Association between pace of biological aging and cancer and the modulating role of physical activity: a national cross-sectional study

Abstract

Background: Cancer remains a serious public health problem impeding gains in life expectancy. Epigenetic clocks, derived from sets of DNA methylation CpGs and mathematical algorithms, have demonstrated a remarkable ability to indicate biological aging and age-related health risks. Dunedin(P)ace(o)f(A)ging(m)ethylation is a single-timepoint DNA methylation clock. It is an aging speedometer rather than a state measure. The association between the DunedinPoAm-measured pace of biological aging and cancer risk based on a nationally non-institutionalized sample remains to be elucidated. Physical activity, a modifiable lifestyle factor, is associated with delayed biological aging and lower risks of developing cancer. We hypothesized that DunedinPoAm-measured pace of biological aging is positively associated with cancer risk, and physical activity moderates this association.

Results: In total, 2,529 participants aged 50 or older from the National Health and Nutrition Examination Survey (NHANES) 1999-2002 were included. Weighted logistic regression calculating odds ratios (OR) and 95% confidence intervals (CI) showed that when scaled per 1-SD increase, DunedinPoAm was positively associated with cancer risk (OR, 95% CI) (1.21, 1.05-1.39) in the crude model and adjusted for age and sex (1.19, 1.01-1.40). Individuals of high DunedinPoAm tertile had a 68% (95% CI 1.16-2.43) increase in cancer risk compared with the low tertile (P trend < 0.001). As hypothesized, effect modification by physical activity was significant (P interaction = 0.013). The association was apparent in physically inactive participants (1.52, 1.16-2.00), whereas insignificant in physically active individuals (1.08, 0.89-1.32). Exploratory interaction analyses for other covariates showed significant effect modification by age (> 65 years, 1.38, 1.08-1.77 vs 50-65 years, 1.00, 0.79-1.27).

Conclusion: The study supported the hypothesis by demonstrating a positive association between the DunedinPoAm-measured pace of biological aging and cancer risk and a modulating role of physical activity. Physically inactive individuals or participants over 65 years showed increased susceptibility to this association. These findings suggest that incorporating the DunedinPoAm-measured pace of biological aging into cancer screening strategies may benefit those with physically inactive lifestyles and older individuals. Whether physical activity can mitigate the increased risk of cancer in individuals with a faster pace of biological aging needs to be validated in further interventional cohort studies.

Keywords: Aging; Biological aging; Cancer; Epigenetic alteration; Methylation; Physical activity; Risk.

Fig. 1

Correlation between DunedinPoAm-measured pace of biological aging and chronological age. In the bottom left panel, each dot represents each individual with (red) or without cancer (blue), the X-axis indicates age, and the Y-axis indicates the pace of biological aging. The histograms in the top left and bottom right panels present the distributions of chronological age and the pace of biological aging estimated using DunedinPoAm in cancer (red) and non-cancer participants (blue). The Pearson correlation coefficient (Pearson’s r) was computed using R function cor

Fig. 2

Restricted cubic spline showed a linear association between DunedinPoAm-measured pace of biological aging and cancer. The OR (solid line) was adjusted for age, sex, race and ethnicity, education level, smoker, drinker, and physical activity. Shaded areas represent 95%CIs. The model was conducted with 3 knots

Fig. 3

Stratified analyses of the association between DunedinPoAm-measured pace of biological aging and cancer. The weighted multivariate logistic regression computes OR adjusted for age (50–65, > 65), sex, race and ethnicity, education level, smoking, drinking, and physical activity, except for the stratification variable itself. Effect modification for physical activity, race, and age was statistically significant after applying the Benjamini–Hochberg FDR correction. *Exploratory due to the small sample size (n = 83).

Fig. 4

Stratified analyses of the association between DunedinPoAm and cancer by physical activity (a) and age (b). The weighted multivariate logistic regression computes OR adjusted for age (50–65, > 65), sex, race and ethnicity, education level, smoking, drinking, and physical activity, except for physical activity (in 4a) or age (in 4b) and the stratification variable itself. *Exploratory due to the small sample size (n = 83). PA, physically active; NPA, physically inactive.