PO:06:093 | Clinical and ultrasound phenotyping comparison between Difficult-To-Treat Rheumatoid Arthritis (D2TRA) and non-D2TRA patients: experience from a monocentric cohort

Silvia Perrone¹, Claudio Cruciani¹, Silvia Mascolo¹, Elisa Bellis¹, Gloria Crepaldi¹, Valeria Data¹, Claudia Garulli¹, Claudia Lomater¹, Elena Marucco¹, Marta Saracco¹, Mariele Gatto¹, Annamaria Iagnocco¹. | ¹Academic Rheumatology Centre, Department of Clinical and Biological Sciences, University of Turin, AO Mauriziano, Turin, Italy.

Background. To investigate clinical and ultrasound (US) findings as potential predictors of difficult-to-treat rheumatoid arthritis (D2TRA).1,2

Methods. Patients diagnosed with D2TRA assessed at our clinic between 2020 and 2024 were enrolled in the study and compared with a group of consecutive non-D2TRA patients. Demographic, US and clinical data were retrospectively collected. The first US available before developing D2TRA was analyzed. All US assessments were performed by four expert rheumatologists from our center using an Esaote MyLabX8 US machine equipped with a high-frequency (15–24 MHz) linear probe. US evaluation included the detection of synovitis, tenosynovitis and bone erosions. Assessed joints encompassed wrists, metacarpophalangeal (MCP), proximal interphalangeal (PIP), shoulders, knees and ankles. Regarding synovitis, GS and PD were considered in accordance with EULAR and OMERACT criteria.3,4,5 GLOESS grading system was also applied.6 Logistic and Cox regression models adjusted for baseline confounders was used for multivariable analysis.

Results. One hundred thirty-six patients were recruited, median (±IQR) follow-up 18.75 (11.42-25.92) years, 80 of them fulfilling EULAR criteria for D2TRA. No significant differences were found between D2TRA and non-D2TRA groups regarding baseline demographics and comorbidities except for fibromyalgia, which was significantly more frequent in D2T cohort (Table 1). On survival univariate analysis, exposure to tsDMARDs was associated to longer time-to-D2T development (HR [CI] : 0.51, [0.26 - 0.99], p=0.046) (Figure 1). 136 US were included in the study. At the time of US examination, patients further-to-be D2T displayed higher DAS28 scores (median [IQR]: 4.30 [3.67-5.40] vs 2.70 [2.02-4.41], p<0.01). A significantly higher prevalence of wrist synovitis (n [%]: 62 [77.5%] vs. 31 [57.4%], p=0.023) and higher wrist GLOESS scores (mean [sd]: 2.67 [1.91] vs 1.77 [1.94]; p=0.007) were observed in furhter-to-be D2T patients (Figure 2). At the multivariable logistic analysis, wrist synovitis (OR 2.79, 95%CI 1.27-6.33, p=0.01) and fybromialgia (OR 4.14, 95%CI 1.22-19.3, p=0.03) independently predicted D2TRA.

Conclusions. In our cohort, fybromialgia and US-documented wrist synovitis independetly predicted D2TRA. Our results highlight the critical role of targeted US beside clinical assessment for risk stratification across RA patients. REFERENCES 1. Bellis et al. The role of musculoskeletal ultrasound in difficult-to-treat RA: Insights from a systematic literature review. Autoimmun Rev. 2025 2. Nagy G et al. EULAR definition of difficult-to-treat rheumatoid arthritis. Ann Rheum Dis. 2021 3. Möller I et al. The 2017 EULAR standardised procedures for ultrasound imaging in rheumatology. Ann Rheum Dis 2017. 4. Costantino et al. EULAR recommendations for the reporting of ultrasound studies in rheumatic and musculoskeletal diseases. Ann Rheum Dis 2021. 5. Bruyn et al. OMERACT Definitions for Ultrasonographic Pathologies and Elementary Lesions of Rheumatic Disorders. J. Rheumatol 2019. 6. Naredo E et al.The OMERACT ultrasound task force - status and perspectives. J Rheumatol 2011.