RMD Open最新文献

Objectives: To assess the effect of treatment on haemostatic parameters in patients with early rheumatoid arthritis (RA).

Methods: Patients with newly diagnosed RA started methotrexate and were randomised to additional conventional treatment, certolizumab pegol, abatacept or tocilizumab. Several biomarkers for haemostasis were analysed including parameters of the two global haemostatic assays-overall haemostatic potential (OHP) and endogenous thrombin potential (ETP), as well as single haemostatic factors-fibrinogen, prothrombin fragment 1+2 (F1+2), D-dimer, thrombin activatable fibrinolysis inhibitor (TAFI) and clot lysis time (CLT) in 24 patients at baseline, 12 and 24 weeks after the start of the treatment.

Results: At baseline, patients had elevated levels of the following biomarkers compared with reference values: fibrinogen, F1+2, D-dimer and parameters of the two global haemostatic assays, that is, ETP and OHP. After 24 weeks we observed a significant reduction in F1+2 (p<0.01), fibrinogen (p<0.01), D-dimer (p<0.01), OHP (p<0.01), ETP (p<0.01), CLT (p<0.01), TAFI (p<0.01) and an increase of OFP (p<0.01). Tocilizumab treatment resulted in the most significant reduction of global haemostatic assays after 24 weeks, that is, a reduction of OHP 73% (p<0.01) compared with certolizumab pegol arm 32% (p<0.01), abatacept arm 24% (p=0.25) or conventional treatment arm 7% (p=0.66).

Conclusion: Newly diagnosed RA patients have enhanced coagulation activation and impaired fibrinolysis as demonstrated by our results. Effective antirheumatic treatments during the first 24 weeks after diagnosis improved this haemostatic imbalance, with prominent effects of biological drugs and especially tocilizumab, compared with conventional treatment.

Objective: Excess cholesterol loading on arterial macrophages is linked to foam cell formation, atherosclerosis and cardiovascular risk in rheumatoid arthritis (RA). However, the effect of changes in cholesterol loading on coronary plaque trajectory and the impact of RA therapies on this relationship are unknown. We investigated the association between variations in cholesterol loading capacity (CLC) over time and atherosclerosis progression.

Methods: In a prospective observational cohort study, coronary CT angiography evaluated atherosclerosis (non-calcified, partially calcified or fully calcified plaques and coronary artery calcium (CAC) score) in 100 patients with RA without cardiovascular disease at baseline and 6.9±0.4 years later. The presence of ≥5 plaques and lesions rendering >50% stenosis was considered an extensive and obstructive disease, respectively. Serum CLC was measured on human THP-1 monocyte-derived macrophages with a fluorometric assay.

Results: Mean CLC change (follow-up CLC-baseline CLC) was 1.54 (SD 3.69) μg cholesterol/mg protein. In models adjusting for atherosclerotic cardiovascular disease risk score, baseline plaque and other relevant covariates, CLC change (per SD unit increase) is associated with a higher likelihood of progression of non-calcified (OR 2.55, 95% CI 1.22 to 5.35), fully calcified plaque (OR 3.10, 95% CI 1.67 to 5.76), CAC (OR 1.80, 95% CI 1.18 to 2.74) and new extensive or obstructive disease (OR 2.43, 95% CI 1.11 to 5.34). Exposure to prednisone unfavourably influenced, while biologics and statins favourably affected the relationship between CLC change and atherosclerosis progression (all p-for-interactions ≤0.048).

Conclusion: CLC change is associated with atherosclerosis progression in a dose-dependent manner, including lipid-rich non-calcified plaques and extensive or obstructive disease that yield the greatest cardiovascular risk.

