Op0022 Genetic Risk Markers Are Associated With Clinical Measures in a Cohort of Jia Patients

Background: Juvenile idiopathic arthritis (JIA) is an inflammatory arthritis defined by a symptom onset before the age of 16. The disease is clinically heterogenous and clinical phenotypes are used to define relatively homogenous subgroups, to aid clinical decision making. For example, current International League of Associations for Rheumatology (ILAR) classification define seven mutually exclusive subtypes. Ongoing research aims to further revise these clinical subgroups with the inclusion of additional clinical measures including age at onset (AAO) of the disease and anti-nuclear antibody status (ANA) in the form of the PRINTO classification criteria. Currently, little is known about the genetic contribution to these observed clinical phenotypes. Objectives: To investigate whether there is a genetic role in ANA status and AAO for JIA. Methods: A total of 3304 JIA samples were genotyped using the Illumina Infinium CoreExome array. All JIA ILAR subtypes were included in this analysis. AAO data was available for 2976 cases and ANA status was available for 1227 cases, 560 JIA cases were ANA negative and 667 were ANA positive. A total of 219768 SNPs were analysed, throughout the genome, after quality control. PLINK was used to conduct a linear regression for AAO and a logistic regression for ANA status. The top three principal components were included in both analyses to account for population stratification. To further investigate the role HLA markers in ANA status, a total of 7773 amino acids and classical 2 two and four digit HLA alleles were imputed using SNP2HLA. A genome-wide significance threshold of 5x10⁻⁸ was implemented to define significant markers. Results: All genome-wide significant variants for AAO and ANA were located with the HLA region of the genome. The top associated SNP for AAO was rs2395185, located within HLA-DRB9 (P=6.35x10⁻³⁶, Beta=1.65). This finding suggests that each copy of rs2395185 is associated to an increase of AAO of JIA by approximately 1.65 years. This indicates that AAO of JIA is influenced by genetics. The top associated SNP for ANA status was rs9405108, located within HLA-DRB9 (P=1.8x10⁻¹⁴, OR=0.45). Further analysis on the HLA region was conducted for ANA status and found that several HLA-DRB1 amino acids were associated with ANA status. The top associated imputed HLA marker for ANA status was amino acid 233 of HLA-DRB1 (P=2.91x10⁻²⁶), where the amino acid threonine was strongly associated with ANA status at this position (P=1.06x10⁻¹⁸, OR=0.49). Conclusion: This analysis has revealed that genetic risk markers within the HLA region are associated to AAO of JIA and ANA status in a JIA cohort. These clinical variables are used in the proposed modified PRINTO JIA classification criteria to define JIA subgroups, specifically early onset, ANA positive JIA. This study begins to help us understand the biological basis for the observed clinical phenotypes on which clinical classification systems are based. This research suggests that further definition of the role of genetic risk markers in clinical measures for JIA can begin to explain the heterogenous phenotypes and could aid classification criteria for JIA in the future. REFERENCES: NIL. Acknowledgements: On behalf of the CLUSTER consortium. Disclosure of Interests: None declared.