Objectives To find causal genes for rheumatoid arthritis (RA) and its seropositive (RF and/or ACPA positive) and seronegative subsets. Methods We performed a genome-wide association study (GWAS) of 31 313 RA cases (68% seropositive) and ~1 million controls from Northwestern Europe. We searched for causal genes outside the HLA-locus through effect on coding, mRNA expression in several tissues and/or levels of plasma proteins (SomaScan) and did network analysis (Qiagen). Results We found 25 sequence variants for RA overall, 33 for seropositive and 2 for seronegative RA, altogether 37 sequence variants at 34 non-HLA loci, of which 15 are novel. Genomic, transcriptomic and proteomic analysis of these yielded 25 causal genes in seropositive RA and additional two overall. Most encode proteins in the network of interferon-alpha/beta and IL-12/23 that signal through the JAK/STAT-pathway. Highlighting those with largest effect on seropositive RA, a rare missense variant in STAT4 (rs140675301-A) that is independent of reported non-coding STAT4 -variants, increases the risk of seropositive RA 2.27-fold (p=2.1×10 −9 ), more than the rs2476601-A missense variant in PTPN22 (OR=1.59, p=1.3×10 −160 ). STAT4 rs140675301-A replaces hydrophilic glutamic acid with hydrophobic valine (Glu128Val) in a conserved, surface-exposed loop. A stop-mutation (rs76428106-C) in FLT3 increases seropositive RA risk (OR=1.35, p=6.6×10 −11 ). Independent missense variants in TYK2 (rs34536443-C, rs12720356-C, rs35018800-A, latter two novel) associate with decreased risk of seropositive RA (ORs=0.63–0.87, p=10 −9 –10 −27 ) and decreased plasma levels of interferon-alpha/beta receptor 1 that signals through TYK2/JAK1/STAT4. Conclusion Sequence variants pointing to causal genes in the JAK/STAT pathway have largest effect on seropositive RA, while associations with seronegative RA remain scarce.

Objectives Pulmonary manifestations in RA are common comorbidities. Interstitial lung disease (ILD), both idiopathic and in RA, has been associated with several genetic variants. We assessed pulmonary fibrosis (PF) in an inception cohort of RA patients in relation to genetic variants and disease-related factors. Methods A total of 1466 early RA patients were consecutively included and followed prospectively from the index date until death or 31 December 2016. Clinical and laboratory data and treatment were continuously registered according to the Swedish Rheumatology Quality Register. DNA was available from 1184 patients and 571 151 genome-wide single-nucleotide polymorphisms (SNPs) were analysed. Thirteen identified genetic variants were extracted. At follow-up, the patients answered a questionnaire regarding disease progression and lung involvement that was validated by reviewing medical records and analysing radiological examinations. Results The prevalence of PF was 5.6% and the annualized incidence rate was 5.0/1000 (95% CI 3.80, 6.54). Four SNPs were associated with PF in RA: rs35705950 [MUC5B; OR 2.5 (95% CI 1.5, 4.0), adjusted P-value = 0.00016, q-value = 0.0021]; rs111521887 [TOLLIP; OR 1.9 (95% CI 1.3, 2.8), adjusted P-value = 0.0014, q-value = 0.0092]; rs2609255 [FAM13A; OR 1.7 (95% CI 1.1, 2.5), adjusted P-value = 0.013, q-value = 0.055] and rs2736100 [TERT; OR 1.5 (95% CI 1.0, 2.2), adjusted P-value = 0.046, q-value = 0.15]. Older age and RF positivity were associated with increased risk, while MTX treatment was associated with a lower risk of PF. Conclusions Development of PF in an inception cohort of RA patients was associated with 4 of 12 ILD risk genes. RA-related factors except for age at diagnosis and RF positivity were of limited importance in PF development.

