Follistatin is a protein that binds and blocks myostatin, the body's brake on muscle growth, and activin, a hormone signal with roles from the pituitary to the liver. Cattle and mice that lack myostatin are visibly heavily muscled, which is why follistatin is sold to people who want the same effect.
This page answers "what are follistatin's side effects" from the only place an answer can come from: the human record. It is short, and it is not about the product people buy. The dose units and the registry detail are on our follistatin dosage page; MuscleLedger covers follistatin for training readers on its follistatin compound page, and Peptide Lexicon explains the molecule in its follistatin entry. Nothing here is a recommendation.
The first fact: no one has injected the protein in a trial
Every human administration of follistatin on record is gene transfer. A vector — an engineered virus (AAV) or a circle of DNA (plasmid) — carrying the follistatin gene is injected into named muscles, and the patient's own muscle cells then make the protein. No registered trial has injected follistatin protein into a person (ClinicalTrials.gov, read 26 October 2026; the full census is on the dosage page).
That matters for side effects in two ways. Gene transfer carries its own risks — immune reactions to the vector, and effects that cannot be switched off once the gene is expressing. And none of the trial findings describes what happens when someone injects a lyophilised protein labelled "follistatin 344", which is a different substance given a different way.
Every human study, and what it reported about harms
| Study | Who | What was given | Participants | Safety reporting |
|---|---|---|---|---|
| Mendell 2015, Mol Ther (part of NCT01519349) | Becker muscular dystrophy | AAV1.CMV.FS344, both quadriceps, 3 or 6 × 10¹¹ vg/kg per leg | 6 | "No adverse effects were encountered" |
| Mendell 2017, Mol Ther (part of NCT01519349) | Sporadic inclusion body myositis | rAAV1.CMV.huFS344, 6 × 10¹¹ vg/kg, both quadriceps | 6 | No adverse-event statement in the abstract |
| NCT01519349 (registry) | Becker and inclusion body myositis | as above, three cohorts | 15 | Completed 2017; no results posted |
| NCT02354781 | Duchenne muscular dystrophy (boys) | 2.4 × 10¹² vg/kg, gluteal, quadriceps and tibialis anterior | 3 (of 6 planned) | Adverse events posted — table below |
| NCT06411366 | Commercial plasmid (Minicircle), phase 1 | Non-viral plasmid carrying FST344 | 43 | Completed August 2023; no results posted |
| NCT07285629 | Healthy adults, klotho + follistatin plasmid | Combined plasmid gene therapy | 14 | Completed April 2026; no results posted |
| NCT07443826 | Age-related muscle loss | AAV9-follistatin | 12 planned | Recruiting; completion estimated 2028 |
Registry records read through the ClinicalTrials.gov API on 26 October 2026; abstracts read the same day.
Counted honestly, that is 87 people given or due to be given a follistatin gene across five registered studies (the two published cohorts are 12 of the first trial's 15), of whom three have a posted adverse-event record. The two largest studies — 43 and 14 volunteers, run by a commercial sponsor — have published nothing on harms.
The one posted adverse-event table
NCT02354781, led by Jerry Mendell at Nationwide Children's Hospital, gave the follistatin gene to three boys with Duchenne muscular dystrophy and posted its results:
| Event | Boys affected (of 3) |
|---|---|
| Serious: head injury from a fall | 1 |
| Pain | 3 |
| Bruising | 2 |
| Increased muscle weakness | 1 |
| Compression fracture | 1 |
| Anxiety | 1 |
| Behavioural changes / agitation | 1 |
| Insomnia | 1 |
| Pharyngitis, rhinorrhoea, nasal congestion, influenza, stye, abrasion | 1 each |
| GERD, constipation | 1 each |
These are every event the trial recorded, not events judged to be caused by follistatin — the registry does not attribute them. Pain and bruising fit multiple intramuscular injections across several muscles. Falls, fractures and increasing weakness are part of Duchenne itself, and three boys with no control group cannot separate the disease from the treatment. This is the whole posted human side-effect record for follistatin.
The hormone concern that shaped the gene
Follistatin's other job is to bind activin, and activin helps drive the pituitary's release of FSH, the hormone that supports sperm production in men and follicle development in women. A follistatin that circulated widely could, in principle, suppress it.
The trial designers addressed this before the first patient. A 2015 review by the Nationwide Children's investigators (Al-Zaidy 2015, J Neuromuscul Dis) explains that the interaction with the pituitary "called for caution", and that they chose the FS344 variant because it is processed to FS315, which has "a 10-fold lower affinity to activin compared to FS288". In monkeys given the same AAV1-FS344 gene in the quadriceps, the original study reported that long-term expression "did not produce any abnormal changes in the morphology or function of key organs" (Kota 2009, Sci Transl Med); that paper received an erratum in August 2026, whose content we did not review.
Whether FSH or other hormones actually changed in the human patients is not stated in the abstracts available to us. It is the most obvious safety question the published record leaves open.
What a research vial is not
Follistatin sold online as "follistatin 344" or "FST-344" is a protein powder labelled for research. Three things separate it from the trial record:
- The trials gave a gene, not a protein. A gene expressed in muscle produces protein continuously, locally, for months; an injected protein would circulate briefly. Neither the effects nor the side effects transfer.
- The dose unit does not convert. Trial doses are vector genomes per kilogram; there is no milligram equivalent (the dosage page explains why).
- Content is unverified. Follistatin is a glycoprotein of more than 300 amino acids; whether a given vial contains correctly folded, active protein is a question for a laboratory, and our page on how to read a certificate of analysis applies.
FDA's adverse-event database returned no reports naming follistatin when queried through openFDA on 26 October 2026 — which reflects how rarely it is reported, not that it is harmless. Other compounds sold for muscle growth with the same thin record include IGF-1 LR3.
What is not established
- Any side effect of injected follistatin protein in humans — never studied.
- Hormonal effects in treated patients, including FSH, despite the concern that shaped the gene's design.
- Harms in the two largest studies, which posted no results.
- Long-term effects, including any effect on cancer risk; follistatin is studied in cancer biology, but no human follow-up addresses it.
- Effects in healthy people seeking muscle growth, beyond the unpublished commercial plasmid studies.
Anyone considering an experimental gene or protein product for muscle growth should take the question to a clinician, ideally one connected to a registered trial.
Sources and dates
- Mendell JR et al. A phase 1/2a follistatin gene therapy trial for Becker muscular dystrophy. Mol Ther. 2015;23(1):192–201. PMID 25322757.
- Mendell JR et al. Follistatin gene therapy for sporadic inclusion body myositis improves functional outcomes. Mol Ther. 2017;25(4):870–9. PMID 28279643.
- Al-Zaidy SA et al. Follistatin gene therapy improves ambulation in Becker muscular dystrophy. J Neuromuscul Dis. 2015;2(3):185–92. PMID 27858738.
- Kota J et al. Follistatin gene delivery enhances muscle growth and strength in nonhuman primates. Sci Transl Med. 2009;1(6):6ra15. PMID 20368179 (erratum 2026).
- ClinicalTrials.gov API v2 records NCT01519349, NCT02354781 (posted adverse events), NCT06411366, NCT07285629, NCT07443826 — read 26 October 2026.
- openFDA drug adverse events, "follistatin": no matching reports — 26 October 2026. Abstracts read 26 October 2026.