black and white photo of doctor in full protective gear taking cheek swab sample from older male patientShare on Pinterest
Researchers are saying that a simple cheek swab sample could hold key information about how a person is aging. Image credit: Courtney Hale/Getty Images.
  • Scientists have long known that aging varies widely among individuals, influenced by genetics and lifestyle choices.
  • Now, a new tool called CheekAge offers a simple, noninvasive way to predict mortality by analyzing epigenetic changes in cheek cells.
  • In their recent study, researchers found that CheekAge can accurately assess mortality risk, outperforming earlier models based on blood samples, and offering a new method of monitoring biological aging.

It is well known that not everyone ages at the same rate. While some supercentenarians may age more slowly due to favourable genetics, many behavioral and lifestyle factors — such as stress, poor sleep, poor nutrition, smoking and alcohol — can accelerate aging.

These environmental influences leave their mark on our genome in the form of epigenetic changes, allowing scientists to measure molecular aging by examining specific genomic sites.

Epigenetics is a branch of genetics that studies how gene activity can be changed without altering the DNA sequence itself.

Essentially, it looks at how our environment, lifestyle and even experiences can influence which genes are turned “on” or “off.”

Over the last decade, scientists have created various tools known as “epigenetic clocks” to estimate biological age by analysing DNA methylation patterns, collected from blood samples.

Earlier this year, scientists in the United States introduced a second-generation epigenetic clock, CheekAge, which uses methylation data from easily collected cheek cells.

Recently, the team published their research in Frontiers in Aging, showing that CheekAge can accurately predict mortality risk, even when epigenetic data from other tissues is used.

Maxim Shokhirev, PhD, the study’s first author and Head of Computational Biology and Data Science at the company Tally Health in New York, explained the key findings to Medical News Today.

He told us that:

“CheekAge is a computational model that predicts your epigenetic age using methylation marks on the DNA. We previously showed that the predicted CheekAge is significantly associated with lifestyle factors such as sleep, exercise, and alcohol consumption as well as health factors such as COVID infection, cancer treatment, and BMI.”

“In this work, we tested if CheekAge is also associated with mortality risk using a blood dataset collected from a group of older adults,” Shokhirev explained.

“We found that even though CheekAge was trained on cheek cells collected using a painless swab test, it was nevertheless highly associated with mortality risk in this blood longitudinal dataset,” he noted

“This further supports that CheekAge picks up on important functional health signals across tissues,” Shokhirev told us.

CheekAge was developed, or “trained,” by analyzing the methylation levels at around 200,000 sites and linking them to an overall health and lifestyle score, which represents presumed variations in physiological aging.

DNA methylation refers to the process through which gene expression — whether a gene is “on” or “off” — is altered.

The researchers applied statistical programming to evaluate how accurately CheekAge predicted mortality from any cause in a cohort of 1,513 men and women, born in 1921 and 1936, who were tracked throughout their lives as part of the Lothian Birth Cohorts (LBC) program at the University of Edinburgh in the United Kingdom.

One of the LBC’s goals was to explore how lifestyle, psychosocial factors, and biomedical, genetic, epigenetic and brain imaging data are linked to cognitive aging.

Every 3 years, the volunteers had their blood cell methylome (collection of gene-modifying chemicals in the blood) analyzed at about 450,000 DNA methylation sites.

The researchers used the most recent available methylation data, along with mortality status, to calculate CheekAge and assess its association with mortality risk.

They obtained mortality data from the Scottish National Health Service Central Register.

The results suggest that CheekAge is significantly associated with mortality in a longitudinal dataset and outcompetes first-generation clocks trained in datasets containing blood data concluded the authors.

Specifically, for every increase by a single standard deviation in CheekAge, the hazard ratio of all-cause mortality increased by 21%.

The researchers suggest that their epigenetic clock, which was trained using cheek cell data, can predict mortality even when the methylome is measured in blood cells, indicating that common mortality signals exist across different tissues.

This suggests that a noninvasive cheek swab could serve as an effective alternative for studying and monitoring the biology of aging.

Brittany Ferri, PhD, an occupational therapist with the National Council on Aging, who was not involved in this research, told MNT that this study “has an innovative perspective on aging.”

”Using cheek cells to predict aging and related health issues is fascinating and could be a promising method. It’s especially intriguing because it is less invasive than traditional blood tests, potentially making it a better choice for routine health checks and keeping us healthy.”

– Brittany Ferri, PhD

“With such a straightforward, noninvasive method to assess aging and potential health concerns, preventive healthcare could become more personalized and proactive,” she explained.

In this study, the researchers examined the methylation sites most strongly linked to mortality in more detail, identifying specific genes located around these sites as potential candidates that may strongly influence lifespan or the risk of age-related diseases.

For instance, the PDZRN4 gene, which may function as a tumor suppressor, and ALPK2, a gene associated with cancer and heart health in animal models, stood out as potential markers of aging.

Other genes identified have previously been linked to cancer, osteoporosis, inflammation and metabolic syndrome.

Ferri told us that “it would be beneficial to know and manage [the] aging process early with tailored health strategies, possibly delaying or preventing age-linked diseases.”

“Plus, making this technology widely accessible might lead to more equitable health practices, benefiting people across different backgrounds,” she added. “This doesn’t just promise a longer life but a better quality of life as we age. However, more research is needed to definitively make any conclusions.”

Future research could explore what other associations CheekAge might capture beyond all-cause mortality, such as the incidence of age-related diseases or the length of a person’s healthspan — the period of life free from chronic age-related diseases and disability.

Shokhirev told us that “CheekAge is the inspiration for our TallyHealth TallyAge(TM) test.” For him and his team, the current findings “further [validate] that our test is more than just an accurate age predictor, but that it also encodes important information about health and mortality in particular.”

According to Shokhirev:

“This means that CheekAge is a useful aging biomarker that can be used for academic research, clinical trials, and as a health tracking tool.”

“We look forward to future studies associating CheekAge with other important health and lifestyle factors,” he added.

However, the test has yet to be validated by the Food and Drug Administration (FDA), which means there may be a long road ahead for this experimental aging predictor.