Objective Recent studies have demonstrated an altered expression of certain microRNAs in patients with rheumatoid arthritis (RA) as well as their first-degree relatives (FDRs) compared to healthy controls (HCs), suggesting a role of microRNA in the progression of the disease. To corroborate this, a set of well-characterized RA families originating from northern Sweden were analyzed for differential expression of a selected set of microRNAs. Method MicroRNA was isolated from frozen peripheral blood cells obtained from 21 different families and included 26 RA patients, 22 FDRs, and 21 HCs. Expression of the selected microRNAs miR-22-3p, miR-26b-5p, miR-34a-3p, miR-103a-3p, miR-142-3p, miR-146a-5p, miR-155, miR-346, and miR-451a was determined by a two-step quantitative real-time polymerase chain reaction (qRT-PCR). Statistical analysis including clinical variables was applied. Results Out of the nine selected microRNAs that previously have been linked to RA, we confirmed four after adjusting for age and gender, i.e., miR-22-3p ( p = 0.020), miR-26b-5p ( p = 0.018), miR-142-3p ( p = 0.005), and miR-155 ( p = 0.033). Moreover, a significant trend with an intermediate microRNA expression in FDR was observed for the same four microRNAs. In addition, analysis of the effect of corticosteroid use showed modulation of miR-103a-3p expression. Conclusions We confirm that microRNAs seem to be involved in the development of RA, and that the expression pattern in FDR is partly overlapping with RA patients. The contribution of single microRNAs in relation to the complex network including all microRNAs and other molecules is still to be revealed. Key Points • Expression levels of miR-22-3p, miR-26b-5p, miR-142-3p, and miR-155 were significantly altered in RA patients compared to those in controls. • In first-degree relatives, a significant trend with an intermediate microRNA expression in FDR was observed for the same four microRNAs.

Background: Patients with rheumatoid arthritis (RA) have increased risk of osteoporosis and low-energy fractures. Several genes associated with bone mineralization, osteoporosis or risk of fracture in the general population have been identified. Objectives: To analyse the association between nine selected SNPs and the risk of low-energy fracture, taking clinical patient characteristics into account. Methods: We identified a cohort of patients (n=896, 70% women, age at inclusion 60.0±14.8 years) with RA according to ACR criteria from the catchment area of the register of Umeå injury database, Umeå, Sweden, which enabled identification of low-energy fractures (n=254). The follow-up (mean 8.8±6.1 years, total 7928 person-years) started two years after RA diagnosis but not earlier than January 1, 1993 and ended at the first of December 31, 2011, death or the first low-energy fracture. Nine SNPs were analysed in all patients with available DNA-samples (n=667) using KASPTM genotyping assays (LGC genomics Ltd, Hoddesdon, UK): rs3801387 (WNT16), rs6666455 (SOAT), rs3736228 (LRP5), rs4796995 (FAM210A), rs4792909 (SOST), rs2062377 (TNFRSF11B/OPG), rs884205 (TNFRSF11A/RANK), rs9533090 (TNFSF11/RANKL), and rs1373004 (DKK1). Anti-CCP was analysed and clinical patient characteristics (duration of RA, ever smoking, disease activity the first two years after RA diagnosis, and joint erosions) were extracted from patient files. Associations between the risk of fracture and risk alleles in the cohort were evaluated using Kaplan-Meier curves (K-M) and Cox proportional hazards models: crude, adjusted for age and sex, and for clinical patient characteristics. Results: The SNPs: rs1373004, rs4792909, and rs2062377 were associated with the risk of fracture in K-M analyses (Figure 1). For the other genes no significant associations were observed. Patients carrying the risk allele of rs1373004 (22.6% of the patients), or who were homozygous for the risk allele of SNP rs4792909 (38.6%), had a >50% higher risk of low-energy fracture compare to other patients, irrespectively of disease characteristics (Table 1). The association between rs2062377 and the risk of fracture was not independent of clinical patient characteristics (Table 1).Abstract THU0143 –Table 1 Cox proportional hazards models estimating the risk of low-energy fracture associated with SNPs in patients with RA Freq HR1* 95% C.I. HR2* 95% C.I. HR3* 95% C.I. rs1373004, T-carrier 22.6% 1.61 1.16;2.22 1.61 1.16;2.22 1.68 1.17;2.41 rs4792909, GG vs. TT+TG 38.6% 1.59 1.19;2.14 1.53 1.13;2.05 1.54 1.12;2.13 r2062377, AA vs. TA+TT 35.7% 1.36 1.01;1.83 1.28 0.81;1.59 1.19 0.86;1.65 *Hazard ratio (HR)1=crude model, HR2= adjusted for age at inclusion and sex, HR3=adjusted for age at inclusion, sex, duration of RA, anti-CCP positivity, ever smoking, disease activity the first two years after RA diagnosis, and erosive disease.Abstract THU0143 – Figure 1 Conclusion: Genes related to bone metabolism may have a considerable contribution to the already high risk of low-energy fractures in RA. References: [1] Estrada K, et al. Nat Genet2012 [2] Baron R & Kneissel M. Nat Med2013 Disclosure of Interests: Lotta Ljung: None declared, Kristina Wiberg: None declared, Lisbeth Ärlestig: None declared, Solbritt Rantapää Dahlqvist Consultant for: Member of the advisory board, Lipum AB, Umeå, Sweden.

