Abstract
BACKGROUND: Tobacco smoking remains a major global health burden and is a leading risk factor for cardiovascular disease and cancer. Accumulating evidence suggests that tobacco smoke induces widespread alterations in DNA methylation, which may contribute to smoking-related morbidity and mortality. METHODS: The Ludwigshafen Risk and Cardiovascular Health (LURIC) study is a monocentric prospective cohort including 3316 patients referred for coronary angiography. Genome-wide DNA methylation was assessed in 2423 participants using the Illumina HumanMethylationEPIC BeadChip. A discovery-replication design was applied (discovery n = 1262; replication n = 1161). Associations between smoking status (never/former vs. current) and CpG-specific methylation levels were evaluated using multivariable linear regression models. Cox proportional hazards models were used to assess associations with all-cause and cardiovascular mortality. Mediation was examined within a counterfactual framework using natural effect models. RESULTS: In the discovery sample, 14,403 CpG sites were significantly associated with smoking after false-discovery-rate correction (differentially methylated probes, DMPs). Of these, 3,000 were replicated in the independent sample at an FDR-adjusted p value < 0.05. In random-effect meta-analysis of both samples, 24,930 DMPs remained significant after multiple testing correction, of which 11,907 had not been reported in the largest published smoking EWAS to date. Among former smokers, a subset of DMPs remained differentially methylated more than 10 years after smoking cessation, indicating long-term persistence of smoking-associated epigenetic alterations. A CpG score constructed from mortality-associated DMPs was strongly associated with all-cause mortality. Inclusion of this score in Cox regression models attenuated the association between smoking status and mortality. Mediation analysis demonstrated a statistically significant natural indirect effect of smoking on all-cause mortality via the CpG score. CONCLUSIONS: Tobacco smoking is associated with widespread, exposure-dependent alterations in DNA methylation, many of which persist for years after cessation. These epigenetic changes are strongly linked to mortality risk and may represent an important biological pathway underlying the association between smoking and adverse health outcomes.