The 12 Hallmarks of Aging Explained (2026): Science, Mechanisms, and Interventions

Photo: Aysegul Aytoren / Pexels
Educational disclaimer: This article is for informational and educational purposes only. It is not medical advice. Do not start, stop, or alter any supplement, medication, or treatment based on this content. Consult a qualified healthcare provider for any health decisions.
In January 2023, Carlos López-Otín, Maria Blasco, Linda Partridge, Manuel Serrano, and Guido Kroemer updated the best-known framework in aging biology: the hallmarks of aging, now numbering 12. Each hallmark is a distinct molecular process, from DNA damage to gut microbiome drift, that worsens with age, speeds up aging when pushed experimentally, and might be slowed by intervention. This article sorts the 12 into the paper's three tiers, explains the mechanism behind each, and separates what has been shown in mice from what has been shown in people.
Why aging is a biological process and not just wear and tear
Anyone who has owned a car knows what wear and tear looks like: parts degrade at a rate set by use. Human aging does not behave that way. In adults, the risk of dying roughly doubles about every eight years, a pattern first described by the British actuary Benjamin Gompertz in 1825 and still holding across populations (Gompertz-Makeham law of mortality). A pattern that regular suggests an underlying process with its own rules, not a random pile-up of broken parts.
The first direct evidence that cells carry an internal clock came in 1961, when Leonard Hayflick and Paul Moorhead reported in Experimental Cell Research that cultured human fetal cells stopped dividing after roughly 50 population doublings, even though they were still metabolically alive. That ceiling, later named the Hayflick limit, contradicted the then-dominant view that cells could divide forever if fed properly (Embryo Project, Arizona State University). The mechanism behind it, telomere shortening, was worked out over the following decades and earned the 2009 Nobel Prize in Physiology or Medicine for Elizabeth Blackburn, Carol Greider, and Jack Szostak (Nobel Prize press release).
A second landmark came in the early 1990s. Cynthia Kenyon's group reported that mutations in a single gene, daf-2, let the roundworm C. elegans live roughly twice as long as normal (Kenyon et al., Nature, 1993). Aging, it turned out, could be tuned by changing a gene. That result moved the field from asking why bodies fail to asking which molecular dials set the pace.
What counts as a hallmark, and how the 2023 update reorganized them
The original framework, published in Cell in June 2013, named nine hallmarks. The 2023 update expanded it to 12 by adding disabled macroautophagy, chronic inflammation, and dysbiosis, and by widening "DNA damage" into the broader category of genomic instability. The authors kept the same three tests for what qualifies. A hallmark must appear with age, speeding it up experimentally must accelerate aging, and treating it must offer a chance to slow, halt, or reverse aging (López-Otín et al., Cell, 2023).
That third test is why the framework matters outside the lab. A hallmark is not only a description of what goes wrong; it is a candidate drug target. The test is also strict, which is why many popular "aging factors" never make the list.
The 12 are arranged in three tiers, which the Jackson Laboratory summarizes as damage, response, and consequence (Jackson Laboratory). Think of a city after a storm. Primary hallmarks are the storm damage itself: broken pipes and downed lines. Antagonistic hallmarks are the emergency repairs, which help at first but cause new problems if they stay in place for years. Integrative hallmarks are the citywide outages that result when the damage and the patchwork accumulate faster than crews can respond.
| Tier | Hallmarks | Role in the framework |
|---|---|---|
| Primary | Genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, disabled macroautophagy | Causes of cellular damage |
| Antagonistic | Deregulated nutrient sensing, mitochondrial dysfunction, cellular senescence | Responses to damage that turn harmful when chronic |
| Integrative | Stem cell exhaustion, altered intercellular communication, chronic inflammation, dysbiosis | Tissue-level outcomes of the other two tiers |
Tier assignments follow the grouping in Wikipedia's "Hallmarks of aging". Secondary descriptions, including the Jackson Laboratory page, differ on a few placements, so treat the tiers as a teaching aid rather than a strict taxonomy.
