Is it possible to "refresh" biological age with multivitamins? COSMOS answers

In a large randomized trial, COSMOS, researchers report that daily multivitamin/mineral (MVM) supplementation in older adults over 2 years favorably alters blood metabolite profiles. The effect is particularly noticeable in obese participants, and a reduction in a number of metabolomic “scores” of biological aging is also observed, according to the abstract. The work appears in a supplemental issue of Current Developments in Nutrition.

Background

Why study multivitamins (MVMs) in older adults at all?
As people age, many accumulate hidden micronutrient deficiencies, even with a “normal” diet. Reviews note a significant proportion of vitamin and mineral deficiencies in older adults, with deficiencies occurring more frequently in some risk groups (e.g., obesity or low protein intake). Against this background, MVMs are viewed as a simple way to “plug holes” in the diet. At the same time, large reviews on “hard” outcomes (cancer, CVD, mortality) show either minimal or no benefit, so the focus shifts to functions (brain, metabolism) and molecular mechanisms.

What has the COSMOS project shown so far?
COSMOS is a large, randomized, double-blind trial in ~22,000 older adults testing a daily multivitamin and cocoa extract. In the cognitive outcomes substudies (COSMOS-Mind, COSMOS-Clinic, COSMOS-Web), daily MVM for 2–3 years improved memory and global cognitive functioning compared with placebo; meta-analysis of the substudies confirmed this signal. There were almost no effects on “severe” events (e.g. cocoa extract did not reduce total CV events, although it was associated with reduced CV mortality). The COSMOS protocol also included biomarkers of “inflammaging” and epigenetic aging.

Why Metabolomics Now?
Metabolomics captures hundreds of small molecules in the blood that are sensitive to diet, microbiota, inflammation, and age. In recent years, metabolomic “clocks” and age-scores have emerged that predict mortality and health better than many classic indicators; the metabolome is particularly sensitive to lifestyle and body weight. So a logical step is to test whether long-term MVM supplementation shifts the metabolome to a “more favorable/younger” profile, and how this relates to the cognitive benefits previously shown in COSMOS.

The role of obesity as a moderator of the effect
Obesity not only alters the plasma metabolome, but is also often associated with deficiencies in a number of micronutrients; therefore, the potential effect of MVM on metabolomic signatures may be greater in obese individuals. This makes stratification by BMI and metabolic profile essential.

Context and expectations of the current abstract
The abstract from COSMOS (2-year analysis) published in Current Developments in Nutrition addresses precisely this gap: assessing how daily MVM changes the blood metabolome and associated metabolomic estimates of biological age, with additional interest in subgroups (e.g., obese participants). It continues the COSMOS line on mechanisms (inflammaging/epigenetics) and helps to link the “molecular fingerprint” in the blood to previously shown functional improvements in cognitive tests.

What exactly did they do?

The study is a randomized, double-blind, placebo-controlled study; the analysis pertains to 2-year data from the COSMOS (COcoa Supplement and Multivitamin Outcomes Study) substudy, which compared daily MVM versus placebo in older men and women. Participants underwent periodic plasma metabolomics analysis and calculated integrated metabolomic risk scores (MRS) associated with cardiometabolic health and/or biological aging.

Key Results

• MVM is associated with favorable metabolomics shifts in older adults overall and with a more pronounced effect in obese participants.
• The reduction in 5 of 7 MRS reflecting biological aging is another important signal that regular multivitamin intake can shift the “biological clock” (as measured by metabolomic markers) to a more youthful profile.

Why is this important?

Metabolomics captures hundreds of small molecules—metabolites—that are sensitive to nutrition, inflammation, and aging. If MVM consistently improves these signatures, it will strengthen the overall COSMOS data set, which has previously found benefits for cognitive performance and biological age (across independent marker panels). It also shows mechanistic “fingerprints” in the blood that help us understand “how” the supplement might work.

Important Disclaimers

• This is a conference abstract, not a full paper: design details, precise effects, and statistics are limited. We are waiting for a full publication with methodology and breakdown by metabolite class.
• MVM ≠ nutritional replacement: vitamins fill the “gaps” in the diet, but do not replace vegetables, fish, fiber and physical activity. It is a supplement, not a “cure-all”. (Context of the COSMOS results on the aims and limitations of interventions.)

What does this mean in practice (subject to reservations)

• If you are 60+, rarely meet your daily micronutrient requirements, and are considering MVM in consultation with your physician, new data suggest potential benefits for metabolomic markers and possibly for the rate of biological aging.
• Choose adequate dosages and quality certification (third parties), avoid megadoses of fat-soluble vitamins.
• The effect may be more noticeable in obese people - but this is a signal from the abstract, not a final clinical conclusion. Confirmatory publications are needed.

What's next?

We expect a full-text article with a detailed set of results: which classes of metabolites change (lipids, amino acids, oxidative stress markers, etc.), the persistence of effects and their relationship with clinical outcomes (memory, blood vessels, sugar). It will also be interesting to compare the metabolomic changes with the cognitive improvements previously shown in COSMOS against the background of MVM.

Source: Abstract “Effects of Multivitamin Supplementation on Metabolomic Profiles: 2-Year Findings From the COSMOS Randomized Clinical Trial” in Current Developments in Nutrition (May 2025, Suppl 2); journal issue page; COSMOS program materials and context. DOI: 10.1016/j.cdnut.2025.106058