Comparison of Biologic Disease-Modifying Antirheumatic Drug Therapy Persistence Between Biologics Among Rheumatoid Arthritis Patients Switching from Another Biologic

INTRODUCTION To compare biologic disease-modifying antirheumatic drug therapy persistence between biologics among patients with rheumatoid arthritis (RA) who previously used ≥1 other biologic. METHODS Using a large United States administrative claims dataset, we identified adult patients with RA initiating abatacept, adalimumab, certolizumab, etanercept, golimumab, infliximab, or tocilizumab between January 1, 2010 and January 1, 2012 (initiation date = index). Patients were required to have used ≥1 other biologic before index. Outcomes were biologic persistence, defined in two alternative ways: (1) time from initiation until switching to a different biologic (time to switch) and (2) time from initiation until switching or the first occurrence of a 90-day gap in treatment with the initiated biologic (time to switch/discontinuation). Rituximab was excluded from analyses due to retreatment based on clinical evaluation, which complicates the measurement of persistence. Multivariable survival analyses compared persistence outcomes between tocilizumab and the other biologics, adjusting for patient characteristics. RESULTS The sample comprised 9,782 biologic initiations; mean age 54 years and 82% female. Compared with tocilizumab, the hazards of switching biologic therapy were significantly higher for abatacept [hazard ratio (HR) = 1.19, P = 0.041], adalimumab (HR = 1.39, P < 0.001), certolizumab (HR = 1.39, P < 0.001), golimumab (HR = 1.20, P = 0.047), and infliximab (HR = 1.33, P < 0.001), but not significantly different for etanercept (HR = 1.19, P = 0.095); the hazards of switching/discontinuing biologic therapy were significantly higher for adalimumab (HR = 1.16, P = 0.014) and certolizumab (HR = 1.15, P < 0.012), but not significantly different for abatacept (HR = 1.08, P = 0.229), etanercept (HR = 0.97, P = 0.644), golimumab (HR = 0.99, P = 0.829), and infliximab (HR = 0.97, P = 0.721). CONCLUSIONS This is one of the first studies of biologic persistence to focus specifically on patients with RA who are not naïve to biologic treatment. Among patients with RA who previously used ≥1 other biologic, tocilizumab-treated patients had similar or significantly better biologic persistence compared with other biologics.