Background/Purpose: Systemic lupus erythematosus (SLE) disproportionately affects African Americans, and the development of SLE is believed to be triggered by exposure to one or more specific environmental factors in genetically susceptible individuals. Tobacco smoke has been implicated as such a potential agent. In this study, we sought to discover whether smoking and/or secondhand smoke inhalation interacts with certain candidate genes, resulting in elevated risk of SLE in African Americans.

Methods: A case control design was used, and all subjects were African Americans of Gullah ancestry, from the Sea Islands of South Carolina or Georgia. All cases met ACR classification criteria for SLE, and all controls were evaluated by a rheumatologist and determined unaffected by autoimmune disease at the time of enrollment. All subjects completed questionnaires to assess smoking history and exposure to secondhand smoke. Candidate genes included ones previously identified in the literature as interacting with tobacco smoke in rheumatic diseases, and genes in the Comparative Toxicogenomics Database that were relevant to tobacco smoke or in an inference network between tobacco smoke and SLE. Genotypic data was available on subjects genotyped on the Illumina Immunochip genotyping array. After standard GWAS quality control, the following were available for analyses: NAT2 (4 single nucleotide polymorphisms [SNPs]), HLA-DRB1 (6 SNPs), APOE (2 SNPs), IL6 (17 SNPs), CXCL8 (1 SNP), IRF5 (20 SNPs), ITGAM (67 SNPs), and ITGAX (31 SNPs). Logic forest, a relatively novel machine learning algorithm we developed, was used to highlight interactions between the environment (i.e. tobacco smoke) and the variants in the candidate genes.

Results:  The study population included 204 participants (n=100 cases, n=104 controls), 86% of whom were female. The logic forest model identified secondhand smoke exposure during childhood as the most important predictor of SLE status. Consistent with the literature, the model identified several SNPs in the IRF5, ITGAM, and ITGAX genes as being risk factors for SLE. Although passive smoke exposure during childhood was of high importance and identified as a main effect, the logic forest detected weak to moderate interactions (odds ratios ranging from 2.3 to 2.5) between it and SNPs in the ITGAM gene.

Conclusion:  In this case-control study, both genetic and environmental risk factors for SLE were identified, with evidence suggesting potential interactions between exposure to secondhand smoke as a child and genetic variation in the ITGAM gene.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/genetic-variation-and-tobacco-smoke-exposure-in-systemic-lupus-erythematosus-a-case-control-study-among-african-americans/