An international team of scientists has developed a blood test that measures how fast we “age” at the cellular level. According to the researchers, this test can determine how experimental treatments and environmental changes might change one’s pace of aging.

Disease is often a product of aging. As we grow older, our bodies accrue more damage than they can effectively repair. This leads to serious conditions like Alzheimer’s disease and many cancers, as well as to more mundane conditions such as arthritis. 

Some people seem to feel the effects of aging differently than others. Two people born on the same date will both turn 60 at the same time. However, one might still run miles every day, while the other’s arthritis keeps him or her from doing much more than walking around the block. 

To scientists, this implies that there exist two ways to measure age. The first, called chronological age, simply measures the passage of the years. The second, called biological age, is more loosely defined, but represents an effort to describe the difference between a population’s average life expectancy and that of an individual person. 

The group behind the current study, composed of researchers from the United States, New Zealand and the United Kingdom, already had some idea of the molecular indications of aging that they should look for.

In 2015, a study was published that analyzed 18 biomarkers, or biological measures of health, in 954 subjects born in Dunedin, New Zealand, between 1972 and 1973. Taking these 18 measurements in subjects at 26, 32, and 38 years old, revealed that subjects’ aging rates varied significantly, despite not having yet developed chronic disease. The results of these 18 tests became known as one’s “Pacing of Age.”

A drawback to this technique, however, was that it required long follow-up times and in-depth clinical assessments. “It wasn’t very useful for studies that need to test the impact of a new drug or lifestyle intervention in a matter of a few years,” said Daniel Belsky, the current study’s lead author, in a related press release

In the current study, the team used machine learning, an artificial intelligence technique, to analyze chemical markers on DNA taken from subjects’ white blood cells. These markers, called DNA methylation, are attached to DNA and alter how any underlying genes are expressed. If you think of our DNA as a recipe for a dish, DNA methylation is akin to herbs and spices. The meal is still there, but what’s on it changes the flavor. 

The group analyzed over 400,000 different DNA methylation marks to find the ones that related to the physiological changes captured in their Pace of Aging measure. Their machine learning algorithm identified 46 DNA methylation markers that provided the same diagnostic measure as their Pace of Aging. The advantage in this case is that all marks can be measured in one single blood test. They named this measure “DunedinPoAm” for Dunedin (P)ace (o)f (A)ging in (m)ethylation.

According to this measure, a Dunedin PoAm value of one equals one year of chronological aging. The Dunedin Study participants had values ranging from 0.6 to 1.4, indicating that in terms of biological measurements, some people seemed to age more than twice as fast as others. 

This is not the first time that people have tried to measure age via DNA methylation. Previous studies have generally compared people at different chronological ages. According to Belsky, however, this approach was limited by the different conditions, under which people born at different times have lived. This, explains Belsky, changes their exposure to factors such as childhood diseases, tobacco smoke, airborne lead, antibiotics, and quality of nutrition, all of which affect DNA methylation.

Belsky and his collaborators hope that this new measure, which can be administered as a single blood test at the beginning and end of a trial among subjects of the same age, will reveal greater details, regarding how things like medications, lifestyle changes and the environment alter human rates of biological aging.