The Longevity Revolution

The Twelve Hallmarks — A Unified Theory of Why We Die

Until relatively recently, ageing was treated less as a biological process and more as a philosophical inevitability — entropy given flesh. The field lacked a unified framework that could tell researchers where to intervene. That changed dramatically in 2013 when Carlos López-Otín and colleagues published "The Hallmarks of Aging" in Cell, cataloguing nine interconnected molecular processes that collectively drive biological decline. In 2023, an updated paper expanded the framework to twelve hallmarks, adding disabled macroautophagy, chronic inflammation, and dysbiosis to the original nine.

The twelve hallmarks 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 crucial insight this framework delivers is not merely descriptive — it is actionable. Each hallmark is, in principle, a target for intervention. The race to address them, simultaneously and systematically, has become the central project of modern longevity science.

Yamanaka's Nobel — and the Discovery That Changed Everything

In 2006, Shinya Yamanaka made one of the most consequential discoveries in the history of biology: introducing just four transcription factors — OCT4, SOX2, KLF4, and c-MYC, now known simply as the Yamanaka factors — into an adult skin cell was sufficient to revert it to a pluripotent stem cell state, erasing decades of accumulated cellular age. Yamanaka won the Nobel Prize in Physiology or Medicine in 2012 for this work. But the implications went further than regenerative medicine. If cells could be fully reprogrammed to youthfulness, could they also be partially reprogrammed — rejuvenated without losing their identity? This question became the founding hypothesis of an entire field.

The answer, it turns out, is yes. In 2016, Juan Carlos Izpisua Belmonte and colleagues at the Salk Institute showed for the first time that partial expression of the Yamanaka factors — cycled on and off rather than left permanently activated — could reverse epigenetic age markers in mice with a premature ageing disease, extending their lifespan by approximately 30% without inducing tumours or causing cells to lose their function. The era of in vivo partial reprogramming had begun.

Altos Labs — $5.5 Billion and the Most Ambitious Ageing Programme in History

In 2022, Altos Labs launched with $3 billion in seed funding — the largest funding round in biotech history at that time — backed by Jeff Bezos, Yuri Milner, and a constellation of institutional investors who had become convinced that biological rejuvenation was an addressable problem, not a philosophical fantasy. With Izpisua Belmonte as Scientific Founder and a roster of Nobel laureates and leading ageing scientists assembled from across the world, Altos set its sights on a single goal: understanding and reversing cellular ageing through partial reprogramming. By 2025, total funding had reached $5.5 billion, according to Pitchbook, making it the most well-resourced longevity research programme ever mounted.

The science has begun to produce results. In 2024, Altos researchers published work demonstrating that targeted partial reprogramming of age-associated cell states successfully extended lifespan in mouse models. In October 2025, a paper in Cell from Belmonte and colleagues showed that selectively silencing specific genes responsible for "mesenchymal drift" — the tendency of cells to lose their specialised identity with age — could restore a more youthful and regulated cellular state. By August 2025, the company had reportedly begun early human safety testing, marking the first tentative steps toward clinical translation of partial reprogramming in healthy humans.

David Sinclair and the Information Theory of Ageing

Harvard geneticist David Sinclair has proposed one of the most influential frameworks for understanding epigenetic ageing: the Information Theory of Ageing. Sinclair's hypothesis is that ageing is fundamentally a loss of epigenetic information — the intricate methylation and histone modification patterns that tell each cell which genes to express and which to silence. The genome itself remains largely intact, Sinclair argues; what degrades is the cell's ability to read it correctly. In his analogy, the DNA is a scratch-free DVD, but the player is broken. Reprogramming, on this view, is not rewriting the disc — it is cleaning the lens.

Sinclair's lab has produced a series of landmark results consistent with this framework, including a 2020 Nature paper showing that damaged optic nerve cells in mice could be rejuvenated using a modified set of Yamanaka factors (OSK, excluding c-MYC), restoring vision in aged and glaucoma-model mice. The implications — that specific tissues might be targeted for epigenetic rejuvenation without systemic intervention — point toward a generation of precision reprogramming therapies aimed at individual diseases of ageing: vision loss, neurodegeneration, cardiac decline.

