Systemic lupus erythematosus (SLE) is the prototypical multi-system autoimmune disease with diverse clinical features in persons with disease. SLE is also unified by characteristic autoimmunity directed against nucleic acid or nucleoprotein complexes. SLE is both more prevalent and typically exhibits a more severe clinical course in persons with African-American ancestry than in persons with European ancestry. The reasons for this discrepancy remain incompletely understood. GWAS studies of SLE in cohorts of individuals with Amerindian, East Asian and European ancestry have identified > 180 risk loci for SLE across the genome. These loci act in several pathways: clearance of autoantigens, innate immune response to nucleic acids, and lymphocyte activation. Despite these advances in understanding the genetic basis of SLE, a genome-wide association scan (GWAS) of SLE in a cohort of individuals with African-American ancestry has not yet been reported. Here, we report preliminary results of GWAS in 1494 SLE cases and 6076 matched controls with African-American ancestry. Illumina Infinium Omni 1, Omni 1S, Omni 2.5 and OmniExpress platforms were used for genotyping. Genotypes were imputed using the TOPMed reference panel at NHLBI. By defining the contribution common genetic variants to disease risk across the genome, our study represents a step towards understanding the genetic basis of SLE and the increased prevalence and severity of SLE in African ancestry populations. To fully define the relative contribution of environment and genetics to the discrepant SLE severity and prevalence observed in African-American populations, future studies will be necessary. These studies should focus both on comprehensive understanding of the environmental influences and comprehensive assessment of genome- wide genetic variation (i.e. whole-genome sequencing) that impact SLE risk and disease severity.

Our results confirm genome-wide significant (P < 5E-8) association with loci ascertained in other populations (STAT4-STAT1, TNIP1, MIR146A, HLA-C4A-C4B, IRF5, BLK, PLAT-IKBKB, RELA-RNASEH2C-OVOL1, ITGAM, and IRF8) and identify several novel genome-wide significant risk loci that are newly described in our study (ENSA, IKBKB/Chr8: Centromere and PCMTD1-ST18). Further, we compared associated variants in our study with those from three large SLE GWAS studies in cohorts of individuals with European, East Asian and Amerindian ancestry. This comparison of SLE risk loci revealed pervasive sharing of SLE genetic risk across ancestral groups. For 70% of the risk loci, the lead marker exhibited nominal association (P < 0.05) with SLE in our GWAS of SLE in African- American persons. Importantly, the association of all such variants cohered with the reported direction in other ancestries.

Overall, our findings are consistent with a polygenic contribution to SLE in African-American individuals that is largely shared across populations. We also find association with a rare variant (MAF < 1%) of large effect (OR = 3.91) near a locus previously identified via ImmunoChip (Illumina), PLAT-IKBKB. The lead variant at this locus is non-polymorphic in populations with ancestry outside of Africa. This association explains the increased risk of SLE in ~4% of cases in our cohort. On the one hand, our GWAS of SLE in persons African-American ancestry provides insights that reinforce the conclusions concerning the known risk loci from other ancestries. On the other hand, this mechanism uniquely raises SLE risk in a small proportion of African ancestry individuals with SLE. Together our findings reveal both uniformity and diversity of genetic risk factors impacting SLE development across populations.