Background Vitamin D has been implicated as being involved in the aetio-pathogenesis of several autoimmune diseases including rheumatoid arthritis (RA). Previous studies present contradictory results. Vitamin D binding protein (DBP), the major transport protein, is also involved in various inflammatory processes. Objectives The aim of this study was to investigate the relationship between circulating levels of 25-hydroxyvitamin D [25(OH) D], DBP and polymorphisms in group-specific component (GC) in pre-symptomatic individuals and matched controls within prospective cohorts of the Northern Sweden. Methods Blood samples donated to the Medical Biobank prior to the onset of symptoms of RA (n=515, mean [SD] time before the onset of symptoms 6.2 [9.3] years) and from matched (2:1) population-based controls (n=267) were used. Plasma 25(OH) vitamin D levels were analysed using liquid chromatography tandem-mass spectrometry and DBP levels were analysed using enzyme-linked immunosorbent assay. GC polymorphisms (rs4588 and rs7041) were analysed with TaqMan assays (Applied Biosystems). Results Levels of 25(OH) D or DBP were not statistically different between pre-symptomatic individuals and controls in a crude, or a multiple-adjusted logistic regression model. However, an increased risk for future RA was found in females of DBP (OR 1.0001 [95%CI 1.000–1.0003]), adjusted for carriage of the minor allele of rs4588, in a multiple-adjusted model (p<0.05). Conclusions This study indicated that vitamin D is not associated with the future risk of RA although increasing levels of DBP were however, associated with an increased risk of disease in females carrying the minor allele of a DBP encoding SNP. Acknowledgements Samples and data for the cohort were obtained through the Västerbotten Intervention Program, the Northern Sweden MONICA Study, and the mammary screening program of Västerbotten. Staff of the Department of Biobank Research, Umeå University, aided in acquisition of samples and data. Disclosure of Interest None declared

Objectives:RANK ligand (RANKL) is involved in destruction and osteoporosis in RA. In this study, the relationships between RANKL and ACPA, anti-carbamylated protein antibodies (anti-CarP), cytokines and chemokines were analysed in individuals before the onset of RA symptoms, and their associations with radiological findings at disease onset were assessed.Methods:This was a case-control study performed within the Medical Biobank of Northern Sweden that included 470 pre-symptomatic individuals [334 women and 136 men; mean (s.d.) age 52.3 (9.4) years] using blood samples donated before symptom onset (pre-dating time; 5.0 years) and 96 controls (60 women and 36 men). Plasma was analysed for RANKL (BioVendor, Karasek, Brno, Czech Republic), anti-CCP2 antibodies (Eurodiagnostics, Malmö, Sweden), anti-CarP antibodies (in-house ELISA), ACPA specificities (ISAC-platform, Phadia AB, Uppsala, Sweden) and cytokines/chemokines (Meso Scale Discovery methods, Rockville, MD, USA). Radiographs of hands and feet were graded using the Larsen score.Results:The concentration of RANKL was higher in the pre-symptomatic individuals compared with controls; mean (s.e.m.): 0.50 (0.03) vs 0.22 (0.02) nmol/l (P < 0.001). The concentration increased gradually over time until symptom onset but appeared later than ACPA/RF/anti-CarP antibodies. Positivity for these antibodies yielded higher levels of RANKL compared with seronegativity (P < 0.001). RANKL concentrations were significantly associated with IL-6 and IL-10 concentrations. The combination of positivity for RANKL and anti-CarP antibodies resulted in a higher Larsen score at diagnosis β = 6.18 (95% CI: 0.93, 11.43; P = 0.022).Conclusion:RANKL concentrations were increased several years before symptom onset for RA, particularly in ACPA/RF/anti-CarP-positive individuals, all detectable earlier than RANKL. Positivity for RANKL and anti-CarP antibodies yielded the highest Larsen score at disease onset.

Anti-citrullinated-peptide antibodies (ACPA) have been detected in individuals with developing rheumatoid arthritis (RA) before the onset of symptom, with an initially limited spectrum of reactivities that gradually broadens. The aim was to analyze the evolution of ACPA response pre-dating symptom onset, using four selected citrullinated exogenous and endogenous antigens.

