Background Despite the well documented effectiveness of anti-TNF treatment in rheumatoid arthritis (RA), some patients can be refractory to treatment (primary [1ry] failure) or may lose responsiveness (secondary [2ry] failure). In such cases, switching to another anti-TNF or a different biologic class can often restore therapeutic response.
Objectives The aim of this analysis was to assess the rate of non-response among RA patients treated with anti-TNFs in Canadian routine clinical practice.
Methods BioTRAC is an ongoing, prospective registry of patients initiating treatment for RA, ankylosing spondylitis, or psoriatic arthritis with infliximab (IFX) or golimumab (GLM). RA patients treated with IFX since 2002 or with GLM since 2010 who had at least one post-baseline assessment were eligible. Patients with available information on DAS28 or EULAR response in at least one post-baseline visit were included in the respective analyses.
Results 1,127 patients (75.6% female) were included with mean (SD) age of 56.1 (13.4) years and disease duration of 8.4 (8.9) years at baseline. The majority were biologic naïve (93.2%), treated with IFX (76.0%), and received a concomitant DMARD (86.2%) at baseline. Mean (SD) disease parameters at baseline were: CDAI: 33.9 (17.6); HAQ: 1.6 (0.7); SJC: 9.8 (6.8); TJC: 11.5 (8.0).
After a mean (SD) follow-up of 35.5 (36.8) mos, 764 (67.8%) patients were discontinued overall and 226 (20.1%) due to effectiveness reasons (lack of response: n=67; loss of response: n=83; disease progression: n=76). Among the patients discontinued due to effectiveness reasons, the majority were discontinued after 12 months (54.4%) and had achieved prior good EULAR response (66.2%). Among patients discontinued due to lack of response, 17.7% and 45.5% had previously achieved DAS28 low disease activity (LDA) and good EULAR response, respectively; whereas, among patients discontinued due to loss of response, 46.9% and 76.8% had a previously documented achievement of the two targets, respectively.
Conclusions The results of this analysis have shown a low rate of failure during treatment with IFX and GLM. Non achievement of DAS28 LDA was a good predictor of lack of response and more predictive than good EULAR response; non-achievement of good EULAR response, on the other hand, was a better predictor of loss of response. Overall, significant variation exists depending on each investigator's definition of 1ry and 2ry failure, which highlights the importance of establishing standardized definitions of these terms.
Disclosure of Interest E. Keystone: None declared, P. Baer: None declared, W. Olszynski: None declared, M. Baker: None declared, B. Haraoui: None declared, W. Bensen: None declared, R. Faraawi: None declared, E. Rampakakis: None declared, J. Sampalis: None declared, A. Lehman Employee of: Janssen Inc., F. Nantel Employee of: Janssen Inc., B. Osborne Employee of: Janssen Inc., C. Tkaczyk Employee of: Janssen Inc., K. Maslova Employee of: Janssen Inc.