First Biologic Drug Persistence in Patients With Ankylosing Spondylitis and Nonradiographic Axial Spondyloarthritis

Axial spondyloarthritis (axSpA) includes ankylosing spondylitis (AS) and nonradiographic axial spondyloarthritis (nr-axSpA). Both are managed with biologic therapies; however, there is a lack of evidence for nr-axSpA therapies. The primary objective was to compare persistence to first biologic between AS and nr-axSpA patients in a longitudinal cohort. Secondary objectives were to examine disease activity markers over time and to evaluate predictors for drug discontinuation. METHODS Data were obtained from persons enrolled in the SpondyloArthritis Research Consortium of Canada registry between 2003 and 2018. Kaplan-Meier curves were constructed from the time of biologic initiation until discontinuation and compared using the log-rank test. Subanalyses were performed according to calendar year and disease activity. Cox proportional hazards models were used to identify factors associated with discontinuation. RESULTS We identified 385 biologic-naive persons. Overall, the 349 AS participants had longer persistence to their first biologic than the 36 nr-axSpA subjects (p < 0.01). The Bath Ankylosing Spondylitis Disease Activity Index and Bath Ankylosing Spondylitis Functional Index decreased by 2.3 points (95% confidence interval [CI], 1.9-2.7) and 3.2 points (95% CI, 2.6-3.7), respectively, in the first year and were stable thereafter. Adjusting for sex, human leukocyte antigen B27, and smoking status, nr-axSpA patients were more likely to discontinue their biologic than AS patients (hazards ratio, 1.65; 95% CI, 1.03-2.62). CONCLUSIONS In this real-world study, AS patients had longer persistence to their first biologic compared with nr-axSpA, with disease subtype being the most significant predictor of treatment persistence. Future studies should be targeted at assessing long-term clinical outcome of axSpA in the real-world setting.