Introduction
New randomized trial data have thrust multivitamin ageing back into the spotlight: across 958 participants with an average age near 70, daily multivitamin–multimineral supplementation for two years produced measurable slowing on some epigenetic clocks. The effect was modest — roughly four months less biological ageing across two years — yet it raises urgent questions about who might benefit and whether biomarkers translate to real clinical gains.
Multivitamin Ageing and the COSMOS randomized trial
The COcoa Supplement and Multivitamin Outcomes Study (COSMOS) provided the randomized framework for assessing multivitamin impacts on biological ageing. The ancillary analysis published in Nature Medicine evaluated 958 participants (482 women, 476 men) randomized into four groups: multivitamin and cocoa extract; multivitamin and cocoa placebo; cocoa extract and multivitamin placebo; or two placebos. Blood samples collected at baseline, year one and year two were analyzed for changes in five DNA methylation measures — often described as epigenetic clocks — used to estimate biological age.
What the epigenetic clocks reveal
The trial tested five methylation-based aging metrics: PCHannum, PCHorvath, PCPhenoAge, PCGrimAge and DunedinPACE. Compared with placebo, daily multivitamin supplementation produced statistically significant but small reductions in the rate of increase for two second‑generation clocks linked to mortality risk: PCGrimAge and PCPhenoAge. The Nature Medicine report quantified the between‑group yearly differences as −0. 113 years for PCGrimAge and −0. 214 years for PCPhenoAge, and the investigators summarized the two‑year effect as approximately four months less biological ageing for those taking the multivitamin. Cocoa extract showed no effect on any of the five clocks and did not modify the multivitamin result.
Notably, the multivitamin effect on PCGrimAge was larger among participants who had accelerated biological ageing at baseline, suggesting baseline nutritional status or preexisting deficits may mediate responsiveness. The trial team highlighted that these epigenetic markers may reflect broader age‑related risk factors beyond nutrition alone.
Expert perspective and implications
Dr Howard Sesso, epidemiologist, Mass General Brigham Department of Medicine and senior author of the work, emphasized caution: “There are no known risks for taking a multivitamin in our two large clinical trials. At the same time, we do not know for sure who benefits, and how. ” He added that determining the clinical relevance of these biomarker changes is critical.
The study authors and accompanying commentary in the published journal underscore two linked imperatives: first, to test whether modest shifts in epigenetic clocks predict meaningful reductions in age‑related illness; second, to identify subgroups — for example, those with accelerated baseline biological ageing or nutritional deficits — for whom supplementation may yield larger benefits. The trial also references prior COSMOS analyses that associated multivitamin use with improvements in domains such as cognition and reductions in select conditions, but the current analysis stops short of asserting direct clinical outcomes from the epigenetic changes alone.
Broader consequences and the path ahead
These findings complicate public health messaging around routine supplement use. On one hand, a readily available daily multivitamin produced modest, reproducible shifts on validated epigenetic measures in a large randomized sample. On the other, prior large trials have not demonstrated clear life‑extension benefits and in some analyses raised concerns about potential harms, stressing the need for targeted research.
For clinicians and policymakers, the study frames a precise question: can inexpensive, low‑risk interventions that alter biomarkers of aging also reduce incidence or severity of age‑related disease? Answering that will require longer follow‑up, larger samples, and preplanned analyses linking epigenetic change to hard clinical endpoints. COSMOS data access procedures were described in the published trial documentation to allow further investigation by qualified investigators.
Conclusion
After two years, multivitamin ageing shows a measurable but modest signal on select epigenetic clocks, concentrated among those with faster baseline biological ageing. The finding prompts a pragmatic question for researchers and clinicians: will this biomarker slowing translate into more years of healthy life, and who should be prioritized for next‑phase trials?
