Explore the vast range of titles available.

In view of the growth of clinical risk prediction models using genetic data, there is an increasing need for studies that use appropriate methods to select the optimum number of features from a large number of genetic variants with a high degree of redundancy between features due to linkage disequilibrium (LD). Filter feature selection methods based on information theoretic criteria, are well suited to this challenge and will identify a subset of the original variables that should result in more accurate prediction. However, data collected from cohort studies are often high-dimensional genetic data with potential confounders presenting challenges to feature selection and risk prediction machine learning models. Patients with psoriasis are at high risk of developing a chronic arthritis known as psoriatic arthritis (PsA). The prevalence of PsA in this patient group can be up to 30% and the identification of high risk patients represents an important clinical research which would allow early intervention and a reduction of disability. This also provides us with an ideal scenario for the development of clinical risk prediction models and an opportunity to explore the application of information theoretic criteria methods. In this study, we developed the feature selection and psoriatic arthritis (PsA) risk prediction models that were applied to a cross-sectional genetic dataset of 1462 PsA cases and 1132 cutaneous-only psoriasis (PsC) cases using 2-digit HLA alleles imputed using the SNP2HLA algorithm. We also developed stratification method to mitigate the impact of potential confounder features and illustrate that confounding features impact the feature selection. The mitigated dataset was used in training of seven supervised algorithms. 80% of data was randomly used for training of seven supervised machine learning methods using stratified nested cross validation and 20% was selected randomly as a holdout set for internal validation. The risk prediction models were then further validated in UK Biobank dataset containing data on 1187 participants and a set of features overlapping with the training dataset.Performance of these methods has been evaluated using the area under the curve (AUC), accuracy, precision, recall, F1 score and decision curve analysis(net benefit). The best model is selected based on three criteria: the ‘lowest number of feature subset’ with the ‘maximal average AUC over the nested cross validation’ and good generalisability to the UK Biobank dataset. In the original dataset, with over 100 different bootstraps and seven feature selection (FS) methods, HLA_C_*06 was selected as the most informative genetic variant. When the dataset is mitigated the single most important genetic features based on rank was identified as HLA_B_*27 by the seven different feature selection methods, consistent with previous analyses of this data using regression based methods. However, the predictive accuracy of these single features in post mitigation was found to be moderate (AUC= 0.54 (internal cross validation), AUC=0.53 (internal hold out set), AUC=0.55(external data set)). Sequentially adding additional HLA features based on rank improved the performance of the Random Forest classification model where 20 2-digit features selected by Interaction Capping (ICAP) demonstrated (AUC= 0.61 (internal cross validation), AUC=0.57 (internal hold out set), AUC=0.58 (external dataset)). The stratification method for mitigation of confounding features and filter information theoretic feature selection can be applied to a high dimensional dataset with the potential confounders.

Background A recent systematic review identified four candidate serum-soluble bone-turnover biomarkers (dickkopf-1, Dkk-1; macrophage-colony stimulating factor, M-CSF; matrix metalloproteinase-3, MMP-3; osteoprotegerin, OPG) showing possible association with psoriatic arthritis (PsA). We aimed to: (i) confirm and determine if these four biomarkers are associated with PsA; (ii) differentiate psoriasis cases with and without arthritis; and (iii) differentiate PsA cases with and without axial arthritis. Methods A prospective cross-sectional comparative two-centre study recruited 200 patients with psoriasis without arthritis (PsC), 127 with PsA without axial arthritis (pPsA), 117 with PsA with axial arthritis (psoriatic spondyloarthritis, PsSpA), 157 with ankylosing spondylitis (AS) without psoriasis, and 50 matched healthy controls (HC). Serum biomarker concentrations were measured using ELISA. Multivariable regression and receiver operating characteristic analyses were performed. Results MMP-3 concentrations were significantly higher and M-CSF significantly lower in each arthritis disease group compared with HC ( p  ≤ 0.02). MMP-3 concentrations were significantly higher (adjusted odds ratio, OR adj 1.02 per ng/ml increase in concentration; p  = 0.0004) and M-CSF significantly lower (OR adj 0.44 per ng/ml increase; p  = 0.01) in PsA (pPsA and PsSpA combined) compared with PsC. Dkk-1 concentrations were significantly higher (OR adj 1.22 per ng/mL increase; p  = 0.01), and OPG concentrations significantly lower (OR adj 0.20 per ng/mL increase; p  = 0.02) in patients with axial arthritis (PsSpA and AS combined) than in those without (pPsA). Furthermore, Dkk-1 concentrations were significantly higher along a spectrum of increasing axial arthritis; Dkk-1 concentrations were higher in AS compared with PsSpA (OR adj 1.18 per ng/mL increase; p  = 0.02). Receiver operating characteristic analysis showed MMP-3 to be the best single biomarker for differentiating PsA from PsC (AUC 0.70 for a cut-off of 14.51 ng/mL; sensitivity 0.76, specificity 0.60). Conclusions MMP-3 and M-CSF are biomarkers for the presence of arthritis in psoriatic disease, and could therefore be used to screen for PsA in psoriasis cohorts. Dkk-1 and OPG are biomarkers of axial arthritis; they could therefore be used to screen for the presence of axial disease in PsA cases, and help differentiate PsSpA from AS. High concentrations of Dkk-1 in AS and PsSpA compared with HC, support previous reports that Dkk-1 is dysfunctional in the spondyloarthritides.

