Background One of the principal findings of genome wide association studies in autoimmune diseases has been the substantial overlap of genetic susceptibility loci identified. This has underpinned fine mapping initiatives such as the Immunochip (IC) project in which custom chips were designed to fine-map regions of associations common to a number of autoimmune diseases. Samples from patients with rheumatoid arthritis (RA), juvenile idiopathic arthritis (JIA) and psoriatic arthritis (PsA) have been genotyped using IC; given that all are types of inflammatory arthritis (IA) which exhibit common pathology and some overlap of associated loci, the data emerging from IC provides a unique opportunity to identify novel overlapping regions and determine whether the causal variants within overlapping loci are the same or different for each disease.
Objectives To investigate IC fine mapping data to identify genetic regions which are associated with susceptibility to more than one type of IA.
Methods As part of the IC project, genotyping, quality control and association analysis was performed for RA (14056 cases and 9583 controls), JIA (1614 polygoarticular and oligoarticular JIA cases and 7153 controls) and PsA (929 cases and 4537 controls). Some controls were shared between cohorts. For this analysis all common SNPs (MAF>0.05) with p<10-3 in any of the 3 diseases were first selected. IC regions were defined as including all SNPs from the 1000 Genomes Project CEU population (September 2009 release) that were in 0.1-cM (HapMap3 CEU) recombination blocks around each GWAS region lead marker. IC Regions which contained SNPs associated with more than one type of IA were considered overlapping.
Results 26 regions showed association at p<10-3 with more than one type of IA, of these, in RA and JIA two loci (PTPN22 and STAT4) reached genome wide significance (5×10-8), with a further 8 associated at p<10-4in both diseases. Interestingly 2 of these loci are also associated with PsA at p<10-3 (RUNX1 and PTPN2). 4 of these loci have been previously associated with both RA and JIA (PTPN22, STAT4, PTPN2, and IL2RA) whilst 6 are novel overlapping regions (ANKRD55, RUNX1, IL2RB, DNASE1L3, UBE2L3 and COG6). For the PTPN22, STAT4, ANKRD55, DNASE1L3, PTPN2 and UBE2L3 regions the RA and JIA associated SNPs are either identical or highly correlated (r2>0.8), the direction of effect is the same in both diseases and odds ratios similar. For RUNX1 and COG6 the associated SNPs are different and weakly correlated (r2>0.4<0.8) but the direction of effect is similar for each disease. An interesting finding is that for IL2RA and IL2RB the pattern of association in the region is completely different.
Conclusions These findings add to the body of evidence that there are shared susceptibility genes for inflammatory arthritis. Further investigation is required to fully explore whether the regions contain the same or different causal effect, including functional studies to determine whether, in overlapping loci with the same causal variant, the variant predisposes to disease by the same mechanism.
Disclosure of Interest None Declared