NEW YORK – Through a genome-wide association study of genomic data from US veterans, researchers have identified novel loci associated with peripheral artery disease risk, including ones that suggest potential treatment options.
For their analysis, Scott Damrauer from the Corporal Michael J. Crescenz VA Medical Center, Philip Tsao from the VA Palo Alto Health Care System, and their colleagues relied on data collected by the Million Veteran Program, which was started in 2011 to study how gene variants influence disease, using genetic and electronic medical record data.
As they reported today in Nature Medicine, the researchers identified 18 novel loci associated with peripheral artery disease (PAD), a condition in which plaque builds up in arteries to restrict blood flow, especially in the legs. A number of these loci have also been implicated in other vascular conditions, suggesting that certain existing treatment approaches might have broader applicability. Other variants, like one linked to clotting, were PAD-specific.
"These results are demonstrative of how large biobanks that couple genetic variation with dense EHR data can be leveraged for biological insights that can inform clinical care," the researchers wrote in their paper.
In their two-phase GWAS, the researchers first conducted separate discovery analyses in 24,009 white veterans with PAD, 5,373 black veterans with PAD, and 1,925 Hispanic veterans with PAD before combining them into a meta-analysis. All the cases had a median minimum ankle-brachial index, a measure of blood flow to the limbs, of less than 0.9, while the 211,753 controls had a median mABI of about 1.
More than 1,200 variants reached suggestive association with PAD in this cohort. Of those, the researchers could test 552 in a UK Biobank dataset of 5,117 cases and 389,291 controls. This enabled them to replicate 19 PAD risk variants, 18 of which were novel.
The researchers also conducted a phenome-wide association study to examine whether any of these 19 lead SNPs were also linked to other disease phenotypes. Of the 1,101 disease phenotypes queried, the researchers found ties to known risk factors for PAD, such as type 2 diabetes, hypercholesterolemia, and hypertriglyceridemia, as well as to smoking.
By combining evidence from prior studies, these PheWAS results, eQTL, and pQTL data, the researchers prioritized their PAD loci into candidate risk genes, including F5, LPA, SORT1, LPL, and LDLR.
Some of these potential PAD risk genes, the researchers noted, also influence the risk for other diseases. In particular, 14 PAD risk variants were also associated with coronary artery disease and 12 with large artery stroke.
Eleven PAD risk variants affected the three vascular beds: coronary, cerebral, and peripheral. This suggested to the researchers that modulation of LDL cholesterol, the LPL pathway, or circulating lipoprotein(a) could treat PAD.
Four variants, though, appear to be PAD specific. One of these is the factor V Leiden gene mutation, F5 p.R506Q, which is a leading cause of inherited thrombophilia, a clotting disorder. When they further tested the relationship between factor V Leiden and increasing PAD severity — as indicated by rest pain, tissue loss, or amputation — they found significant associations with PAD subtypes. Carriers of the mutation, for instance, had a 62 percent increased risk of a PAD-linked amputation.
This suggested to the researchers that treatments targeting the coagulation cascade could also work for PAD, something they noted is supported by the recent COMPASS trial, which found that adding low-dose rivaroxaban, an inhibitor of factor Xa, to aspirin treatment prevented major adverse limb events.
"Our results highlight mechanistic similarities and differences among coronary, cerebral, and peripheral atherosclerosis and provide therapeutic insights," the researchers wrote.