Machine Learning Pipeline for Predicting Bone Marrow Edema Along the Sacroiliac Joints on Magnetic Resonance Imaging.

OBJECTIVE: We aimed to develop and validate a fully automated machine learning (ML) algorithm that predicts bone marrow edema (BME) on a quadrant level in sacroiliac (SI) joint magnetic resonance imaging (MRI). METHODS: A computer vision workflow automatically locates the SI joints, segments regions of interest (ilium and sacrum), performs objective quadrant extraction, and predicts presence of BME, suggestive of inflammatory lesions, on a quadrant level in semicoronal slices of T1/T2-weighted MRI scans. Ground truth was determined by consensus among human readers. The inflammation classifier was trained using a ResNet18 backbone and five-fold cross-validated on scans of patients with spondyloarthritis (SpA) (n = 279), postpartum individuals (n = 71), and healthy subjects (n = 114). Independent SpA patient MRI scans (n = 243) served as test data set. Patient-level predictions were derived from aggregating quadrant-level predictions, ie, at least one positive quadrant. RESULTS: The algorithm automatically detects the SI joints with a precision of 98.4% and segments ilium/sacrum with an intersection over union of 85.6% and 67.9%, respectively. The inflammation classifier performed well in cross-validation: area under the curve (AUC) 94.5%, balanced accuracy (B-ACC) 80.5%, and F1 score 64.1%. In the test data set, AUC was 88.2%, B-ACC 72.1%, and F1 score 50.8%. On a patient level, the model achieved a B-ACC of 81.6% and 81.4% in the cross-validation and test data set, respectively. CONCLUSION: We propose a fully automated ML pipeline that enables objective and standardized evaluation of BME along the SI joints on MRI. This method has the potential to screen large numbers of patients with (suspected) SpA and is a step closer towards artificial intelligence-assisted diagnosis and follow-up.

Gut Inflammation in Axial Spondyloarthritis Patients is Characterized by a Marked Type 17 Skewed Mucosal Innate-like T Cell Signature.

OBJECTIVE: Patients with spondyloarthritis (SpA) often present with microscopic signs of gut inflammation, a risk factor for progressive disease. We investigated whether mucosal innate-like T cells are involved in dysregulated interleukin-23 (IL-23)/IL-17 responses in the gut-joint axis in SpA. METHODS: Ileal and colonic intraepithelial lymphocytes (IELs), lamina propria lymphocytes (LPLs), and paired peripheral blood mononuclear cells (PBMCs) were isolated from treatment-naive patients with nonradiographic axial SpA with (n = 11) and without (n = 14) microscopic gut inflammation and healthy controls (n = 15) undergoing ileocolonoscopy. The presence of gut inflammation was assessed histopathologically. Immunophenotyping of innate-like T cells and conventional T cells was performed using intracellular flow cytometry. Unsupervised clustering analysis was done by FlowSOM technology. Serum IL-17A levels were measured via Luminex. RESULTS: Microscopic gut inflammation in nonradiographic axial SpA was characterized by increased ileal intraepithelial γδ-hi T cells, a γδ-T cell subset with elevated γδ-T cell receptor expression. γδ-hi T cells were also increased in PBMCs of patients with nonradiographic axial SpA versus healthy controls and were strongly associated with Ankylosing Spondylitis Disease Activity Score. The abundance of mucosal-associated invariant T cells and invariant natural killer T cells was unaltered. Innate-like T cells in the inflamed gut showed increased RORγt, IL-17A, and IL-22 levels with loss of T-bet, a signature that was less pronounced in conventional T cells. Presence of gut inflammation was associated with higher serum IL-17A levels. In patients treated with tumor necrosis factor blockade, the proportion of γδ-hi cells and RORγt expression in blood was completely restored. CONCLUSION: Intestinal innate-like T cells display marked type 17 skewing in the inflamed gut mucosa of patients with nonradiographic axial SpA. γδ-hi T cells are linked to intestinal inflammation and disease activity in SpA.

Distinct immune modulatory roles of regulatory T cells in gut versus joint inflammation in TNF-driven spondyloarthritis.