The five primary hallmarks: where the damage starts
1. Genomic instability
Every cell sustains DNA damage constantly. Published estimates run to tens of thousands of lesions per mammalian cell per day, with single-strand breaks the most common type, from oxidative chemistry, replication errors, and environmental exposure (rates compiled on Wikipedia). Almost all of it is fixed within hours by repair systems with names like base excision repair and homologous recombination.
The trouble is the small remainder. Repair is not perfect, and its fidelity appears to decline with age, so unrepaired errors become permanent mutations that get copied into every daughter cell. In tissues that renew constantly, such as the gut lining and blood, mutations accumulate and raise cancer risk. In cells that rarely divide, such as neurons and heart muscle, damage accumulates as dysfunction.
Genomic instability also covers damage outside the nucleus. Mitochondria carry their own small genome with more limited repair options, a point that returns under mitochondrial dysfunction below. For a deeper look at the repair machinery, see our explainer on how DNA replication works.
2. Telomere attrition
Telomeres are stretches of repeated DNA (the sequence TTAGGG in humans) that cap the ends of chromosomes, like the plastic tips on shoelaces. Every time a cell divides, the copying machinery cannot quite reach the very end of the chromosome, so the cap gets a little shorter. Most adult cells have switched off telomerase, the enzyme that rebuilds telomeres, so the caps wear down. When they get critically short, the cell stops dividing or dies, which is the mechanism behind the Hayflick limit described earlier. Our guide to telomeres and telomerase covers the biochemistry in more detail.
The 2023 paper treats telomere attrition as a hallmark that can be separated from general DNA damage, because mouse experiments tie telomere length directly to lifespan. Mice with shortened or lengthened telomeres show decreased or increased lifespan, respectively. In 2012, Blasco's group at the Spanish National Cancer Research Centre (CNIO) delivered the telomerase gene (TERT) to adult and old mice using a viral vector. Treatment at one year of age raised median lifespan by 24 percent, and at two years of age by 13 percent, with no increase in cancer (Bernardes de Jesus et al., EMBO Molecular Medicine, 2012). In 2019 the same lab reported that mice derived from stem cells with naturally longer telomeres lived about 13 percent longer at the median, with a lower tumor incidence that did not reach statistical significance (Muñoz-Lorente et al., Nature Communications, 2019).
Mouse telomeres are not human telomeres, and telomerase is also switched on in most cancers, which is why lengthening telomeres in people is not a simple proposition. Nothing in the human literature currently shows that lengthening telomeres extends lifespan.
3. Epigenetic alterations
If DNA is the score for a symphony, the epigenome is the set of performance markings written over it: which passages to play loudly, which to skip. Chemical tags such as DNA methylation and modifications to the histone proteins that DNA wraps around decide which genes are active in a given cell type. With age, those markings drift. Some regions lose methylation, others gain it, and the clean pattern that distinguishes a liver cell from a skin cell blurs.
That drift is measurable. Steve Horvath's 2013 paper in Genome Biology, "DNA methylation age of human tissues and cell types," showed that methylation levels at a defined set of sites predict chronological age across many tissues, and later "clocks" were trained to predict health outcomes instead of birthdays. Two examples are PhenoAge, built on a composite of clinical measures, and GrimAge, built on methylation proxies for plasma proteins and smoking history, which has outperformed other clocks at predicting mortality in several cohorts (PubMed record for the GrimAge comparison).
The 2023 authors are careful here. They state that there is no definitive evidence that methylation changes cause aging; the clocks may read the symptoms of aging rather than a cause. They also note that transient expression of reprogramming factors rejuvenates aging markers in cells and in some tissues, and has extended the lifespan of mice with accelerated-aging disease, but that "extension of longevity by OSKM has not yet been reported for wild-type mice" (as of the paper's writing).
The strongest data come from the eye. In a 2020 Nature paper, David Sinclair's group delivered three reprogramming factors (Oct4, Sox2, and Klf4, without the cancer-linked fourth factor c-Myc) to retinal ganglion cells in mice. Treatment reversed vision loss in a glaucoma model and in aged mice. Of 464 genes that changed with age, roughly 90 percent returned to youthful levels, and the benefit depended on the DNA-demethylating enzymes TET1 and TET2 (Lu et al., Nature, 2020). This is partial reprogramming at its most convincing, and it is still a mouse eye.
