Identification of DNA methylation signatures of smoking in women and associations with cardiovascular disease mortality.

Tobacco smoking is critical risk factor for cardiovascular disease (CVD) and mortality, with epigenetics implicated as a key biological mechanism. Smoking disproportionately impacts CVD risk in women; however, limited research has examined smoking-associated DNA methylation (DNAm) changes in women. We measure whole blood DNAm (EPICv2) and self-reported smoking (three-category, pack-years, and years since cessation) via questionnaires in 6,005 Women's Health Initiative Memory Study participants. We perform epigenome-wide association studies and develop epigenetic predictors of each smoking phenotype, and compare DNAm and self-reported pack-years for CVD-related and all-cause mortality prediction. Here we find 20,388 FDR-significant sites related to current smoking, 2149 for former smoking, 14,709 for pack-years, and 753 for smoking cessation. The epigenetic predictors of smoking activity are highly correlated with their respective phenotypes and DNAm pack-years outperforms self-reported pack-years in mortality prediction. We report robust validation of previous findings, novel smoking associations in women, and introduce epigenetic predictors of smoking activity.