Association of blood cadmium levels with epigenetic age acceleration in U.S. adults aged > 50 years

Panpan Mi, Xu Cao, Haixia Feng, Huijie WangFrontiers in Public Health 2025 · DOI

TL;DR

This study reveals a significant link between blood cadmium levels and accelerated biological aging in U.S. adults over 50. Using data from the NHANES database, researchers found that individuals with higher cadmium exposure showed "older" biological profiles according to advanced epigenetic clocks, specifically PhenoAge and GrimAge, even after adjusting for lifestyle factors like smoking.

What Problem Does This Paper Address?

We all know someone who "looks young for their age" or, conversely, someone whose health seems to decline faster than their birth certificate would suggest. This discrepancy is the difference between chronological age (years lived) and biological age (the functional state of your body).

While we know that environmental toxins are bad for us, quantifying exactly how they "age" us at a cellular level is difficult. Cadmium (Cd) is a heavy metal and a ubiquitous pollutant found in tobacco smoke, air pollution, and certain foods (like leafy greens and organ meats). While its toxicity is well-documented, its specific impact on epigenetic age acceleration (EAA)—a measure of how much faster your "biological clock" is ticking compared to your chronological years—has remained under-explored in older U.S. populations. This paper seeks to bridge that gap.

Methods

The researchers conducted a cross-sectional study using data from the National Health and Nutrition Examination Survey (NHANES) from 1999–2002. They focused on a cohort of 1,958 adults aged 50 and older.

Key technical steps included: * Cadmium Measurement: Blood Cd levels were measured using atomic absorption spectrometry. * Epigenetic Clocks: The team utilized DNA methylation (DNAm) data to calculate four distinct "clocks": * First-Generation: HorvathAge and HannumAge (primarily trained to predict chronological age). * Second-Generation: PhenoAge and GrimAge (trained to predict clinical biomarkers and mortality). * Statistical Modeling: They used weighted linear regression to calculate "Age Acceleration" (the residual difference between epigenetic age and chronological age). The models were adjusted for a wide array of confounders, including BMI, race, education, poverty index, and—crucially—smoking status.

Datasets

The study relied entirely on the NHANES (1999–2002) database. This is a high-quality, nationwide survey designed to assess the health and nutritional status of adults and children in the U.S. The researchers specifically extracted the subset of participants who had both DNAm-based epigenetic biomarkers and blood Cd data available.

Key Insights and Ideas

  • Not All Clocks Are Equal: The most striking finding was that cadmium exposure didn't affect all epigenetic clocks the same way. It significantly accelerated PhenoAge and GrimAge, but had little to no effect on HorvathAge. This suggests that cadmium impacts the specific biological pathways related to morbidity and mortality (which the second-gen clocks track) rather than just general cellular aging.
  • The Smoking Factor: Cadmium is a major component of cigarette smoke. While smokers naturally had higher Cd levels, the association between Cd and age acceleration remained significant even in non-smokers. This indicates that environmental and dietary exposure to cadmium is a standalone risk factor for accelerated aging.
  • Threshold Effects: The study used a threshold of 0.5 µg/dl to define "high" cadmium. Even at these relatively low levels, the biological impact was measurable, suggesting there may be no "safe" level of exposure when it comes to epigenetic integrity.

Key Results

  • PhenoAge Acceleration: Participants in the high cadmium group (≥0.5 µg/dl) showed an age acceleration of 1.37 years ($\beta = 1.37, P = 0.017$) compared to the low-exposure group.
  • GrimAge Acceleration: The high cadmium group showed an acceleration of 1.31 years ($\beta = 1.31, P = 0.003$).
  • HannumAge: Showed a significant association in the binary model ($\beta = 0.94, P = 0.016$), but this lost significance when cadmium was treated as a continuous variable.
  • HorvathAge: No significant association was found, likely because this clock is less sensitive to environmental stressors that drive age-related disease.

Limitations and Future Work

The authors acknowledge several important caveats:

  1. Cross-Sectional Design: Because the data is a snapshot in time, the study cannot prove that cadmium causes the aging; it only proves they are strongly associated.
  2. Single Measurement: Blood cadmium reflects relatively recent exposure. Since cadmium has a long half-life in the body (stored in kidneys and liver), a single blood test might not capture the full lifetime "burden" of the metal.
  3. Residual Confounding: While they adjusted for many factors, other unmeasured environmental exposures could still be influencing the results.

Future Work: The researchers call for longitudinal studies to track individuals over decades to see if reducing cadmium exposure can actually "slow down" the epigenetic clock and prevent age-related diseases.

Summary updated

Abstract

In this study, after adjusting for confounders, blood Cd levels were positively associated with PhenoAge acceleration and GrimAge acceleration in people over 50 in the United States. These results may be useful in proposing interventions in environmental exposures to slow the aging process and prevent age-related diseases.
BibTeX
@inproceedings{mi2025association,
  author = {Panpan Mi and Xu Cao and Haixia Feng and Huijie Wang},
  title = {{Association of blood cadmium levels with epigenetic age acceleration in U.S. adults aged \> 50 years}},
  booktitle = {Frontiers in Public Health},
  year = {2025},
  doi = {10.3389/fpubh.2025.1504830},
}
Figure