E. Korendowych

A modified sharp score demonstrates disease progression in established psoriatic arthritis

To use a modified Sharp score (MSS) to measure radiologic progression and to assess its relationship to other radiologic features, peripheral joint disease, and physical function in psoriatic arthritis (PsA).

SAT0292 Facilitating a modular approach to the assessment of psoriatic arthritis

Background Psoriatic Arthritis (PsA) is a complex condition involving both axial and peripheral joints, together with enthesitis, dactylitis and skin. A complete assessment of PsA is challenging within the short time available in routine clinics. Assessment of these components is important as they all contribute significantly to disease burden, and may require specific treatments for optimal outcome. Objectives As part of an initiative programme promoting the “Treat-to-Target” concept, a group of rheumatologists, dermatologists and nurse specialists considered ways of assessing all components of psoriatic disease in routine clinics. A modular approach for assessing psoriatic disease was proposed, with skin and axial disease initially assessed by questionnaire. Patients identified with significant symptoms could then be further clinically assessed. This study reports patient acceptability and utility of this proposal in a routine clinic setting. Methods A trained receptionist handed two questionnaires to sequential patients with psoriatic arthritis attending routine clinics at Guy’s Hospital, asking two questions:Questionnaire 1: Have you been suffering from any neck or back pain recently? Questionnaire 2: Do you have psoriasis at the moment? If patients answered yes to question 1, the questionnaire invited them to complete the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), on the reverse side. If they answered yes to question 2, they completed the Dermatology Life Quality Index (DLQI). Questionnaires had a tick-box format, and were completed in the waiting area prior to their appointment, then reviewed at their consultation with the doctor. Patients who had BASDAI scores >4 were seen again in a follow-up consultation at 3 months. Results 70 pairs of questionnaires were distributed and completed. All patients found them easy to understand and to answer. 47 patients (67.1%) identified co-existing neck or back pain, with 28 (59.6%) scoring BASDAI >4. On further review, 19 had axial pain due to degenerative spinal disease, and 9 cases had active spondylitis. 45 patients (62.6%) identified themselves as having co-existing psoriasis, with 24 patients (51%) scoring DLQI >5 and 8 patients (17%) a DLQI >10. Conclusions This preliminary use of screening questionnaires to facilitate a modular approach to the assessment of PsA showed high patient acceptability, and clinicians found the scores useful to indicate the extent of skin and axial involvement. We closely monitored our cohort with active spondylitis, and initially optimized NSAID therapy if possible. At three month follow up, 2 patients had repeat BASDAI >4, making them eligible for tumour necrosis factor inhibitor therapy. This was started in these 2 patients with good effect. DLQI >5 indicates significant skin involvement. 62.6% of our patient group reported skin involvement, of whom over a third had a DLQI >5, but only 8 of 70 patients had a score >10 suggesting more severe psoriasis activity. We propose the use of these questionnaires in routine practice. This patient acceptable approach helps identify skin and axial skeleton involvement early, so that treatment can be optimized quickly. In the future this approach will assist new composite disease activity measures for the assessment and treatment of psoriatic arthritis. Acknowledgements Abbott Immunology supported the Psoriatic Arthritis Assessment Academy programme. Disclosure of Interest None Declared

AB0908 Ability of the reductive x-ray score for psoriatic arthritis (REXSPA) to detect change in an observational cohort of patients with psa