OBJECTIVE:The early diagnosis of rheumatoid arthritis (RA) is desirable to install treatment to prevent disease progression and joint destruction. Autoantibodies and immunological markers pre-date the onset of symptoms by years albeit not all patients will present these factors, even at disease onset. Additional biomarkers would be of high value to improve early diagnosis and understanding of the process, leading to disease development.METHODS:Plasma samples donated before the onset of RA were identified in the Biobank of Northern Sweden, a collection within national health survey programs. Thirty samples from pre-symptomatic individuals and nineteen from controls were subjected to liquid chromatography-mass spectrometry (LCMS) metabolite and lipid profiling. Lipid and metabolite profiles discriminating samples from pre-symptomatic individuals from controls were identified after univariate and multivariate OPLS-DA based analyses.RESULTS:The OPLS-DA models including pre-symptomatic individuals and controls identified profiles differentiating between the groups that was characterized by lower levels of acyl-carnitines and fatty acids, with higher levels of lysophospatidylcholines (LPCs) and metabolites from tryptophan metabolism in pre-symptomatic individuals compared with controls. Lipid profiling showed that the majority of phospholipids and sphingomyelins were at higher levels in pre-symptomatic individuals in comparison with controls.CONCLUSIONS:Our LCMS based approach demonstrated that there are changes in small molecule and lipid profiles detectable in plasma samples collected from the pre-symptomatic individuals who subsequently developed RA, which point to an up-regulation of levels of lysophospatidylcholines, and of tryptophan metabolism, perturbation of fatty acid beta-oxidation and increased oxidative stress in pre-symptomatic individuals' years before onset of symptoms.

OBJECTIVE:To investigate the gene-environment interaction between smoking and single nucleotide polymorphisms (SNPs), using Immunochip material, on the risk of developing either of two serologically defined subsets of RA.METHODS:Interaction between smoking and 133,648 genetic markers from the Immunochip was examined for two RA subsets, defined by the presence or absence of ACPA. A total of 1590 ACPA-positive and 891 ACPA-negative cases were compared with 1856 controls in the Swedish Epidemiological Investigation of RA (EIRA) case-control study. Logistic regression models were used to determine the presence of interaction. The proportion attributable to interaction was calculated for each smoking-SNP pair. Replication was carried out in an independent dataset from northern Sweden. To further validate and extend the results, interaction analysis was also performed using genome-wide association studies data on EIRA individuals.RESULTS:In ACPA-positive RA, 102 SNPs interacted significantly with smoking, after Bonferroni correction. All 102 SNPs were located in the HLA region, mainly within the HLA class II region, 51 of which were replicated. No additional loci outside chromosome 6 were identified in the genome-wide association studies validation. After adjusting for HLA-DRB1 shared epitope, 15 smoking-SNP pairs remained significant for ACPA-positive RA, with 8 of these replicated (loci: BTNL2, HLA-DRA, HLA-DRB5, HLA-DQA1, HLA-DOB and TAP2). For ACPA-negative RA, no smoking-SNP pairs passed the threshold for significance.CONCLUSION:Our study presents extended gene variation patterns involved in gene-smoking interaction in ACPA-positive, but not ACPA-negative, RA. Notably, variants in HLA-DRB1 and those in additional genes within the MHC class II region, but not in any other gene regions, showed interaction with smoking.

The aim of this study was to analyze relationships between receptor activator of nuclear factor kappa-B (RANKL), sclerostin and their gene polymorphisms with radiological progression in patients with early rheumatoid arthritis (RA). Patients with early RA (n = 407, symptomatic <1 year) (ARA criteria) examined radiologically at inclusion and after 24 months were consecutively included. Disease activity score and C-reactive protein were regularly recorded. Sclerostin, RANKL, and anti-CCP2 antibodies were analyzed in plasma at baseline using ELISAs. Data on gene polymorphism for sclerostin and RANKL were extracted from Immunochip analysis. Sex- and age-matched controls (n = 71) were identified from the Medical Biobank of Northern Sweden. The concentration of RANKL was significantly higher in patients compared with controls, median (IQR) 0.56 (0.9) nmol/L and 0.20 (0.25) nmol/L (p < 0.001), and in anti-CCP2-positive patients compared with sero-negative individuals. Sclerostin was significantly increased in female patients 0.59 (0.47-0.65) ng/mL compared with female controls 0.49 (0.4-0.65) ng/mL (p < 0.02). RANKL concentration was related to the Larsen score at baseline (p < 0.01), after 24 months (p < 0.001), and to radiological progression at 24 months (p < 0.001). Positivity of RANKL and anti-CCP2 yielded significant risk for progression with negativity for both as reference. No single nucleotide polymorphism encoding TNFSF11 or SOST was associated with increased concentrations of the factors. The concentration of RANKL was related to the Larsen score at baseline, at 24 months, and radiological progression at 24 months particularly in anti-CCP2-positive patients, while the concentration of sclerostin was unrelated to radiological findings.