Objectives To investigate a shared genetic aetiology for skin involvement in psoriasis and psoriatic arthritis (PsA) by genotyping single-nucleotide polymorphisms (SNPs), reported to be associated in genome-wide association studies of psoriasis, in patients with PsA. Methods SNPs with reported evidence for association with psoriasis were genotyped in a PsA case and control collection from the UK and Ireland. Genotype and allele frequencies were compared between PsA cases and controls using the Armitage test for trend. Results Seven SNPs mapping to the IL1RN, TNIP1, TNFAIP3, TSC1, IL23A, SMARCA4 and RNF114 genes were successfully genotyped. The IL23A and TNIP1 genes showed convincing evidence for association (rs2066808, p = 9.1×10−7; rs17728338, p = 3.5×10−5, respectively) whilst SNPs mapping to the TNFAIP3, TSC1 and RNF114 genes showed nominal evidence for association (rs610604, p = 0.03; rs1076160, p = 0.03; rs495337, p = 0.0025). No evidence for association with IL1RN or SMARCA4 was found but the power to detect association was low. Conclusions SNPs mapping to previously reported psoriasis loci show evidence for association to PSA, thus supporting the hypothesis that the genetic aetiology of skin involvement is the same in both PsA and psoriasis.

Objective A number of rheumatoid arthritis (RA) susceptibility genes have been identified in recent years. Given the overlap in phenotypic expression of synovial joint inflammation between RA and psoriatic arthritis (PsA), the authors explored whether RA susceptibility genes are also associated with PsA. Methods 56 single nucleotide polymorphisms (SNPs) mapping to 41 genes previously reported as RA susceptibility loci were selected for investigation. PsA was defined as an inflammatory arthritis associated with psoriasis and subjects were recruited from the UK and Ireland. Genotyping was performed using the Sequenom MassArray platform and frequencies compared with data derived from large UK control collections. Results Significant evidence for association with susceptibility to PsA was found toa SNP mapping to the REL (rs13017599, ptrend=5.2×104) gene, while nominal evidence for association (ptrend<0.05) was found to seven other loci including PLCL2 (rs4535211, p=1.7×10−3); STAT4 (rs10181656, p=3.0×10−3) and the AFF3, CD28, CCL21, IL2 and KIF5A loci. Interestingly, three SNPs demonstrated opposite effects to those reported for RA. Conclusions The REL gene, a key modulator of the NFκB pathway, is associated with PsA but the allele conferring risk to RA is protective in PsA suggesting that there are fundamental differences in the aetiological mechanisms underlying these two types of inflammatory arthritis.

