From Epigenetic Speedometers to AI Drug Design, Science Is Learning to Reverse Aging at Its Source

We are living through a rare moment in the history of medicine. The question is no longer whether biological aging can be measured with precision. It can. The question is no longer whether the aging brain retains any capacity for regeneration. It does. And the question is no longer whether artificial intelligence can discover drugs that human researchers have missed. It has. The convergence of these three realities, documented in landmark research over the past several months, is reshaping the science of longevity at a speed that felt theoretical just a few years ago.

Presented By Our Partners

For those tracking the path toward what futurist Ray Kurzweil calls Longevity Escape Velocity, the moment when medicine advances faster than we age, three new studies deserve careful attention. One rewrites the clinical significance of epigenetic clocks by showing that how fast your biological age is changing matters more than where it starts. Another demonstrates, for the first time in humans, that targeted cognitive training can physically reverse cholinergic decline in the aging brain. And a third marks a turning point for AI in medicine, as the first clinical validation of a drug discovered and designed entirely by artificial intelligence arrives in the pages of Nature Medicine. Together, they do not just advance the literature. They shift it.

The Rate of Aging, Made Visible

Science has long understood that chronological age tells only part of the story. Two people born in the same year can be biologically years apart, their cells aging at strikingly different rates depending on diet, exercise, environmental exposures, and genetics. The field of epigenetics has spent the last decade building tools to measure this gap, using patterns of chemical marks called DNA methylation to construct what researchers call epigenetic clocks: mathematical models that read the genome’s surface to estimate how quickly a person is biologically aging.

For years, a critical question remained open: do changes in these clocks over time actually predict mortality, or are they simply correlated with other known risk factors like smoking, obesity, and cardiovascular disease? A study published in Nature Aging in 2026 by Kuo, P.-L. and colleagues provides one of the clearest answers yet. The InCHIANTI study is a population-based cohort of adults from two towns near Florence, Italy, followed for up to 24 years. Researchers evaluated the longitudinal trajectories of multiple generations of epigenetic clocks in 699 participants, including first-generation tools such as the Hannum clock, second-generation tools such as DNAmPhenoAge and DNAmGrimAge, and third-generation pace-of-aging measures including DunedinPACE.

The finding is both intuitive and clinically significant: it is not just where your biological age starts, but how fast it is moving, that determines survival. Faster longitudinal increases in DNAmGrimAge, DNAmGrimAge version 2, DunedinPACE, and several related clocks were robustly and independently associated with higher mortality risk, even after accounting for baseline epigenetic age and all major confounders. Two individuals with similar epigenetic ages today can face dramatically different prognoses depending on the trajectory of their biological aging over time.

The implication is significant. Epigenetic clocks are not just a snapshot of where you are; they are a readout of where you are going and how fast. The field has spent years asking which clock is the most accurate. This research suggests the more important question is whether the clock is accelerating. Modifiable cardiovascular behaviors, including smoking, body mass index, blood glucose, and blood pressure, are known to shape both the level and the rate of epigenetic aging. Which means the trajectory is, at least in part, a choice.

The Brain’s Hidden Regenerative Reserve

At the center of cognitive aging lies a neurotransmitter that most people outside of neuroscience have never heard of: acetylcholine. This chemical messenger coordinates the speed and accuracy of information processing across brain regions critical to attention, memory consolidation, and executive function. The cholinergic system, the network of neurons that produce, release, and respond to acetylcholine, declines progressively with age, by an estimated 2.5 percent per decade across the lifespan. In Alzheimer’s disease, this decline accelerates catastrophically, and the cholinergic hypothesis of Alzheimer’s pathology has anchored a generation of drug development.

For decades, the clinical assumption has been that once cholinergic terminal density begins to fall, medication is the only meaningful lever. A randomized clinical trial from McGill University has now challenged that assumption at its foundation.

The INHANCE trial (Improving Neurological Health in Aging via Neuroplasticity-based Computerized Exercise) enrolled 92 community-dwelling healthy older adults aged 65 and above between July 2021 and December 2023. Participants were randomized to one of two conditions: a speed-based cognitive training program designed to improve the accuracy and rapidity of visual information processing, or an active control consisting of entertainment games that posed no particular cognitive challenge. The researchers used FEOBV-PET imaging, a highly specialized positron emission tomography technique that directly measures vesicular acetylcholine transporter density in living brain tissue, to assess whether the intervention produced any detectable neurochemical change.

