Effectiveness of switching between TNF inhibitors in patients with axial spondyloarthritis : is the reason to switch relevant?

Abstract

Background: To investigate whether the reason to discontinue the first TNF inhibitor (TNFi) affects the response to the second TNFi in axial spondyloarthritis (axSpA). Methods: Patients with axSpA from the Rheumatic Diseases Portuguese Register (ReumaPt), who discontinued their first TNFi and started the second TNFi between June 2008 and May 2018, were included. Response was assessed by the Ankylosing Spondylitis Disease Activity Score (ASDAS) clinically important improvement (ASDAS-CII), major important improvement (ASDAS-MI), low disease activity (ASDAS-LDA), and inactive disease (ASDAS-ID). The reason for discontinuation of the first TNFi was defined, according to ASDAS-CII as primary failure (no response ≤ 6 months), secondary failure (response ≤ 6 months but lost thereafter), adverse events, and others. The association between the reason for discontinuation of the first TNFi and response to the second TNFi over time was assessed in multivariable generalized equation (GEE) models. Results: In total, 193 patients were included. The reason for discontinuation of the first TNFi did not influence the response to the second TNFi, according to the ASDAS-CII. However, a difference was found with more stringent outcomes, e.g., there was a higher likelihood to achieve ASDAS-ID with the second TNFi for patients discontinuing the first TNFi due to secondary failure (OR 7.3 [95%CI 1.9; 27.7]), adverse events (OR 9.1 [2.5; 33.3]), or other reasons (OR 7.7 [1.6; 37.9]) compared to primary failure. Conclusion: Patients with axSpA with secondary failure to their first TNFi, compared to those with primary failure, have a better response to the second TNFi according to stringent outcomes.