Cross-sectional and Longitudinal Relationship between Sex Hormones and Six Epigenetic Clocks in Older Adults: Results of the Berlin Aging Study II (BASE-II)

Hannah Schmid,Valentin Max Vetter,Jan Homann,Vivien Bahr,Christina M. Lill,Vera Regitz-Zagrosek,Lars Bertram,Ilja Demuth

DOI: https://doi.org/10.1101/2024.11.15.24317371

2024-11-15

Abstract:Background Beyond their essential roles in regulating reproduction and development, sex hormones play a crucial role in the aging processes. Observational studies have indicated that low sex hormone concentrations in older age are associated with adverse health events. DNA methylation age acceleration (DNAmAA) estimated from epigenetic clocks quantifies differences in biological aging. DNAmAA was previously shown to be associated with age at menopause, ovariectomy, hormone replacement therapy and testosterone level. Methods We analysed the relationship between estradiol, dehydroepiandrosterone sulfate (DHEAS) and the Free Androgen Index (FAI) with DNAmAA estimators from six epigenetic clocks (Horvath's, Hannum's, 7-CpG clock, PhenoAge, GrimAge, DunedinPACE) in 1,404 participants of the Berlin Aging Study II (BASE-II, mean age at baseline 68.7 +/-3.7 years, 48% women). The relationship was investigated in multiple linear regression models cross-sectionally at two time points and longitudinally over on average 7.3 years of follow-up. Results We did not observe any consistent associations between the sex hormones and DNAmAA estimators investigated. However, we found several nominal associations (alpha=0.05) of unclear relevance. For instance, we identified an inverse association between DHEAS and Horvath's DNAmAA, i.e. a reduced biological age with higher DHEAS levels in men at baseline. In women we found an inverse association between estradiol and DunedinPACE (baseline) and a positive association with GrimAge (follow-up). In longitudinal analyses, Delta DHEAS and Delta DunedinPACE were inversely associated in both sexes. Conclusions Our results suggest that sex hormones play at best a minor role with respect to biological aging in the older population studied here.

Geriatric Medicine

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to explore the cross - sectional and longitudinal relationships between sex hormones (such as estradiol, dehydroepiandrosterone sulfate (DHEAS) and free androgen index (FAI)) and DNA methylation age acceleration (DNAmAA) estimated by six different epigenetic clocks. Specifically, the study aims to analyze the relationships between these sex hormones and the biological age acceleration estimated by six epigenetic clocks (Horvath’s, Hannum’s, 7 - CpG clock, PhenoAge, GrimAge, DunedinPACE) at two time points (with an interval of about 7 years), and to study the changes of these variables during the observation period.

### Research Background
1. **The Role of Sex Hormones**: In addition to their crucial roles in regulating reproduction and development, sex hormones also play important roles in the aging process. Observational studies have shown that low sex hormone concentrations in old age are associated with adverse health events.
2. **DNA Methylation Age Acceleration (DNAmAA)**: DNAmAA is calculated from DNA methylation data through epigenetic clock algorithms and is used to quantify the difference between biological age and chronological age. Previous studies have shown that DNAmAA is related to menopausal age, oophorectomy, hormone replacement therapy and testosterone levels.

### Research Hypothesis
Based on previous reports, the authors hypothesized that sex hormone concentrations are negatively correlated with DNAmAA, and this association can also be reflected in longitudinal data.

### Research Methods
- **Research Subjects**: 1,404 participants in the Berlin Aging Study II (BASE - II) (average age 68.7 ± 3.7 years, 48% female).
- **Data Analysis**: Multiple linear regression models were used for cross - sectional and longitudinal analyses, adjusting for potential confounding factors (such as chronological age, BMI, smoking history and alcohol intake).

### Main Findings
1. **Cross - sectional Analysis**:
- In men, DHEAS was negatively correlated with Horvath’s DNAmAA (β = - 0.0007, p = 0.032), that is, for every one - standard - deviation increase in DHEAS, Horvath’s DNAmAA decreased by 0.45 years.
- In women, estradiol was negatively correlated with DunedinPACE (β = - 0.0004, p = 0.032), that is, for every one - standard - deviation increase in estradiol, DunedinPACE decreased by 0.01 years.

2. **Longitudinal Analysis**:
- ΔDHEAS was negatively correlated with ΔDunedinPACE in both men (β = - 0.00002, p = 0.022) and women (β = - 0.00003, p = 0.005), that is, for every one - standard - deviation increase in DHEAS, DunedinPACE decreased by 0.009 years and 0.01 years respectively.

### Conclusion
Overall, this study did not find a consistent and clinically meaningful relationship between sex hormones and DNAmAA. Nevertheless, the study found several nominal associations, but their clinical significance is unclear. The results suggest that sex hormones may play only a secondary role in the biological aging of the studied elderly population.

### Formula Presentation
- The calculation formula for **Free Androgen Index (FAI)** is:
\[
\text{FAI}=\frac{\text{TT (nmol/l)} \times 100}{\text{SHBG (nmol/l)}}
\]
- The calculation formula for **DNAmAA** is:
\[
\text{DNAmAA}=\text{DNAmA}-\text{Chronological Age}
\]

Through these analyses, this study provides new insights into understanding the relationship between sex hormones and biological aging, but also points out the directions for further exploration in future research.