Featured Partner

Invest in the Infrastructure Behind Modern Medicine

As healthcare expands beyond hospital walls, the buildings and campuses supporting that shift are generating compelling returns for investors who move early. The Healthcare Real Estate Fund offers qualified investors direct access to a curated portfolio of medical office, outpatient, and specialty care facilities.

Learn More →

Results published in JMIR Serious Games showed that after ten weeks of speed training, participants demonstrated a statistically significant increase in FEOBV binding in the anterior cingulate cortex, a region central to attention, decision-making, and executive control, and a region known to be among the first to show cholinergic deterioration in aging (SUVR change mean +0.044, 95% CI 0.006 to 0.082, P=.03, medium effect size). The researchers noted that this gain of approximately 2.3 percent may offset the estimated 2.5 percent cholinergic decline typically observed over a full decade of natural aging.

This is the first time in human subjects that any behavioral intervention has been shown to directly reverse, rather than merely slow, the loss of cholinergic terminal density in the aging brain. INHANCE is the largest FEOBV-PET trial ever conducted, lending its findings unusual statistical weight. The anterior cingulate cortex does not operate in isolation; its upregulation through speed training appears to cascade into wider networks supporting memory encoding and executive function, precisely the domains most threatened by age-related cognitive decline and most disrupted in early Alzheimer’s disease.

What makes this finding particularly compelling is its accessibility. The intervention required no surgery, no pharmaceutical compound, and no specialized medical facility. It required ten weeks of directed cognitive effort. The neuroplasticity dividend was measurable, neurochemically verified, and significant.

When Artificial Intelligence Designs the Drug

The third piece of this emerging picture arrives from the frontier of AI-accelerated medicine, and it may be the most consequential of the three for what it reveals about the future of disease treatment.

Idiopathic pulmonary fibrosis is a progressive, fatal scarring of the lung tissue that affects roughly 3 million people worldwide. Its median survival from diagnosis is 3 to 5 years. Existing approved therapies can slow the rate of lung function decline, but none has ever demonstrated the ability to stabilize or reverse it. IPF has long been considered one of the most stubborn targets in respiratory medicine, in part because its molecular drivers were not fully understood.

Rentosertib, a small-molecule inhibitor of the protein TNIK (TRAF2- and NCK-interacting kinase), was not discovered by a team of biologists sifting through the literature. It was discovered and designed by a generative AI platform developed at Insilico Medicine. The system identified TNIK as a previously unrecognized disease-relevant target in IPF, then designed candidate drug molecules from scratch to inhibit it. The process from target identification to an optimized clinical candidate took approximately 18 months, a fraction of the 10 to 15 years typical of conventional drug development.

On June 3, 2025, results from the Phase IIa GENESIS-IPF trial (Generative AI Enabled Novel Experimental Study of ISM001-055 in Subjects with Idiopathic Pulmonary Fibrosis) were published in Nature Medicine by researchers at Peking Union Medical College Hospital and Insilico Medicine. The double-blind, placebo-controlled trial enrolled 71 patients across 22 sites in China, randomizing participants to placebo or one of three doses of rentosertib over 12 weeks.

The results at the highest dose were striking. Patients in the placebo group showed a mean forced vital capacity (FVC) decline of 62.3 mL over the trial period, consistent with the expected natural progression of the disease. Patients receiving 60 mg of rentosertib once daily showed a mean FVC gain of +98.4 mL over the same period, a difference of more than 160 mL in lung function trajectory between treated and control groups. Safety was favorable, with adverse event rates similar across all treatment groups and all events resolving upon discontinuation.

This is the first clinical validation of a drug in which artificial intelligence identified not just the molecule but the biological target itself. Previous AI drug programs have used machine learning to optimize known targets or accelerate screening; rentosertib represents something categorically different, with AI reasoning about disease biology at the level of target discovery. Insilico Medicine has announced plans to initiate Phase III trials in late 2026, which will determine whether the lung function improvements seen in Phase IIa translate into meaningful survival benefits at scale.

