Ask most people when their body starts to age and they will say something in the region of forty. Ask a thymus researcher and the answer is closer to eighteen months.
The thymus sits behind your breastbone, roughly the size of a small deck of cards, and it does one job that no other organ can do: it is the school where newly made T cells learn to distinguish your own tissue from a threat. Fail that training and you get autoimmunity. Skip it entirely and you get almost no functioning adaptive immunity at all.
It is also, by a wide margin, the fastest-ageing organ you own. And after decades as an immunology curiosity, it has become one of the more heavily funded targets in longevity biotech. In September 2023 the US government's health research agency ARPA-H made its first-ever industry award under its open funding call — up to $37 million — to a Cambridge, Massachusetts company called Thymmune Therapeutics, to grow thymic tissue from stem cells.[7] In January 2026 a Basel startup, TECregen, raised $12.6 million led by Boehringer Ingelheim's venture arm for engineered proteins designed to regenerate thymic epithelial cells.[8] In July 2026 a paper in Nature Communications put a specific thymic hormone at the centre of age-related inflammation.[6]
So it is worth asking the question plainly. How much of this is established, and how much is a very good story?
The organ that starts shrinking before you can walk
The foundational measurement here is forty years old and still unmatched. In 1985, a group in Germany obtained 136 thymuses removed at autopsy from people who had died suddenly, added biopsy material from immunologically healthy cardiac patients, and ended up with tissue from 204 individuals ranging in age from one month to 107 years. They then measured what was actually inside them.[1]
Two findings from that paper deserve to be much better known.
The first is that the thymus does not really shrink. Its median volume across the whole human lifespan stays around 19.5 cubic centimetres, with individual maximum size reached in the first year of life. What changes is the composition. The thymic epithelium — the tissue where T cell training physically happens — involutes continuously from the first year to the end of life, following a smooth negative logarithmic curve. Adipose tissue moves in to fill the space, extensively after age fifteen. The organ keeps its shape while the functional part quietly empties out.
The second finding is the one that should have changed the textbooks. The velocity of thymic epithelial involution does not change at puberty. The authors state it directly: the age-related involution of the human thymus is not related to puberty.[1] Sex steroids do influence the thymus, and blocking them can produce short-term regrowth, but they are not the driver of the underlying decline. That matters, because the intuitive model — hormones flip a switch in adolescence and the thymus winds down — is the wrong shape. This is a developmental programme running from infancy, not a midlife failure.
Why an empty thymus shows up everywhere else
The consequence is a slow contraction in the supply of naive T cells: fresh cells that have not yet met their target and can therefore respond to something genuinely new. A February 2026 review in Science Advances summarises where that leads — reduced T cell production raises infection and cancer risk, and produces the poor vaccine responses that are one of the most reliable features of an ageing immune system.[2] That review adds a newer claim worth flagging: age-related T cell defects have recently been implicated in the loss of tissue integrity and function more generally, which would extend the stakes well beyond infection.
Mechanistically the picture is one of a collapsing conversation. Developing T cells and thymic epithelial cells depend on constant cross-talk; hormonal shifts and chronic inflammation degrade both sides of it, the cortical and medullary epithelial cells are lost, fibroblasts expand and the perivascular space widens.[2][3] This is also where inflammaging enters, a loop we have covered in our piece on why your immune defences slow down with age.
Is thymic decline actually causing immune ageing, or merely accompanying it? A 2025 experiment gets unusually close to an answer. Researchers used mice engineered to lose FOXN1, a transcription factor whose decline is one of the earliest events in normal thymic involution, and so induced premature involution in young animals. Those young mice developed aged-like thymic epithelial cells, severe lymphopenia, reduced IL-2 secretion, exaggerated early interferon-gamma responses and blunted responses to influenza infection — a peripheral T cell profile closely resembling genuinely old mice.[4] Take the thymus away early and immunosenescence arrives early. That is a causal claim with an experiment behind it, in mice.
The nine men everyone quotes
Now the headline result. In 2019, Gregory Fahy and colleagues published in Aging Cell what remains the most cited human evidence that thymic regeneration is possible — and, separately, the first report of epigenetic age reversal in people.[5]
The TRIIM trial enrolled ten healthy men aged 51 to 65 for one year on a combination of recombinant human growth hormone at 0.015 mg/kg, DHEA at 50 mg daily and metformin at 500 mg daily, with vitamin D3 and zinc, individualised after week three. The DHEA and metformin were there to offset growth hormone's known tendency to push blood sugar in the wrong direction.
The results were striking. Thymic fat-free fraction on MRI increased significantly, with seven of nine analysed volunteers showing restored thymic functional mass. Across four epigenetic clocks, mean epigenetic age came in about 1.5 years below baseline after twelve months — a 2.5-year change relative to chronological ageing over the same period. The rate accelerated markedly in the final quarter, from −1.6 years per year over months 0–9 to −6.5 years per year over months 9–12. GrimAge, the clock most strongly tied to mortality, showed a two-year decrease that persisted six months after treatment stopped.
Here is the part that gets lost in the retelling. There was no control group. Ten men, one year, all white, all male, one arm. The authors themselves call it a one-year pilot trial involving nine volunteers and write that it will be necessary to verify the results by replicating them in an appropriately powered follow-up study.[5] They also note that epigenetic clocks do not capture every feature of ageing — a caveat we unpack in our explainer on what epigenetic clocks actually measure.
