk of fractures, including serious fractures, and warns of possible fetal harm and effects on reproductive function.

Follistatin's own biology adds a reproductive consideration. The UniProt record describes follistatin as a specific inhibitor of the production and release of follicle-stimulating hormone (FSH), the pituitary hormone that drives egg and sperm development, because it traps activin A; the gene therapy team chose the FS344 construct, in its own words, "to avoid potential binding to off target sites." Neither point has been measured in a person injecting follistatin protein. A research vial adds the risks of its own production: only one of the seven certificates on the follistatin lab-test page reports an endotoxin result, the test for bacterial residue that causes fever when injected.

Stop and get checked for any sign of an allergic reaction or an infection after an injection, and tell the clinician exactly what was injected.

Stop and seek care: swelling of the face or throat, difficulty breathing or hives after an injection are signs of a serious allergic reaction; spreading redness, warmth or pus at an injection site with fever is an infection; severe or persistent abdominal pain with vomiting on a GLP-1 drug can be pancreatitis; confusion, sweating or shaking on an IGF-1 or insulin-adjacent compound can be hypoglycaemia. Call your local emergency number (911 in the US, 112 in the EU, 999 in the UK, 000 in Australia) or go to an emergency department, and bring the vial and its label. Research-use vials are not approved products in any jurisdiction, even where a prescription product of the same molecule exists; no article on this site replaces a clinician who knows your history.

How do you verify a follistatin vial?

A follistatin vial cannot be fully verified by the tests on the certificates on file: an HPLC purity figure shows how uniform the main material is, not which follistatin form it is, whether it carries its sugars or whether its 18 disulfide bonds are paired correctly. A mass-spectrometry result that states the measured mass is the minimum identity check, since a mature FS-315 chain weighs about 34.7 kDa before sugars and the full precursor 38.0 kDa.

HPLC, high-performance liquid chromatography, separates a sample into its components and reports the share in the main peak. Mass spectrometry (MS) weighs the molecules, which is how identity is confirmed. The FS-315 figure is our arithmetic from the UniProt sequence, for the 315-amino-acid chain with its 18 disulfide bonds and no sugars attached; sugars add weight that varies from molecule to molecule.

Which checks matter for a glycoprotein this size?

Six checks matter for a follistatin certificate, and only two appear on every follistatin certificate on file: HPLC purity and measured content. A protein-scale mass, a glycosylation analysis, a disulfide analysis and an endotoxin result are what separate a characterised protein from an unknown one; Peptigrity's records describe one of the nine follistatin certificates as an HPLC-MS report, and none reports a glycosylation or disulfide analysis (verified September 2026).

That one certificate states a molecular weight of 31–39 kDa and reports identity as "Confirmed via MALDI-MS," but its printed spectrum spans a mass-to-charge ratio (m/z) of 400 to about 880 and labels no matched fragments. MALDI, the mass-spectrometry method named, normally shows a protein as a peak near its own mass, so the identity call cannot be checked from the document itself.

For a small peptide, Mass Spectrometry for Peptides sets out how a mass reading confirms the molecule; follistatin needs the same reading at protein scale. Because sugar chains vary from molecule to molecule, a real glycoprotein is a family of glycoforms, the same protein at slightly different weights, rather than one sharp peak. Net content is how many milligrams of protein a vial actually holds against its label, and Peptide Net Content explains why the two differ. Endotoxin is measured by the Limulus amebocyte lysate (LAL) test, which uses an extract of horseshoe-crab blood cells.

A vial that arrived without these results can still be sent to a lab, and how to test your peptides explains how to send one for independent testing and what each test costs.

What do Peptigrity's lab records show for follistatin?

Peptigrity holds 7 certificates of analysis on its follistatin lab-test page, from 6 shops and 6 labs, and 2 more on its follistatin-344 page (verified September 2026). All nine are for 1 mg vials. On the follistatin page, HPLC purity runs from 97.57% to 99.80%, and measured content from 60% to 132% of the label, with a median of 108%; one vial came in under 90% of its label, and one certificate reports endotoxin.