Interstitial lung disease (ILD) associated with rheumatoid arthritis or with connective tissue diseases such as systemic sclerosis can be collectively named systemic autoimmune rheumatic disease-associated ILDs (SARD-ILDs) or rheumatic musculoskeletal disorder-associated ILDs. SARD-ILDs result in substantial morbidity and mortality, and there is a high medical need for effective therapies that target both fibrotic and inflammatory pathways in SARD-ILD. Phosphodiesterase 4 (PDE4) hydrolyses cyclic AMP, which regulates multiple pathways involved in inflammatory processes. PDE4 is overexpressed in peripheral blood monocytes from patients with inflammatory diseases. However, clinical data on pan-PDE4 inhibition in fibrotic conditions are lacking. The PDE4B subtype is highly expressed in the brain, lungs, heart, skeletal muscle and immune cells. As such, inhibition of PDE4B may be a novel approach for fibrosing ILDs such as idiopathic pulmonary fibrosis (IPF) and SARD-ILD. Preclinical data for PDE4B inhibition have provided initial evidence of both anti-inflammatory and antifibrotic activity, with reduced potential for gastrointestinal toxicity compared with pan-PDE4 inhibitors. In a proof-of-concept phase II trial in patients with IPF, nerandomilast (BI 1015550), the only PDE4B inhibitor currently in clinical development, prevented a decline in lung function over 12 weeks compared with placebo. The potential clinical benefit of PDE4B inhibition is now being investigated in the phase III setting, with two trials evaluating nerandomilast in patients with IPF (FIBRONEER-IPF) or with progressive pulmonary fibrosis other than IPF (FIBRONEER-ILD). Here, we review the preclinical and clinical data that provide rationale for PDE4B inhibition as a treatment strategy in patients with SARD-ILD.

Purpose: To examine whether incorporating anatomy-centred deep learning can improve generalisability and enable prediction of disease progression.

Methods: This retrospective multicentre study included conventional pelvic radiographs of four different patient cohorts focusing on axial spondyloarthritis collected at university and community hospitals. The first cohort, which consisted of 1483 radiographs, was split into training (n=1261) and validation (n=222) sets. The other cohorts comprising 436, 340 and 163 patients, respectively, were used as independent test datasets. For the second cohort, follow-up data of 311 patients was used to examine progression prediction capabilities. Two neural networks were trained, one on images cropped to the bounding box of the sacroiliac joints (anatomy-centred) and the other one on full radiographs. The performance of the models was compared using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity and specificity.

Results: On the three test datasets, the standard model achieved AUC scores of 0.853, 0.817, 0.947, with an accuracy of 0.770, 0.724, 0.850. Whereas the anatomy-centred model achieved AUC scores of 0.899, 0.846, 0.957, with an accuracy of 0.821, 0.744, 0.906, respectively. The patients who were identified as high risk by the anatomy-centred model had an OR of 2.16 (95% CI 1.19, 3.86) for having progression of radiographic sacroiliitis within 2 years.

Conclusion: Anatomy-centred deep learning can improve the generalisability of models in detecting radiographic sacroiliitis. The model is published as fully open source alongside this study.

Introduction: The study aimed to identify and describe disease activity trajectories over 10 years in patients with recent-onset axial spondyloarthritis (axSpA) and determine their impact on long-term outcomes.

Methods: This prospective, multicentre study (Devenir des Spondylarthropathies Indifférenciées Récentes cohort, ClinicalTrials.gov NCT) followed patients with early axSpA for 10 years. Only patients with at least three Axial Spondylitis Disease Activity Score (ASDAS) values were included. Long-term outcomes assessed were TNF inhibitors (TNFi) exposure, structural progression, function (Bath Ankylosing Spondylitis Functional Index (BASFI) and Bath Ankylosing Spondylitis Metrology Index), quality of life (36-items Short Form Survey), sick leave days and cardiovascular (CV) events. ASDAS trajectories were identified using k-means clustering. Multinomial multivariable regression estimated associations between baseline characteristics and trajectories. Long-term outcomes for each trajectory were described.

Results: Among 601 patients, five ASDAS trajectories were identified: persistent low disease activity/remission (tA), clinically important improvement (tD) and persistent moderate (tB), high (tC) or very high (tE) disease activity. Patients in tA were more likely to be male, have a university degree, have white-collar jobs, have positive HLA B27 status and have less fibromyalgia. Trajectory tE was linked to poorer function (BASFI 50/100 vs 7/100 for lower ASDAS trajectory), higher TNFi exposure (74% vs 29%) and more CV events (5.7% vs 0). Structural progression was low but comparable across trajectories, except for higher sacroiliac joint progression in tB.

Conclusion: The k-means method revealed distinct disease activity trajectories in axSpA. Higher disease activity trajectories were associated with a higher prevalence of fibromyalgia and poorer outcomes, except for structural progression, which was comparable across trajectories.

Objectives: We propose and test a framework to detect disease diagnosis using a recent large language model (LLM), Meta's Llama-3-8B, on French-language electronic health record (EHR) documents. Specifically, it focuses on detecting gout ('goutte' in French), a ubiquitous French term that has multiple meanings beyond the disease. The study compares the performance of the LLM-based framework with traditional natural language processing techniques and tests its dependence on the parameter used.

