411-P: Potential to Utilize DNA Methylation as a Biomarker to Predict Major Coronary Artery Disease (CAD) Risk in Type 1 Diabetes (T1D)

DNA methylation (DNAme) has been associated with risk of future CAD in the general population and is a potential target for pharmacological intervention. In T1D, DNAme has been associated with microvascular complications but its association with CAD has not been examined in this high risk population. Thus, we performed an epigenome-wide association study (EWAS) of DNAme and 28-year major CAD (CAD death, nonfatal myocardial infarction) incidence in an observational T1D cohort. Whole blood DNAme was assayed via Illumina EPIC beadchip in the Pittsburgh Epidemiology of Diabetes Complications (EDC) study of childhood onset (<17 years) T1D (n=400; mean age 29, T1D duration 21 yrs) . After quality control DNAme at 683,597 CpGs was analyzed in Cox models for time-to-CAD, adjusting for T1D duration, sex, smoking pack-years, estimated blood cell composition, and technical/batch covariates. Over 28 years 102 (25%) developed major CAD. DNAme at two CpGs was associated with lower CAD risk: cg06125996 (Chr 5, NLN; log HR per 5% DNAme= -1.10, p=6.0e-11) and cg051638 (Chr 9, CARD9; log HR= -0.97, p=1.2e-08) . Other associations with FDR<5% included CpGs in MRPS21, ZNF428, PCMTD1, and LYRM4. KEGG pathway analysis of all CpGs with FDR<10% (n=44) suggested enrichment for renin-angiotensin-aldosterone system, lipid metabolism, and immune system pathways. In GWAS we identified methylation quantitative trait loci (QTL) associated with DNAme at cg06125996 and cg05163804, three of which were annotated as expression QTLs in GTEx including SNPs in genes involved in insulin receptor signaling (SIK2) and cytokine activity (CMTM2, CMTM4) , and calcium ion binding (EFHB) . DNAme was not correlated with cross-sectional HbA1c at either CpG and adjusting for baseline HbA1c did not alter CpG-CAD associations. Our results demonstrate that DNAme may be an early marker of CAD risk in T1D, but the potential role of glycemic exposure in this association is unclear and warrants further study.