August 2023 to November 2023
Authors: Aleksandra Opinc-Rosiak and Giacomo Cafaro
RMD Open
Machine learning identifies phenotypic clusters of patients with Still’s disease
Ruscitti et al (doi.org/10.1136/rmdopen-2023-003419) analyzed a large cohort of Still’s disease patients by applying a machine learning clustering algorithm, including a large set of clinical, laboratory and imaging features gathered at the time of diagnosis. They were able to identify four main clusters. Cluster 1 mostly included younger patients with pharyngodynia, skin rash and myalgia, cluster 2 included patients with milder disease phenotype, cluster 3 was mostly composed of patients with very high levels of acute-phase reactants and ferritin, and cluster 4 mostly included older patients with higher systemic activity and mortality rate.
Risk prediction of VTE in patients with SLE
You et al (doi.org/10.1136/rmdopen-2023-003568) developed and validated a model allowing prediction of risk of venous thromboembolism at 6 months in systemic lupus erythematosus (SLE). They included 4,502 patients, and the final model included 11 variables, including age, sex, anti-phospholipid antibodies and renal involvement. The area under the curve of the model was 0.808 in an external validation cohort of 3,780 SLE subjects. The model also suggested a cut-off for annual screening in patients with increased risk (≥1.03% risk).
Investigation of non-traditional antibodies in patients with seronegative RA
Van den Beukel et al (doi.org/10.1136/rmdopen-2023-003480) investigated the presence of autoantibodies against advanced glycation end-products (anti-AGE) and malondialdehyde-acetaldehyde (anti-MAA) in rheumatoid arthritis (RA) and non-RA arthritis. They found a slightly higher prevalence of anti-AGE and anti-MAA in RA. However, the most interesting finding was that their presence in seronegative RA was associated with HLA-DRB1*03 and with radiographic progression.
Risk of major cardiovascular events in patients with RA
Two interesting studies evaluating the risk of major cardiovascular events (MACE) in real-life settings comparing RA patients treated with Janus-kinase inhibitors (JAKi), TNF inhibitors (TNFi) and other biological DMARDS (bDMARDs) have been published.
Bower et al (doi.org/10.1136/rmdopen-2023-003630) used a Swedish registry including 13,492 patients from 2016 to 2021. The authors demonstrated no increased rate of incident MACE in patients treated with JAKi compared to TNFi (HR: 0.71, 95% CI: 0.51 to 0.99, JAKi vs TNFi), even in a selected high-CV risk sub-population (HR: 0.65, 95% CI: 0.45 to 0.93, JAKi vs TNFi).
Meissner et al (doi.org/10.1136/rmdopen-2023-003489) analysed 21,218 patient-years and found no increased rate of MACE in subjects treated with JAKi compared to TNFi (IR: 0.89, 95% CI: 0.52 to 1.52). No increased rates of MACE were found, even when cDMARDs were used as a reference or when high-risk patients were evaluated.
Aleksandra Opinc-Rosiak
Aleksandra is a research and teaching assistantat the Medical University of Lodz, Poland, and a trainee in rheumatology at the USK-WAM Hospital in Lodz. Her main research interests are idiopathic inflammatory myopathies and autoantibodies. Aleksandra is a member of the Newsletter sub-committee.
Giacomo Cafaro
Giacomo is an Assistant Professor of Rheumatology and Consultant Rheumatologist at University of Perugia, Italy. His main research interests are T cell biology in autoimmune diseases, psoriatic arthritis, Sjögren’s syndrome and ultrasound in rheumatology. Giacomo is a member of the Newsletter sub-committee.