That follow-up, TRIIM-X, has been running since November 2020: a phase 2 study of 85 participants aged 40 to 80, this time including women and control arms, with GrimAge, thymic density on MRI or CT, and safety as co-primary endpoints. Its listed primary completion date was December 2025.[9] As of mid-August 2026, no results have been posted to the registry. Until they are, the honest summary of human thymic regeneration is: one uncontrolled pilot, seven years old.
The July 2026 result that put this back in the news
The newest piece is a different kind of finding. Researchers at the University of Southern California, publishing in Nature Communications in July 2026, went looking for why inflammatory myeloid cells accumulate with age in both mice and humans, pumping out IL-1α, IL-1β, IL-6 and TNF-α.[6]
Using heterochronic parabiosis — surgically joining the circulation of a young and an old mouse — they showed that something circulating in young blood suppresses that inflammatory activation, and that it does not come from bone marrow. The factor they identified was thymulin, a zinc-dependent peptide hormone secreted by thymic epithelial cells that declines with age. Thymulin suppressed inflammatory cytokine production by inhibiting NF-κB signalling, improved anti-tumour T cell immunity, and made tumours more responsive to anti-PD-L1 immunotherapy — in an age-dependent way.
Read carefully, this reframes the thymus. It is not only a T cell training school whose closure leaves you short of new recruits; it is also an endocrine organ whose decline actively removes a brake on inflammation elsewhere in the body. That is a genuinely new argument for why thymic involution might matter to ageing broadly.
The caveats are equally real. This is mouse work with supporting human cell data, framed primarily as a cancer immunotherapy paper rather than a longevity one. Thymulin is not a drug you can buy, and no human trial of thymulin supplementation for ageing exists.
Where the evidence actually stands
| Approach | Best evidence | Status |
|---|---|---|
| GH + DHEA + metformin (TRIIM protocol) | Uncontrolled 1-year pilot, 10 men; thymic fat-free fraction up, 7/9 responders[5] | Suggestive, unreplicated. Phase 2 results overdue. |
| Thymulin restoration | Mouse parabiosis + human myeloid data, July 2026[6] | Preclinical. No human trial. |
| Stem-cell-derived thymic epithelial cells | ARPA-H funded to restore T cell development in animals[7] | Animal stage. Transplantation protocols still being developed. |
| Engineered thymic growth factors | $12.6m seed, preparing IND-enabling studies[8] | Preclinical. No trial started. |
| Sex hormone blockade | Produces regrowth, but involution rate is puberty-independent[1] | Mechanistically incomplete. |
| Any supplement marketed for “thymus support” | None showing thymic regrowth in humans | Unsupported. |
What this means if you are not a mouse
There is no version of this article that ends with a shopping list, and any page that gives you one is selling something.
The TRIIM protocol should not be self-administered. Recombinant growth hormone is a prescription medicine with a real side-effect profile — the TRIIM-X exclusion criteria alone rule out anyone with a cancer history or familial risk, pre-existing carpal tunnel syndrome, significant joint problems, diabetes, elevated CRP or a BMI of 35 or above — and it is precisely the drug most heavily counterfeited in the grey longevity market. Metformin is a separate discussion we have had in our piece on the existing drugs being tested to slow aging. Neither is approved for thymic regeneration anywhere.
What you are left with is unglamorous and honest. Nothing on a shelf has been shown to regrow thymic tissue. The interventions with the best evidence for immune ageing generally — exercise, sleep, not being chronically inflamed, keeping vaccinations current precisely because responses weaken — do not act on the thymus specifically, and we cover them in our review of the habits with actual longevity data. If you want a defensible number for where your own biology sits, our biological age calculator uses validated blood markers rather than any thymic measure.
What makes the thymus genuinely interesting is not that it is treatable yet. It is that it is the clearest example in human biology of an organ whose decline is programmed, measurable, and demonstrably upstream of things we care about. Most of the field is still arguing about whether ageing has discrete causes. Here is one you can photograph on an MRI, quantify as a fat-free fraction, and watch fall on a smooth curve from the age of one. Whether anyone can bend that curve back is, as of August 2026, still an open question with roughly $50 million riding on it.
Common questions
Can you regrow your thymus?
Partially, and only under drug treatment not approved for that purpose. Seven of nine men in the TRIIM pilot showed significant restoration of thymic functional mass on MRI after a year of growth hormone, DHEA and metformin.[5] There was no control group and the finding has not been replicated at scale. Nothing sold as a supplement has been shown to do it.
At what age does the thymus start to shrink?
The working tissue starts involuting in the first year of life and continues on a smooth curve to the end of it. The gland keeps its size — a median 19.5 cm³ — because fat replaces the functional tissue, mostly after age fifteen.[1] Notably, the rate does not change at puberty.
Why does a shrinking thymus matter for aging?
Fewer naive T cells means weaker responses to new infections and vaccines, reduced tumour surveillance and more autoimmunity.[2] The July 2026 thymulin work suggests a second route: the thymus also secretes a hormone that suppresses inflammatory signalling elsewhere, so losing it may remove a brake on inflammaging.[6]