The study, led by Yeonju Kim and Habyeong Kang of Hanyang University in Seoul, drew on data from 1,911 adults aged 50 to 84 who participated in the National Health and Nutrition Examination Survey (NHANES) between 1999 and 2002. NHANES is a long-running program run by the National Center for Health Statistics that combines interviews, physical examinations, and laboratory testing on a sample designed to represent the entire US population. For this analysis, the researchers took advantage of a relatively recent addition to the survey: DNA methylation measurements, which capture patterns of methyl groups — small chemical modifications — attached to cytosine bases across the genome.

DNA methylation is central to what scientists call epigenetic aging clocks. These are mathematical models, trained on large datasets, that read methylation patterns at specific sites in the genome and convert them into an estimate of biological age. Because methylation patterns shift in predictable ways with stress, disease, and lifestyle exposures, the estimated epigenetic age can diverge from chronological age. When it runs ahead, researchers call the difference epigenetic age acceleration, or EAA — a measure that has repeatedly been linked to elevated risks of cardiovascular disease, cancer, and earlier death.

Participants in the study were classified as never smokers, former smokers, or current smokers, with former smokers further divided by how long they had quit: less than 10 years, 10 to just under 20 years, or 20 or more years. Rather than relying on a single clock, the team evaluated six validated DNA methylation clocks spanning three generations of development. The first-generation clocks — HorvathAge, HannumAge, and SkinBloodAge — were the original models introduced in 2013 and 2018 that estimate chronological age from methylation patterns. The second-generation clocks, PhenoAge and GrimAge2, were designed to predict clinical outcomes such as mortality and disease rather than birthday count alone. The third-generation clock, DunedinPoAm, estimates the pace of biological aging itself, capturing how quickly a person’s body is deteriorating relative to peers.

The results were strikingly consistent. Across all six clocks, current smokers showed the highest epigenetic age acceleration of any group, confirming that active smoking is associated with a measurably faster biological aging profile. But the most important finding concerned former smokers. The researchers observed a dose–response relationship: the longer a person had been quit, the lower their epigenetic age acceleration. This pattern held across the panel of clocks but was strongest for GrimAge2 and DunedinPoAm, the two clocks most closely tied to mortality risk and the pace of aging.

The magnitude of the effect among long-term quitters was remarkable. Former smokers who had abstained for 20 years or more showed a reduction in GrimAge2 acceleration of 12.66 years compared with current smokers, with a 95 percent confidence interval spanning 14.56 to 10.75 years. That figure approached the level observed among never smokers, whose GrimAge2 acceleration was 13.88 years lower than current smokers, with a confidence interval of 15.70 to 12.05 years. In other words, after two decades of cessation, the epigenetic aging profile of former smokers was statistically close to that of people who had never picked up the habit.

To understand the timing of this recovery, the researchers used restricted cubic spline analysis, a statistical technique that models the relationship between cessation duration and epigenetic aging without assuming a straight-line pattern. The splines revealed a rapid initial decline in epigenetic age acceleration during the first decade after quitting, followed by continued, more gradual improvement beyond the 20-year mark for the GrimAge2 and DunedinPoAm clocks. This shape suggests that the body begins unwinding the epigenetic damage from smoking relatively quickly, but that full recovery is a long-term process measured in decades rather than months.

The findings fit with a broader body of molecular evidence. Cigarette smoke leaves a well-documented signature on the methylome, including changes at sites in the AHRR gene that are among the most robustly replicated epigenetic markers of smoking exposure. Previous work has shown that some of these methylation changes begin to revert after cessation, and studies have linked accelerated epigenetic aging to cardiometabolic abnormalities, kidney dysfunction, and other hallmarks of biological decline. What the new study adds is a nationally representative, survey-weighted analysis that quantifies how the reversal unfolds over time across multiple generations of aging clocks.

The technical rigor of the analysis matters for interpreting the results. Because NHANES uses a complex sampling design, the researchers applied survey weights in their regression models, ensuring that the estimates reflect the US population rather than the specific sample. The use of six independent clocks guards against the possibility that the association is an artifact of any single algorithm, and the consistency of the dose–response pattern across first-, second-, and third-generation clocks strengthens the biological interpretation. Still, as with any cross-sectional study, the data capture a single moment in time; the researchers measured different people at different cessation durations rather than following the same individuals over the years after they quit.

That limitation leaves open some questions that longitudinal studies will need to address, including whether individual trajectories of epigenetic recovery mirror the population-level pattern and how factors such as smoking intensity, duration of prior smoking, sex, and genetic background modify the pace of reversal. Even so, the implications are significant. Tobacco smoking remains one of the leading modifiable risk factors for age-related disease and premature mortality worldwide, and health campaigns have long emphasized the clinical benefits of cessation. This study adds a biological dimension to that message: quitting does not merely halt further damage, but appears to set in motion a measurable molecular recovery that brings the aging clock back toward baseline. For the millions of former smokers wondering whether the harm is already done, the answer, at the level of DNA methylation, appears to be that much of it can be undone — given enough time smoke-free.

Subject of Research: The relationship between smoking cessation duration and DNA methylation-based epigenetic age acceleration in US adults

Article Title: Duration of smoking cessation and DNA methylation aging clocks among US adults: potential reversibility of smoking-induced biological aging

Article References: Kim, Y., & Kang, H. (2026). Duration of smoking cessation and DNA methylation aging clocks among US adults: potential reversibility of smoking-induced biological aging. GeroScience. https://doi.org/10.1007/s11357-026-02535-x

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02535-x

Keywords: smoking cessation, epigenetic clocks, DNA methylation, biological aging, epigenetic age acceleration, NHANES, GrimAge2, DunedinPoAm, tobacco smoking, GeroScience, healthy aging, epigenetics