Effects of daily multivitamin-multimineral and cocoa extract supplementation on epigenetic aging clocks in the COSMOS randomized clinical trial

Why it matters

Epigenetic clocks read chemical marks (methyl groups) on DNA to estimate how old the body seems. Many products claim to slow these clocks, but few claims have been tested in a proper randomized trial. This study is a large, placebo-controlled test of two widely used supplements, and it ties directly to nutrition: can something as ordinary as a multivitamin change a DNA-based measure of aging? A clock that moves more slowly is only a stand-in for health, so this kind of trial is a first step: it tells us whether the measure can be shifted at all, before anyone asks whether people actually stay healthier.

What they did

The researchers used blood samples from 958 older adults (482 women and 476 men) in COSMOS, a randomized trial in which people took a daily multivitamin-multimineral (Centrum Silver), a cocoa extract (500 mg cocoa flavanols a day, including 80 mg epicatechin), both, or placebos. This was a prespecified extra study within the trial. Over two years they tracked five DNA-methylation measures of aging: PCHannum, PCHorvath, PCPhenoAge, PCGrimAge and DunedinPACE.

What they found

Compared with placebo, the multivitamin modestly slowed the rise of the two "second-generation" clocks, which are built to predict health and death rather than just age. The yearly difference was about 0.11 years for PCGrimAge and about 0.21 years for PCPhenoAge. For PCGrimAge, the effect was larger in people whose biological age was already running ahead of their real age at the start (about 0.24 years a year) and close to zero in people who were aging normally or slowly. Cocoa extract did not change any of the five clocks. The authors call the effects statistically significant but small, and say more studies are needed to show whether they matter clinically.

Limits

The changes are small, and only two of the five clocks moved. A slower clock is not the same as better health or a longer life; the trial did not show that. The measurements are from blood only, the participants were older adults, and the abstract does not tell us whether the effect lasts after people stop taking the supplement.

Key terms

DNA methylation

Small chemical tags (methyl groups) attached to DNA that help switch genes on or off without changing the letters.

Epigenetic clock

A formula that turns methylation at many DNA sites into an estimate of biological age or pace of aging.

Second-generation clock

A clock trained to predict health outcomes or death, such as GrimAge or PhenoAge.

DunedinPACE

A clock that estimates how fast someone is aging right now, in biological years per calendar year.

Randomized controlled trial

A study in which chance decides who gets the treatment and who gets placebo, so the groups can be fairly compared.

Placebo

A dummy pill that looks like the real one, used so expectations do not bias the result.

Confidence interval

The range of values the true effect probably lies in.