Exercise May Slow the Aging Clock, Especially When Sleep Is Good

Biological aging clocks have become one of the most talked-about tools in modern longevity science, promising to reveal how fast a body is truly aging beneath the surface. Now a new study from France suggests that the lifestyle levers we already know about, namely physical activity, sedentary behavior, and sleep quality, do not act on these clocks in isolation. Instead, they interact, and the right combination may blunt some of the biological wear that poor sleep or long hours of sitting would otherwise inflict. The findings, published in the journal GeroScience, come from a team led by Jérémy Raffin of the IHU HealthAge Gérontopôle in Toulouse, working with Yves Rolland, Sophie Guyonnet, Bruno Vellas, and Philipe de Souto Barreto.

The research drew on the INSPIRE-T cohort, a large translational geroscience platform based in Toulouse that spans animal models, human cohorts, and the healthcare system. Eight hundred and ninety adults aged between 20 and 102 years were included in the analysis, each with a measurement of at least one of five biological aging clocks: PhenoAge, GrimAge, DunedinPACE, iAge, and a blood-based epigenetic clock for intrinsic capacity, referred to as the IC clock. This unusually wide age range, covering more than eight decades of adult life, allowed the researchers to examine how lifestyle factors relate to biological aging across the entire adult lifespan rather than only in older adults.

Biological aging clocks are statistical models that estimate biological age from molecular signatures, most commonly DNA methylation patterns across the genome. GrimAge, for example, embeds estimates of plasma proteins such as cotinine and CRP into its methylation model and strongly predicts lifespan and healthspan. DunedinPACE estimates the pace of aging, essentially how many years of biological change occur per chronological year. PhenoAge captures systemic inflammation and metabolic decline, while iAge is a deep-learning-derived inflammatory aging clock that tracks multimorbidity, immunosenescence, frailty, and cardiovascular aging. The IC clock, developed by the same Toulouse group, predicts mortality and is associated with clinical, immunological, and lifestyle factors, making it a functional complement to the purely predictive clocks.

To capture lifestyle, the team used validated questionnaires. Physical activity levels were estimated with the International Physical Activity Questionnaire, sitting time was recorded as a proxy for sedentary behavior, and sleep deterioration was assessed through a sleep quality measure. The researchers then explored the associations of age acceleration or aging rate with each lifestyle factor and, crucially, with their interactions, using robust regression models adjusted for age, sex, education, and comorbidities. Robust regression is a statistical technique that limits the influence of outliers, and the false discovery rate was controlled across the many tests performed, a standard safeguard when examining multiple clocks and multiple lifestyle factors simultaneously.

The main-term analyses, which looked at each lifestyle factor on its own, produced a relatively modest result. Higher levels of physical activity were associated with reduced age acceleration for the IC clock in women. This sex-specific finding is notable because epigenetic clocks have previously been shown to differ between men and women, and twin studies have suggested that clocks may partly explain sex differences in lifespan. The absence of broad independent associations for the other clocks and factors is itself informative, hinting that the real story lies in how these lifestyle components combine rather than in any single behavior acting alone.

That is precisely what the interaction analyses revealed. In older adults, the adverse association between sleep deterioration and biological aging was attenuated with higher levels of physical activity for GrimAge, iAge, DunedinPACE, and the IC clock. In other words, among older participants, being physically active appeared to soften the biological damage that poor sleep would otherwise register on multiple aging clocks. A balanced combination of physical activity and sitting time produced a similar buffering effect for DunedinPACE and GrimAge, suggesting that the ratio between movement and stillness matters as much as either behavior alone.

The findings on sitting time were equally striking. In the total population and in men, the detrimental association between sitting time and GrimAge was attenuated, or even reversed, with higher levels of physical activity, especially in individuals with good sleep quality. This mirrors a large body of mortality research, including a harmonised meta-analysis of more than one million men and women published in The Lancet, which asked whether physical activity attenuates or even eliminates the detrimental association of sitting time with mortality. The new study extends that question to the molecular level, showing that the sitting-versus-activity balance registers on epigenetic aging clocks, and that good sleep appears to amplify the protective effect of activity against sedentary time.

From a mechanistic perspective, the results fit with what is known about how exercise modulates the hallmarks of aging. Physical activity has been shown to attenuate major aging hallmarks including chronic inflammation, mitochondrial dysfunction, and cellular senescence, while sedentary behavior has been linked to the biological hallmarks of aging in a 2023 review by members of the same Toulouse group. Sleep deterioration, meanwhile, has been associated with accelerated biological aging in multiple cohorts, including Mendelian randomization analyses suggesting that sleep traits causally affect epigenetic age acceleration, and UK Biobank studies showing U-shaped associations between sleep duration and biological aging. The interaction effects observed here suggest that these pathways are not additive but compensatory, with activity partially offsetting the inflammatory and metabolic signatures that poor sleep and prolonged sitting imprint on DNA methylation.

The study has limitations that the authors and readers should keep in mind. The design is cross-sectional, meaning that lifestyle and clock measurements were taken at the same time, so causality cannot be established. It is possible that people whose clocks show slower aging are simply more able to be active, sleep better, and sit less, rather than the reverse. Lifestyle factors were measured by questionnaire rather than objective devices, and recent work using in-home electroencephalography has revealed discrepancies between subjective and objective sleep assessments, so misclassification is possible. The clocks themselves also differ in what they capture, from pure mortality prediction to inflammatory state to pace of aging, which may explain why associations varied across clocks and population subgroups.

Even so, the practical message is clear and encouraging. The findings suggest that maximizing sleep quality remains the first priority, but that increasing physical activity and limiting sedentary behavior, or achieving an optimal balance between the two, can buffer some of the biological aging associated with deteriorated sleep, particularly in older adults. For a field racing to translate aging clocks from the laboratory into clinical practice, the study adds an important nuance: the aging clock is not just a readout of time, but a molecular ledger that records how we move, sit, and sleep, and the entries interact in ways that lifestyle choices can still influence well into old age.

Subject of Research: Physical activity, sedentary behavior, sleep quality and the biological aging clocks

Article Title: Physical activity, sedentary behavior, sleep quality and the biological aging clocks

Article References: Raffin, J., Rolland, Y., Guyonnet, S., Vellas, B., de Souto Barreto, P., I. H. U. HealthAge INSPIRE/Open Science study group, Brigitte, L., Milhet, A., Paez, E., Muller, E., Le Floch, S., Takeda, C., Faisant, C., Lala, F., Abellan Van Kan, G., Steinmeyer, Z., Comte, M., Daniaud, N., Boissou-Parachaud, F., … Champigny, C. (2026). Physical activity, sedentary behavior, sleep quality and the biological aging clocks. GeroScience. https://doi.org/10.1007/s11357-026-02592-2

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02592-2

Keywords: Physical, activity, sedentary, behavior, sleep, quality, biological, aging, clocks, scientific research