Epigenetic Clocks Tell You If You’ll Live — Not If You’ll Remember

Key Takeaway: In a study of nearly 6,000 women, 15 different epigenetic aging clocks reliably predicted who would survive to age 90. Yet none of these clocks could distinguish between those who reached this age cognitively healthy and those who arrived with severe memory impairment. The finding exposes a critical blind spot in one of modern aging science’s most lauded tools.

The Promise and the Blind Spot

Imagine two women, both 90 years old. One manages her own finances, reads novels, and debates politics with her grandchildren. The other cannot recall what she had for breakfast. By any meaningful measure, these are two radically different outcomes of aging. Yet according to today’s most advanced biological aging tests, these two women might have looked nearly identical decades earlier.

This unsettling reality is at the heart of striking new findings from the Women’s Health Initiative Memory Study (WHIMS), one of the largest and most rigorously designed investigations into aging and cognition in women. The study tested whether epigenetic clocks—molecular tools that estimate “biological age” by reading chemical tags on a person’s DNA—could predict not only how long a woman would live, but also whether she would arrive in old age with her memory and thinking skills intact^[1]. The answer to the first question was a definitive yes. The answer to the second was a disappointing no.

What Did the Researchers Do?

The study included 5,844 women from the WHIMS cohort who had blood samples available for DNA methylation analysis. From these samples, researchers calculated 15 different epigenetic clocks, spanning three generations of increasingly sophisticated algorithms. First-generation clocks, like the Horvath and Hannum clocks, predict chronological age from methylation patterns^[2]. Second-generation clocks, such as PhenoAge, incorporate clinical biomarkers to better capture disease risk^[3]. Third-generation tools like GrimAge and DunedinPACE go even further, trained on mortality data and the pace of biological deterioration^[4].

Participants were then followed to determine three outcomes: whether they survived to age 90, whether they survived with preserved cognitive function, or whether they survived with measurable cognitive decline. The key question was whether any of the 15 clocks could distinguish the latter two groups from one another.

The Findings

Confirming decades of previous work, every clock tested was associated with survival to age 90. The newer-generation clocks performed best. For AgeAccelGrim2, each standard deviation increase in epigenetic age acceleration—meaning a person’s biology is aging faster than their driver’s license suggests—was associated with a 34% decrease in the odds of reaching 90 (OR = 0.66). DunedinPACE, which measures the rate of aging rather than a static snapshot, showed similarly strong associations.

But this is where the findings become sobering. Not a single one of the 15 clocks showed a stronger association with cognitively healthy survival compared to survival with cognitive impairment. In other words, the clocks could predict whether a woman would live a long life, but they were effectively blind to the quality of that life as measured by cognitive function.

Why Do the Clocks Miss the Brain?

To understand this result, one must grasp what epigenetic clocks actually measure. DNA methylation—the addition of small methyl groups to cytosine bases in DNA—is one of the body’s fundamental mechanisms for regulating gene expression^[5]. As we age, methylation patterns shift in predictable ways across thousands of genomic regions. Epigenetic clocks aggregate these changes into a single number: your biological age.

The problem is that these clocks are built from blood samples. They capture systemic aging as reflected in circulating white blood cells—the cumulative burden of inflammation, metabolic dysfunction, cardiovascular wear-and-tear, and immune senescence. These are powerful predictors of mortality because cardiovascular disease, cancer, and organ failure are the leading causes of death.

But brain aging plays by a partially different set of rules. Neurodegeneration involves processes like amyloid-beta accumulation, tau protein misfolding, synaptic loss, and neuroinflammation driven by microglia, the brain’s resident immune cells^[6]. The blood-brain barrier isolates much of this biology from the peripheral circulation. A methylation signature in a white blood cell may faithfully reflect arterial stiffness or chronic inflammation, but it can say very little about what is happening in the hippocampus, the brain region most vulnerable to Alzheimer’s disease^[7].

This isn’t a failure of the technology itself. It is a lack of specificity. The current clocks are trained to predict death, and they do it well. They were never designed to predict whether the brain, specifically, will endure the journey.

Limitations to Consider

Several caveats are worth noting. Because the cohort consisted entirely of postmenopausal women, the findings may not generalize to men or younger populations. While the cognitive assessment was standardized, it was performed at intervals and may not have captured the full spectrum of decline. Furthermore, the clocks were measured from blood drawn at a single time point, meaning they could not track how biological aging trajectories changed over the years. And no single study, however large and well-designed, is the final word on the matter.

Implications for the Future

For anyone who has taken or is considering a commercially available biological age test, this study offers a necessary dose of realism. A favorable result from an epigenetic clock may indeed suggest your body is aging well systemically, and that is genuinely useful information. But it tells you almost nothing about whether your brain is on a similar trajectory. The two can diverge dramatically.

The practical takeaway is that cognitive health requires its own dedicated pursuit. The lifestyle interventions with the strongest evidence for protecting the aging brain—aerobic exercise, staying mentally active, managing vascular risk factors like hypertension and diabetes, sufficient sleep, and social connection—remain indispensable precisely because no blood test can yet replace them.

For the scientific community, this study is a clear call to build the next generation of clocks: tools trained not just on survival, but on the preservation of what makes survival meaningful. Brain-specific methylation signatures, cerebrospinal fluid biomarkers, and multi-tissue models of aging represent promising new frontiers. Until those tools are developed, the gap between living long and living well will remain a void our most advanced molecular technologies cannot yet bridge.

Scientific Sources

1. LaCroix AZ, et al. Epigenetic Clocks of Biological Aging and Cognitively Healthy Longevity: The Women’s Health Initiative Memory Study. The journals of gerontology. Series A, Biological sciences and medical sciences. 2026. PubMed: https://pubmed.ncbi.nlm.nih.gov/42679383/
2. Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013. DOI: 10.1186/gb-2013-14-10-r115
3. Levine ME, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 2018. DOI: 10.18632/aging.101414
4. Belsky DW, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 2022. DOI: 10.7554/eLife.73420
5. Jones PA. Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat Rev Genet. 2012. DOI: 10.1038/nrg3230
6. Heneka MT, et al. Neuroinflammation in Alzheimer’s disease. Lancet Neurol. 2015. DOI: 10.1016/S1474-4422(15)70016-5
7. Braak H, et al. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 1991. DOI: 10.1007/BF00308809