close
close

This simple cheek swab can accurately predict how long you will live

This simple cheek swab can accurately predict how long you will live

(Source: Andrey_Popov/Shutterstock)

NEW YORK — A quick cheek swab could one day reveal how long you have left to live. Scientists have found that a DNA test originally designed to measure the biological aging of cells found in the cheek can accurately predict the risk of death, even when applied to blood samples. The findings suggest that common biological markers of aging exist in different body tissues, potentially opening new opportunities to assess health risks and develop anti-aging interventions.

The study focuses on a tool called CheekAge, a next-generation “epigenetic clock” developed earlier this year. Our genes contain instructions for building and maintaining our bodies, but the way these genes are read and expressed can change over time. One way this happens is through a process called DNA methylation, in which tiny chemical tags attach to our DNA. These markers act like switches, turning genes on and off.

Most previous epigenetic clocks relied on blood samples, making them less practical for widespread use. CheekAge, as the name suggests, was designed to work with easily harvested cheek cells. What makes this new study particularly intriguing is the fact that CheekAge proved effective in predicting risk of death, even when applied to blood sample data.

To test this, they used a unique dataset from Scotland called Lothian Birth Cohorts. This long-term study followed two groups of people born in 1921 and 1936, collecting detailed health information and biological samples over many years.

Using blood samples from 1,513 participants (712 men and 801 women) aged 67 to 90, the team applied the CheekAge algorithm. Even though CheekAge was designed to analyze cheek samples and almost half of its DNA markers were not present in blood data, it still showed strong ability to predict mortality risk.

Specifically, for each standard deviation increase in the difference between a person’s buccal age and their actual age, the risk of death increased by 21%. To put this into perspective, researchers divided participants into three groups based on CheekAge scores. The “oldest” biological age group reached 50% mortality approximately 7.8 years earlier than the “youngest” biological age group.

3D rendering showing the structure of the DNA double helix. Even though CheekAge was designed to analyze cheek samples and almost half of its DNA markers were not present in blood data, it still showed strong ability to predict mortality risk. (Source: Unsplash/THAVIS 3D)

What’s particularly impressive is that CheekAge did better at predicting mortality than several other established epigenetic clocks. It even competed with a specialized clock called DNAm PhenoAge, which was specifically designed to predict mortality from blood samples.

“The fact that our epigenetic clock trained on cheek cells predicts mortality when measuring the methylome in blood cells suggests that there are common mortality signals in tissues,” explains Dr. Maxim Shokhirev, first author of the study and head of computational biology and data science at Tally Health in the statement. “This means that a simple, non-invasive cheek swab could provide a valuable alternative in studying and tracking the biology of aging.”

Study published in the journal The limits of agingalso identified specific DNA markers that seemed particularly important in predicting mortality. One of the differentiators was a marker linked to a gene called ALPK2. When this marker was removed from the analysis, the ability to predict mortality decreased significantly. Interestingly, ALPK2 has been linked to heart development in animal studies and may play a role in some cancers.

“It would be intriguing to determine whether genes such as ALPK2 influence lifespan and health in animal models. Future research is also needed to determine what other associations beyond all-cause mortality can be detected using CheekAge,” says Dr. Adiv Johnson, the study’s final author and chief science and education officer at Tally Health. “For example, other possible associations may include the occurrence of various age-related diseases or the duration of ‘health’, that is, a period of healthy life free from age-related chronic diseases and disabilities.”

Other important markers have been linked to genes involved in bone health, metabolism and cellular processes associated with aging. This suggests that CheekAge captures a variety of biological factors that influence overall health and longevity.

Although more research is needed to fully understand how CheekAge works and confirm its predictive power in larger, more diverse populations, this study opens exciting possibilities. Imagine being able to assess health risk with a simple cheek swab, potentially allowing for earlier interventions and personalized health strategies.

Of course, keep in mind that these tools provide probabilities, not certainties. A high Cheek Age does not mean you are doomed to an early grave, any more than a low Cheek Age guarantees a long life. However, as our understanding of the aging process increases, tools like CheekAge can become valuable resources in our quest to live healthier, longer lives.

Paper summary

Methodology

The researchers used DNA methylation data from blood samples collected as part of the Lothian Birth Cohorts study. They applied the CheekAge algorithm to these blood data, which was originally designed for cheek swab samples. Despite the lack of some DNA markers, the algorithm still performed effectively. They then used statistical models to analyze how well CheekAge predicted the risk of death, taking into account factors such as age, gender and the composition of cell types in the blood samples.

Key results

The CheekAge study showed a significant association with mortality risk. For each standard deviation increase in the difference between CheekAge and chronological age, the risk of death increased by 21%. The researchers also identified specific DNA markers that were particularly important for these predictions, with markers related to genes such as ALPK2, B4GALNT3 and SAT1 standing out.

Limitations of the study

The study used blood samples instead of cheek swabs, which is what CheekAge was originally designed for. About half of the DNA markers used in the full CheekAge model were also missing from the blood data. The study population was limited to older adults in Scotland, so the results may not apply equally to other age groups or populations. Additionally, while CheekAge predicts mortality risk, it does not explain the underlying causes of this risk.

Discussion and takeaways

This study suggests that epigenetic changes measurable in easily accessible tissues, such as cheek cells, may provide valuable information about overall health and mortality risk. CheekAge’s performance, even when applied to blood samples with missing data, indicates that it captures basic aging processes. The identification of specific DNA markers associated with mortality risk opens new avenues for understanding the biology of aging and age-related diseases.

Financing and disclosure

The study was funded by Tally Health, a company with which some of the authors are affiliated. It has also received support from a range of research councils and charities including the Wellcome Trust, the Biotechnology and Life Sciences Research Council, the Economic and Social Research Council and Age UK. When interpreting the results, the connections between the researchers and the company that may be interested in the research results should be taken into account.