OBJECTIVES: Gut and joint inflammation commonly co-occur in spondyloarthritis (SpA) which strongly restricts therapeutic modalities. The immunobiology underlying differences between gut and joint immune regulation, however, is poorly understood. We therefore assessed the immunoregulatory role of CD4+FOXP3+ regulatory T (Treg) cells in a model of Crohn's-like ileitis and concomitant arthritis. METHODS: RNA-sequencing and flow cytometry was performed on inflamed gut and joint samples and tissue-derived Tregs from tumour necrosis factor (TNF)∆ARE mice. In situ hybridisation of TNF and its receptors (TNFR) was applied to human SpA gut biopsies. Soluble TNFR (sTNFR) levels were measured in serum of mice and patients with SpA and controls. Treg function was explored by in vitro cocultures and in vivo by conditional Treg depletion. RESULTS: Chronic TNF exposure induced several TNF superfamily (TNFSF) members (4-1BBL, TWEAK and TRAIL) in synovium and ileum in a site-specific manner. Elevated TNFR2 messenger RNA levels were noted in TNF∆ARE/+ mice leading to increased sTNFR2 release. Likewise, sTNFR2 levels were higher in patients with SpA with gut inflammation and distinct from inflammatory and healthy controls. Tregs accumulated at both gut and joints of TNF∆ARE mice, yet their TNFR2 expression and suppressive function was significantly lower in synovium versus ileum. In line herewith, synovial and intestinal Tregs displayed a distinct transcriptional profile with tissue-restricted TNFSF receptor and p38MAPK gene expression. CONCLUSIONS: These data point to profound differences in immune-regulation between Crohn's ileitis and peripheral arthritis. Whereas Tregs control ileitis they fail to dampen joint inflammation. Synovial resident Tregs are particularly maladapted to chronic TNF exposure.

Extent of axial damage in psoriatic arthritis and spondyloarthritis: comparative data from the BEPAS and (Be-)GIANT multicentre cohorts.

BACKGROUND: To examine radiographic axial damage of the sacroiliac joints and spine in patients with psoriatic arthritis (PsA) and spondyloarthritis (SpA) in private and academic Belgian practices. METHODS: Patients with PsA with clinical diagnosis of PsA and fulfilling the Classification Criteria for Psoriatic Arthritis from the prospective Belgian Epidemiological Psoriatic Arthritis Study and patients with SpA fulfilling the Assessment of SpondyloArthritis international Society classification criteria for SpA originate from the Ghent and BelGian Inflammatory Arthritis and spoNdylitis cohorTs were included in this study. Baseline pelvic and spinal radiographs were analysed by two calibrated readers. Blinded for the origin of the cohort or clinical data readers assessed the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) and modified New York criteria on spinal and pelvic radiographs, respectively. Data were compared between both patient groups. RESULTS: Of the 525 patients included (312 PsA and 213 SpA), most patients showed normal spinal radiographs: 87.5% of the patients with PsA and 92.0% of the patients with SpA. Patients with SpA with spinal damage show higher mSASSS than the patients with PsA (p<0.05). In patients with PsA, cervical spine is more often affected; 24/33 patients (72.7%) compared with lumbar spine 11/33 (33.3%). While in patients with SpA, syndesmophyte location was more evenly distributed; cervical 9/14 (64.3%) and lumbar 10/14 (71.4%). CONCLUSION: Minimal radiographic spinal damage was observed in Belgian patients with PsA or SpA. Patients with SpA tend to have higher mSASSS values and more syndesmophytes compared with PsA. Syndesmophytes were more often located in the cervical spine of patients with PsA, while the location was equally distributed in axSpA.

Anti-CD74 IgA antibodies show diagnostic potential for axial spondyloarthritis but are not associated with microscopic gut inflammation.

OBJECTIVES: Gut inflammation commonly occurs in axial SpA (axSpA), and is linked to disease activity and outcome. Given the role of IgA in mucosal immunity, we explored the association between anti-CD74 IgA antibodies, gut inflammation and axSpA. METHODS: Anti-CD74 IgA was measured by ELISA in serum samples of axSpA patients, fulfilling the 2009 Assessment of SpondyloArthritis international Society classification criteria. A group of fibromyalgia (FM) and RA patients served as non-inflammatory and inflammatory controls. Newly diagnosed axSpA patients underwent ileocolonoscopy; mucosal biopsies were histopathologically assessed as normal, acute or chronically inflamed. Optimal anti-CD74 IgA cut-off values were determined with a receiver operating characteristics curve. RESULTS: axSpA patients (n = 281) showed higher anti-CD74 IgA levels [mean (s.d.) 18.8 (12.4) U/ml] compared with 100 FM patients [10.9 (5.0) U/ml, P < 0.001] and 34 RA patients [13.7 (9.6) U/ml, P = 0.02]. The area under the receiver operating characteristics curve for diagnosis (axSpA vs FM) was 0.70, providing a sensitivity of 60% and specificity of 87% (cut-off 15 U/ml). Antibody concentrations were not significantly different between axSpA patients with (n = 40) and without (n = 69) gut inflammation (P = 0.83), yielding an area under the receiver operating characteristics curve of 0.51. Anti-CD74 IgA levels were not associated with degree of bone marrow oedema on MRI of the sacroiliac joints, CRP or any other disease-specific feature such as the use of NSAIDs or biological treatment. CONCLUSION: Serum anti-CD74 IgA is a potentially useful diagnostic biomarker for axSpA. However, antibody levels do not correlate with any phenotypical feature, including microscopic gut inflammation, suggesting this to be a disease-specific rather than an inflammatory marker.

