POS0936 DISEASE TRAJECTORIES OF AXIAL SPONDYLOARTHRITIS PATIENTS INITIATED ON BIOLOGIC DMARDs: PRELIMINARY ANALYSIS FROM A REAL-WORLD COHORT

Abstract

BackgroundThe number of biologic/targeted synthetic (b/ts) disease-modifying anti-rheumatic drugs (DMARDs) available for the treatment of axial spondyloarthritis (axSpA) are increasing. However, 1 in 4 axSpA patients may discontinue their first bDMARD within the first 12-months (1). A greater understanding of real-world axSpA disease/treatment trajectories pre- and post-bDMARD initiation is needed to inform optimal treatment for patients.ObjectivesTo explore the use of biologic therapies in patients living with axSpA, and estimate the natural progression of disease and patient reported measures pre- and post- initiation of first bDMARD.MethodsData was extracted from the Bath SpA Research Biobank for all bio-naïve axSpA patients initiated on bDMARDs between 7th March 2000 and 1st February 2021. Patients were grouped based on first bDMARD. Linear regression models were used to explore changes in Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), and Bath Ankylosing Spondylitis Functional Index (BASFI) over time; from 36-months before first bDMARD initiation, to 36-months after initiation. Changes in BASDAI/BASFI were estimated using piecewise mixed-effects linear regression. Models included fixed terms for age, sex, HLA-B27 status and disease duration. Switches to a second bDMARD and reasons for switching were captured.ResultsIn total, 282 axSpA patients were included, and grouped into 3 cohorts based on first bDMARD: adalimumab (46.1%), etanercept (30.9%), other (23.1%; 6.4% certolizumab pegol, 5.0% golimumab, 5.3% infliximab, 0.7% rituximab, 5.7% secukinumab). Mean age at diagnosis was 30.7 (SD 11.3), 69.9% of patients were male, 11.0% diagnosed with non-radiographic axSpA and 85.3% HLA-B27 positive. In the 36-month period before bDMARD initiation, patients on average had a BASDAI score of 4.50 (95% CI 4.21-4.79) and BASFI of 4.02 (95% CI 3.68-4.38), with a linear and statistically significant worsening of 0.56 (95% CI 0.43-0.68) per year for the BASDAI and 0.55 (95% CI 0.41-0.69) for the BASFI (minimally clinically important difference defined as 1.1 and 0.6 for BASDAI and BASFI, respectively (2)). Following bDMARD initiation, patients indicated a statistically significant reduction in BASDAI to 3.13 (95% CI 2.82-3.45), 3.23 (95% CI 2.83-3.63) and 3.46 (95% CI 3.04-3.88) and reductions in the BASFI to 3.22 (95% CI 2.86-3.60), 3.06 (95% CI 2.61-3.53) and 3.64 (95% CI 3.16-4.12) for those treated with adalimumab, etanercept or other biologics respectively. Reduced scores were maintained over 36-months post initiation (Figure 1). Details of first and second bDMARD are outlined in Table 1. Reasons for switching were switch to biosimilar (32.4%), intolerance (25.1%), inefficacy (22.2%), malignancy (1.9%) or other (18.4%).Table 1.First and second bDMARDFirst bDMARDSecond bDMARDAdalimumabEtanerceptOtherAdalimumab59 (57.3%)22 (21.4%)22 (21.4%)Etanercept28 (40.6%)37 (53.6%)4 (5.8%)Other15 (48.4%)8 (25.8%)8 (25.8%)Figure 1.Predictive margins of BASDAI and BASFI 36-months pre- and post- bDMARD initiationA-E, Predictive margins of the BASDAI score (overall – all components, A); BASDAI Q1 fatigue (B); BASDAI average of Q2-Q4 spinal pain, joint pain, enthesitis (C); BASDAI average of Q5-Q6 morning stiffness (D); and BASFI (E).ConclusionIn our study population, there was clinically meaningful worsening of disease activity over 36-months prior to initiation of first bDMARD, despite baseline (36-months prior to first bDMARD initiation) BASDAI exceeding the current threshold for bDMARD treatment (BASDAI≥4). A clinically meaningful and maintained improvement in disease activity was reported across all cohorts in the 36-months following first bDMARD initiation.