OP0064 FREQUENCY AND PREDICTORS OF MULTIPLE TREATMENT SWITCHING IN RHEUMATOID ARTHRITIS

BackgroundDespite the significant improvements in the field of rheumatoid arthritis (RA) treatment, a significant and troubling minority of patients remain refractory to multiple disease modifying antirheumatic drugs (DMARDs). The consequence of this ‘’difficult-to-treat’’ state is irreversible damage, risk of co-morbid conditions, and substantial loss of quality of life. Early, precise, and actionable identification of this challenging group of RA patients is crucial for optimal prevention and management.ObjectivesTo assess the frequency and to identify predictors of switching between multiple biological and targeted synthetic DMARDs (b/tsDMARDs) in RA patients in a large national register.MethodsObservational cohort study including RA patients starting a first-ever b/tsDMARD 2009-2018, based on data from the Swedish Quality Rheumatology register. Comorbidities were identified through linkage to the national Patient Register. Baseline (time of RA diagnosis) characteristics of the population were described. Three groups were investigated: A) Patients starting ≥3 treatment courses; B) Patients starting ≥4 treatment courses; and C) Patients starting ≥5 treatment courses. Predictors of multi-switching were explored using univariate and multivariable logistic regression analyses.Results23,908 RA patients were identified. Proportions of patients starting ≥3, ≥4 or ≥5 b/tsDMARDs treatment courses were 7%, 3.2% and 1.6%, during a mean (95% CI) of 3.6 (3.5-3.7), 4.3 (4.2-2.5) and 4.9 (4.7-5.2) years from RA diagnosis, respectively. In Table 1 baseline characteristics for each multi-switching group are summarized. For definition A, the following baseline univariate predictors were identified: female sex (OR=1.57, 95% CI=1.39-1.76), younger age (OR=0.96, 95% CI=0.95-0.96), positive RF (OR=1.36, 95% CI=1.22-1.53) and ACPA (OR=1.40, 95% CI=1.24-1.58), higher DAS28 (OR=1.21, 95% CI=1.15-1.26), HAQ (OR=1.46, 95%=1.33-1.61), pain (OR=1.014, 95% CI=1.012-1.017) and fatigue (OR=1.017, 95% CI=1.014-1.021). In the multivariable logistic regression model, female sex, younger age, higher HAQ, pain and fatigue at baseline were independent predictors of multiple treatment switching. Similar results were found for all three multi-switch definitions. Several comorbidities (i.e. heart failure, ischemic heart disease, malignancy, renal failure) were associated with a lower risk for multiple treatment switching, suggestive of medical contraindications for b/tsDMARDs.Table 1.Baseline (time of RA diagnosis) characteristicsAll patientsMulti-switching definitionA: ≥3 b/tsDMARDsB: ≥4 b/tsDMARDsC: ≥5 b/tsDMARDsN, %23 908 (100)1677 (7)755 (3.2)385 (1.6)Age (years), mean (SD)59.5 (15.2)50.3 (14.5)50.0 (14.6)47.6 (14.7)Sex (male) %30.7%22.5%22.4%21.6%CRP, mg/L (median, IQR)8.9 (4-22)10 (4-24)10 (4.2-25)10 (4-29)ESR (median, IQR)23 (12-40)23 (25-39)23 (11-37.5)22 (11.8-36.3)Patient global, VAS 0-100 (mean, SD)48.8 (26.8)58.4 (25.3)59.8 (24.5)60.8 (24.0)Patient pain, VAS 0-100 (mean, SD)50.2 (26.8)59.2 (25.4)59.8 (25.2)59.9 (25.2)Fatigue, VAS 0-100 (mean, SD)48.5 (28.8)61.2 (26.5)63.1 (25.9)64.6 (23.6)HAQ (mean, SD)0.97 (0.66)1.1 3(0.64)1.15 (0.64)1.13 (0.63)Swollen joint count (median, IQR)5 (2-9)6 (3-10)6 (3-10)6 (3-10)Tender joint count (median, IQR)5 (2-9)6 (3-12)6 (3-12)7 (3-11.25)Concomitant csDMARD, %52.5%50.4%51.7%51.4%Reumatoid factor positive, %63.1%69.6%69.8%67.8%ACPA positive, %67.0%73.4%74.3%72.7%DAS28-ESR (mean, SD)4.62 (1.51)5.00 (1.41)5.10 (1.37)5.12 (1.33)VAS general health physician (mean, SD)42.28 (23.02)45.30 (22.19)45.68 (20.69)51.40 (25.83)ConclusionIn this large national observational cohort, multiple treatment switching, indicative of difficult to treat RA, was observed in a significant proportion of patients, ranging between around 2 to 7% during the first 5 years from time of diagnosis. Risk factors include female gender, younger age, higher HAQ, pain and fatigue at the time of RA diagnosis, suggesting increased attention to this challenging group of patients.Disclosure of InterestsEmma Birgersson Wettersand: None declared, Daniela Di Giuseppe: None declared, Johan Askling Grant/research support from: Karolinska Institutet has entered into agreements between Karolinska Institutet (JA as principal investigator) with AbbVie, BMS, MSD, Eli Lilly, Pfizer, Roche, Samsung Bioepis, Sanofi and UCB, mainly regarding safety monitoring of anti-rheumatic therapies., Katerina Chatzidionysiou Consultant of: consultancy fees from Eli Lilly, AbbVie and Pfizer.