Multivitamins modestly slow epigenetic aging in two-year COSMOS trial
A two-year randomized study of 958 older adults in the COSMOS trial found that taking multivitamins daily modestly slowed certain DNA methylation “epigenetic clock” measures, trimming about four months of biological aging over two years. The finding points to small but measurable shifts in biomarkers tied to mortality risk, though the clinical meaning remains unsettled.
COSMOS randomized trial details
In the COcoa Supplement and Multivitamin Outcomes Study (COSMOS), 482 women and 476 men with an average age of about 70 years were assigned to one of four groups: daily cocoa extract plus a multivitamin; cocoa extract plus a multivitamin placebo; cocoa extract placebo plus a multivitamin; or two placebos. Blood samples were collected at baseline, one year, and two years. Researchers then assessed five DNA methylation measures—PCHannum, PCHorvath, PCPhenoAge, PCGrimAge, and DunedinPACE—to estimate changes in biological aging. The figures point to a design capable of isolating the independent and combined effects of each supplement.
Multivitamins and PCGrimAge slowdown
Compared with placebo, a daily multivitamin (Centrum Silver) modestly reduced the yearly increase of two second-generation clocks: PCGrimAge by −0. 113 years (95% CI −0. 205 to −0. 020; P = 0. 017) and PCPhenoAge by −0. 214 years (−0. 410 to −0. 019; P = 0. 032). By contrast, effects were not observed across all five measures. The pattern suggests that later-generation epigenetic metrics tied to mortality risk are more responsive to nutritional inputs than older first-generation clocks.
Across the two-year follow-up, the slowdown equated to roughly four months less biological aging for participants taking multivitamins. Effects on PCGrimAge were stronger among individuals who entered the trial with accelerated biological aging (−0. 236) than among those with normal or decelerated aging (−0. 013; P = 0. 018 for interaction). Researchers indicated that greater responses among faster-aging participants could reflect more pronounced baseline nutritional deficits, though this explanation requires further testing.
Centrum Silver and cocoa extract
The tested multivitamin was Centrum Silver. The cocoa arm provided 500 mg of cocoa flavanols per day, including 80 mg (−)-epicatechin. Cocoa extract did not alter any of the five epigenetic clocks and showed no interaction with the multivitamin. This result narrows the likely driver of the observed clock changes to the multivitamin formula at the doses and duration tested.
Funding for the research included support from Mars, a confectionery manufacturer. Authors emphasized that while the multivitamin effects were statistically significant, they were small, and further work is needed to determine whether modest epigenetic shifts translate into health benefits that matter to patients. A separate large study published last year found daily multivitamins did not extend lifespan and might raise early-death risk, a reminder that changes in biomarkers do not automatically predict clinical outcomes.
What remains unresolved is whether the documented improvements in PCGrimAge and PCPhenoAge correspond to fewer age-related diseases or longer survival. If that link is confirmed in future trials, the data suggests targeting multivitamins to people with accelerated baseline aging could prove more effective than universal supplementation.
