Abstract 18660: CVSN Best Abstract Award: Genome-wide Candidates Unique to Females With Coronary Artery Disease Significantly Predict Mortality Risk

Introduction: Women are disproportionately affected by coronary artery disease (CAD). Among major factors such as age and anatomical differences, biologic contributions are purported; yet, little is understood about the underlying genomic architecture that may contribute to female-specific disparities. We explored whether genome-wide variants could predict survival in females with clinically-appreciable CAD, unique to that of male cases. Methods: A sex-stratified genome-wide association analysis was conducted for survivorship in Caucasian patients with CAD in a discovery cohort ( N =1,099; 589 females) and a replication cohort ( N = 404; 198 females), selected from a cardiovascular biorepository consisting of sequential patients referred for cardiac catheterization. CAD cases were defined as having a CAD index >32 (at least one vessel CAD) and indication for catheterization was concern for ischemic heart disease. GWAS genotyping was done using the Illumina Omni Quad v1.0 GWAS chip with stringent QC metrics. Sex-stratified, Cox multivariable regression models for each SNP (additive), adjusted for age and four Eigenstrat-defined principal components, were constructed (R statistical package), with models censored on time from cardiac catheterization to all-cause mortality/last follow-up (med 5.2 yrs). Results: Fourteen SNPs met nominal significance (p<10 -5 ) for increased hazards of all-cause mortality in women; none of these SNPs or genes were even nominally significant at p<0.05 in male cases. Three top SNPs were biologically relevant: rs17591646 ( SLC9A9 ; HR 6.2, 95%CI 2.9-13.3, p discovery =3.2x10 -6 , p replication =0.68, p male =0.87); rs12145981 ( NEK2/LPGAT1 ; HR 3.3, 95%CI 2.0-5.6, p d =4.2x10 -6 , p r =0.05, p m =0.92); and rs7217169 ( RAP1GAP2 ; HR 4.1, 95%CI 2.3-7.5, p d =4.3x10 -6 , p r =0.02, p m =0.44), as they are involved in endosome recycling ( SLC9A9 ), cardiolipin synthesis ( LPGAT1 ), and regulating dense granule secretion from platelets at endothelial damage sites ( RAP1GAP2 ). Conclusions: We present preliminary evidence of genome-wide candidates that appear to uniquely predict mortality risk in women with CAD. Further investigation of genomic variation may help improve understanding of sex differences in CAD disparities.