ObjectivesTo compare the prevalence, clinical and radiographic characteristics of psoriatic spondyloarthritis (PsSpA) in psoriatic arthritis (PsA), with ankylosing spondylitis (AS).MethodsA prospective single-centre cross-sectional observational study recruited consecutive PsA and AS cases. Participants completed outcome measures, and underwent clinical examination, axial radiographic scoring and HLA-sequencing. Multivariable analyses are presented.ResultsThe 402 enrolled cases (201 PsA, 201 AS; fulfilling classification criteria for respective conditions) were reclassified based upon radiographic axial disease and psoriasis, as: 118 PsSpA, 127 peripheral-only PsA (pPsA), and 157 AS without psoriasis (AS) cases. A significant proportion of patients with radiographic axial disease had PsSpA (118/275; 42.91%), and often had symptomatically silent axial disease (30/118; 25.42%). Modified New York criteria for AS were fulfilled by 48/201 (23.88%) PsA cases, and Classification of Psoriatic Arthritis criteria by 49/201 (24.38%) AS cases. pPsA compared with PsSpA cases had a lower frequency of HLA-B*27 (OR 0.12; 95% CI 0.05 to 0.25). Disease activity, metrology and disability were comparable in PsSpA and AS. A significant proportion of PsSpA cases had spondylitis without sacroiliitis (39/118; 33.05%); they less frequently carried HLA-B*27 (OR 0.11; 95% CI 0.04 to 0.33). Sacroiliac joint complete ankylosis (adjusted OR, ORadj 2.96; 95% CI 1.42 to 6.15) and bridging syndesmophytes (ORadj 2.78; 95% CI 1.49 to 5.18) were more likely in AS than PsSpA. Radiographic axial disease was more severe in AS than PsSpA (Psoriatic Arthritis Spondylitis Radiology Index Score: adjusted incidence risk ratio 1.13; 95% CI 1.09 to 1.19).ConclusionsIn a combined cohort of patients with either PsA or AS from a single centre, 24% fulfilled classification criteria for both conditions. The pattern of axial disease was influenced significantly by the presence of skin psoriasis and HLA-B*27.

Objective To identify predictors of poorer physical function in established psoriatic arthritis (PsA). Methods PsA patients with disease duration of ≥10 years were identified from the Bath longitudinal cohort. Physical function was assessed using the Stanford Health Assessment Questionnaire (HAQ). Sex, age at diagnosis, duration of symptoms prior to diagnosis, smoking, treatment and year of diagnosis were included in a multivariable regression analysis to identify associations with HAQ. Results 267 patients were identified for inclusion. The median age was 56 years (IQR 45–63), median disease duration was 13 years (IQR 10–18) and median HAQ score was 0.63 (IQR 0.13–1.25). The model predicted significant increases in HAQ related to smoking (0.23, 95% CI 0.04 to 0.42), age >50 years at diagnosis (0.27, 95% CI 0.03 to 0.51), symptom duration of ≥1 year before diagnosis (0.22, 95% CI 0.02 to 0.42), female sex (0.39, 95% CI 0.20 to 0.57) and history of treatment with an anti-TNF agent (0.63, 95% CI 0.32 to 0.93) at follow-up. Conclusions Smoking, delay to diagnosis, older age at diagnosis, female sex and a history of anti-TNF treatment are associated with worse physical function in established PsA.

ObjectivesThe Psoriatic Arthritis Impact of Disease (PsAID) Questionnaire is a recently developed patient-reported outcome measure (PROM) of disease impact in psoriatic arthritis (PsA). We set out to assess the validity in an independent cohort of patients, estimate the minimally important difference for improvement and explore the potential of individual components of the PsAID in clinical practice.MethodsData were collected prospectively for a single-centre cohort of patients with PsA. Construct validity was assessed by Spearman correlation with other PROMs and reliability by intraclass correlation coefficient (ICC) at 1 week. Sensitivity to change at 3 months was determined by the standardised response mean (SRM) in those patients with active disease requiring a change in treatment.ResultsA total of 129 patients (mean ±SD age 52.1±13.3, 57% women, disease duration 10.2±8 years) completed the baseline questionnaires and assessments. The mean baseline PsAID12 score was 3.92±2.26 with an ICC of 0.91 (95%CI 0.87 to 0.94). The SE of measurement was 0.51 and the minimal detectable change was 1.41. There was strong correlation (r≥0.70) with most of the PROMs studied and moderate correlation with clinical outcomes (r=0.40–0.57). The SRM of the PsAID12 was 0.74 (95%CI 0.45 to 0.97). There was strong correlation with individual PsAID items and their corresponding PROM questionnaires (r≥0.67).ConclusionThe PsAID is a reliable, feasible and discriminative measure in patients with PsA. The good responsiveness of the PsAID and strong correlation of individual items with other PROMS represent an opportunity to reduce questionnaire burden for patients in studies and clinical practice.

