Blood-based measures of biological aging do not all point to the same thing in the brain. In a study published April 7, 2026, in Volume 18 of Aging-US, researchers tested five widely used epigenetic clocks against MRI measures of brain structure collected roughly eight years later in 1,196 older women from the Women's Health Initiative Memory Study. Only one clock, AgeAccelGrim2, showed a significant association with the Alzheimer's Disease Pattern Similarity Score, a validated imaging biomarker tied to elevated dementia risk. None of the five clocks were linked to SPARE-BA, a composite MRI index of accelerated brain aging.
The work was led by first and corresponding author Linda K. McEvoy of the Kaiser Permanente Washington Health Research Institute, with collaborators at institutions across the United States and Europe. Epigenetic clocks estimate biological age from chemical marks on DNA, and they have drawn intense interest as low-cost, blood-based indicators of how fast someone is aging. Whether they say anything specific about the brain has been much less clear.
That distinction is the useful part of these results. General brain aging and the particular pattern of atrophy seen in Alzheimer's disease appear to be separable processes, and the epigenetic clocks tested here did not capture the former at all. The single positive signal for AgeAccelGrim2 suggests that clocks built to predict mortality and disease risk, rather than chronological age alone, may be the more promising place to look for markers relevant to neurodegeneration.
Because the analysis drew on a cohort of older women with a single later MRI assessment, the findings will need testing in more varied populations before epigenetic measures could be considered informative about an individual's brain health. For now, the study narrows the field, indicating which clocks and which imaging outcomes are worth pursuing in future research.