Biotech companies such as Altos Labs and Retro Biosciences have publicly pursued reprogramming-based therapies. Human efficacy data for partial reprogramming had not been published as of this writing.
4. Loss of proteostasis
Proteostasis is a cell's quality-control system for proteins. A protein is a long chain that must fold into an exact three-dimensional shape to work. Chaperone proteins help new chains fold, and two disposal systems, the ubiquitin-proteasome system and autophagy, remove the ones that fail. It is like a factory with an inspection line and two kinds of waste removal.
With age, every stage degrades. Chaperones become less abundant, the proteasome loses capacity, and misfolded proteins clump into aggregates. Those aggregates define several late-life neurological diseases: amyloid-beta and tau in Alzheimer's disease, alpha-synuclein in Parkinson's disease, huntingtin in Huntington's disease, and TDP-43 in amyotrophic lateral sclerosis. Whether aggregation is the cause or a symptom differs by disease and remains debated. Our overview of Alzheimer's disease biology walks through that debate.
5. Disabled macroautophagy
Macroautophagy ("self-eating") is the cell's recycling program. A membrane wraps around worn-out organelles or protein clumps, forms a sac called an autophagosome, and delivers its cargo to the lysosome, the cell's digestive compartment, where it is broken into reusable parts. Proteostasis was listed as a single hallmark in 2013; the 2023 update separated autophagy because its failure affects far more than proteins, including damaged mitochondria and invading microbes.
Animal data link autophagy to lifespan directly. Overexpressing the autophagy gene Atg5 extended lifespan in mice, while knocking it down caused premature degeneration, and oral spermidine, a polyamine that induces autophagy, extended mouse longevity by up to 25 percent (both findings as summarized in López-Otín et al., 2023).
Human data are thinner. In the SmartAge trial, 100 older adults with subjective cognitive decline took a wheat-germ spermidine supplement (0.9 mg per day) or placebo for 12 months. The authors concluded that spermidine did not change memory performance or biomarkers compared with placebo; an exploratory signal in one inflammation marker would need replication (JAMA Network Open, 2022).
The three antagonistic hallmarks: responses that overstay
6. Deregulated nutrient sensing
Cells monitor what they have to spend. Four linked pathways do the watching: mTOR (which senses amino acids and growth signals), AMPK (which senses low energy), the sirtuins (which depend on the molecule NAD+), and insulin/IGF-1 signaling (which tracks glucose and growth hormone). Plenty of food and growth signals push cells toward building and dividing. Scarcity pushes them toward maintenance and repair.
The oldest result in the field is that turning down the growth side extends lifespan. Kenyon's daf-2 worms were the first example, and the same logic holds in flies and mice. Rapamycin, a drug that inhibits mTOR, is the best-studied case in mammals. In the 2009 study by Harrison and colleagues, mice started on rapamycin at 600 days of age (roughly late middle age for a mouse) lived longer, with the age at 90 percent mortality up 14 percent in females and 9 percent in males, and the effect was replicated at three independent test sites (Harrison et al., Nature, 2009). Those figures are well below the 25 percent sometimes quoted in popular coverage.
Metformin, a diabetes drug that activates AMPK, has a weaker mouse record. In the National Institute on Aging's Interventions Testing Program, metformin alone at 0.1 percent of the diet did not significantly extend lifespan; combined with rapamycin it did, though that comparison relied on historical rapamycin-only controls (Strong et al., Aging Cell, 2016). The 2023 hallmarks paper itself says only that metformin "has been discussed as a possible anti-aging drug."
7. Mitochondrial dysfunction
Mitochondria burn fuel with oxygen to make ATP, the cell's energy currency, and they also help trigger programmed cell death and manage calcium signals. They carry their own DNA, a legacy of their origin as free-living bacteria, and they are more exposed to damage than nuclear DNA. Our explainer on how mitochondria became the cell's powerhouses tells that story, and how ATP works covers the chemistry.
