Follistatin, Klotho, Telomerase: Gene Therapy You Can Buy on an Island

The full story

Three genes dominate the longevity gene-therapy pitch: follistatin for muscle, klotho for cognition, telomerase for lifespan. All three rest on real laboratory work with results that would be exciting if they held up in people. And all three are now being injected into paying volunteers in jurisdictions chosen for what they do not require - which is the part that guarantees we will not find out whether they hold up.

The preclinical science is legitimate. The offshore industry on top of it is not producing data - it is producing announcements, and the two look similar enough on a phone screen to be worth telling apart.

Follistatin: strong animal data, one unblinded human study

The foundation is a 2009 review in Muscle Nerve describing AAV delivery of an alternatively spliced follistatin gene to muscle in mice and monkeys, which increased muscle size and strength with no organ-system pathology or change in reproductive capability detected. That is the entire preclinical basis - and it is in animals.

The human data amount to one study, reported in 2017 in Mol Ther, in six patients with sporadic inclusion body myositis. Annualized six-minute walk distance improved by 56.0 metres per year in the treated group versus a decline of 25.8 metres per year in eight matched untreated subjects, at p=0.01. Read the design before the number: it was unblinded, and the comparator was matched rather than randomized. Two of the six improved by only 5 to 23 metres. Six unblinded patients against a hand-picked comparison group is a signal worth chasing, not a substitute for a randomized trial.

Klotho: a monkey dose-response curve that runs backwards

A 2014 study in Cell Rep found that transgenic systemic klotho overexpression improved multiple learning and memory tests in mice and enhanced long-term potentiation through synaptic GluN2B enrichment. The same paper's human contribution is an association only - carriers of one copy of the KL-VS variant scored better on cognition - with no intervention at all.

The result the sales pitch never mentions came in 2023 in Nat Aging, in aged rhesus monkeys: a single administration of low-dose klotho protein enhanced memory, and the high dose did not. A non-monotonic dose-response in a small nonhuman-primate study is a warning label: more is not better, and nobody can back-calculate a human dose from it. The authors say so directly: whether klotho could enhance cognition in humans is unknown. A clinic with a dose ready for you knows something those researchers said they do not.

Telomerase: one mouse lifespan result, using a mouse gene

The telomerase claim traces to a single 2012 paper in EMBO Mol Med: one AAV injection carrying mouse TERT at one or two years of age raised median lifespan by 24% and 13% respectively and improved insulin sensitivity, osteoporosis and neuromuscular coordination without increasing cancer. Striking - and the sole lifespan finding behind the entire concept, obtained in mice with a mouse transgene. There is no human counterpart at all.

The delivery vehicle has its own problem

The vector is not a neutral courier either. A 2022 meta-analysis in Front Immunol covering 255 rAAV clinical trials, with 7,289 patients planned to be dosed, recorded 11 patient deaths across eight trials, treatment-emergent serious adverse events in 30.6% of trials, and 18 of 30 clinical holds attributable to toxicity - most prominently hepatotoxicity and thrombotic microangiopathy with systemic delivery. A companion 2022 review adds complement activation, dorsal-root-ganglia toxicity and severe hepatotoxicity, with most critical complications occurring at very high vector doses.

Effect scales with dose - and so does the toxicity that has halted trials run by companies with monitoring committees and regulatory reporting obligations. A clinic in a special economic zone has neither, and no obligation to tell anyone what happened.

What an offshore n-of-1 can and cannot produce

MIT Technology Review reported in February 2023 that Minicircle was running a follistatin gene-therapy study in Prospera, Honduras - a special economic zone - recruiting through the purchase of an NFT, with an unknown number of participants. Gene-therapy researchers quoted there were unsentimental: nothing has yet proven to work in human clinical trials the way it does in animal models, and follistatin would not be an obvious first choice for a fountain-of-youth drug. In December 2025 the same publication reported Unlimited Bio running follistatin and VEGF studies in the same zone with only 12 to 15 volunteers in its initial muscle study, volunteers paying their own travel and treatment costs, and follistatin's evidence base remaining primarily rodent research. Bioethicist Holly Fernandez Lynch, quoted there, said the technology still has serious questions about its safety and effectiveness.

Call these what they are: unblinded, uncontrolled, unregistered n-of-1 experiments, self-funded by the subject. Each can generate a feeling, a photograph and a testimonial. None can generate a result. There is no placebo arm to subtract expectation, no blinding to keep measurement honest, no registry entry fixing an endpoint in advance, and no route by which any of it enters the literature. If a dozen volunteers pay to be dosed and the strongest output is a founder's post, the study was staged rather than conducted.

Regulators have been unusually direct. In a statement dated 21 November 2017 covering self-administration gene-therapy products and do-it-yourself kits, the FDA said flatly that the sale of these products is against the law, and that consumers should confirm any gene therapy is FDA-approved or being studied under appropriate regulatory oversight. Its consumer page on regenerative medicine, current as of 8 April 2024, goes further: if you are being charged for these products, or offered them outside a clinical trial, you are likely being deceived. It also disposes of the standard legitimising move - appearing in a clinical-trials registry, or being FDA-registered, does not make unauthorised marketing lawful - and lists blindness, tumour formation, neurological events and life-threatening bloodstream infections among documented harms.

None of this is inevitable. A 2024 analysis in Cell Stem Cell found that targeted action against direct-to-consumer advertising of unproven interventions worked in Australia and Canada, concluding that fears governments cannot regulate this market are unfounded; a 2019 paper in BMC Med Ethics notes Canada's existing tools already cover it, from a ban on materially false promotional representations to every medical college's requirement of evidence-based practice. Offshore is a jurisdictional choice, not a law of nature.

What the evidence does not show

No gene therapy has been shown to extend human lifespan, healthspan, or any aging-related outcome. Nothing in this literature establishes a human dose, a duration of effect, or a long-term safety profile. The one human follistatin study was unblinded, in a muscle disease, with a matched comparator. Klotho's human evidence is an association. Telomerase has none. A 2023 position paper in Cytotherapy describes the industry that grew in that vacuum: a global direct-to-consumer trade in prematurely commercialized cell- and gene-based products with unknown safety and efficacy profiles, deploying what the authors call tokens of scientific legitimacy - the citations, credentials and registry entries that decorate a claim without testing it.

How to read the next headline about this

• Find the species in the first sentence. Most of what sounds like a breakthrough here is a mouse, occasionally a monkey. If the species is missing, that omission is the message.
• Ask for the control arm. No blinding and no controls means the finding cannot separate biology from expectation, however many volunteers signed up.
• Registration is not approval. Being listed on a registry proves only that someone filled in a form; look for a prespecified endpoint.
• Look for the dose-response curve. Klotho's monkey data show the high dose failing. Anyone confident about a human dose is extrapolating from animals whose own investigators declined to.

The science here deserves better than what is being done in its name. The honest version of this field is slow: dose-ranging, monitoring committees, control arms, adverse events published whether or not they flatter anyone. The offshore version compresses that into a flight and a wire transfer and returns nothing anyone can learn from - least of all the volunteer, who has no control arm to compare himself against. You are allowed to find the biology thrilling and the industry selling it indefensible. Those are not in tension; they are the same judgement, applied consistently.