The Science of Cellular Senescence — Why Zombie Cells Are So Dangerous

When a cell sustains damage that cannot be repaired — from radiation, oxidative stress, telomere shortening, or oncogenic signals — it can enter a state called cellular senescence: a kind of permanent arrest in which it stops dividing but refuses to die. In youth, the immune system clears these cells efficiently. With age, the clearance slows, and senescent cells accumulate across every tissue in the body. The problem is not simply that these cells are non-functional. It is that they are actively harmful. Senescent cells secrete a toxic cocktail of inflammatory signals, proteases, and growth factors — collectively known as the SASP (Senescence-Associated Secretory Phenotype) — that damage neighbouring healthy cells, promote tumour growth, accelerate atherosclerosis, and drive the chronic low-grade inflammation underlying virtually every major age-related disease.

Animal studies have been dramatic. When researchers at the Mayo Clinic genetically engineered mice to selectively eliminate senescent cells throughout their lifespan, the animals showed significant delays in the onset of age-related diseases — cataracts, muscle wasting, fat loss, kidney dysfunction — and lived measurably longer than controls. This was not slowing ageing at the edges; it was removing one of its core drivers. The question became: could a drug do what genetics had demonstrated in mice?

Dasatinib + Quercetin — The First Senolytic Drug Combination Reaches Human Trials

The answer came from an unlikely pairing: dasatinib, a leukaemia drug already approved by the FDA, and quercetin, a plant-derived flavonoid found in red wine and onions. Researchers at Mayo Clinic discovered in 2015 that together, these two compounds selectively induced apoptosis — programmed cell death — in senescent cells while leaving healthy cells largely unharmed. The combination, abbreviated D+Q, became the first senolytic drug cocktail to enter human clinical trials.

Results have been cautiously encouraging. A Phase 2 trial at Mayo Clinic, published in Nature Medicine in July 2024, enrolled 60 women over 65 with osteoporosis and administered D+Q over 20 weeks. Women with the highest burden of senescent cells showed modest but measurable improvements in bone formation and wrist bone density. A parallel safety trial in patients with early Alzheimer's disease found that dasatinib penetrated the cerebrospinal fluid — meaning it can reach the brain — opening the door to testing senolytics as a treatment for neurodegeneration. Results from the larger Alzheimer's cognition trial are expected to publish through 2025 and 2026.

Unity Biotechnology and the Eye — Senolytics Enter the Clinic for Real-World Disease

While systemic senolytic therapy faces the challenge of delivering drugs safely across the entire body, Unity Biotechnology has pursued a more targeted strategy: senolytic injection directly into the eye for diabetic macular edema, one of the leading causes of blindness in the developed world. Their compound foselutoclax (UBX1325) targets BCL-xL, a protein that helps senescent cells survive. In Phase 2 clinical trial results, a single intraocular injection of foselutoclax produced approximately five additional letters gained on a standard eye chart at 11 months versus placebo — a clinically meaningful improvement in visual acuity in a condition that had previously been treated exclusively with frequent, costly anti-VEGF injections. The result is significant not only for what it achieved in the eye, but as proof of concept that a senolytic approach works in human disease tissue — not just in mice.

Rapamycin — The Most Reproducible Lifespan-Extension Result in Biology

In 2009, the National Institute on Aging's Interventions Testing Program published a landmark result in Nature: rapamycin, a drug originally isolated from soil bacteria on Easter Island and used as an immunosuppressant in organ transplant patients, extended the lifespan of middle-aged mice by 28% in females and 38% in males — even when treatment began late in life, at an age equivalent to roughly 60 in human terms. This result has since been replicated in yeasts, worms, flies, and multiple strains of mice. Among longevity researchers, the rapamycin effect is sometimes described as the closest thing the field has to a law of nature.

Rapamycin works by inhibiting mTOR (mechanistic Target Of Rapamycin), a master nutrient-sensing kinase that acts as the cell's primary throttle between growth and maintenance. When nutrients are abundant, mTOR promotes cell growth and suppresses autophagy — the cellular recycling process that clears damaged proteins and organelles. When nutrients are scarce, mTOR is suppressed and autophagy activates, essentially putting the cell into a high-maintenance, low-growth state that appears to be more resistant to ageing. Rapamycin mimics the molecular signature of caloric restriction — the most robustly validated lifespan-extending intervention in biology — without requiring the organism to actually starve.