With age, mitochondria become less efficient, their DNA accumulates mutations, and the process that clears damaged ones (mitophagy, a selective form of autophagy) slows down. The older "free radical theory" blamed aging on damage from reactive oxygen species, but reactive oxygen also works as a signal inside cells, which is one reason the 2023 paper frames the hallmark as dysfunction and not as oxidative damage alone.
Exercise is the most widely recommended behavior that acts on mitochondria, and it is discussed further below. For urolithin A, a gut-microbe metabolite of compounds in pomegranate, the human data are mixed. A 2022 randomized trial in 66 adults aged 65 to 90 gave 1,000 mg daily for four months. Muscle function improved in some measures, but six-minute walk distance and muscle ATP production measured by magnetic resonance spectroscopy did not differ significantly from placebo (Liu et al., JAMA Network Open, 2022).
8. Cellular senescence
A senescent cell is one that has permanently stopped dividing in response to stress such as telomere shortening, DNA damage, or oncogene activation. That is a useful safety feature, since it blocks damaged cells from becoming tumors. The problem is what senescent cells do next. They stay alive and release a mix of inflammatory proteins, enzymes, and growth factors known as the senescence-associated secretory phenotype (SASP). Imagine a retired employee who refuses to leave the building and keeps sending urgent emails to everyone, which unsettles the whole office.
Young bodies clear senescent cells efficiently. With age, clearance falters. The 2023 paper reports that senescent cells accumulate 2-fold to 20-fold when comparing healthy donors under 35 with those over 65.
The causal evidence comes from mice. In 2011, Darren Baker and Jan van Deursen's lab built a transgene called INK-ATTAC that kills cells expressing the senescence marker p16, and showed that removing them delayed age-related disorders in fast-aging mice. In 2016 they repeated the approach in normally aging mice, starting at one year of age, and found extended median lifespan in both sexes, along with delayed tumors and less deterioration in kidney, heart, and fat (Baker et al., Nature, 2016). The authors noted the trade-off: senescent cells also help with wound repair, so clearing them is not free.
Drugs that kill senescent cells selectively are called senolytics. The first combination, dasatinib plus quercetin (D+Q), came from James Kirkland's group at the Mayo Clinic in 2015, based on the finding that senescent cells depend on pro-survival pathways that these drugs block (Zhu et al., Aging Cell, 2015). Dasatinib is approved for certain leukemias, quercetin is a plant flavonoid, and the combination is not approved for any aging-related use. Human evidence so far is small. A 2019 open-label pilot gave 14 people with idiopathic pulmonary fibrosis intermittent D+Q for three weeks; walking distance, gait speed, and chair-stand time improved, but lung function, frailty index, and self-reported health did not change, and the SASP biomarkers were inconclusive (Justice et al., EBioMedicine, 2019). It had no control group, so a placebo effect cannot be ruled out.
The four integrative hallmarks: when the system-level symptoms appear
9. Stem cell exhaustion
Tissues renew themselves from small reserves of stem cells: blood stem cells in the marrow, satellite cells in muscle, crypt cells in the intestine, and neural stem cells in the hippocampus. With age these pools shrink or shift in character. Blood stem cells turn out a narrower range of immune cells, muscle repair slows, and the gut lining regenerates less readily. For background on how these cells decide their fates, see how stem cells know what to become and what stem cells can actually treat.
Stem cell exhaustion sits in the integrative tier because it is mostly a downstream effect. DNA damage, epigenetic drift, inflammation, and senescent neighbors that poison the local environment all feed into it. That also makes it hard to treat directly: replacing stem cells does little if the niche around them stays hostile.
The research behind reprogramming goes back to Shinya Yamanaka's discovery that four factors can turn adult cells into stem-like cells, covered in our explainer on Yamanaka's iPSC experiments. The factors used in partial reprogramming, described earlier, are borrowed from that work.
10. Altered intercellular communication
Cells talk through hormones, nerve signals, and molecules carried in the blood. With age that conversation changes: levels of growth hormone and sex hormones fall, stress hormones and inflammatory molecules rise, and the blood carries a different mix of proteins. Our explainers on cortisol and the stress response and neurotransmitters cover two of those channels.
