Stem-cell exhaustion

Your body keeps a reserve of repair cells — stem cells — that renew worn-out tissue throughout life, and with age that reserve shrinks and works less well, so healing slows, muscle rebuilds poorly, and the immune and blood-forming systems falter. The reassuring part: resistance training is the one everyday lever shown to boost muscle stem cells in older people, and it, along with real sleep and sensible eating, rests on far firmer ground than the stem-cell injections and reprogramming therapies that get the headlines.

Stem-cell exhaustion is one of the hallmarks of aging — it appeared in the original nine (2013) and again in the expanded twelve (2023) — and is an integrative one: less a root cause than a convergence point, where the damage accumulated by the upstream hallmarks finally overwhelms the body's capacity to repair itself.[1][2] Adult (somatic) stem cells are the dedicated repair engines of each tissue: blood-forming cells in the bone marrow, satellite cells in muscle, neural stem cells in the brain, and the versatile mesenchymal stem cells that support many tissues. Most of the time they sit dormant in a protected resting state called quiescence; when tissue is damaged they wake, divide, and rebuild.[3] With age both their numbers and their function decline, and the consequences are exactly the syndromes that define frailty — muscle loss (sarcopenia), a weakened immune system, slow recovery from illness or injury, and failing tissue maintenance across the body.[4]

Why the reserve runs down

Stem cells wear out for two kinds of reasons: damage inside the cells themselves, and decay of the neighbourhood they live in.

The intrinsic drivers are the upstream hallmarks playing out inside the stem cell. Decades of DNA damage, telomere shortening, and epigenetic erosion corrupt the cell's instructions and blur its identity, so it loses the precise control over when to rest and when to divide.[5] That balance is governed by the same nutrient-sensing machinery — the growth-sensing mTOR pathway and the energy-sensing AMPK/sirtuin pathways — that runs the rest of metabolism; when nutrient sensing is deregulated, stem cells are pushed out of their protective resting state.[6] Failing mitochondria compound the problem, draining the cell's energy and leaking reactive byproducts that damage it further.[7]

A subtle but important point is that quiescence is the asset. The resting state is what protects the stem-cell pool over a lifetime — when researchers force stem cells to divide continuously by removing the brakes that keep them dormant, the cells proliferate in a brief burst and then fail to sustain repair, exhausting the pool.[8] Aging tends to erode quiescence, and that erosion is itself a route to exhaustion.

The clearest picture of how that erosion happens comes from muscle stem cells in very old animals. In geriatric mice — not merely old ones — these cells stop returning to reversible rest and instead slide into an irreversible pre-senescent state, driven by the switching-on of a cell-cycle brake (the p16 gene); silencing that brake restored their dormancy and repair capacity, and the same gene is switched on in satellite cells from very old people (Moderate; mouse, corroborated in human cells).[9] Part of what keeps the resting state intact is the cell's own self-cleaning machinery: basal autophagy clears worn-out mitochondria and misfolded proteins, and when it fails in aged satellite cells they tip from quiescence into senescence — the mechanistic bridge from failing protein and organelle housekeeping to stem-cell exhaustion (Moderate; mouse plus human geriatric cells).[10]

The niche: stem cells fail because their neighbourhood fails

A stem cell is only as healthy as the micro-environment that houses it — the niche, a specialized local structure of supporting cells, signals, and scaffolding that tells the stem cell when to rest and when to act.[11] With age the niche degrades, and one of the largest culprits is the build-up of senescent cells — damaged cells that stop dividing but linger, pouring out an inflammatory secretion (the senescence-associated secretory phenotype). Inside a niche, that secretion impairs the stem cell's ability to home and self-renew, and can even tip healthy neighbours into senescence themselves.[12] This is the direct mechanical link between stem-cell exhaustion and chronic inflammation: the inflamed tissue environment of older bodies is actively hostile to repair.

