E-news Exclusive

Geroscience: Epigenetic Markers Can Predict Aging and Health

By Keith Loria

Estimating age based on DNA changes can provide insight into health outcomes and mortality in older adults. After all, biological age, assessed through epigenetic changes like DNA methylation, can be indicative of age-related health outcomes, including the onset of chronic diseases, cognitive decline, functional limitations, and mortality.

Nevertheless, other variables such as demographics, socioeconomic status, mental health, and lifestyle choices offer equally reliable, if not stronger, predictions of late-life health outcomes.

These conclusions stem from a National Institute on Aging–funded study published in the Proceedings of the National Academy of Sciences led by a group of researchers from the University of Michigan, Yale, and the University of Southern California (USC). In the study, the team investigated whether epigenetic aging markers known as epigenetic clocks could also predict health outcomes and mortality, looking to discover if these clocks offer additional predictive value for health outcomes beyond traditional surveys of social and behavioral factors and conventional biomarkers.

“I wanted to understand underlying mechanisms that cause changes of health with age,” says lead author Eileen M Crimmins, PhD, a professor and AARP chair in gerontology at the USC Leonard Davis School of Gerontology. “Epigenetic age acceleration is linked to many social circumstances, stressors, and behaviors.”

Prior research indicated that socioeconomic and behavioral factors play a significant role in predicting health outcomes in older adults, and has shown that factors like education, poverty, race, access to medical care, and certain health behaviors can influence aging rates.

Crimmins explains that historically, scientists measure biological age by analyzing the epigenome, which records alterations in a cell’s DNA and associated proteins. Epigenetic clocks are tools researchers use to assess biological age by examining a set of biological markers. These clocks operate on the principle that specific epigenetic changes, such as DNA methylation patterns, evolve in a consistent manner over time within cells.

For this study, the research team examined data from more than 3,500 participants in the Health and Retirement Study, a comprehensive and long-term survey of Americans aged 51 and older. The scientists explored the link between epigenetic age acceleration and age-related health outcomes, including mortality, using three types of epigenetic clocks: first-generation (Horvath and Hannum), second-generation (GrimAge and PhenoAge), and third-generation (DunedinPACE clock).

The team found that different generations of epigenetic clocks (Horvath, Hannum, GrimAge, PhenoAge, DunedinPACE) differ in their ability to predict health outcomes.

“Horvath and Hannum do not do a very good job of predicting health outcomes; the other three do better,” Crimmins says. “Still, even doing better means that the amount of the variation explained in population health outcomes is fairly small.”

In the study, the team assessed participants’ age acceleration and studied its correlation with cognitive decline and functional limitations, as well as its ability to predict the onset of multiple chronic conditions after two years and mortality after four years.

“The second- and third-generation clocks, which were developed to enhance the earlier models, effectively predicted functional limitations in daily activities and cognitive impairment,” Crimmins says.

The team also compared these measures to established social and behavioral predictors of health and mortality, including health habits, demographics, mental health, socioeconomic status, and other blood-based biomarkers.

“The newer measures of epigenetic age acceleration consistently aligned with key aging-related health outcomes, although results varied somewhat across clock types,” Crimmins says. “Social and behavioral factors continued to be more reliable predictors of aging-related outcomes than epigenetic age acceleration measures, except in one significant instance: multimorbidity, or the presence of multiple chronic diseases.”

When combined with other biomarkers, epigenetic age was still predictive of mortality and multimorbidity, but it did not predict functional difficulties or cognitive dysfunction, she explains.

The study’s results indicate that epigenetic age measures, in addition to conventional social and behavioral health predictors, offer a useful approach for aging research and forecasting health outcomes in later life.

“Our findings indicate that while DNA methylation provides a valuable tool for predicting health outcomes in later life, other factors such as demographics, socioeconomic status, mental health, and health behaviors continue to be equally, if not more, reliable predictors of health,” Crimmins says. “We found that while epigenetic age does predict certain health outcomes later in life, it does little to explain important differences related to social factors.”

With that in mind, Crimmins believes further research is necessary to better understand the relationship between epigenetic age acceleration and various aging-related health outcomes.

“We need to better understand what epigenetic age acceleration is and is caused by,” she says. “Future research could investigate how age acceleration predictions vary across different population groups. I think there is promise in the future with measures that are understood to reflect changes that affect health outcomes.”

— Keith Loria is a D.C.-based award-winning journalist who has been writing for major publications for nearly 20 years on topics as diverse as real estate, travel, Broadway, and health care.