The classic experiment here is heterochronic parabiosis, in which the circulatory systems of a young and an old mouse are joined surgically. Old tissues often look healthier after sharing blood with a young partner, and young tissues look worse after sharing with an old one. That suggests circulating factors matter, though the specific molecules proposed as the active ingredients remain disputed.
Human trials of the idea have been modest. The AMBAR study randomized 347 people with mild-to-moderate Alzheimer's disease to plasma exchange with albumin replacement, which swaps out the patient's plasma, or a sham procedure. At 14 months the treated groups declined less on a daily-living scale, with benefit seen in moderate but not mild disease, which is a signal worth testing again, not proof (Boada et al., Alzheimer's & Dementia, 2020). It is a different intervention from infusing young plasma, and it should not be mistaken for one.
11. Chronic inflammation (inflammaging)
Acute inflammation is the body's fire alarm: loud, local, and temporary. Inflammaging is a smoke detector stuck on low, a persistent, low-grade, sterile (no infection required) inflammatory state that rises with age. Its sources overlap with other hallmarks: SASP from senescent cells, DNA released from stressed mitochondria, a leaky gut barrier, and an immune system that regulates itself less precisely. For the immune side of the story, see how the immune system works.
Blood levels of markers such as interleukin-6 and C-reactive protein predict disease and mortality in older people, which is why inflammation is treated as a hallmark, not a side effect. The strongest causal clues come from targeted blocking. In mice, blocking TNF-alpha with etanercept prevented muscle wasting and extended lifespan in females. In people, the CANTOS trial found that canakinumab, which blocks interleukin-1 beta, reduced recurrent cardiovascular events in heart-attack survivors with high C-reactive protein (both as summarized in the 2023 paper). That shows inflammation is a modifiable driver of at least one age-related disease. It does not show that dampening inflammation slows aging overall, and blocking immune signals carries infection risk.
12. Dysbiosis
The gut hosts trillions of microbes, and their balance shifts with age. The 2023 paper notes that the human gut microbiota changes significantly during aging, with lower ecological diversity. A less diverse, more inflammatory community and a more permeable gut lining let microbial products reach the bloodstream and feed inflammaging. Our overview of the gut microbiome explains the basic ecology.
The cleanest causal test is again in mice. In a 2022 study from the Quadram Institute, fecal transfers between young and old mice worked in both directions: old-donor microbes accelerated brain inflammation, eye changes, and gut permeability in young recipients, and young-donor microbes reversed those changes in old ones. Mouse transfer results may not carry over to people (Parker et al., Microbiome, 2022). For people, the best-supported microbiome-related habits remain unglamorous ones, such as eating plenty of dietary fiber, but those have not been shown to slow aging itself.
What the human evidence shows, intervention by intervention
Mouse lifespan data are plentiful and human lifespan data are almost nonexistent, because a trial that waits for people to die would run for decades. Researchers use surrogate endpoints instead: epigenetic clocks, inflammatory markers, strength, and function. The table summarizes the best-documented candidates and where each stands, with sources and years.