The damage is niche-specific. In muscle, the loss is fibre-type specific and measurable in people: older men carry roughly half as many satellite cells per fast-twitch fibre as young men, while slow-twitch fibres are largely spared. That was first seen in a small study of 16 men and then confirmed in a larger study of 165 people, where fast-twitch satellite-cell numbers fell steadily with age across adulthood up to 86 (Strong; human cross-sectional).[13][14] These fast-twitch satellite cells also sit farther from their blood supply: in older men they sat about 27% farther from the nearest capillary than in young men (about 22 vs 17 micrometres).[15] That seems to matter for training. Over 24 weeks of resistance training, the older men with relatively few capillaries around their fast-twitch fibres gained neither satellite cells nor muscle-fibre size, while those with better capillary supply gained both (Moderate; small human studies).[16] In the brain's memory centre, the neural-stem-cell niche appears to thin out, though how much new-neuron formation continues in adult humans is genuinely contested, with careful post-mortem studies reaching opposite conclusions (see below), so the link to cognitive decline is weaker than it is often presented (Weak-preliminary; conflicting human data). And in the bone marrow, the blood-forming system shifts with age toward making more inflammatory (myeloid) white cells and fewer lymphocytes.[17] One measurable form of this is clonal hematopoiesis, the age-related expansion of mutated blood-cell clones (detailed below). People who carry such clones have more coronary heart disease, and in mice, bone marrow lacking one of the commonly mutated genes produced larger artery plaques.[18]

The one you can actually measure: clonal hematopoiesis

Most of stem-cell exhaustion is invisible in a living person. The exception is in the blood, where it leaves a fingerprint you can read off a blood sample. In adults under about 65, blood is made evenly by a large, diverse pool of roughly 20,000 to 200,000 blood-forming stem cells, each of which picks up about 17 new DNA mutations a year from birth. After about 75 that diversity collapses: in each older person studied, 30 to 60% of all blood came from just 12 to 18 expanded clones. Most of those clones had started growing before age 40, and only about 1 in 5 carried a known cancer-driver mutation (Moderate; whole-genome sequencing in 10 people).[19]

When one of those expanding clones happens to carry a mutation in a growth-driver gene — most often the epigenetic regulators DNMT3A, TET2, or ASXL1 — the condition is called clonal hematopoiesis of indeterminate potential (CHIP): a detectable mutant clone in someone with normal blood counts and no cancer. It becomes common with age: standard sequencing finds it in about one in ten people over 65, rising to nearly one in five in their 90s.[20][21] In a small study using ultra-sensitive sequencing, tiny clones turned up in 95% of the healthy 50- to 60-year-olds tested, so how common it looks depends on how small a clone the test can detect.[22] Pooled across 88 studies, carriers have about four times the risk of a later blood cancer (hazard ratio, HR, 4.28, a measure of relative risk over time; the first two large studies put it at 11 to 13 times), but the absolute risk stays small, about 0.5% to 1% a year, at the higher end with large clones.[23][24][25] The same pooled data put the risk of death from any cause about a third higher (HR 1.34). The risk depends on the gene and the clone's size: clones carrying only a DNMT3A mutation showed no clear extra risk of myeloid blood cancer or of death (Strong; large human cohorts and meta-analysis).[26]

The striking part is cardiovascular. Pooled across studies, people with CHIP have about 75% higher risk of coronary heart disease (HR 1.76) and about 40% more cardiovascular events overall (HR 1.40); a higher risk of dying from heart disease shows up only in people with larger clones.[27] In the original studies, CHIP carriers had about four times the risk of a heart attack at a young age after accounting for age, sex, diabetes and smoking, apparently because the mutant immune cells stoke arterial inflammation (Strong; human observational, mouse mechanism).[28] The American Heart Association now also links CHIP to heart failure, atrial fibrillation and blood clots, with the size of the risk depending on which gene is mutated and how big the clone is.[29] CHIP is the clearest example on this page of a hallmark you can actually quantify — the concrete human readout of stem-cell exhaustion, and a direct bridge to chronic inflammation and cardiovascular aging.

