Hormone replacement therapy through the lens of epigenetic clock: estrogen signals, glycemic status, and lifespan
摘要
Epigenetic age acceleration (EAA) is a critical biomarker of aging. Hormone replacement therapy (HRT) is commonly used to alleviate age-related diseases, while glycemic status also influences the aging process. We aimed to investigate the association between estrogen and EAA, and the effects of HRT, EAA, and glycemic status on mortality.
MethodsMendelian randomization (MR) established causal relationships between genetically predicted estrogen receptor expression and EAA markers (HorvathAA, HannumAA, PhenoAA, and GrimAA). Using data from the National Health and Nutrition Examination Survey, we employed Cox proportional hazards models to analyze the association between HRT and mortality, exploring whether it was modified by EAA and glycemic status.
ResultsElevated estrogen receptor α (ESR1) expression significantly decelerated PhenoAA (β -1.26; 95% CI: -2.16, -0.37; P = 0.005). HRT significantly correlated with a reduced risk of all-cause (HR: 0.65, 95% CI: 0.47–0.88; P = 0.006) and cardiovascular (HR: 0.52, 95% CI: 0.28–0.95; P = 0.033) mortality. Additionally, there was a significant three-way interaction among HRT, two types of EAA (HorvathAA and PhenoAA), and abnormal glucose status. Stratified analyses showed no significant interaction in the normal glucose subgroup; however, a significant antagonistic interaction emerged between HRT and both HorvathAA and PhenoAA in the abnormal glucose subgroup. Consequently, although HRT exhibited a protective main effect in patients with abnormal glucose, its survival benefits were significantly attenuated as their epigenetic aging accelerated.
ConclusionsEstrogen may decelerate PhenoAA via the ESR1 pathway. In postmenopausal women, HRT is significantly associated with a reduced risk of mortality; however, this survival benefit is substantially modified by glycemic status. Future studies should evaluate whether incorporating EAA and glycemic status into HRT decision-making could optimize clinical outcomes.