During the previous reporting periods (Z01 AR041185-01, HG200370-01) we established an international consortium for the purpose of identifying and obtaining DNA specimens from at least 1000 well-characterized cases of systemic onset JIA and 1000 ethnically-matched controls. This consortium includes investigators from several of the major pediatric rheumatology centers in North America, including the Childrens Hospital of Philadelphia, Cincinnati Childrens Hospital Medical Center, the Hospital for Sick Children in Toronto, Stanford University, and the University of Utah;from several of the major pediatric rheumatology centers in Europe, including Great Ormand Street Hospital in London, Hopital Necker-Enfants Malades in Paris, the Gaslini Institute in Genova, and the Wilhelmena Hospital in Utrecht;and other centers in Turkey, Argentina, Brazil, and Australia. In early May, 2010, we hosted a meeting of international collaborators at the NIH, in which there were discussions regarding recruitment strategies, appropriate informed consent for genome-wide association studies (GWAS), sample shipment, genotyping, and data analysis. Because of the difficulties for some centers to recruit appropriate ethnically matched controls, where necessary we planned to obtain in silico controls from existing online databases, using bioinformatic tools to ensure appropriate matching by ancestry. All genotyping has been performed in the Inflammatory Disease Section, using Illumina Human Omni 1M-Quad, v1.0 bead chips and an Illumina iScan Beadarray scanner. By the end of the previous reporting period, we had genotyped a total of 785 cases and 436 controls. During the past year, we have sought to expand the size and diversity of the discovery collection of the sJIA GWAS. We have, for example, genotyped samples from 159 children with sJIA and 205 healthy children from Munich, Germany, bringing the number of directly genotyped samples in our discovery collection to 982 sJIA and 651 control subjects. We have obtained ethnically-matched control genotype datasets in silico for the United States, Canadian, Spanish, and Argentine collections, bringing our total number of control samples to 4688 healthy controls. Subdividing the collection into nine strata based on ethnicity, we have performed stringent quality control procedures to exclude problematic samples and SNPs, and to exclude from each stratum individuals of dissimilar ethnicity. After data conditioning, we were left with 785 sJIA cases and 3870 healthy controls. Across the nine strata, the number of SNPs carried forward for subsequent analysis ranged from 156,136 SNPs to 740,509 SNPs. We have undertaken several iterations of analysis on our discovery collection, ultimately performing SNP imputation followed by genomewide meta-analysis of results from nine strata. For each stratum, we performed pre-phasing of our data using the Segmented HAPlotype Estimation and Imputation (ShapeIT) software and SNP imputation with IMPUTE2 software, using the haplotypes from the full HapMap3 collection (release 2) as our reference panel (1,581,224 SNPs). After merging the direct and imputed genotype data, and excluding both SNPs with minor allele frequencies below 0.05 and those with poor imputation quality scores, we performed association testing using SNPTESTv2 in each stratum. To account for the uncertainty of SNP imputation, we performed the association testing using the genotype probability data. Finally, we performed Cochran-Mantel Haenszel meta-analysis of 1,447,416 high-quality markers using Genomewide Association Meta-Analysis (GWAMA) software to perform both fixed-effects and random-effects meta-analyses of the nine strata in our discovery collection. The genomewide association meta-analysis of nine case-control strata, under the additive model, identified 126 markers with p less than 5 times 10 to the negative fifth, which represent 49 different genes. Twenty-six of these SNPs and fourteen of these genes reside within a 3 Mb segment of the major histocompatibility complex (MHC) that spans the MHC class II and class III gene clusters. This sJIA-associated region contains two markers that exceeded the threshold of genomewide significance , rs615672 (p = 3.04 times 10 to the negative eighth;OR = 0.715 95CI 0.636, 0.805) and rs17208888 (p = 3.42 times 10 to the negative eighth, OR = 0.61 95CI 0.51, 0.73), which are located between HLA-DRB1 and HLA-DQA1, and 4.6 Kb upstream of BTNL2, respectively. The strongest regional association signal outside of the MHC locus is located nearest to AJAP1, which encodes the adherens-junction associated protein-1. This region contains 6 SNPs with p less than 5 x 10 to the negative sixth, the most significant of which, rs16838195, nearly meets the threshold for genomewide significance (p = 7.73 times 10 to the negative eighth, OR = 1.97 95CI 1.54 - 2.53). It is noteworthy that we have observed an almost complete absence of overlap between our 49 candidate loci and the known autoimmune and autoinflammatory loci. If this observation is corroborated by additional investigations, then we may expect this study to identify novel pathways and mechanisms involved in sJIA, which may also represent novel therapeutic targets. Our objectives for the upcoming year will include a range of follow-up studies to validate the results of our GWAS and understand their functional implications, with the immediate goal of publishing a manuscript reporting these results. To examine the MHC regional association, we will first perform imputation of the classical MHC types with collaborators from the Broad Institute. We will perform regression and haplotype analysis of the regional SNP associations, as well as those of the imputed HLA types, to determine the number and source of independent MHC-regional associations. Should the MHC class III gene cluster prove to harbor an independent association signal, we will perform fine-map genotyping on the Sequenom platform to specify the source of that association, a strategy that we will also employ for the investigation of all associations outside of the MHC locus. To specifically examine the group of genes associated with autoimmunity, we will genotype the majority of our samples using the Immunochip, a custom SNP array that deeply interrogates over 200 genes of documented importance in inflammatory and autoimmune diseases. Finally, we plan to undertake targeted, deep resequencing of all GWAS-identified susceptibility loci to examine for rare or novel genetic variants that may influence sJIA susceptibility but were not captured by the GWAS approach.

Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
National Human Genome Research Institute
Zip Code
Ombrello, Michael J; Remmers, Elaine F; Tachmazidou, Ioanna et al. (2015) HLA-DRB1*11 and variants of the MHC class II locus are strong risk factors for systemic juvenile idiopathic arthritis. Proc Natl Acad Sci U S A 112:15970-5