Measurable does not yet mean actionable. No treatment aimed at CHIP has been shown to prevent heart disease, and testing healthy people is not part of routine care. CHIP is increasingly found by chance during genetic testing, especially in cancer care, and specialists are still working out when deliberate testing helps, for example for persistent unexplained low blood counts or before cancer chemotherapy.[30][31] Early hints come from heart trials of anti-inflammatory drugs. In heart-attack survivors with raised inflammation, those with TET2-mutant clones had about 60% fewer cardiovascular events on the antibody canakinumab (HR 0.38), though not clearly more benefit than other patients got. In people with coronary disease, low-dose colchicine slowed the growth of TET2 clones but not of clones overall. Both are exploratory subgroup findings that need confirmation (Weak-preliminary; subgroup analyses of randomised trials).[32][33] Smoking, obesity and chronic inflammation help these clones expand, so the practical response is the usual one: don't smoke and keep cardiovascular risk factors in check.[34]

One finding cuts the other way. Several studies have linked CHIP, particularly larger clones, to a lower risk of Alzheimer's dementia (about a third lower odds in one analysis; odds ratio, OR, 0.64), and in the brains of a handful of carriers studied after death, the same mutant blood-derived cells made up a large share of the brain's own immune cells (microglia), although what they do there is not yet known.[35][36] The picture is not settled: in about 465,000 UK Biobank participants CHIP showed no link to Alzheimer's and a modestly higher risk of brain disease linked to blood-vessel damage (Weak-preliminary; conflicting human observational data).[37]

Is human brain neurogenesis really declining? A genuine controversy

The tidy story, that the brain's stem-cell pool dwindles and takes memory with it, is shakier in humans than in mice. Post-mortem studies from 2018–2019 disagreed sharply about whether adults form new hippocampal neurons at all,[38][39] and one reported fewer young neurons in Alzheimer's disease.[40] Newer studies that read the genetic activity of single brain cells now find dividing neural stem cells and young neurons in adult brains,[41] including in older adults, with a distinct pattern in older people whose memory is unusually sharp.[42] Some new-neuron formation probably persists into old age. How much there is, and whether it matters for memory, is still unknown (Weak-preliminary; human post-mortem data).

A genuine surprise: exhaustion is partly reversible

For a long time the decline of stem cells looked like a one-way street. The more hopeful recent picture is that a meaningful share of it reflects a suppressed rather than destroyed cell — and suppression can sometimes be lifted. Transplant experiments show that an aged stem cell placed in a young environment, or an old niche restored to better working order, can recover much of its lost function, which means the niche's signals are doing a lot of the aging.[43] This reframes the goal from "replace the worn-out cells" to "fix the conditions that are holding them back," and it is the conceptual basis for most of the rejuvenation research now moving toward the clinic.[44] A second route has emerged in the blood. The aging marrow's tilt toward inflammatory white cells (described above) comes from a subset of blood-forming stem cells that favour them over the lymphocytes that drive immune memory. In aged mice, using antibodies to remove that subset restored a younger-looking immune system and improved responses to viral infection (Weak-preliminary; mouse only).[45]

Much of this thinking traces to parabiosis — surgically joining the circulations of an old and a young mouse, which restored the repair capacity of aged muscle and liver stem cells (Moderate; mouse).[46] That result launched a hunt for youthful "rejuvenation factors" in blood, and the cautionary tale that followed is worth knowing. A circulating protein called GDF11 was reported to reverse age-related decline in mouse muscle,[47] but an independent team could not reproduce this. They found that the tests used could not tell GDF11 apart from a near-identical protein (myostatin), and that extra GDF11 actually hindered muscle repair in mice.[48] Even whether GDF11 levels change with age is still disputed: a later human study using a test specific to GDF11 found that its levels did not fall with age (Moderate; mouse, with conflicting human measurements, and a textbook case of a young-blood claim that did not hold up).[49] Later work reframed the benefit again: much of what parabiosis achieves may come from diluting the pro-aging factors in old blood rather than adding youthful ones — simply exchanging old plasma for a neutral saline-albumin solution reproduced part of the effect (Moderate; mouse).[50] This is why the commercial "young plasma" infusions marketed for anti-aging are not the same idea, and are unproven — a distinction covered below.

