P90 Genetic risk, smoking and the development of systemic autoimmune rheumatic disease

• Published in: Lupus science & medicine 2020-03, Vol.7 (Suppl 1), p.A71
• Main Authors: Bidstrup Leffers, Henrik Christian, Westergaard, David, Banasik, Karina, Jacobsen, Søren
• Format: Article
• Language: English
• ISSN: 2053-8790
• DOI: 10.1136/lupus-2020-eurolupus.134

PurposeEnvironmental and genetic factors have individually been extensively researched in the pathogenesis of systemic autoimmune rheumatic diseases (SARDs). Notably, more than half of the associations are shared by at least two autoimmune diseases and SLE exhibits association to almost all. Smoking is an overall risk factor for the development of any SARD. We aim to determine a polygenic risk score for SARD including smoking as a covariate which could advocate for gene-environment interaction.MethodsA case-control study will be conducted to determine a polygenic risk score including smoking.Data will be retrieved from the UK Biobank, which is a general-population cohort of roughly 0.5 million participants recruited across the UK during 2006–2010. Participants have been genotyped. A genome wide association study (GWAS) will be performed using the Scalable and Accurate Implementation of GEeneralized mixed model (SAIGE). Cases are defined to have SARD if having one of the following diagnoses: SLE, rheumatoid arthritis, polymyositis/dermatomyositis, systemic sclerosis or primary Sjögrens syndrome. Controls comprise UK Biobank participants not fulfilling the case definition or having other autoimmune diseases.ResultsWe have identified 1,048 patients in the UK Biobank with at least one SARD and more than 300,000 matched controls. Results from the initial GWAS calculations are shown in figure 1. Based on a top-100 list of susceptibility SNPs, we will derive a polygenic risk core and determine to which extent smoking, gender and age increase the risk of SARD among subjects which are highly genetically susceptible. External validity is tested by replication in independent cohorts from the Danish Blood Donor Study and Copenhagen Biobank.Abstract P90 Figure 1ConclusionClinical tools based on genetics predictive of SARDs are in wanting but have generally been judged of little to no useful information. In this study, we will provide a validated predictive model of SARD based on multiple genes and interaction with non-genetic factors.AcknowledgementSupported by the Danish Rheumatism Association