Background A preponderance of females develop autoimmune disease, including juvenile idiopathic arthritis (JIA), yet the reason for this female bias remains elusive. Evidence suggests that genetic risk of disease may be influenced by sex. The JIA genetic risk variant, PTPN22 rs2476601, has been reported to show female-specific association in type 1 diabetes.
Objectives We sought to determine whether PTPN22 rs2476601 exhibits a female-specific association with JIA.
Methods The discovery sample was drawn from the Australian CLARITY JIA biobank. rs2476601 was genotyped in 413 cases (67% female, mean age 9.1 years) and 690 healthy controls (42% female, mean age 7.1 years). We used logistic regression to assess the association of rs2476601 with JIA in the total sample (males and females). We then explored the role of sex in this association: we adjusted the regression analysis for sex, stratified by sex, and performed a genotype x sex interaction analysis. Sensitivity analyses were performed by restricting to individuals of European descent, subtypes of oligoarticular and polyarticular RF negative disease course, and exclusion of systemic JIA. We sought to replicate the sex-specific associations in an independent JIA case-control sample collected from the USA and Norway and held at the Children's Hospital of Philadelphia. Analyses were performed using Stata or PLINK.
Results The outcomes of analyses are shown in the Table. rs2476601 was associated with all JIA in both the discovery and replication samples, and the association remained when adjusted for sex. When stratified by sex, a strong association was observed for females, but not for males. There was evidence for genotype-by-sex interaction (p=0.0087). Similar patterns were observed in all sensitivity analyses. In the replication sample, we also observed association of rs2476601 with JIA in females, but not in males.
Conclusions We have demonstrated, in both a discovery and replication sample, that the association between PTPN22 rs2476601 and JIA appears restricted to females, helping to explain the greater number of females that develop this disease.
Disclosure of Interest None declared