The experimental frontier

Several strategies aim to restore the stem-cell compartment directly. All are early, and none is established for healthy adults.

  • Stem-cell transplantation. Mesenchymal stem cells are attractive because they are well tolerated and work largely by secreting repair-promoting factors rather than by permanently engrafting — they do not take up residence in the tissue. The most advanced human data is in physical frailty. A placebo-controlled trial in 148 frail older adults tested four doses of donor bone-marrow stem cells (laromestrocel), each given as a single drip. At six months, the main comparison of each dose against placebo on the walking test was not statistically significant, although the gains rose with dose. By nine months the highest dose added about 63 metres to a six-minute walk, and a lower dose added about 49 metres. But the highest-dose group had only 16 people, the other two doses showed no clear benefit, and gait speed, grip strength and self-reported function did not improve. No serious side effects were blamed on the cells. The trial was run by the company that makes the product (with a US government research grant), and its corresponding author is the company's chief science officer and an equity holder (Weak-preliminary; one small human RCT on a walking test).[51] A 2026 review of the three placebo-controlled trials so far (about 208 people) found the cells well tolerated but judged the trials too few and too varied to say the treatment works.[52] A later trial of umbilical-cord stem cells in 147 frail adults reported a gain of about 1 point on a 12-point physical-performance score, but it was open-label, ran at a single centre and compared the cells with supplements alone, with no placebo.[53] This is the strongest stem-cell-for-aging evidence to date, but it is not a proven anti-aging therapy, and intravenous cell delivery carries its own risks.
  • Partial reprogramming. Briefly switching on a subset of the "Yamanaka factors" can roll back a cell's epigenetic age while keeping its tissue identity: cyclic, short-pulse induction ameliorated signs of aging in prematurely-aged mice, and a version of the approach restored vision in aged and glaucoma-damaged mouse eyes — the basis for the first human trial, confined to the eye (Weak-preliminary; mouse).[54][55] The central safety caveat: keeping all four factors switched on continuously killed mice within days, and earlier studies found it caused teratomas (stem-cell tumours).[56] The eye version leaves out the most cancer-linked factor (c-Myc), and in mice it caused no tumours even after many months of continuous use.[57] That is why the first human attempt is restricted to a small, controllable tissue like the eye rather than the whole body: in June 2026 its developer announced that the first participant had been dosed in a small safety trial for optic-nerve damage from glaucoma or from a type of eye stroke. No results have been reported yet.[58] The mechanism is covered further under epigenetic alterations.
  • Rejuvenating immune cells. Engineered T-cells used in cancer therapy lose potency in older patients as they become exhausted and senescent; researchers are tuning culture conditions to coax them toward a long-lived, stem-cell-like memory state, though the cytokine signals involved can be dangerous at high doses.[59]

The small-molecule angle is less dramatic but better grounded — with the important caveat that almost all of the stem-cell data is preclinical. Rapamycin restored the self-renewal of aged mouse blood stem cells and improved their immune response (Weak-preliminary; mouse).[60] At the high doses used after organ transplants, drugs that block mTOR suppress blood counts and raise infection risk, but low, intermittent doses behave differently. In 264 older adults, a 6-week course of low-dose mTOR blockers was followed by fewer infections over the next year.[61] A larger phase 3 trial in about 1,000 people aged 65 and over did not reproduce the benefit for respiratory illness.[62] In a company-run trial, about a year of weekly low-dose rapamycin in healthy adults looked safe but missed its main outcome.[63] None of these human trials measured stem cells. Metformin acts on the same nutrient-sensing pathways and improves the "stemness" of human stem cells grown in the lab by damping mTOR signalling (Weak-preliminary; cell and animal work).[64][65] In people, though, it can work against training. In a 14-week trial in 94 healthy adults aged 65 and older, those taking 1,700 mg of metformin a day gained less lean mass and thigh muscle from resistance training than those on placebo, although metformin did not change the training-driven rise in muscle stem cells (Moderate; human RCT).[66] Both drugs are covered, with their human evidence and caveats, under geroprotectors.