| Intervention | Hallmark(s) targeted | Best evidence (source, year) | Human status |
|---|---|---|---|
| Rapamycin | Nutrient sensing, autophagy | Mice started at 600 days: age at 90% mortality up 14% (females), 9% (males) (Harrison, 2009) | PEARL trial (2025): 114 adults aged 50 to 85, 5 or 10 mg weekly for 48 weeks; missed its primary endpoint (visceral fat); secondary gains in lean mass and pain in women |
| Metformin | Nutrient sensing (AMPK), inflammation | No lifespan gain alone in the NIA Interventions Testing Program (Strong, 2016) | TAME trial planned for about 3,000 people aged 65 to 79; still only partially funded as of an April 2024 report |
| Dasatinib + quercetin | Cellular senescence | Pilot, 14 people with pulmonary fibrosis; function improved, lung function did not (Justice, 2019) | Small, open-label studies; not approved for aging |
| NMN or other NAD+ precursors | Genomic instability, mitochondria, sirtuins | 10-week trial, 25 prediabetic women; muscle insulin sensitivity rose, other metabolic markers did not (Yoshino, 2021) | Raises NAD+; no proof of anti-aging effect |
| Urolithin A | Mitochondrial dysfunction | 66 adults aged 65 to 90, four months; mixed results; walk distance and muscle ATP not significant (Liu, 2022) | Sold as a supplement; larger trials needed |
| Spermidine | Autophagy | Mouse longevity up to 25%; SmartAge trial, 100 adults: no significant cognitive effect (2022) | Unproven in people |
| Calorie restriction | Several | CALERIE, 220 adults aged 21 to 50: DunedinPACE slowed 2 to 3%, PhenoAge and GrimAge unchanged (Waziry, 2023) | Participants achieved about 12% restriction, not the 25% target |
| Exercise | Mitochondria, inflammation, stem cells | WHO: insufficiently active people have a 20% to 30% higher risk of death | Recommended at least 150 minutes of moderate activity per week |
| Partial reprogramming | Epigenetic alterations | Mouse retina: vision loss reversed in a glaucoma model and in aged mice (Lu, 2020) | No human trials |
A few patterns stand out. First, the results in humans are modest even when the mouse results are striking. Second, the best-supported intervention is the least glamorous: exercise has the broadest evidence and no patent holder to market it. The WHO fact sheet on physical activity recommends at least 150 minutes of moderate-intensity activity per week for adults and reports a 20 to 30 percent higher risk of death among insufficiently active people (WHO). Our piece on the science of exercise explains why it touches so many hallmarks at once.
What the calorie-restriction trial did and did not show
Calorie restriction is the oldest lifespan intervention, and CALERIE is the longest controlled test in humans. It randomized 220 healthy, non-obese adults aged 21 to 50 to a prescribed 25 percent reduction or to eating as they pleased for two years. A 2023 analysis in Nature Aging found that the restricted group's DunedinPACE score, a methylation-based estimate of the pace of aging, slowed by about 2 to 3 percent, while two other clocks (PhenoAge and GrimAge) did not differ between groups. Participants managed an average reduction of 11.9 percent, about half the target, and individual responses overlapped widely between groups (Waziry et al., Nature Aging, 2023). The authors cautioned that it remains unknown whether such changes translate into less disease or longer life, and several of them are inventors of DunedinPACE.
That is a good illustration of how to read this field. A real effect on one measure, a null result on two others, a short follow-up, and a conflict-of-interest disclosure. None of that makes the result useless, but none of it makes it a longevity protocol either. Our article on intermittent fasting applies the same scrutiny to a related approach.
Lifespan versus healthspan, and why clocks are not clinical tools yet
Most researchers in the field no longer aim to push maximum age upward at any cost. The goal is healthspan: more years free of serious disease and disability, with the unhealthy period at the end compressed. A drug that added two years of life but all of them in frailty would count as a failure by that standard.
The trouble is measuring it. Epigenetic clocks are the most prominent biomarkers, and they do track mortality risk in large cohorts. But they differ in what they measure, they can disagree with each other, as the CALERIE analysis shows, and none has been validated as a stand-in for lifespan in a drug trial. Until one is validated, a company can show that a drug moves a clock without having shown that the drug makes anyone healthier.
That is also why the question of classifying aging as a disease keeps surfacing. Some researchers argue that recognizing aging as a treatable process would speed drug development, while others hold that labeling normal aging a pathology is a mistake. The argument is as much about regulation and drug approval as about biology. For more on that debate, see why aging is a disease and whether it could be cured and the longevity interventions that actually work.
What people get wrong about the hallmarks
"The hallmarks are 12 independent causes." They are not. The authors stress that the hallmarks are interconnected, and the integrative tier exists precisely because the others feed into it. Senescent cells drive inflammation, inflammation damages stem cell niches, and mitochondrial damage releases DNA that triggers inflammation. Treating one in isolation may help less than expected.
"If it extends mouse lifespan, it will extend yours." Mouse results are a starting point. Rapamycin extends mouse lifespan reliably, but the human PEARL trial missed its primary endpoint. Mice live only a couple of years and are housed in conditions very unlike human life, so many compounds that work in mice never translate to people.
