Background Among patients who have failed a prior TNF agent, there is little data to inform prescribing decisions for the next biologic.
Objectives We sought to perform a head to head comparison of the effectiveness of tocilizumab (TCZ) versus abatacept (ABA) among RA patients with previous anti-TNF exposure using data from a multi-center observational registry within the United States (Corrona, LLC).
Methods We identified RA patients between 4/2/2009 through 5/2/14 who had discontinued at least 1 anti-TNF and initiated either TCZ or ABA, with no prior use of either of these two drugs, were not in remission or low disease activity based on the Clinical Disease Activity Index (CDAI) at the time of initiation, and had follow-up at 6 months (TCZ=266; ABA=663). Propensity score (PS) matching (1:1 match) based on demographics, comorbid conditions and RA disease characeristics was utilized to control for imbalances between treatment groups stratified by the number of prior anti-TNF agents. This resulted in 266 matched (TCZ:ABA) pairs of which 45% had 1 and 55% ≥2 prior TNFs. Treatment response was measured at 6 months after initiation based on mean change in disease activity using the CDAI, and change in mHAQ. Secondary analyses included achievement of low disease activity (CDAI ≤10). For any patients who switched from ABA or TCZ to a new biologic, last observation carried forward (LOCF) prior to the switch was applied. Change in prednisone use was well as initiation or discontinuation of a nonbiologic disease modifying anti-rheumatic drug (DMARD) over the study period was also examined. Regression analysis was performed with the matched pairs as a random effect.
Results There were 266 TCZ initiators matched to 266 ABA initiators. Most patients were female (74-75%) with a mean age of 57 years, mean disease duration of 10-11 years and mean CDAI of 28. There were no standardized differences >0.10 between the 2 groups for demographics, insurance, comorbidities, RA disease characteristics, disease activity, functional status, medication use, concomitant prednisone use, and number of prior biologics. There were no significant differences between the two groups in terms of the primary and secondary outcomes (Table). Additionally there were no significant differences between TCZ vs. ABA in terms of prednisone use (dose escalation [14.0% vs. 12.5%]) and dose reduction [18.9% vs. 18.6%]) and nonbiologic DMARD use (rates of initiation [11.3% vs. 14.3%] and discontinuation [13.1% vs. 15.0%]).
Conclusions Both TCZ and ABA were associated with substantial improvements in disease activity at 6 months and had comparable effectiveness in this difficult to treat population of RA patents. Longer term outcomes will be needed in order to determine best strategies in TNF inadequate responder patients with RA in the US.
Acknowledgements This study is sponsored by Corrona, LLC. The Corrona RA registry has been supported through contracted subscriptions in the last two years by AbbVie, Amgen, Astra Zeneca, BMS, Genentech, Horizon Pharma USA, Janssen, Eli Lilly, Novartis, Pfizer, and UCB.
Disclosure of Interest L. Harrold Grant/research support from: Corrona, LLC., G. Reed Employee of: Corrona, LLC., D. Solomon Grant/research support from: Amgen, Lilly and Corrona, J. Curtis Grant/research support from: Roche/Genentech, UCB, Janssen, Corrona, Amgen, Pfizer, BMS, Crescendo, Abb Vie, Consultant for: Roche/Genentech, UCB, Janssen, Corrona, Amgen, Pfizer, BMS, Crescendo, Abb Vie, M. Liu Employee of: Corrona, LLC., J. Greenberg Shareholder of: Corrona, LLC., Consultant for: AstraZeneca, Celgene, Novartis and Pfizer, Employee of: Corrona, LLC., J. Kremer Shareholder of: Corrona, LLC., Consultant for: AbbVie, Amgen, BMS, Genentech, Lilly, Pfizer, Employee of: Corrona, LLC.