Segmental duplications (SDs) comprise about 5% of the human genome and are enriched for immune genes. SD loci often show copy numbers variations (CNV), which are difficult to tag with genotyping methods. CNV in the Fcγ receptor region (FCGR) has been suggested to be associated with rheumatic diseases. The objective of this study was to delineate association of FCGR-CNV with rheumatoid arthritis (RA), coeliac disease and Inflammatory bowel disease incidence. We developed a method to accurately quantify CNV in SD loci based on the intensity values from the Immunochip platform and applied it to the FCGR locus. We determined the method’s validity using three independent assays: segregation analysis in families, arrayCGH, and whole genome sequencing. Our data showed the presence of two separate CNVs in the FCGR locus. The first region encodes FCGR2A, FCGR3A and part of FCGR2C gene, the second encodes another part of FCGR2C, FCGR3B and FCGR2B. Analysis of CNV status in 4578 individuals with RA and 5457 controls indicated association of duplications in the FCGR3B gene in antibody-negative RA (P=0.002, OR=1.43). Deletion in FCGR3B was associated with increased risk of antibody-positive RA, consistently with previous reports (P=0.023, OR=1.23). A clear genotype–phenotype relationship was observed: CNV polymorphisms of the FCGR3A gene correlated to CD16A expression (encoded by FCGR3A) on CD8 T-cells. In conclusion, our method allows determining the CNV status of the FCGR locus, we identified association of CNV in FCGR3B to RA and showed a functional relationship between CNV in the FCGR3A gene and CD16A expression.

OBJECTIVEA highly polygenic aetiology and high degree of allele-sharing between ancestries have been well elucidated in genetic studies of rheumatoid arthritis. Recently, the high-density genotyping array Immunochip for immune disease loci identified 14 new rheumatoid arthritis risk loci among individuals of European ancestry. Here, we aimed to identify new rheumatoid arthritis risk loci using Korean-specific Immunochip data.METHODSWe analysed Korean rheumatoid arthritis case-control samples using the Immunochip and genome-wide association studies (GWAS) array to search for new risk alleles of rheumatoid arthritis with anticitrullinated peptide antibodies. To increase power, we performed a meta-analysis of Korean data with previously published European Immunochip and GWAS data for a total sample size of 9299 Korean and 45,790 European case-control samples.RESULTSWe identified eight new rheumatoid arthritis susceptibility loci (TNFSF4, LBH, EOMES, ETS1-FLI1, COG6, RAD51B, UBASH3A and SYNGR1) that passed a genome-wide significance threshold (p<5×10(-8)), with evidence for three independent risk alleles at 1q25/TNFSF4. The risk alleles from the seven new loci except for the TNFSF4 locus (monomorphic in Koreans), together with risk alleles from previously established RA risk loci, exhibited a high correlation of effect sizes between ancestries. Further, we refined the number of single nucleotide polymorphisms (SNPs) that represent potentially causal variants through a trans-ethnic comparison of densely genotyped SNPs.CONCLUSIONSThis study demonstrates the advantage of dense-mapping and trans-ancestral analysis for identification of potentially causal SNPs. In addition, our findings support the importance of T cells in the pathogenesis and the fact of frequent overlap of risk loci among diverse autoimmune diseases.

B-regulatory cells are functionally impaired in patients with rheumatoid arthritis and in their first degree relatives

Objective.Inflammasomes are intracellular protein complexes important for the production of pro-inflammatory cytokines. Studies have suggested that the NLRP3 inflammasome influences both the severity of rheumatoid arthritis (RA) and development of atherosclerosis. Therefore, we investigated whether functional genetic variants related to the NLRP3 inflammasome influence the risk of cardiovascular (CV) disease (CVD) in patients with RA.Methods.The incidence of CVD was assessed in 522 patients with established RA by a retrospective survey of medical records in combination with a 6-year prospective followup. NLRP3-Q705K and CARD8-C10X genotypes were analyzed in relation to CVD by logistic regression, adjusting for traditional risk factors, antirheumatic treatment, and age at the onset of RA.Results.Carriage of the NLRP3-Q705K minor allele was associated with an increased risk of stroke/transient ischemic attack (TIA; OR 2.01, 95% CI 1.0–4.1, p = 0.05), while CARD8-C10X was not associated with any type of CV event. Patients with ≥ 1 variant allele in both polymorphisms had an increased risk of CVD when compared with patients without variant alleles present in both polymorphisms (adjusted OR 3.05, 95% CI 1.42–6.54, p = 0.004). Stratification showed that this risk was confined to stroke/TIA (adjusted OR 5.09, 95% CI 2.27–11.44, p < 0.0001) and not to myocardial infarction (MI)/angina pectoris (adjusted OR 1.58, 95% CI 0.67–3.73). Risk estimates were consistently higher among female patients.Conclusion.Genetic variants of the NLRP3 inflammasome influence the risk of stroke/TIA, but not of MI/angina pectoris in Swedish patients with established RA.