A Convergence at the Frontier of Longevity Science

Epigenetic clocks, cholinergic brain training, and AI-designed therapeutics are separated by discipline, methodology, and therapeutic context. But they share a common and increasingly urgent message: aging is not a fixed trajectory, and the rate at which it unfolds is far more mutable than medicine has historically assumed.

Viewed through the lens of the six longevity pillars, these findings are remarkable in their breadth. The InCHIANTI epigenetic clock research speaks directly to Cellular Health, offering the clearest evidence yet that the velocity of biological aging is a meaningful clinical signal and a modifiable one. The INHANCE trial advances our understanding of Neurology, demonstrating that the decline of the cholinergic system, long considered one of aging’s most reliable one-way trajectories, is in fact reversible with targeted cognitive practice. The rentosertib story sits at the intersection of Pulmonary Health and AI-accelerated medicine, showing what happens when the discovery engine itself is transformed by a different kind of intelligence.

The broader arc is visible: science is not just finding more facts about aging; it is building the tools to intervene at its source. Epigenetic measurement is transitioning from research instrument to clinical biomarker. Neuroplasticity is proving to be biochemically verifiable, not just behaviorally observable. And artificial intelligence is identifying therapeutic targets that decades of human-led biology missed entirely.

What This Means for You

The practical takeaway from this triad of findings is more concrete than the research setting might suggest. If you are invested in understanding your biological age, the InCHIANTI epigenetic clock data validates tracking not just a single measurement but its rate of change over time; the velocity of aging matters as much as its starting point, and that velocity responds to lifestyle choices. If cognitive health is a priority, the INHANCE trial offers compelling evidence that ten weeks of directed speed-based cognitive training can produce real, measurable, neurochemically verified change in the cholinergic system of the aging brain. And for anyone affected by idiopathic pulmonary fibrosis, or following the broader arc of AI in medicine, rentosertib’s Phase IIa results represent a credible reason for cautious optimism as the field advances toward late-stage trials.

The Road Ahead

Ray Kurzweil has predicted that Longevity Escape Velocity, the point at which medical science extends human life expectancy by more than a year for every year that passes, will arrive sometime in the 2030s. Whether or not that specific threshold materializes on schedule, the research of 2026 is advancing in its direction with a momentum that would have seemed implausible a decade ago. Epigenetic clocks are moving from the research lab into clinical practice. Brain training is proving it can alter neurochemistry, not just behavior. Artificial intelligence is discovering drugs that human biology never identified on its own.

The biology of aging is, increasingly, a problem that science knows how to study, how to measure, and how to interrupt. The years ahead will determine how much of it can ultimately be solved.

Sources:
Insilico Medicine and Peking Union Medical College Hospital Researchers, “A generative AI-discovered TNIK inhibitor for idiopathic pulmonary fibrosis: a randomized phase 2a trial,” Nature Medicine, June 2025
INHANCE Trial Investigators, McGill University, “Effects of Computerized Cognitive Training on Vesicular Acetylcholine Transporter Levels in Healthy Older Adults: Results from the INHANCE Randomized Clinical Trial,” JMIR Serious Games, 2025
Kuo, P.-L. et al., “Longitudinal changes in epigenetic clocks predict survival in the InCHIANTI cohort,” Nature Aging, 2026
InCHIANTI Study Group, “A blood-based epigenetic clock for intrinsic capacity predicts mortality and is associated with clinical, immunological and lifestyle factors,” Nature Aging, 2025
Researchers at Multiple Institutions, “An unbiased comparison of 14 epigenetic clocks in relation to 174 incident disease outcomes,” Nature Communications, 2025
PubMed, Rentosertib Phase IIa Trial Entry, PMID 40461817, 2025

Free Daily Briefing

The Latest Longevity Science.
Delivered Every Morning.

Join researchers, physicians, and health professionals getting daily breakthroughs in AI-driven medicine, epigenetics, and longevity research.

Support the research that powers this editorial

Subscribe Free

No spam. Unsubscribe anytime. We respect your inbox.