Introduction: Peripheral artery disease (PAD) is the third leading cause of cardiovascular morbidity and mortality worldwide. Despite significant evidence of an underlying genetic component to PAD susceptibility, only a minimal number of loci have reached genome-wide significance (GWS). Previous PAD GWAS have been limited by statistical power due to sample size; the large number of PAD cases within the Million Veteran Program (MVP), a mega-biobank established by the United States Department of Veterans Affairs (VA), is the ideal setting to identify novel risk loci.
Hypothesis: DNA sequence variants affect PAD risk.
Methods: Using ICD-9 and Current Procedural Terminology (CPT) based algorithms, we identified individuals with and without clinical PAD from among the 353,323 MVP participants genotyped on a customized Affymetrix Axiom Biobank Array. We tested the association of 379,916 directly genotyped DNA sequence variants with clinical PAD separately in participants of European (EUR), African (AFR), and Hispanic (HIS) ancestry using logistic regression models controlling for age, age2, sex and population structure. The results were combined using inverse variance weighted meta-analysis.
Results: Using 3 million person-years of prevalent EHR data, we identified 20,287 individuals (15,708 EUR, 3,492 AFR, 1,187 HIS) with, and 296,158 individuals without, clinical PAD; the remaining individuals failed to meet either case or control definitions and were excluded from analyses. PAD cases were 97% male and had mean age of 69 ± 9 years; controls were 92% male and had a mean age of 62 ± 14 years. Trans-ethnic meta-analysis identified 6 loci associating with PAD at GWS. We replicated known associations at LPA (p=3.0x10-24), 9p21 (p=1.6x10-22) and HDAC9 (p=2.1x10-12), and identified novel associations at F5 (p=3.6x10-11), LPL (p=2.0x10-9) and WWP2/NFAT5 (p=3.0x10-9). Notably, variants at the F5 and WWP/NFAT5 loci have not been previously reported to associate with any atherosclerotic cardiovascular disease.
Conclusions: Using the VA MVP, we assembled the largest reported cohort of individuals with PAD and genetic data, and identified 3 novel PAD risk loci. These loci highlight elements of both common and distinct genetics between CAD and PAD.