Leukocyte Composition Change Accounts for Most Apparent Ten-Year DNA Methylation Drift in Blood: A Two-Cohort Longitudinal Analysis
This longitudinal study demonstrates that in older adults, the apparent ten-year drift in blood DNA methylation is primarily driven by time-varying changes in leukocyte composition rather than intrinsic epigenetic aging, as adjusting for cell type shifts largely eliminates the observed methylation velocity.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Blood is a bustling city of cells, a fluid mixture where different types of white blood cells patrol the body, fighting infection and maintaining health. Over time, as people age, the population of this city shifts. The numbers of certain cell types, particularly those that fight new threats, tend to decline, while others expand. Scientists have long used chemical tags on DNA, known as methylation, to track the aging process and detect diseases like cancer. These tags act like a record of a cell's history. Because blood is easy to draw, researchers often measure these tags in blood samples to create a "biological clock" or a surveillance tool for health. The standard assumption has been that if you measure the same person twice, years apart, and look at the difference, you cancel out all the unique quirks of that individual. You are left with a pure signal of change over time, free from the noise of comparing one person to another.
However, a new analysis challenges this assumption by pointing out a flaw in the logic: the "noise" itself changes. If the mix of white blood cells in a person's body shifts significantly over a decade, and those different cell types carry different chemical tags, then the measurement of change will capture that shift in population rather than a change in the cells themselves. It is like trying to measure the average height of a crowd by taking a photo, waiting ten years, and taking another. If the crowd has changed from mostly children to mostly adults, the average height will rise, but not because the individuals grew taller; it is because the composition of the group changed. This study asks whether the apparent aging signal in blood DNA is actually just a reflection of this shifting cellular population.
The researchers investigated this question using data from two large groups of older adults, one from Denmark and one from the United States. They looked at blood samples taken from the same people over long periods—ten years for the Danish group and five years for the American group. They focused on a specific set of forty-eight locations on the DNA strand that are known to change with age and are used to calculate a "burden score," a number that represents how much a person's DNA methylation pattern deviates from a healthy standard. First, they calculated how much this score changed over time without making any corrections. In the Danish group, the score rose significantly, suggesting a steady drift in the biological markers of aging.
Next, the researchers applied a mathematical adjustment to account for the changing mix of white blood cells. They estimated how much the proportions of different cell types, such as T-cells and neutrophils, had shifted for each person between the first and second blood draw. When they removed the effect of these cellular shifts from the DNA score, the picture changed dramatically. In the Danish cohort, the apparent speed of aging dropped by nearly two-thirds. The remaining change was so small that it could no longer be distinguished from zero, meaning it was statistically indistinguishable from random fluctuation. In the American group, the initial change was not statistically significant to begin with, but the adjustment still explained a large portion of the variation in the data. The study found that the shift in cell populations alone accounted for roughly thirty-six percent of the change in the Danish group and nearly sixty-four percent in the American group.
The researchers also checked to ensure that the results were not caused by errors in the laboratory or by the samples degrading over time in storage. They compared samples that were run on different machines and found that the technical error was too small to explain the large changes they observed. They also tested whether the DNA was simply moving toward a middle ground or following a specific directional path. The data showed that the changes were not following a coordinated path but were instead moving toward a state of greater disorder, a pattern consistent with the natural randomness that increases as cells age. Crucially, the study ruled out the idea that the results were an artifact of how long the samples had been stored in freezers, as the pattern of change did not match what would be expected from degradation.
The conclusion is that the standard method of comparing a person's current blood sample to their own past sample does not automatically remove the influence of changing cell types. Because the mix of white blood cells shifts in a predictable direction as people age, this shift gets baked into the measurement of DNA change. The study suggests that for blood-based DNA tests to accurately measure the aging of individual cells, they must explicitly adjust for the changing composition of the blood. Without this adjustment, a large part of what looks like a biological aging signal is actually just a census of the changing cellular population. The researchers emphasize that this does not prove that no biological aging occurs in the cells themselves, but rather that the current methods often fail to separate the signal of cellular aging from the noise of cellular population shifts. The findings imply that future tests need to report results that have been corrected for these cell counts to be truly meaningful.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.