Background The measurement of radiographic joint damage is important in characterising disease severity, progression, and prognosis in psoriatic arthritis (PsA). Existing radiographic measures are time-consuming to perform, leading to limited data collection from existing longitudinal observational studies. Objectives We have previously proposed a Reductive X-ray Score for Psoriatic Arthritis (ReXSPA)1 as more feasible method, and in this study set out to examine the sensitivity of ReXSPA in a new cohort of patients. Methods A retrospective sample of 28 patients who had hand and feet radiographs at 3 time points (5 years before [T0], at the time of [T1], and 5 years post [T2] commencement of anti-TNF treatment) were taken from the Bath longitudinal PsA cohort. Radiographs were scored for erosion, joint space narrowing and proliferation to calculate the Sharp-van der Heijde modified method (VDH) and ReXSPA scores. A sample of 9 radiographs were scored by all assessors (WT, AA and AA) to determine inter- and intra-rater reliability using intra-class correlation coefficients (ICC). Sensitivity to change was determined from timepoint T0 to T2 using the Standardised Response Mean (SRM) and Smallest Detectable Change (SDC). Results The patients’ mean age (SD) at T0 was 61 years (13.4), the mean disease duration was 11.2 years (11.14). Patients were followed up for a mean (SD) of 10.2 years (2.76). Overall inter- and intra-rater reliability for ReXSPA and VDH were 0.80 and >0.92 and 0.91 and >0.90 respectively. The median (IQR) of ReXSPA score was 8.5 (1–14), 12.5 (5–20) and 14.5 (8–36) at T0, T1, and T2 respectively. The percentage SDC was 0.91 for the ReXSPA method and 0.77 for the VDH method, and the SRMs were 0.92 and 0.87 respectively (table 1), demonstrating the sensitivity of both methods in detecting change. There was a trend towards slowing in radiographic progression following the initiation of TNF-inhibitors, but ReXSPA was less sensitive compared to the VDH and was not able to detect a significant change in the rate of progression post-TNF inhibition (p 0.08) (Graphic 1).Abstract AB0908 – Table 1 Sensitivity to change of each scoring method Method Mean Change SD of change SEM SRM SDC SDC as% of total score VDHErosionJSN 22.117.2916.96 19.1410.6215.17 3.622.012.87 0.871.460.89 4.092.273.24 0.770.711.56 ReXPSAErosionJSNProliferation 10.93.216.360.96 10.05.315.112.32 1.891.000.970.44 0.921.650.802.40 2.141.131.090.50 0.911.031.241.39 Sharp-van der Heijde modified method (VDH), Standard deviation (SD), Standard error of mean (SEM), Standardised response mean (SRM), Smallest detectable change (SDC).Abstract AB0908 – Figure 1 Cumulative probability plot demonstrating ReXSPA progression pre- and post- TNF inhibition Conclusions The RexSPA is a reliable and sensitive alternate scoring method for the detection of radiographic progression in an observational cohort of patients with PsA, but not as sensitive to change as the VDH method. Reference [1] Tillett W, et al. (2016) Novel Composite Radiographic Score for Longitudinal Observational Studies of Psoriatic Arthritis: A Proof-of-concept Study. J Rheum; 43(2):367–370 Disclosure of Interest None declared

SAT0442 Functional disease status and healthcare costs in psoriatic arthritis: a similar relationship to that seen in rheumatoid arthritis

Background The Health Assessment Questionnaire-Disability Index (HAQ-DI), an important, validated measure of functional status in patients with psoriatic arthritis (PsA), has been used as an intermediate variable in modelling of costs and quality-adjusted life-years. The relationship between HAQ-DI and costs has been well documented for rheumatoid arthritis (RA),1,2 but less so for PsA. The HAQ-DI study data in PsA patients from the British Society for Rheumatology Biologics Registers was mapped to resource use data in the Health Improvement Network dataset.3 A drawback of this approach relates to the HAQ-DI and resource use data being derived from separate patient cohorts. Objectives Estimate the HAQ-DI and cost relationship in PsA within a single cohort of patients. Methods Functional disease status, patient demographic, disease history and healthcare resource use data were extracted from a cohort of patients at the Royal National Hospital for Rheumatic Diseases. Resource data were available for primary and secondary care consultations, prescriptions, accident and emergency attendance, hospital admissions and tests, and collected for 6 months before and after HAQ-DI measurement. Medication costs were excluded from the modelling as it was not possible to specify which were PsA-related. Linear regression models were used to predict costs as a function of HAQ-DI and age at HAQ-DI measurement. As cost data were not normally distributed, standard errors (SEs), 95% confidence intervals (CIs) and P values were estimated by bootstrap resampling (10,000 samples). Results Data were available for 95 PsA patients (female: n=43). Mean HAQ-DI score was 0.81 (SE 0.08); mean age at HAQ-DI measurement was 58.6 (SE 1.04). Mean total annual healthcare costs, excluding medication costs, were £ 1,588 (SE 172.3). Regression modelling indicated that a 1-point increase in HAQ-DI score was associated with an increase in total costs, excluding medications, of £ 547.49 (SE 222.7; 95% CI 191.7–1,103.8; P=0.004) (Table). Subgroup analyses suggested trends for higher cost increases within the lower HAQ-DI cutoff subgroup (HAQ-DI 0–1) and for those with greatest disease duration (>10 years). Costs associated with secondary care consultations seemed to be the primary factor in the association between HAQ-DI and total costs. Conclusions Models incorporating total healthcare costs were highly significant; this association appears to be driven mainly by secondary care consultation costs. Costs in this study were similar to previous studies in RA populations. A study limitation was that only direct medical costs were considered, which may underestimate the true burden of PsA on healthcare systems and the wider society. References Lajas C, et al. Arthritis Rheum. 2003;49:64–70. Kobelt G, et al. Joint Bone Spine. 2008;75:408–15. Poole CD, et al. Rheumatology (Oxford). 2010;49:1949–56. Disclosure of Interest A. Maguire Grant/research support from: Cogentia Healthcare Consulting Ltd, UK, I. Handel Consultant for: Visible Analytics Limited, W. Tillett: None declared, F. Mughal Employee of: Celgene Ltd, J. Morris Grant/research support from: Celgene Ltd, N. Hawkins Grant/research support from: Celgene Ltd, C. Cavill Grant/research support from: Celgene Ltd, E. Korendowych Grant/research support from: Abbvie, Celgene and Pfizer, Consultant for: Janssen, Abbvie, Novartis, Pfizer and Celgene, Speakers bureau: Janssen, Abbvie, Novartis, Pfizer and Celgene, N. McHugh Grant/research support from: Celgene Ltd

