LAUSR

OBJECTIVES Genetics play an important role in SLE risk, as well as osteonecrosis (ON), a significant and often debilitating complication of SLE. We aimed to identify genetic risk loci for ON in people with childhood-onset (cSLE) and adult-onset SLE (aSLE). METHODS We enrolled participants from two tertiary care centres who met classification criteria for SLE. Participants had prospectively collected clinical data and were genotyped on a multiethnic array. Un-genotyped SNPs were imputed, and ancestry was inferred using principal components (PCs). Our outcome was symptomatic ON confirmed by imaging. We completed time-to-ON and logistic regression of ON GWASs with covariates for sex, age of SLE diagnosis, five PCs for ancestry, corticosteroid use, and selected SLE manifestations. We conducted separate analyses for cSLE and aSLE and meta-analyzed results using inverse-variance weighting. Genome-wide significance was P < 5x10-8. RESULTS The study included 940 participants with SLE, 87% female and 56% with cSLE. ON was present in 7.6% (n = 71). Median age of SLE diagnosis was 16.9 years (IQR: 13.5, 29.3), with median follow-up of 8.0 years (IQR: 4.2, 15.7). Meta-GWAS of cSLE and aSLE time-to-ON of 4,431 911 SNPs identified a significant Chr.2 SNP, rs34118383 (MAF = 0.18), intronic to WIPF1 (HR = 3.2, [95% CI: 2.2,4.8], P = 1.0x10-8). CONCLUSION We identified an intronic WIPF1 variant associated with a 3.2 times increased hazard for ON ([95% CI: 2.2,4.8], P = 1.0x10-8) during SLE follow-up, independent of corticosteroid exposure. The effect of the SNP on time-to-ON was similar in cSLE and aSLE. This novel discovery represents a potential ON risk locus. Our results warrant replication.