NAD⁺ precursors (NMN, nicotinamide riboside) are the cleanest illustration of the mouse-to-human gap. Restoring NAD⁺ — a coenzyme that falls with age — rescued muscle, neural, and pigment stem-cell function and modestly extended lifespan in aged mice (Moderate; mouse).[67] But when 32 adults aged 55 to 80 took a daily combination of nicotinamide riboside (1,000 mg) and pterostilbene (200 mg) or a placebo before an experimental muscle injury, the supplement was safe yet did not improve muscle stem-cell recruitment or recovery. Muscle NAD⁺ did not even rise, though blood levels did (Moderate; small human RCT).[68]

What actually helps healthy adults

The levers that help are, once again, the familiar ones, and they fit the biology well: exercise, sleep and diet act on the niches and the nutrient-sensing pathways that govern stem cells, though only exercise has direct stem-cell evidence in people.

Exercise — the best-evidenced lever, mainly for muscle stem cells. Resistance training is the most direct way to mobilize muscle satellite cells out of dormancy. In 51 older adults, 12 weeks of supervised strength training raised satellite-cell numbers in exactly the fast-twitch fibres that age depletes, and a pooled analysis of 19 studies points the same way, although that analysis mixed very different measurements (Moderate; small before-and-after human studies).[69][70] Blood supply seems to matter too. In 14 older men, 12 weeks of mostly resistance training plus one weekly interval session sharpened the fast-twitch satellite-cell response to a workout, and the men who gained the most small blood vessels (capillaries) around those fibres improved the most. That was a small study without a comparison group, so it shows a link rather than proof (Weak-preliminary; small human study).[71] Whether endurance training does the same for older muscle stem cells is untested; the main evidence is that running made satellite cells better at renewing themselves in young mice (Weak-preliminary; mouse).[72] In the brain, a randomised trial in 120 older adults found that a year of aerobic exercise training enlarged part of the memory centre (the hippocampus) by about 2%, roughly reversing one to two years of age-related shrinkage. Bigger gains went with better spatial memory and with higher blood levels of brain-derived neurotrophic factor (BDNF), a growth factor that supports new-neuron formation in animals (Moderate; human RCT, brain-size surrogate).[73] Whether this works through stem cells in people is unproven (see above). This is the cellular rationale behind resistance training, zone 2 and VO₂ max work, and the muscle-and-bone benefits covered under bone density. Exercise may also matter for the blood-forming pool. In a snapshot of about 91,000 adults who wore activity trackers, each extra hour a day of moderate-to-vigorous activity went with about 14% lower odds of carrying a mutant blood-stem-cell clone driven by genes other than DNMT3A, the most common one (odds ratio 0.86); DNMT3A clones showed no link.[74] In a Chinese cohort of about 6,100 adults followed for 12 years, the extra heart-disease risk that comes with clonal hematopoiesis showed up in people with intermediate or poor lifestyles but was not detectable in those with a healthy lifestyle (combining weight, not smoking, sleep, activity and diet), though that estimate was imprecise.[75] Both studies are observational, so they show an association, not proof that exercise shrinks the clones (Weak-preliminary; human observational, mouse mechanism).

