Trans-ancestral ImmunoChip: SLE risk loci show enrichment for NK cytotoxicity and Cell Adhesion Pathways. D. S. Cunninghame Graham1, J. A. Kelly2, C. D. Langefeld3, R. R. Graham4, P. M. Gaffney2, T. J. Vyse1, SLE ImmunoChip Consortium 1) King's College, London, United Kingdom; 2) Oklahoma Medical Research Foundation, OA; 3) Wake Forest School of Public Health Genomes, NC; 4) Genentech Inc, CA.

   Background: The prototypical autoimmune disease Systemic Lupus Erythematosus (SLE) shows variation in prevalence and disease severity across ancestries. We and others have shown differences in the identity and frequencies of susceptibility alleles at associated loci for different ancestries. Multiple genome-wide association analysis in European and in SE Asian samples have identified many susceptibility loci (OR>1.3). However low genotyping density has prevented pinpointing of risk alleles, so the vast majority of moderate risk alleles (OR 1.1-1.3) remain either undiscovered or unconfirmed. Aim: To identify novel susceptibility loci for SLE across multiple ancestries we genotyped large cohorts of affected individuals with shared controls from three different ethnicities on the Immunochip platform and undertook network analyses to discover underlying biological relevance of these loci in SLE pathogenesis. Study Cohort: All SLE cases met the American College of Rheumatology criteria. The ratios of cases and controls (nSLE:nc) in each population were: African American (AA)(2970:2452), European (EA) (6748:11516)and Hispanic American (HA) (1872:2016) were genotyped and passed quality control - a total study cohort of 12147 cases and 16732 controls. Analysis methods: Association was modelled using a logistic regression adjusting for admixture proportions. Trans-racial non-parametric meta-analysis was computed using METAL. In each ancestry functional pathways were investigated using GSA-SNP. False discovery rate (FDR) adjusted p-values were computed to account for the actual number of tests. Results: Numerous regions met Immunochip-wide significance (P<1x10-8) and FDR<0.05: AA (11), EA (52) and HA (25), including novel and fine-mapping of known loci. Heterogeneity of odds ratio plots between populations from the trans-ancestral meta-analysis indicated greatest heterogeneity in regions including the HLA region, chromosomes 8 and 17. Preliminary pathway analysis revealed an enrichment of risk loci implicated in natural killer cell cytoxicity (hsa04650) (including BRAF, VAV2, KRAS, SHC4, GRB2, VAV1 and IFNAR) and in cell adhesion (hsa04514) (including NTXN1, NCAM1, ITGA8, GLG1, PTPRM, CDH4, ICAM1) (Pcorr<0.01 in any of the three ethnicities). Conclusion: Trans-ancestral mapping of known autoimmune loci has revealed novel SLE-associated loci and signalling pathways important in immunological function, further extending our understanding of lupus pathogenesis.

You may contact the first author (during and after the meeting) at