SAT0011 Replication of A Distinct Psoriatic Arthritis Risk Variant at IL23R

Background Psoriatic arthritis (PsA) is an inflammatory arthritis that is associated with psoriasis. Although the majority of PsA genetic risk loci identified also confer risk for psoriasis, we have recently reported evidence of loci that are associated with PsA and not psoriasis, including an association at the IL23R locus which is also independent of a psoriasis variant at the same locus. Methods The PsA-specific SNP rs12044149, and the psoriasis SNP rs9988642 at the IL23R locus were genotyped in a cohort of 914 PsA cases and 6,945 controls from Spain, Crete, Germany and the UK, with the Life Technologies QuantStudio genotyping platform. Association testing was carried out using PLINK, and a meta-analysis was performed with PsA Immunochip data (1,962 cases, 8,923 controls). Genotype data was also available from the psoriasis WTCCC2 study (excluding known PsA, n=1,784) to compare effect sizes between PsA and psoriasis using multinomial logistic regression in Stata, and to directly compare PsA and psoriasis genotypes. Credible SNP sets were calculated for the PsA and psoriasis IL23R associations using the Bayesian refinement method, and functionally annotated. Results We replicated the association rs12044149 with PsA (P=4.03x10–6), which remained significant when conditioning upon the psoriasis variant, rs9988642 (Pcond=4.86x10–6), and reached genome-wide significance during meta-analysis of the combined PsA dataset (Pmeta=4.76x10–20). Only a modest association was found for rs9988642 (P=0.04), with no genome-wide significance found during meta-analysis (P=4.61x10–4). Within the psoriasis dataset, this association remained significant when conditioning upon rs12044149 (P=1.0x10–07 vs. Pcond=1.63x10–5). Effect estimates for rs12044149 were significantly different between PsA and psoriasis (P=2.0x10–3), and direct comparison of genotypes for PsA and psoriasis found the risk allele to be significantly increased in PsA (P=4.52x10–4, OR=1.3). Credible SNPs for rs12044149 mapped to promoter and enhancer regions within memory CD8+ T cells, which we have previously reported to be critical for PsA. Conclusions We have successfully replicated a PsA-specific (associated with PsA but not psoriasis) risk variant at the IL23R locus in an independent cohort, confirming rs12044149 to be distinct from rs9988642 (r2=0.02). This gives five PsA-specific associations that have now been identified. Disclosure of Interest None declared

THU0003 Fine-mapping of autoimmune susceptibility LOCI using immunochip identifies novel susceptibility LOCI for psoriatic arthritis

