An epigenome-wide association study of 3,857 postmenopausal women in the Women's Health Initiative identified 841 stress-related DNA methylation sites — 742 hypomethylated and 99 hypermethylated under psychosocial stress — that collectively predict incident coronary heart disease over a mean 16.7-year follow-up. A condensed 13-site methylation risk score (MRS13) validated independently in the Jackson Heart Study (n=3,053) and MESA (n=870), yielding hazard ratios of 1.33–1.37. Critically, these methylation signatures mediated 16.5–17.7% of the stress-to-CHD pathway. Cell-type deconvolution pinpointed monocytes — innate immune sentinels central to atherosclerosis — as the dominant locus of stress-driven epigenetic change, with 124 hypomethylated sites. The implicated genes include TNF and ALDH2, connecting chronic stress to inflammatory and oxidative-stress pathways long suspected in cardiovascular pathogenesis.
This preprint, not yet peer-reviewed, represents a meaningful methodological advance: moving beyond associating self-reported stress with disease endpoints to quantifying a durable biological imprint stress leaves on the epigenome. Most prior methylation-cardiovascular work has lacked multi-cohort validation and longitudinal depth; this study provides both. The monocyte-specific signal is particularly compelling, mechanistically tying psychological burden to myeloid-driven arterial inflammation. Limitations include the all-female, postmenopausal cohort restricting generalizability, and the observational design prevents causal inference. The MRS13's modest hazard ratio (~1.34) also suggests limited standalone clinical utility, but its potential as a refinement layer atop existing risk calculators is real. If peer review confirms these findings, stress-derived methylation scores could eventually guide early intervention in high-risk populations.