Sleep — protecting the blood-forming pool, so far in mice. In mice, sleep helps keep the bone-marrow stem-cell niche in check. Healthy sleep keeps a wake-promoting brain signal (hypocretin) acting on the marrow, where it restrains the overproduction of inflammatory white cells. Repeatedly broken sleep lowered that signal, made blood stem and progenitor cells divide more (their number roughly doubled), raised inflammatory monocytes and enlarged artery plaques (Moderate; mouse).[76] Whether the same brake works in people is not yet shown. In later mouse work, broken sleep left a lasting mark on how blood stem cells switch their genes on and narrowed the variety of clones making blood; in people, cutting sleep short also altered gene regulation in blood stem cells and stepped up blood-cell production.[77] In mice carrying mutant blood-stem-cell clones, unbroken sleep slowed the clones' growth, but only for some mutations and not for DNMT3A, the most common one in people.[78] No study has shown that poor sleep causes clonal hematopoiesis in people (Weak-preliminary; mostly mouse). Stem cells also follow a daily clock. In old mice, the clock in skin and muscle stem cells keeps ticking, but the genes it switches on and off each day shift away from routine upkeep toward coping with stress such as DNA damage, and long-term calorie restriction held back much of that shift.[79] No study has shown that a regular sleep schedule protects human stem cells, but keeping one aligned to the body's circadian rhythm is a low-cost habit that fits this biology.

Diet and fasting — with an important nuance. Caloric restriction preserves stem cells in part by increasing their protective quiescence and improving the niche, an effect shown across intestinal and muscle stem cells.[80][81] In the gut, the mechanism is well mapped: eating less lowers mTORC1 signalling (the core growth-driving arm of the mTOR pathway) in the support cells surrounding intestinal stem cells, boosting their function — and rapamycin mimics the effect (Moderate; mouse).[82] Intermittent and periodic fasting likewise protect stem cells by lowering inflammation and inducing the cell's self-cleaning autophagy.[83] But there is a real caveat in the refeeding. In mice, eating again after a 24-hour fast triggers a burst of gut stem-cell division driven by mTORC1, which ramps up the cells' protein production through polyamines. When a cancer-starting gene was knocked out during that refeeding window, more tumours formed in the small intestine and colon than in fasted or normally fed mice, and blocking mTORC1, polyamine production or protein production prevented this (Moderate; mouse).[84] This is mouse work and not a reason for healthy people to avoid fasting. It also does not show the safest way to break a fast: the researchers did not compare different amounts or kinds of food at refeeding, and they say fast-refeed cycles need careful testing before they are used to boost regeneration.[85] Separately, a review of cell and animal work warns that very prolonged calorie restriction can push the cell's self-cleaning (autophagy) into a harmful excess that kills cells, which favours moderate regimens over extreme ones.[86]

Supplements aimed at the niche. Several compounds sold as supplements target the same pathways: the senolytics fisetin and quercetin (see cellular senescence), the autophagy-inducing polyamine spermidine, and the mitophagy activator urolithin A (see mitochondrial dysfunction). The rationale is strongest for senolytics, though it rests on mice: injecting a small number of senescent cells into young mice was enough to impair their physical function and spread senescence into the animals' own tissues, and giving very old mice an intermittent dasatinib (a prescription cancer drug) plus quercetin combination improved physical function and lengthened their median remaining lifespan by about a third (36%) (Moderate; mouse).[87] The human trials so far have mostly been disappointing. In a randomised trial in 60 postmenopausal women, 20 weeks of intermittent dasatinib plus quercetin did no better than no treatment on the trial's main bone marker.[88] A year of spermidine did not improve memory in 100 older adults with self-reported memory worries.[89] The urolithin A trials for muscle are small, and pooled together they are inconclusive.[90] In one small 4-week trial funded by the manufacturer, urolithin A shifted blood immune cells toward a younger-looking profile, but that is a lab marker, not a health outcome.[91] None of these trials measured tissue stem cells. At most these compounds are plausible add-ons, not proven ones (Weak-preliminary; mouse rationale, mostly null small human trials).

What this does and doesn't tell you

What it tells you: stem-cell exhaustion is the integrative hallmark where upstream damage becomes visible decline — failing repair in muscle, brain, blood, and gut. Much of it is driven by the niche and by lost quiescence rather than by cells being irreversibly destroyed, which is why an aged stem cell can often be coaxed back toward youthful function. And the levers that help are familiar ones, though their direct stem-cell evidence is uneven. Resistance training measurably increases muscle stem cells in older adults,[92][93] and people who get more moderate-to-vigorous exercise are somewhat less likely to carry certain mutant blood-stem-cell clones, an observational link that cannot show cause and effect.[94] The case for sleep and moderate caloric restriction as stem-cell protectors rests mainly on mouse work. All of them are worth doing for their broader, well-proven health benefits.

