Yale analysis finds drugs and exercise lowered epigenetic age, supplements did not

Yale researchers pooled 51 human studies of longevity interventions and found that metformin, semaglutide and anti-TNF drugs lowered epigenetic age the most, while over-the-counter supplements did not lower it, according to the analysis published in the peer-reviewed journal Nature Medicine. A healthy diet combined with exercise also lowered epigenetic age, and did so consistently.

The comparison covers choices that sit in ordinary routines, from prescriptions and supplement shelves to what people eat and how they train, and it measures all of them against the same kind of DNA marker.

Titled "Responsiveness of epigenetic aging biomarkers to longevity interventions in humans," the paper examined more than 110 DNA methylation biomarkers, 16 of them major epigenetic clocks, according to the ScienceDaily summary of the Nature Medicine paper. The test was whether those markers moved in response to a drug, a change in diet and exercise, a supplement or a medical procedure.

The team worked from the results of studies that had already been completed, and Yale's account of the analysis sorts the interventions into four categories: pharmacological treatments, lifestyle changes, over-the-counter supplements and medical procedures. Pooling of this kind allows interventions to be compared even when no single trial tested them side by side.

Some of the drug studies were carried out in patient groups, according to Medical News Today's report on the analysis, so part of the drug evidence comes from people with a medical condition rather than from healthy volunteers. All three drugs are established prescription medicines: metformin for type 2 diabetes; semaglutide, sold as Ozempic and Wegovy, for type 2 diabetes and weight loss; and anti-TNF therapies, which block the inflammatory protein tumour necrosis factor, for conditions such as rheumatoid arthritis and Crohn's disease.

The lifestyle result covers diet and exercise as a pair. A healthy diet plus exercise consistently decreased epigenetic age, Yale's account says, and the diets it names are Mediterranean, low-carb and low-fat. The three differ in emphasis: the Mediterranean pattern is built around vegetables, legumes, whole grains, fish and olive oil, while low-carb diets limit starches and sugars and low-fat diets limit fats.

The drugs had the largest average effects and the lifestyle results were more consistent, in Medical News Today's comparison. Those are separate measures: an average can be lifted by a few large results, while consistency describes how closely the studies agreed with one another.

Over-the-counter products "didn't have an effect in decreasing epigenetic age," in the words of Yale's account. "The evidence for supplements was limited and inconsistent," Sehgal, a researcher quoted by Medical News Today, said in that coverage, adding that supplements may still help people with deficiencies, a separate purpose from slowing aging.

The result does not mean supplements do nothing, Sehgal said in coverage by NutraIngredients, a trade publication covering the supplement industry. Post-hoc analyses of DO-Health, which ran for three years, and COSMOS, which ran for two, showed modest but statistically significant effects on newer epigenetic clocks, the publication reports.

Neither trial was in the TranslAGE data that the analysis drew on, the publication notes. Post-hoc analyses are ones not planned before a trial began, and because the question is chosen after the data are in, such results are generally treated as leads to test in later trials.

Newer clocks responded more consistently in the analysis, the same publication reports, naming them as DunedinPACE, PCGrimAge, GrimAgeV2, PCPhenoAge and SystemsAge. The clocks are blood-based and estimate biological age from methyl groups, small chemical tags on DNA, according to ScienceDaily.

Methylation is one kind of epigenetic change, which alters how genes are read without changing the DNA sequence. Its pattern shifts as people get older, and a clock turns that pattern into an estimate of biological age, a statistical figure that can sit above or below a person's age in years.

"A biomarker responding to an intervention is not the same as proving that the intervention slows aging," Sehgal said in the Medical News Today coverage, adding: "There is not yet a single intervention proven to broadly slow aging in humans."

On the clocks themselves, Furman, a researcher quoted by NutraIngredients, said: "There is no direct evidence linking change in an epigenetic clock to future outcomes." Outcomes of that kind are events such as the onset of age-related disease, loss of physical function or death, which a change in a blood measure taken during a trial does not itself record.

Several points are missing from the Yale, ScienceDaily, Medical News Today and NutraIngredients coverage: how many people took part in each of the 51 studies and how long they were followed, how many took part in DO-Health and COSMOS, the size of the drops expressed in years, which individual supplements were tested, whether exercise alone moved the clocks, what the medical procedures category showed, and the first names and posts of the researchers quoted as Sehgal and Furman.

What the change means "for our long-term risk of age-related disease and functional decline" is the question the ScienceDaily summary leaves for the clocks to answer.