Pos0683 Synovial Correlates of Refractory Rheumatoid Arthritis According to Non-Inflammatory and Persistent Inflammatory Clinical Phenotypes

Background: Refractory rheumatoid arthritis (RA) is an emerging clinical condition. Clinically, two main patterns of refractoriness have been proposed. Non-inflammatory refractory RA (NIRRA) is characterised by more prevalent comorbidities, obesity, and painful syndromes, whereas persistent inflammatory RA (PIRRA) is characterised by musculoskeletal ultrasound (MSUS)-confirmed active inflammation (power doppler) or elevated inflammatory markers. However, neither MSUS nor inflammatory markers convincingly define NIRRA or PIRRA, and whether such conditions are associated with underlying synovial pathology has yet to be determined. Objectives: The aim of this study was to compare synovial inflammation, immunohistochemistry (IHC), and the three main synovial pathotypes (lymphoid, diffuse myeloid, and fibroid/pauci-immune) in patients with NIRRA, PIRRA, and non-refractory RA. Methods: In this prospective, observational cohort study conducted between January 23 and January 10, 2024, we included patients with established RA who had not responded to standard therapy. All patients had active disease (clinical disease activity index > 10). We performed MSUS-guided synovial biopsies in accordance with EULAR-OMERACT guidelines. Joint inflammation was measured with the OMERACT grading (0 to 3). The Krenn Synovial Score (KSS) was calculated for each patient. To clinically characterise NIRRA and PIRRA phenotypes, we investigated patients' perceptions of the disease through the administration of several patient-reported outcome measures, including short form-36 (SF-36), health disability questionnaire disability index (HAQ-DI), and fatigue (FACIT-F). Results: During the study period, 47 patients had US-guided synovial biopsies of their joints performed, and 38/47 had adequate sampling or no missing data and were analysed (females 68.6%, median age 59 years). The median (interquartile range) disease duration was 5 years (26.5). PIRRA was diagnosed in 5 (13.2%) of the participants, NIRRA in 10 (26.3%), and non-refractory RA in 23 (60.5%). Compared to PIRRA, NIRRA patients had lower MSUS scores, tender joints, inflammatory markers, and physician global assessment scores (Table 1). In terms of SF-36 components, patients with NIRRA and PIRRA differed, with NIRRA having lower values for most domains (8/9) and the mental component and PIRRA having the best scores for emotional domains (even more than non-refractory patients; Figure 1A). FACIT-F and HAQ-DI did not differ across groups. Body weight and comorbidities were balanced except for fibromyalgia and depression, which were present respectively in 10% of NIRRA, 8.7% and 4.3% of non-refractory patients, and in none of PIRRA patients. With regard to synovial histology, KSS was lowest in NIRRA patients. Accordingly, synovial hyperplasia, density, and infiltrates were all lowest in NIRRA patients, while density was highest in PIRRA patients (Figure 1B). PIRRA had the highest CD20+ and CD138+ scores, while NIRRA had lower scores for all IHC markers (Figure 1C). The fibroid/pauci-immune pathotype was commonest in NIRRA patients (50% vs. 0% of PIRRA vs. 17.4% of non-refractory RA patients; p=0.108 for the Fischer Exact test), whereas the lymphoid pathotype was more common in PIRRA (60%) and non-refractory (61%) RA patients (Figure 1D). Conclusion: We found a more prevalent fibroid/pauci-immune pathotype in NIRRA patients and a lymphoid pathotype in PIRRA patients. Our findings provide a mechanistic basis for categorising refractory RA into two groups. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests: None declared.