A person’s birthday tells them how many years they have been alive, but it cannot show whether their heart, brain or kidneys have aged at the same rate.
That is because organs within the same body can show different levels of biological ageing.
A new study published in Nature Medicine suggests that ageing leaves distinct marks on the microscopic structure of different organs.
Scientists have used these marks to develop a method that may predict which organs appear older than expected from a single blood sample.
The method is not yet a blood test available at hospitals, nor can it track an organ over many years to directly measure how quickly it is ageing. Instead, it estimates whether an organ appears biologically older or younger than expected for a person’s chronological age.
A team led by André Rendeiro, with Ernesto Abila, Iva Buljan and Yimin Zheng as co-first authors, examined 25,712 microscopic tissue images from 983 deceased donors. The samples covered 40 tissue types across 29 organs.
The researchers trained an artificial intelligence model using tissue samples from people of different ages. Over time, the system learned to recognise subtle structural changes associated with ageing, creating a kind of “tissue clock” for each organ.
For example, if the clock judged a 60-year-old person’s lung tissue to resemble that of older people, the lung was considered older than expected.
On average, the models’ estimates differed from the donors’ actual ages by about five years.
The estimates also appeared to reflect more than a person’s date of birth. Tissues that looked older than expected tended to show more structural changes and health conditions associated with ageing. These included tissue shrinkage, scar-like stiffening, loss of tiny blood vessels, fat accumulation in skeletal muscle and thinning nerves.
The body did not follow one universal ageing pattern. Some people showed signs of advanced ageing across many organs, while in others, a single organ such as the heart, lung or stomach appeared considerably older than the rest of the body.
Directly assessing an organ’s biological age, however, usually requires an invasive procedure to collect tissue.
To overcome this, the researchers matched age estimates derived from tissue images with patterns of gene activity in blood samples from the same donors.
“A blood sample does not directly image the organ; it estimates the organ’s age gap through a learned correspondence,” said André Rendeiro, a principal investigator at CeMM and the Ludwig Boltzmann Institute for Network Medicine at the University of Vienna.
The researchers then developed a second model that estimated how old different organs appeared using only blood data. Its predictions were strongest for overall body ageing, the digestive system and the spleen.
To assess whether these estimates reflected real illnesses, the researchers tested the method on blood data from healthy people and people with eight different medical conditions. The models generally assigned larger age gaps to organs associated with each condition.
For people who had experienced a stroke, the model assigned the largest age gap to the brain.
In people with Crohn’s disease, the oesophagus, stomach, small intestine and colon were assigned larger age gaps.
In the Alzheimer’s disease group, the brain was the only organ with a significantly elevated age gap.
These findings suggest that blood may carry clues about what is happening deep inside the body. However, an organ appearing older does not mean accelerated ageing caused the disease. The illness may have aged the organ, or both processes may have developed together.
Earlier blood-based methods have estimated organ age by measuring circulating proteins. One such study found that having an older-looking brain was associated with a higher risk of death.
However, existing commercial biological-age tests can produce widely differing results.
The new study also has important limitations. Most of the tissues were collected after death, and the researchers did not follow the same people as they aged. The blood-based method was mainly evaluated in people who had already been diagnosed with a disease.
It remains unknown whether the test could detect illness before symptoms emerge.
Prospective studies, consistent performance across different populations and regulatory approval would still be required. Reaching clinical use will take “realistically several years, possibly close to a decade,” Rendeiro said.
Abila, Buljan, Zheng and Rendeiro declared competing financial interests in the form of a patent related to the work.
For now, the study suggests that reducing biological age to a single number may miss an important part of the story.
The heart, brain and kidneys may share the same body, but they do not necessarily experience time in the same way.