Progressive Increase in Sacroiliac Joint and Spinal Lesions Detected on Magnetic Resonance Imaging in Healthy Individuals in Relation to Age.

OBJECTIVE: Magnetic resonance imaging (MRI) plays a pivotal role in spondyloarthritis (SpA) diagnosis. However, a detailed description of MRI findings of the sacroiliac (SI) joints and spine in healthy individuals is currently lacking. This study was undertaken to evaluate the occurrence of MRI-detected SI joint and spinal lesions in healthy individuals in relation to age. METHODS: Ninety-five healthy subjects (ages 20-49 years) underwent MRI of the SI joints and spine. Bone marrow edema (BME) and structural lesions of the SI joints were scored using the Spondyloarthritis Research Consortium of Canada (SPARCC) method. Spinal inflammatory and structural lesions were evaluated using the SPARCC MRI spine inflammation index and the Canada-Denmark MRI scoring system, respectively. Fulfillment of the Assessment of SpondyloArthritis international Society definition of a positive MRI for sacroiliitis/spondylitis was reviewed. Findings were compared to MRIs of axial SpA patients from the Belgian Inflammatory Arthritis and Spondylitis cohort. RESULTS: Of the subjects ≥30 years old, 17.2% fulfilled the definition of a positive MRI for sacroiliitis, but this occurred rarely in younger subjects. SI joint erosions (20.0%) and fat metaplasia (13.7%) were detected across all age groups. Erosions were more frequently visualized in subjects ages ≥40 years (39.3%). Spinal BME (35.7%) and fat metaplasia (28.6%) were common in subjects older than 40 years. Nonetheless, only 1 subject had ≥3 corner inflammatory lesions. SI joint and spinal SPARCC scores and total structural lesions scores increased progressively with age. CONCLUSION: Contrary to what is commonly believed, structural MRI-detected SI joint lesions are frequently seen in healthy individuals. Especially in older subjects, the high occurrence of inflammatory and structural MRI-detected lesions impacts their specificity for SpA, which has important implications for the interpretation of MRIs in patients with a clinical suspicion of SpA.

The Value of Magnetic Resonance Imaging for Assessing Disease Extent and Prediction of Relapse in Early Peripheral Spondyloarthritis.

OBJECTIVE: This study was undertaken to assess the inflammatory burden in peripheral spondyloarthritis (SpA) by magnetic resonance imaging (MRI) of the legs in an early remission-induction strategy study of tumor necrosis factor (TNF) blockade. Furthermore, we sought to determine the value of MRI to predict disease relapse versus sustained remission after treatment discontinuation. METHODS: Thirty-two patients with early peripheral SpA with involvement of the legs determined on clinical examination and confirmed by ultrasonography (US) participated in a remission-induction trial of a TNF inhibitor (TNFi). Patients underwent MRI of the joints and entheses of the legs at baseline and at clinical remission, after which TNFi treatment was withdrawn. Images were evaluated for joint effusion, joint osteitis, entheseal soft tissue inflammation, and entheseal osteitis. RESULTS: Joint effusion and enthesitis on clinical examination and US correlated well with MRI abnormalities. In addition, a substantial amount of subclinical involvement was seen on MRI, mainly in the ankle joints and heel entheses. Inflammation scores were markedly lower in the subclinically involved joints and entheses versus those that were clinically involved (P values ranged from 0.01 to <0.001). Inflammatory load on MRI decreased significantly upon TNFi treatment (P < 0.001). Whereas 80% of the joints that were clinically involved at baseline showed no effusion on remission MRI, 2 of 3 entheses involved at baseline showed residual inflammation. In addition, patients who experienced a relapse after treatment discontinuation displayed more entheseal soft tissue inflammation on remission MRI compared to those who maintained drug-free remission (P = 0.028). CONCLUSION: Our findings delineate a differential response of synovitis and enthesitis, with enthesitis on MRI being less responsive to TNFi treatment. Furthermore, residual entheseal inflammation might be indicative of the need for continuous therapy.