"Supplements that raise a biomarker are anti-aging." NAD+ precursors raise NAD+ levels reliably, but a higher level of one molecule is not the same as slower aging. The Yoshino trial found a gain in muscle insulin sensitivity in 25 women with prediabetes, and not in other metabolic measures.
"Aging is a single clock we can reset." Partial reprogramming is an exciting result, but it was shown in mouse tissue, mainly the retina, with delivery methods that cannot yet be used safely across the human body. The 2023 authors themselves note that lifespan extension through reprogramming has not been shown in normal mice.
"A longer life means a longer period of illness." The compression-of-morbidity framing argues the opposite is the goal, and the shape of the Gompertz curve supports why it is hard: the risk of dying rises with age regardless of the cause, so slowing the underlying process, not just treating one disease, is what could shift the curve.
Frequently Asked Questions
Q: What are the 12 hallmarks of aging? A: According to López-Otín and colleagues (Cell, 2023), they are genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, disabled macroautophagy, deregulated nutrient sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, altered intercellular communication, chronic inflammation, and dysbiosis. The 2013 version had nine; the 2023 update added macroautophagy, inflammation, and dysbiosis.
Q: What is the difference between primary, antagonistic, and integrative hallmarks? A: Primary hallmarks cause cellular damage, such as DNA lesions and protein aggregation. Antagonistic hallmarks are responses to that damage, such as cellular senescence, that help at first but harm when chronic. Integrative hallmarks, including stem cell exhaustion and inflammation, are tissue-level consequences of the other two groups.
Q: Does rapamycin extend human lifespan? A: Unknown. Rapamycin extended lifespan in mice in the 2009 NIA-funded study, raising the age at 90 percent mortality by 9 percent in males and 14 percent in females. In humans, the 48-week PEARL trial (114 adults) missed its primary endpoint of reducing visceral fat, and no trial has measured human lifespan. Rapamycin is an immunosuppressant with real side effects, so it is not a supplement.
Q: What are senolytics, and have they worked in people? A: Senolytics are drugs that selectively kill senescent cells. Dasatinib plus quercetin cleared senescent cells in mice and improved physical function in a 14-person pilot study in pulmonary fibrosis, but that study had no control group and no lung function benefit. Larger randomized trials are needed, and no senolytic is approved to treat aging.
Q: Are epigenetic clocks reliable measures of biological age? A: They predict mortality and disease risk in large groups, but different clocks measure different things and can disagree, as in the CALERIE trial where one clock moved and two did not. They are research tools, not diagnostic tests, and the authors of the 2023 review say there is no definitive evidence that methylation changes cause aging.
Q: Do NAD+ supplements reverse aging? A: There is no evidence that they do. NAD+ precursors raise NAD+ in people, and a 10-week trial in 25 prediabetic women found improved muscle insulin sensitivity, but no change in blood glucose, blood lipids, or liver measures. Larger and longer trials would be needed to test any effect on aging.
Q: What is the best-supported way to slow aging today? A: Regular physical activity has the broadest evidence. The WHO recommends at least 150 minutes of moderate-intensity activity per week for adults and reports a 20 to 30 percent higher risk of death among people who are insufficiently active. That is an association from observational data and a public-health recommendation, not a guarantee for any individual.
The takeaway
The hallmarks framework does something unusual for a field with so many headlines: it gives a map. Each new story about a longevity drug, a supplement, or a reprogramming result lands on one or more of the 12, and the map shows whether the claim addresses a cause, a response, or a consequence. It also reveals the gap that matters most. For nearly every hallmark, the evidence that moving it extends life comes from worms, flies, and mice, while the human evidence is small trials measuring surrogate endpoints. Rapamycin, senolytics, and reprogramming are real science worth following, and so is the discipline of asking, for each claim, what was measured, in whom, and for how long.
Sources
- López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M., and Kroemer, G. "Hallmarks of aging: An expanding universe." Cell, 186(2), 243-278, January 2023. Full text
- López-Otín, C. et al. "The hallmarks of aging." Cell, 153(6), 1194-1217, June 2013.
- Jackson Laboratory, "Hallmarks of aging" and Wikipedia, "Hallmarks of aging", for the three-tier grouping.