A Dense Mapping Of Human Leukocyte Antigen Region For Study Of Interaction With Smoking In The Development Of Rheumatoid Arthritis

Objectives: Genome-wide association studies have facilitated the identification of over 30 susceptibility loci for rheumatoid arthritis (RA). However, evidence for a number of potential susceptibility genes have not so far reached genome-wide significance in studies of Caucasian RA.Methods: A cohort of 4286 RA patients from across Europe and 5642 population matched controls were genotyped for 25 SNPs, then combined in a meta-analysis with previously published data.Results: Significant evidence of association was detected for nine SNPs within the European samples. When meta-analysed with previously published data, 21 SNPs were associated with RA susceptibility. Although SNPs in the PTPN2 gene were previously reported to be associated with RA in both Japanese and European populations, we show genome-wide evidence for a different SNP within this gene associated with RA susceptibility in an independent European population (rs7234029, P = 4.4x10(-9)).Conclusions: This study provides further genome-wide evidence for the association of the PTPN2 locus (encoding the T cell protein tyrosine phosphastase) with Caucasian RA susceptibility. This finding adds to the growing evidence for PTPN2 being a pan-autoimmune susceptibility gene.

Background Multiplex analysis has demonstrated the presence of several antibodies against cyclic citrullinated peptides (ACPA) preceding the development of rheumatoid arthritis (RA) by several years. Certain HLA-DRB1* alleles are strongly associated with RA. Objectives To relate the presence of different ACPA, in individuals before the onset of symptoms of RA, to single nucleotide polymorphisms (SNPs) within the HLA-DRB1 region in order to identify the SNPs most highly associated with the ACPAs analysed in the disease development. Methods The study group comprised 406 individuals, with 717 samples, who were identified before the onset of symptoms (median (IQR) 7.4 (9.3) years), of 976 population controls, who were identified as donors to the Medical Biobank of Northern Sweden. Analysis of antibodies against 10 different citrullinated peptides in plasma using a microarray system developed in collaboration with Phadia AB/Thermofisher, Uppsala, based on their ISAC platform. The highest frequencies of antibodies were found for fibrinogen (Fib) ß36-52, CEP-1, and filaggrin. Samples from the same individuals (n=370) were genotyped with the Immunochip custom array according to Illumina protocols (SNP&SEQ Technology Platform, Uppsala, Sweden). 81 SNPs covering the HLA-DRB1* to DQA1* region were analysed. The Haploview software version 4.2 was used for haplotype and association analysis with a permutation test and cluster analysis (Bray-Curtis) was performed using Past software. Results Thirty-seven of the SNPs were associated with the pre-diseased individuals compared with controls. After permutation test 22 of them and six haplotypes (7-16 SNPs) remained significantly associated. Anti-CEP-1 antibodies were associated with 15 SNPs although none remained significant after permutation tests. Anti-filaggrin antibodies were related to 4 SNPs, two of which were similar as for anti-CEP-1 (rs9271588 and rs9272363), none remained significant after permutation test. Anti-Fibß36-52 antibodies were associated with 4 different SNPs compared with the other antibodies and none was significant after permutation test. Antibodies against Fibß36-52 associated with the SNPs were all within the same cluster of SNPs, whilst associations of antibodies against CEP-1 were within three different sub-clusters. One cluster of SNPs was similar to where associations to anti- filaggrin were found. Conclusions The presence of ACPA is related to a number of SNPs within the HLA- region albeit with somewhat different patterns between the various antibodies and clusters of SNPs. Disclosure of Interest None Declared

A major challenge in human genetics is to devise a systematic strategy to integrate disease-associated variants with diverse genomic and biological data sets to provide insight into disease pathogenesis and guide drug discovery for complex traits such as r