Background Psoriatic arthritis (PsA) is a chronic inflammatory joint disease that typically accompanies psoriasis vulagris (PsV). Both conditions are considered to be complex diseases with both genetic and environmental susceptibility factors. Genome-wide association studies (GWAS) have identified a number of genetic susceptibility loci for both diseases. Comparing GWAS results performed in many common autoimmune diseases has revealed an extensive overlap in the genetic liability between autoimmune diseases in general. Objectives Based on the observed sharing of susceptibility loci across autoimmune diseases, the aim of this project is to identify novel susceptibility loci for PsA by fine-mapping all the currently confirmed susceptibility loci for 12 autoimmune diseases. Methods Data was available for 929 PsA cases and 4537 healthy controls genotyped using the Immunochip Illumina iSelect array at the Sanger Centre (www.sanger.ac.uk). Control data was sourced from the WTCCC (www.wtccc.org.uk). This custom array was designed to comprehensively fine-map confirmed autoimmune susceptibility loci and contains 196,524 SNPs covering approximately 200 candidate regions. A strict SNP-focused quality control process was applied to the dataset followed by single point analysis using the Armitage test for trend. Results Association analysis of 182,883 high quality SNPs in 862 cases and 4306 controls robustly detected association to previously confirmed PsA risk loci; HLA-C (p=8.2×10-35), IL23R (p=8.4×10-8) and TRAF3IP2 (p=2.8×10-7), IL12B (p=6.5×10-6). In addition we find convincing evidence to support association to a number of novel loci not previously reported for PsA, including; 17q21 (ptrend =3.3×10-05, SMARCE1), 18p11 (ptrend =2.0×10-05, PTPN2), 11q23 (ptrend =1.4×10-05, TREH), and 19p13 (ptrend =9.8×10-05, TYK2). Interestingly, we also find convincing evidence for association to the CARD15 gene, supported by multiple SNPs. Conclusions The preliminary single point analysis of an autoimmune loci fine-mapping dataset identifies a number of potential novel loci for PsA susceptibility. Validation of these results is currently underway in an independent sample collection. Single-point analysis will be followed by analyses to indentify independent effects within the previously identified and newly discovered loci. Disclosure of Interest None Declared

SAT0007 Fine-mapping of the 5Q31 psoriatic arthritis susceptibility LOCI in a british population

Background Psoriatic Arthritis (PsA) is a chronic inflammatory arthritis associated with psoriasis which manifests as a heterogeneous disease with a wide spectrum of disease severity. PsA is a complex disease whereby environmental factors trigger disease in genetically susceptible individuals and multiple genetic factors contribute towards disease susceptibility. The overlap of PsA with psoriasis has resulted in confounding results for genome wide association studies (GWAS) producing difficulty in separating out the 2 diseases. There is however evidence from subgroup analysis of a GWAS to suggest that a locus on chromosome 5q31 is specifically associated with PsA (1). This result has been replicated in two independent studies involving UK and Canadian subjects (2, 3). There is substantial linkage disequilibrium within this locus necessitating the need for fine mapping within the region. Objectives To fine-map the chromosome 5q31 region using a dense set of single nucleotide polymorphisms (SNPs) to refine the association signal to a specific gene or regulatory element. Methods Genotype data for a total of 1518 SNPs spanning the 5q31 region (131.4-132.2Mb) was available for 929 PsA UK cases and 4537 UK healthy controls (www.wtccc.org.uk). Genotyping was performed as part of a wider study targeting confirmed autoimmune susceptibility loci using the Immunochip Illumina iSelect array. A strict quality control (QC) process was applied to the dataset followed by single point analysis using the Armitage test for trend. Results Following stringent QC, the analysis of 1440 high quality SNPs in 862 cases and 4306 controls identified an association to rs715285 (p=8.92×10-5). This localises the association to an intergenic region flanked by the genes: colony stimulating factor 2 (CSF) and prolyl 4-hydroxylase, alpha polpeptide II (P4HA2). Conclusions This preliminary analysis localises the association signal for susceptibility to PsA to an intergenic region approximately 500Kb away from the previously identified IL13 SNP (rs20541), demonstrating the importance of fine-mapping. The study highlights two genes for consideration in disease aetiology: the cytokine CSF2 and P4HA2 which encodes a subunit of the proly 4-hydroxylase; a key enzyme involved in collagen synthesis. References Duffin KC, Freeny IC, Schrodi SJ, Wong B, Feng B-J, Soltani-Arabshahi R, et al. Association between IL13 polymorphisms and psoriatic arthritis is modified by smoking. Journal of Investigative Dermatology. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov’t]. 2009 Dec;129(12):2777-83. Bowes J, Eyre S, Flynn E, Ho P, Salah S, Warren RB, et al. Evidence to support IL-13 as a risk locus for psoriatic arthritis but not psoriasis vulgaris. Annals of the Rheumatic Diseases. 2011 June 1, 2011;70(6):1016-9. Eder L, Chandran V, Pellett F, Pollock R, Shanmugarajah S, Rosen CF, et al. IL13 gene polymorphism is a marker for psoriatic arthritis among psoriasis patients. Annals of the Rheumatic Diseases. [Multicenter Study Research Support, Non-U.S. Gov’t]. 2011 Sep;70(9):1594-8. Disclosure of Interest None Declared