Genomic correlates of glatiramer acetate adverse cardiovascular effects lead to a novel locus mediating coronary risk.
PLoS ONE 2017 ; 12: e0182999.
Brænne I, Zeng L, Willenborg C, Tragante V, Kessler T, CARDIoGRAM Consortium, CARDIoGRAMplusC4D Consortium, Willer CJ, Laakso M, Wallentin L, Franks PW, Salomaa V, Dehghan A, Meitinger T, Samani NJ, Asselbergs FW, Erdmann J, Schunkert H
DOI : 10.1371/journal.pone.0182999
PubMed ID : 28829817
PMCID : PMC5567477
URL : https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0182999
Abstract
Glatiramer acetate is used therapeutically in multiple sclerosis but also known for adverse effects including elevated coronary artery disease (CAD) risk. The mechanisms underlying the cardiovascular side effects of the medication are unclear. Here, we made use of the chromosomal variation in the genes that are known to be affected by glatiramer treatment. Focusing on genes and gene products reported by drug-gene interaction database to interact with glatiramer acetate we explored a large meta-analysis on CAD genome-wide association studies aiming firstly, to investigate whether variants in these genes also affect cardiovascular risk and secondly, to identify new CAD risk genes. We traced association signals in a 200-kb region around genomic positions of genes interacting with glatiramer in up to 60 801 CAD cases and 123 504 controls. We validated the identified association in additional 21 934 CAD cases and 76 087 controls. We identified three new CAD risk alleles within the TGFB1 region on chromosome 19 that independently affect CAD risk. The lead SNP rs12459996 was genome-wide significantly associated with CAD in the extended meta-analysis (odds ratio 1.09, p = 1.58×10-12). The other two SNPs at the locus were not in linkage disequilibrium with the lead SNP and by a conditional analysis showed p-values of 4.05 × 10-10 and 2.21 × 10-6. Thus, studying genes reported to interact with glatiramer acetate we identified genetic variants that concordantly with the drug increase the risk of CAD. Of these, TGFB1 displayed signal for association. Indeed, the gene has been associated with CAD previously in both in vivo and in vitro studies. Here we establish genome-wide significant association with CAD in large human samples.