The Purity Index gives follistatin a recency-weighted score of 99.02 across its 7 tests, counting recent certificates more; follistatin-344, with 2, falls below the index's 5-test minimum and gets no score (verified September 2026). Four of the seven certificates on the follistatin page name the 344 form, two name no variant and one reads "Follistatin-334."

These are certificates shops chose to publish, verified for authenticity before listing, so they reflect the batches shops made public rather than every batch sold. A purity figure between 97.57% and 99.80% says the main peak dominates; it does not say the peak is follistatin, which form it is, or whether it is folded.

Is follistatin approved, and is it banned in sport?

Follistatin is not part of any FDA-approved drug and has no route into a US compounding pharmacy: FDA's 2020 warning letter to a compounding pharmacy named follistatin among bulk substances not eligible under section 503A of the federal drug law. The World Anti-Doping Agency prohibits it by name, at all times, in section S4.3 of its 2026 Prohibited List, as a myostatin-binding protein. Both statements hold as of September 2026.

Section 503A lets a pharmacy compound from a bulk substance only if it has a USP or NF monograph (an official quality standard), is a component of an FDA-approved drug, or is on the 503A bulks list. The letter, Tailor Made Compounding LLC - 594743 - 04/01/2020, lists follistatin among the pharmacy's bulk substances and states that they "are not the subject of an applicable USP or NF monograph, are not a component of an FDA-approved human drug, and do not appear on the 503A bulks list."

Nothing since has changed that. FDA's page Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks, current as of 22 April 2026, holds 14 substances in its active Category 2 table, FDA's list of substances that may present significant safety risks in compounding, and 17 whose nominations were withdrawn; follistatin is in neither. That absence is not clearance: follistatin was not among the twelve peptides FDA took off Category 2 in April 2026 and put before its compounding advisory committee for July 2026 and February 2027. With no approved product and no compounding route, what reaches buyers is research-labelled material from the grey market for peptides, where nobody upstream answers for a batch.

In sport the answer is unambiguous. The World Anti-Doping Code International Standard Prohibited List 2026 places follistatin in S4, hormone and metabolic modulators, prohibited in and out of competition, under S4.3, agents preventing activin receptor IIB activation. Its entries include "Decoy activin receptors (e.g. ACE-031)", "Anti-activin receptor IIB antibodies (e.g. bimagrumab)", "Myostatin-binding proteins (e.g. follistatin, myostatin propeptide)" and "Myostatin- or precursor-neutralizing antibodies (e.g. apitegromab, domagrozumab, landogrozumab, stamulumab)": every myostatin-pathway compound in this article, by name.

Follistatin sits in S4 rather than S2, where most peptide hormones are listed, so an athlete who checks only S2 will miss it, and Do Peptides Show Up on Drug Tests covers how testing works in sport, the military and workplaces.

What trial would settle the follistatin question?

No trial can settle follistatin until the product is characterised, because nobody can currently say which species is in a vial: 344-residue precursor, FS-315 or FS-288, glycosylated or not. The trial that would settle it injects characterised protein rather than delivering a gene, measures whether it reaches muscle, and takes function as its primary endpoint: strength, gait speed, walking distance. Mass is already known to respond.

Pharmacokinetics, in the table below, means how a drug is absorbed, distributed and cleared by the body.

The fourth row is where this category has repeatedly failed, and it is the only row that matters to anyone buying it.

Where does this leave follistatin?

Follistatin fails on two independent grounds, and either one would be sufficient. The first is identity: "Follistatin-344" names a precursor that cells process into FS-315, and no certificate on file shows a protein-scale mass or a sugar analysis that would settle which form, if any, a vial holds. The second is effect: blocking myostatin has added muscle tissue in trial after trial, but usable function in only one trial, in children with spinal muscular atrophy.

