OA16 Core genes for polymyalgia rheumatica and giant cell arteritis and their effects on cardiovascular diseases

Abstract Background/Aims Polymyalgia rheumatica (PMR) and giant cell arteritis (GCA) are age-associated inflammatory diseases with overlapping features, including a risk of ischaemic complications. Glucocorticoids, the mainstay treatment, are effective but carry risks for adverse effects and relapse is common after discontinuation. Our aim was to identify core genes for PMR and GCA using a novel method of genetic analysis, and to evaluate the effect of identified genes on other vascular disorders. Methods We performed genome-wide aggregated trans-effects (GATE) analysis in two case-control cohorts with genotypic data: the UK Biobank (6,200 cases with a mention of either PMR or GCA and 458,000 controls) and the UKGCA Consortium (1,878 GCA cases and 5,520 controls). GATE scores for predicted protein levels were computed as a weighted sum of an individual’s genotypes at trans-protein quantitative trait loci (trans-pQTLs) extracted from the deCODE study and the UK Biobank Pharma Proteomics Project. GATE scores were tested for association with PMR/GCA, and putative core genes were identified using a Bonferroni-corrected p-value threshold (p 2 x 10-5). GATE scores for putative core genes were subsequently tested for association with coronary artery disease (53,000 cases, 411,000 controls) and peripheral vascular disease (14,000 cases, 450,000 controls) in the UK Biobank. Results In the UK Biobank, 4,015 GATE scores, comprising at least 5 trans-pQTLs and corresponding to 3,187 unique genes, were evaluated. Five genes (CXCL9, LECT2, PLG, MST1, CXCL10) were identified as putative core genes for PMR/GCA (p 2 x 10-5). In UKGCA, 3,247 GATE scores, corresponding to 1,993 unique genes and with 5 trans-pQTLs, were evaluated. LECT2 and four additional genes (FGFBP3, SNAP25, RSPO1, LRRN1) were identified as putative core genes for GCA. Five of the putative core genes (LECT2, CXCL10, CXCL9, PLG, MST1) were associated with coronary artery disease at p 10-4 and with consistent direction of effect to PMR/GCA. LECT2 and CXCL10 were also associated with peripheral vascular disease at p 10-4 and consistent effect direction. Conclusion Using GATE analysis, we identified nine putative core genes for PMR and GCA. Four of these (CXCL9, CXCL10, LECT2, MST1) are directly involved in immune regulation and in particular T-cell responses, and two (PLG, FGFBP3) are involved in angiogenesis and tissue remodelling during repair of vascular injury. LECT2 had the strongest evidence in support of a causal role in vascular pathology with robust associations to PMR/GCA, to GCA alone, and to the two cardiovascular outcomes. While LECT2 is a multifunctional protein, a likely mechanism mediating its effect on PMR/GCA is through its involvement in vascular inflammation upon binding to a receptor on endothelial cells. Mechanistic studies of LECT2 in PMR/GCA are warranted to evaluate its potential as a therapeutic target for these diseases. Disclosure A. Nwagbata: None. A. Iakovliev: None. A. Morgan: None. P. McKeigue: None. A. Spiliopoulou: None.