What it doesn't tell you: that any stem-cell injection, reprogramming therapy, or supplement is established for healthy adults. The transplant results are early, small and mixed, reprogramming has no human results yet, and both carry risk; the small-molecule data is largely preclinical; and even fasting carries a mouse-demonstrated refeeding caveat worth respecting.

It also doesn't tell you how much real harm the gap between hype and evidence has already done. In 2021, more than 2,700 clinics in the United States, run by about 1,500 businesses, were selling stem-cell treatments the US Food and Drug Administration (FDA) has never approved, more than four times as many businesses as in 2016.[95] The injuries are documented: a 2021 tally by the Pew Charitable Trusts, a US policy group, found reports of at least 360 people worldwide harmed between 2004 and 2020, including 20 deaths, 9 cases of partial or complete blindness and over 100 hospitalizations, and that is probably an undercount because these clinics rarely report harms.[96] In one widely-reported case, three women were blinded after a clinic injected fat-derived "stem cells" into their eyes for macular degeneration (Caution; regulatory reports and case series).[97] The "young plasma" infusions marketed for anti-aging are a related trap: in 2019 the FDA warned that they have no proven benefit against aging or memory loss and carry infection, allergy, lung and heart risks.[98] Don't confuse them with the research on plasma dilution (therapeutic plasma exchange), in which part of a person's own plasma is removed and replaced with an albumin solution. An uncontrolled study of 8 people reported younger-looking blood-protein profiles after repeated exchanges.[99] The first trial with a fake-treatment (sham) comparison group enrolled 44 healthy adults over 50, of whom 42 finished. The procedure was well tolerated. Midway through, it lowered blood-based biological-age estimates by about 1.3 years on average, or about 2.6 years when combined with an immunoglobulin infusion, but by the end of the study the difference from the sham group had disappeared. Participants were assigned in order of sign-up rather than truly at random, no health or function outcomes were measured, and several authors work for a plasma-exchange company (Weak-preliminary; biomarker-only human data).[100]

A recurring theme runs through all of this: almost every mechanism in stem-cell aging — the signalling pathways, the young-blood factors, reprogramming, NAD⁺, senolytics — rests on mouse data. The robust human evidence is mostly observational (the CHIP cohorts, the human satellite-cell counts) or comes from small trials. As with the other hallmarks, the mechanism is elegant and the proven interventions are the unglamorous ones — and "build muscle, sleep well, don't overeat" remains a better stem-cell strategy than anything currently sold as one.

Further reading

  • Ageing and rejuvenation of tissue stem cells and their niches.[101]
  • Editorial: stem cell exhaustion in aging.[102]
  • Adult stem cells and diseases of aging.[103]
  • Recent clinical trials with stem cells to slow or reverse normal aging processes.[104]
  • From bench to bedside: translating cellular rejuvenation therapies into clinical applications.[105]
  • Recent progress in stem cell and immune cell-based interventions for aging and age-related disorders.[106]
  • The resistance training effects on skeletal muscle stem cells in older adults — a meta-analysis.[107]
  • Prolonged exercise training improves the acute type II satellite cell response in older men.[108]
  • Sleep modulates haematopoiesis and protects against atherosclerosis.[109]
  • Caloric restriction maintains stem cells through niche and regulates stem cell aging.[110]
  • Short-term post-fast refeeding enhances intestinal stemness via polyamines.[111]
  • Clonal dynamics of haematopoiesis across the human lifespan (Nature 2022).[112]
  • Clonal Hematopoiesis and Risk of Atherosclerotic Cardiovascular Disease (NEJM 2017).[113]
  • Randomized phase 2b trial of stem cell therapy (laromestrocel) for aging frailty.[114]
  • Harms linked to unapproved stem-cell interventions (Pew Charitable Trusts 2021).[115]

— § —