Background: We compared the effectiveness of abatacept (ABA) vs tocilizumab (TCA) in tumor necrosis factor inhibitor (TNFi) experienced patients.
Methods: We identified rheumatoid arthritis (RA) patients from a large observational US cohort (1 January 2010-31 May 2014) who had discontinued at least one TNFi and initiated ABA or TCZ in moderate or high disease activity based on the Clinical Disease Activity Index (CDAI) and had no prior exposure to the comparator drug. Using propensity score matching (1:1) stratified by prior TNF use (1 TNFi vs ≥2 TNFis), effectiveness at 6 months after initiation was evaluated. Mean change in CDAI over 6 months following initiation was the primary outcome, with secondary outcomes of achievement of low disease activity/remission (CDAI ≤ 10) and mean change in modified Health Assessment Questionnaire (mHAQ) score.
Results: The 264 pairs of propensity score-matched ABA and TCZ initiators were well matched with no substantial differences in the baseline characteristics, defined as standardized differences >0.1 in the stratification. Both treatment groups had similar mean change in CDAI at 6 months (-11.3 in ABA vs -9.9 in TCZ; mean difference -1.27, 95% CI -3.65, 1.11). Similar proportions of both treatment groups achieved low disease activity/remission (adjusted odds ratio for ABA vs TCZ 0.99, 95% CI 0.69, 1.43). Mean change in mHAQ was -0.12 in ABA initiators vs -0.11 in TCZ initiations (mean difference -0.01, 95% CI -0.09, 0.06).
Conclusions: Patients receiving either ABA or TCZ had substantial improvement in clinical disease activity. In this propensity score-matched sample, similar outcomes were observed for both treatment cohorts.
Keywords: Abatacept; Disease-modifying anti-rheumatic drugs (biologic); Rheumatoid arthritis; Tocilizumab; Treatment.