André Reis and colleagues report a genome-wide association study identifying a susceptibility locus at TRAF3IP2 for psoriatic arthritis and psoriasis. Psoriatic arthritis (PsA) is an inflammatory joint disease that is distinct from other chronic arthritides and which is frequently accompanied by psoriasis vulgaris (PsV) and seronegativity for rheumatoid factor. We conducted a genome-wide association study in 609 German individuals with PsA (cases) and 990 controls with replication in 6 European cohorts including a total of 5,488 individuals. We replicated PsA associations at HLA-C and IL12B and identified a new association at TRAF3IP2 (rs13190932, P = 8.56 × 10 −17 ). TRAF3IP2 was also associated with PsV in a German cohort including 2,040 individuals (rs13190932, P = 1.95 × 10 −3 ). Sequencing of the exons of TRAF3IP2 identified a coding variant (p.Asp10Asn, rs33980500) as the most significantly associated SNP ( P = 1.13 × 10 −20 , odds ratio = 1.95). Functional assays showed reduced binding of this TRAF3IP2 variant to TRAF6, suggesting altered modulation of immunoregulatory signals through altered TRAF interactions as a new and shared pathway for PsA and PsV.

Psoriatic arthritis (PsA) is a chronic inflammatory arthritis associated with psoriasis and, despite the larger estimated heritability for PsA, the majority of genetic susceptibility loci identified to date are shared with psoriasis. Here, we present results from a case–control association study on 1,962 PsA patients and 8,923 controls using the Immunochip genotyping array. We identify eight loci passing genome-wide significance, secondary independent effects at three loci and a distinct PsA-specific variant at the IL23R locus. We report two novel loci and evidence of a novel PsA-specific association at chromosome 5q31. Imputation of classical HLA alleles, amino acids and SNPs across the MHC region highlights three independent associations to class I genes. Finally, we find an enrichment of associated variants to markers of open chromatin in CD8 + memory primary T cells. This study identifies key insights into the genetics of PsA that could begin to explain fundamental differences between psoriasis and PsA. Psoriatic arthritis (PsA) is a chronic inflammatory arthritis with a significant genetic component. Here, the authors analyse immune-related genetic markers in 1,962 PsA patients and 8,923 controls to identify novel PsA risk loci and highlight distinct genetic differences between psoriasis and PsA.

ObjectivesPsoriatic arthritis (PsA) is a chronic inflammatory arthritis, with a strong heritable component, affecting patients with psoriasis. Here we attempt to identify genetic variants within the major histocompatibility complex (MHC) that differentiate patients with PsA from patients with cutaneous psoriasis alone (PsC).Methods2808 patients with PsC, 1945 patients with PsA and 8920 population controls were genotyped. We imputed SNPs, amino acids and classical HLA alleles across the MHC and tested for association with PsA compared to population controls and the PsC patient group. In addition we investigated the impact of the age of disease onset on associations.ResultsHLA-C*06:02 was protective of PsA compared to PsC (p=9.57×10−66, OR 0.37). The HLA-C*06:02 risk allele was associated with a younger age of psoriasis onset in all patients (p=1.01×10−59). After controlling for the age of psoriasis onset no association of PsA to HLA-C*06:02 (p=0.07) was observed; instead, the most significant association was to amino acid at position 97 of HLA-B (p=1.54×10−9) where the presence of asparagine or serine residue increased PsA risk. Asparagine at position 97 of HLA-B defines the HLA-B*27 alleles.ConclusionsBy controlling for the age of psoriasis onset, we show, for the first time, that HLA-C*06:02 is not associated with PsA and that amino acid position 97 of HLA-B differentiates PsA from PsC. This amino acid also represents the largest genetic effect for ankylosing spondylitis, thereby refining the genetic overlap of these two spondyloarthropathies. Correcting for bias has important implications for cross-phenotype genetic studies.