Methods: The framework was developed using a training and testing set of 700 paragraphs assessing 'gout' from a random selection of EHR documents from a tertiary university hospital in Geneva, Switzerland. All paragraphs were manually reviewed and classified by two healthcare professionals into disease (true gout) and non-disease (gold standard). The LLM's accuracy was tested using few-shot and chain-of-thought prompting and compared with a regular expression (regex)-based method, focusing on the effects of model parameters and prompt structure. The framework was further validated on 600 paragraphs assessing 'Calcium Pyrophosphate Deposition Disease (CPPD)'.

Results: The LLM-based algorithm outperformed the regex method, achieving a 92.7% (88.7%-95.4%) positive predictive value, a 96.6% (94.6%-97.8%) negative predictive value and an accuracy of 95.4% (93.6%-96.7%) for gout. In the validation set on CPPD, accuracy was 94.1% (90.2%-97.6%). The LLM framework performed well over a wide range of parameter values.

Conclusion: LLMs accurately detected disease diagnoses from EHRs, even in non-English languages. They could facilitate creating large disease registers in any language, improving disease care assessment and patient recruitment for clinical trials.

Background: The Assessment of SpondyloArthritis international Society Health Index (ASAS HI) is a novel questionnaire of global functioning for patients with axial spondyloarthritis (SpA).

Objective: The objective was to assess the construct validity, discriminatory ability and responsiveness of ASAS HI in relation to patient-reported outcome measures (PROMs), MRI and radiography.

Methods: Data from two longitudinal studies with tumour necrosis factor inhibitor (TNFi) initiation (novel MRI And biomarkers in Golimumab-treated patients with axial spondyloarthritis (MANGO): n=45) respectively tapering (Dose adjustment of Biological treatment in patients with SpA (DOBIS): n=106) were used. Analyses included a wide panel of PROMs, MRI and radiography scores of the spine and sacroiliac joints (SIJs).

Results: In the MANGO study, 30 (68%) patients were clinical responders at week 16. In the DOBIS study, 105 (99%) patients flared after mean (SD; min-max) 31 (17; 2.7-81) weeks. After initiation of TNF inhibitor in MANGO, ASAS HI significantly decreased from baseline to week 4, 16 and 52. In DOBIS, ASAS significantly increased from baseline to the flare visit and significantly decreased from the flare visit to week 96. In multivariate regression models, ASAS HI was independently associated with Spondyloarthritis Research Consortium of Canada MRI SIJ Inflammation score, Canada-Denmark MRI Spine Inflammation score, EuroQol, Ankylosing Spondylitis Disease Activity Index, Bath Ankylosing Spondylitis Disease Activity Index and Patient Global. Patients stratified according to ASAS HI health status groups (good, moderate, poor) at baseline and change categories (absolute and percentage change) from baseline to week 16/flare showed good discriminatory ability for almost all outcome variables (p≤0.001). ASAS HI had a large responsiveness in MANGO (standardised response mean (SRM)=-1.3, effect size (ES)=-1.7) and moderate responsiveness in DOBIS (SRM=0.7, ES=0.6).

Conclusion: ASAS HI showed good construct validity, discriminatory ability and responsiveness.

Trial registration number: ClinicalTrials.gov: NCT02011386.

It is known that 25%-30% of individuals with cutaneous psoriasis (PsC) will develop psoriatic arthritis (PsA). To date, the reasons for the development of PsA in individuals with PsC have not been identified. Furthermore, there are considerable delays in the diagnosis and treatment of PsA, which lead to joint and bone deformation and chronic pain. It is therefore important to develop more precise diagnostic and screening tools. In this narrative review of the literature, clinical risk factors and novel molecular biomarkers (genetic markers, blood and inflammatory markers, lipid, metabolite and protein biomarkers) have been evaluated. The review included 38 publications that were reported between May 2020 and May 2024. Similar to previous reviews, nail involvement was one of the strongest clinical risk factors for the development of PsA, while molecular biomarkers did not provide a clear and robust differentiation between PsC and PsA groups. The seemingly poor performance of molecular markers may be largely attributed to small study populations and heterogeneity in study designs. Data and sample sharing in large consortia such as HIPPOCRATES (Health initiatives in Psoriasis and PsOriatic arthritis ConsoRTium European States) could help to overcome the limitations of small studies and enable the development of more robust diagnostic and screening tools for PsA.