The first ground is what is in the vial. Gene therapy trials deliver the 344 construct as DNA because that is how a cell is made to build, fold and glycosylate the protein. Whether a research vial contains recombinant glycosylated protein, an unglycosylated bacterial product, a fragment or something else cannot be determined from any test on the certificates on file.

The second is whether it would matter if it were real. Novartis, Lilly, Wyeth and Pfizer took myostatin-pathway antibodies into trials of 116 to 251 patients; lean mass rose repeatedly, and six-minute walk distance, gait speed and physical performance did not improve significantly. Domagrozumab was terminated explicitly for lack of efficacy. The exception, apitegromab, improved motor function in children with SMA already on treatment, and even it showed no notable functional difference when used to protect muscle during weight loss. Muscle you cannot use is the most consistent finding in this field, and it was established by people with every incentive to find otherwise.

A reader who already holds a vial can check whether its lot appears among the certificates on file, and ask a lab for a measured intact mass.

Frequently Asked Questions

What is follistatin-344?

Follistatin-344 is the full-length, 344-amino-acid precursor of human follistatin, a glycoprotein of about 38,000 daltons. Cells cut off its first 29 amino acids as they secrete it, so the protein in blood is the 315-amino-acid FS-315, and gene therapy trials use the 344 version as the DNA they deliver.

Can follistatin be made by peptide synthesis?

Not realistically. Solid-phase synthesis works routinely to about 50 amino acids, adds no sugar chains in its standard form, and would leave only about 18% full-length chains after 344 couplings even at 99.5% efficiency per step; a genuine follistatin protein is made in engineered cells. What is actually in a research vial is not something a purity certificate can tell you.

Has injected follistatin been tested in humans?

No. All five registered human studies of follistatin deliver its gene, by virus or plasmid, and none injects the protein, as of September 2026.

Does myostatin inhibition build usable muscle?

Mostly not. Bimagrumab raised lean mass in trials of 251 and 217 patients without significantly improving walking or physical performance, and domagrozumab was stopped for lack of efficacy. The exception is apitegromab, approved by the FDA in September 2026 for spinal muscular atrophy after a motor-function gain in children already on SMN2-targeted treatment.

Is follistatin banned in sport?

Yes, by name. Section S4.3 of the World Anti-Doping Agency's 2026 Prohibited List includes "Myostatin-binding proteins (e.g. follistatin, myostatin propeptide)", prohibited at all times. It sits in S4, the hormone and metabolic modulators, not in S2 with most peptide hormones.

What about the gene therapy study that showed improved walking?

Two open-label reports from Nationwide Children's Hospital described walking gains: in 4 of 6 patients with Becker muscular dystrophy in 2015, and in patients with inclusion body myositis in 2017. The 2017 report was challenged in Molecular Therapy the same year, because the trial had no control group and its patients also received prednisone and an exercise programme, and its lead author replied in the same issue. A 2015–2017 trial in 3 boys with Duchenne muscular dystrophy posted results in which none walked farther.

When should you talk to a clinician about follistatin?

Talk to a clinician before any use, and especially if you are pregnant, planning a pregnancy or trying to conceive, or have a bone condition: follistatin's biology includes suppressing FSH, the pituitary hormone that drives egg and sperm development, and the approved drug on the same pathway carries warnings for fetal harm, reproductive effects and fractures. An unexplained reaction after any injection is a reason to stop and get checked.

This article is for educational and informational purposes only and does not constitute medical advice. Peptides discussed may be investigational compounds not approved by the FDA (or equivalent regulators in your jurisdiction) for human use. Peptigrity does not recommend doses; any dose figure above is an approved-label value, a published-trial value or a community report, and is labelled as such. Always consult a qualified healthcare provider before using any peptide or research compound. Peptigrity is an independent review platform and does not sell, endorse, or recommend specific products or vendors.