- Harrison, D. E. et al. "Rapamycin fed late in life extends lifespan in genetically heterogeneous mice." Nature, 460, 392-395, 2009.
- Strong, R. et al. "Longer lifespan in male mice treated with a weakly estrogenic agonist, an antioxidant, an alpha-glucosidase inhibitor or a Nrf2-inducer." Aging Cell, 15(5), 2016 (NIA Interventions Testing Program, includes metformin results).
- Barzilai, N. et al. "Metformin as a Tool to Target Aging." Cell Metabolism, 23(6), 1060-1065, 2016 (TAME trial background).
- Fight Aging!, "The TAME Trial for Metformin Remains Only Partially Funded," April 2024.
- Lu, Y. et al. "Reprogramming to recover youthful epigenetic information and restore vision." Nature, 588, 124-129, 2020.
- Baker, D. J. et al. "Naturally occurring p16Ink4a-positive cells shorten healthy lifespan." Nature, 2016.
- Zhu, Y. et al. "The Achilles' heel of senescent cells: from transcriptome to senolytic drugs." Aging Cell, 14(4), 644-658, 2015.
- Justice, J. N. et al. "Senolytics in idiopathic pulmonary fibrosis: Results from a first-in-human, open-label, pilot study." EBioMedicine, 2019.
- Waziry, R. et al. "Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial." Nature Aging, 2023.
- Bernardes de Jesus, B. et al. "Telomerase gene therapy in adult and old mice delays aging and increases longevity without increasing cancer." EMBO Molecular Medicine, 4(8), 691-704, 2012.
- Muñoz-Lorente, M. A. et al. "Mice with hyper-long telomeres show less metabolic aging and longer lifespans." Nature Communications, 10, 4723, 2019.
- Boada, M. et al. AMBAR primary results. Alzheimer's & Dementia, 16(10), 1412-1425, 2020.
- Parker, A. et al. "Fecal microbiota transfer between young and aged mice reverses hallmarks of the aging gut, eye, and brain." Microbiome, 2022.
- SmartAge spermidine trial. JAMA Network Open, 5(5), e2213875, 2022.
- Liu, S. et al. Urolithin A trial in older adults. JAMA Network Open, 2022 (summary).
- World Health Organization, "Physical activity" fact sheet.
- Nobel Prize in Physiology or Medicine 2009, press release.
- Horvath, S. "DNA methylation age of human tissues and cell types." Genome Biology, 14, R115, 2013.
- Hayflick, L., and Moorhead, P. S. "The serial cultivation of human diploid cell strains." Experimental Cell Research, 25, 585-621, 1961.
Related Articles
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

How Symbiosis Works: Mutualism, Commensalism, and Parasitism Explained
Symbiosis is one of the most powerful forces in evolution. From the bacteria in your gut to the fungal partners of nearly every land plant, the living world is a web of intimate partnerships, and not all of them are friendly.

How Vaccines Actually Work: The Immune System Story
How vaccines work: from your first encounter with an antigen to lifelong immunity. Covers mRNA tech, adjuvants, herd immunity math, and 2026's cancer vaccine pipeline.

Why We Age — The Biology of Getting Old
The 12 hallmarks of aging, why evolution allowed us to grow old, and what interventions — from rapamycin to reprogramming — actually have evidence.

Longevity Interventions: What Actually Works
Evidence-based review of longevity interventions — from exercise and sleep to metformin, NMN, rapamycin, and what's genuinely overhyped.

What If Humans Could Live to 1,000 Years? The Science of Radical Longevity
Some scientists genuinely believe today's children could live for centuries. Here's what aging actually is, what we'd need to defeat it, and what a 1,000-year lifespan would mean for humanity.

The Longest-Living Animals on Earth: From 500-Year Clams to the Immortal Jellyfish
In 2006, researchers dredged a clam from the seabed off Iceland and counted the rings on its shell. The number stopped them cold: 507 years. The animal, later nicknamed Ming, had been alive since before Columbus sailed. It is the oldest individual animal ever reliably aged — but it isn't even the strangest entry on this list. One creature appears to cheat death entirely, reversing its own aging on demand.