TNFi-induced sustained clinical remission in peripheral spondyloarthritis patients cannot be maintained with a step-down strategy based on methotrexate.

OBJECTIVES: Treatment with golimumab monotherapy in early peripheral SpA (pSpA) results in higher rates of clinical remission compared with treatment in more longstanding disease. When reaching remission, treat-to-target recommendations suggest tapering of treatment. We therefore explored whether addition of MTX would permit discontinuation of golimumab in patients with pSpA in sustained clinical remission. METHODS: After a 2-year extension phase with golimumab treatment, patients with pSpA reaching clinical remission in the CRESPA trial were offered a tapering strategy leading to discontinuation of golimumab and replacement by MTX monotherapy. Patients were prospectively followed to assess the rate of sustained biologic-free clinical remission. In case of relapse of arthritis, enthesitis or dactylitis under MTX monotherapy, golimumab was restarted. RESULTS: Of the original 60 pSpA patients, 25 entered the step-down strategy. Currently, only 4 patients (16%) are still in sustained remission under MTX monotherapy. In 21 patients (84%), golimumab was reinstalled because of relapse of disease activity (n = 19) or development of adverse events related to MTX (n = 2). Restarting golimumab treatment promptly restored clinical remission in all patients within 12 weeks. CONCLUSION: In patients with early pSpA achieving clinical remission after 2 years of golimumab treatment, step-down therapy to monotherapy with MTX led to high rates of clinical relapse. This underscores the overall weak efficacy of MTX in maintaining clinical remission in pSpA. TRIAL REGISTRATION: ClinicalTrials.gov, https://clinicaltrials.gov, NCT01426815.

Axial involvement in patients with early peripheral spondyloarthritis: a prospective MRI study of sacroiliac joints and spine.

OBJECTIVES: To assess axial involvement on MRI in early peripheral spondyloarthritis (pSpA) and to evaluate whether axial inflammation predicts relapse on treatment withdrawal. METHODS: Fifty-six patients with early, active, newly diagnosed pSpA underwent MRI of the sacroiliac joints (SIJs) and spine prior to golimumab initiation. At sustained clinical remission of pSpA, treatment was withdrawn and a second MRI was performed. Bone marrow oedema (BME) was scored by three readers according to the Spondyloarthritis Research Consortium of Canada (SPARCC) method. Scores were compared with an axial spondyloarthritis cohort (Belgian Arthritis and Spondylitis cohort). Structural lesions were assessed using a similar method. Furthermore, fulfilment of the Assessment of Spondyloarthritis International Society (ASAS) definition of a positive MRI for sacroiliitis was assessed. Spinal images were evaluated for BME and structural lesions using the Canada-Denmark MRI spine scoring system by two readers. RESULTS: Thirty-six per cent showed SIJ BME at baseline, all fulfilling the ASAS definition of sacroiliitis. No association with back pain was found. Twenty-one per cent displayed SIJ structural lesions. Spinal BME was limited: the median inflammation scores were low and no patients had ≥5 inflammatory corner lesions. On clinical remission, a significant decrease in SIJ SPARCC scores was detected. On clinical remission, no significant differences in SIJ SPARCC scores were noted between patients relapsing and those maintaining remission after treatment discontinuation. CONCLUSION: In patients with early pSpA, a surprisingly high prevalence of sacroiliitis on MRI was observed; SPARCC scores decreased significantly on tumour necrosis factor inhibition. Residual inflammation on MRI was not predictive of relapse of peripheral manifestations. No relevant inflammatory spinal involvement was detected. Collectively, our findings suggest a higher inflammatory burden in patients with early pSpA than anticipated.

Low Incidence of Inflammatory Bowel Disease Adverse Events in Adalimumab Clinical Trials Across Nine Different Diseases.