The PEARL Trial — The First Rigorous Human Rapamycin Study

Translating the rapamycin effect to humans has been the central challenge of longevity medicine for over a decade. The drug's immunosuppressive effects at high transplant doses raised legitimate safety concerns, but longevity researchers noted that the doses needed to extend lifespan in mice were far lower — typically intermittent, weekly dosing rather than daily high-dose administration. The PEARL trial (Participatory Evaluation of Aging with Rapamycin for Longevity) became the most rigorous attempt to date to evaluate low-dose rapamycin in healthy adults.

Published in 2025, PEARL was a 48-week, double-blinded, randomised, placebo-controlled trial of 114 participants receiving either placebo, 5 mg or 10 mg compounded rapamycin weekly. The primary finding was reassuring on safety: participants in both rapamycin arms showed no clinically significant adverse changes across the full panel of biomarkers tracked — lipid profiles, inflammatory markers, glucose regulation, and organ function all remained within normal ranges. A parallel Oxford pilot study published the same year showed that eight weeks of daily low-dose rapamycin (1 mg/day) measurably reduced p21, a key senescence marker in immune cells, suggesting that even at minimal doses, mTOR inhibition may have downstream effects on cellular senescence. The results position rapamycin for larger, longer trials powered to detect effects on meaningful healthspan endpoints.

Biological Age Clocks — Measuring What the Calendar Cannot

One of the most consequential innovations in ageing research has been the development of epigenetic clocks — machine learning models trained on DNA methylation patterns that can estimate a person's biological age with striking accuracy, independently of their chronological age. The divergence between biological and chronological age — known as age acceleration — has been shown to predict mortality risk, disease onset, and health trajectories in large prospective cohort studies far better than birthdate alone.

The most predictive of these clocks is GrimAge, developed by Steve Horvath and colleagues at UCLA. Unlike earlier generation clocks that estimated age directly, GrimAge was trained to predict time-to-death — making it a composite biomarker of how much biological ageing has actually occurred, independent of when someone was born. A 2025 retrospective cohort study of 1,942 participants from the NHANES dataset confirmed that GrimAge and its successor GrimAge2 effectively predict all-cause mortality risk, outperforming virtually every other clinical biomarker available. The practical implication is profound: for the first time, longevity scientists have a sensitive, quantitative tool to measure whether an intervention is actually slowing ageing in living humans — without needing to wait decades for mortality data.

Insilico Medicine — AI-Designed Longevity Drugs in Human Trials

Insilico Medicine has emerged as the most prominent example of AI-first drug discovery for ageing-related diseases. The company's generative AI platform, PandaOmics and Chemistry42, identifies novel drug targets by analysing multi-omics datasets — genomics, transcriptomics, proteomics — and then designs candidate molecules from scratch against those targets. Insilico's lead programme, ISM001-055, targets a fibrosis pathway identified by its AI as a driver of idiopathic pulmonary fibrosis — a progressive, fatal lung disease closely associated with ageing. The drug was identified, designed, and advanced to clinical trials in approximately 18 months, a fraction of the typical decade-long timeline. By 2025, the compound had completed Phase 2a trials with a safety and efficacy signal sufficient to justify Phase 2b advancement.

More broadly, Insilico's approach points toward a new paradigm in longevity pharmacology: rather than testing existing drugs for new uses, AI systems can exhaustively explore the space of novel molecules targeting the twelve hallmarks of ageing — designing compounds that have never existed before, optimised simultaneously for target affinity, metabolic stability, safety, and manufacturability.

Deep Aging Clocks and Multi-Modal Biology

The next generation of biological age measurement is moving beyond DNA methylation alone. Deep Aging Clocks — neural network models trained on transcriptomics, metabolomics, microbiome composition, and even medical imaging — are being developed to capture the full spectrum of biological change that occurs with ageing, not merely the epigenetic layer. A 2025 review in Ageing Research Reviews catalogued the current landscape of deep aging clocks, noting that imaging-based clocks trained on retinal photographs and brain MRI scans can estimate biological age from a single non-invasive image — potentially enabling population-scale ageing surveillance at clinical visit frequency rather than research-lab cost. The convergence of multi-modal biological clocks with AI drug design creates a closed loop: measure biological age accurately, intervene with AI-designed compounds, remeasure, and iterate — compressing what would otherwise require decades of observation into years of controlled experimentation.