Persistence and Effectiveness of upadacitinib and TNFi for rheumatoid arthritis using pooled data from Canadian and Swiss registries

OBJECTIVES: The ROSA study evaluated real-world persistence and effectiveness of upadacitinib and tumour necrosis factor inhibitors (TNFi) in patients with rheumatoid arthritis (RA), using data from 2 Canadian registries, RHUMADATA and Ontario Best Practices Research Initiative (OBRI), and the Swiss Clinical Quality Management in Rheumatic Diseases (SCQM) registry. METHODS: This observational study analysed data from 990 patients with RA who initiated upadacitinib or TNFi between 01-Jan-2020 and 30-Jun-2023. Multiple imputation was used for baseline covariates with missing data and results were adjusted for propensity scores derived from pre-specified variables (age, disease duration, prior treatments, comorbidities, etc.). Persistence was assessed using Kaplan-Meier and Cox proportional hazard regression models. Effectiveness was evaluated using observed (non-imputed) disease activity indices over 6 and 12 months. RESULTS: Overall, 174, 216, and 600 patients were included from the RHUMADATA, OBRI, SCQM registries, respectively. Patients using upadacitinib had a lower risk of discontinuing treatment than patients using TNFi (hazard-ratio [HR] [95%CI]: 0.602 [0.469, 0.773]) and this was unaffected by previous exposure to advanced treatment. Concomitant use of any csDMARD (HR [95% CI]; upadacitinib: 0.225 [0.107, 0.471]; TNFi: 0.516 [0.303, 0.881]), and methotrexate (upadacitinib: 0.604 [0.390, 0.934]; TNFi: 0.694 [0.522, 0.925]) were associated with improved persistence for both groups. A similar proportion of patients receiving upadacitinib or TNFi achieved Clinical Disease Activity Index low disease activity or remission at 6 months (70.8% vs 61.0%; p = 0.1541) and 12 months (70.5% vs 75.2%; p = 0.4494). CONCLUSION: Canadian and Swiss patients with RA had longer persistence with upadacitinib than TNFi, independent of previous advanced treatments.