OBJECTIVE: Adalimumab is approved for treatment of Crohn's disease and ulcerative colitis. Thus, we postulated that exacerbation or new-onset of inflammatory bowel disease (IBD) would be rare events in patients treated with adalimumab for non-IBD indications. The objective was to evaluate the incidence of IBD adverse events (AEs) across adalimumab trials. METHODS: IBD AE rates in 75 adalimumab clinical trials in rheumatoid arthritis, polyarticular juvenile idiopathic arthritis, pediatric enthesitis-related arthritis, uveitis, hidradenitis suppurativa, adult and pediatric psoriasis, psoriatic arthritis, nonpsoriatic arthritis peripheral spondyloarthritis (SpA), axial SpA, including nonradiographic axial SpA, and ankylosing spondylitis, were analyzed. Search terms for IBD AEs (new onset or worsening/flare) included IBD, ulcerative colitis, Crohn's disease, and ulcerative proctitis. RESULTS: This analysis included 24,114 patients, representing 36,508 patient-years of adalimumab exposure. The overall rate of IBD AEs in adalimumab-treated patients was 0.1 (95% confidence interval [95% CI] 0.1-0.2)/100 patient-years (41 events), ranging from no events (psoriatic arthritis, uveitis, and pediatric trials) to 0.8 (95% CI 0.2-2.2)/100 patient-years in peripheral SpA. The rate of IBD in axial SpA was 0.6 (95% CI 0.4-1.0)/100 patient-years. During placebo-controlled trials, the overall IBD rate was 0.1 (95% CI 0.0-0.3)/100 patient-years for adalimumab groups (3 events in 6,781 patients; 2,752 patient-years of exposure) and 0.1 (95% CI 0.0-0.4)/100 patient-years for placebo groups (1 event in 3,493 patients; 1,246 patient-years of exposure). IBD rates in axial SpA were 0.5 (95% CI 0.1-1.4)/100 patient-years for adalimumab and 0.6 (95% CI 0.0-3.1)/100 patient-years for placebo. CONCLUSION: The rates of IBD AEs in adalimumab clinical trials were generally low across the evaluated diseases, including axial SpA; all events occurred in adult patients.

Peripheral spondyloarthritis: a neglected entity-state of the art.

Peripheral spondyloarthritis (pSpA) refers to a number of seemingly different spondyloarthritis subsets in which psoriatic arthritis (PsA) is the most common, and symptoms of arthritis, enthesitis or dactylitis predominate the clinical presentation. Although formal classification criteria for pSpA have been introduced in 2011, only a minority of epidemiological and clinical studies addressed this clinical entity as a separate disease. Moreover, research on outcome measures and treatment modalities in pSpA has been mainly focused on PsA. Subsequently, all biological treatments are off-label in patients with non-psoriatic pSpA. Its neglected status has important implications for clinical practice since the emerging group of early-diagnosed non-psoriatic pSpA patients remains poorly characterised and lacks specific treatment recommendations. This review summarises what is currently known regarding pSpA in terms of epidemiology, clinical presentation, diagnosis and therapeutic approach.

High prevalence of spondyloarthritis-like MRI lesions in postpartum women: a prospective analysis in relation to maternal, child and birth characteristics.

OBJECTIVES: Bone marrow oedema (BMO) on MRI of sacroiliac joints (SIJs) represents a hallmark of axial spondyloarthritis (SpA), yet such lesions may also occur under augmented mechanical stress in healthy subjects. We therefore sought to delineate the relationship between pregnancy/delivery and pelvic stress through a prospective study with repeated MRI. Results were matched with maternal, child and birth characteristics. METHODS: Thirty-five women underwent a baseline MRI-SIJ within the first 10 days after giving birth. MRI was repeated after 6 months and, if positive for sacroiliitis according to the Assessment of SpondyloArthritis International Society (ASAS) definition, after 12 months. BMO and structural lesions were scored by three trained readers using the Spondyloarthritis Research Consortium of Canada (SPARCC) method. RESULTS: Seventy-seven per cent of the subjects (27/35) displayed sacroiliac BMO immediately postpartum, 60% fulfilled the ASAS definition of a positive MRI. After 6 months, 46% of the subjects (15/33) still showed BMO, representing 15% (5/33) with a positive MRI. After 12 months, MRI was still positive in 12% of the subjects (4/33). Few structural lesions were detected. Intriguingly, in this study, the presence of BMO was related to a shorter duration of labour and lack of epidural anaesthesia. CONCLUSION: A surprisingly high prevalence of sacroiliac BMO occurs in women immediately postpartum. Our data reveal a need for a waiting period of at least 6 months to perform an MRI-SIJ in postpartum women with back pain. This study also underscores the importance of interpreting MRI-SIJ findings in the appropriate clinical context.