Background/Purpose: Switching of disease-modifying antirheumatic drugs (DMARDs) in rheumatoid arthritis (RA) patient treatment is common in real-world clinical practice. The context for why patients switch has not been well described and various factors may impact switching behavior. The purpose of this analysis is to characterize factors associated with DMARD switching using an US clinical RA registry, JointMan^®. This database provides a practical outcome tool to manage patients with RA in a traditional office-based setting.

Methods: A retrospective analysis using the JointMan^®registry was conducted in RA patients (N = 5,817) diagnosed by rheumatologists. Adult RA patient inclusion parameters were: a physician prescription record of a tumor necrosis factor inhibitor (TNFi) (infliximab, etanercept, adalimumab, golimumab, certolizumab) from April 1, 2010 through March 31, 2015; and a subsequent follow-on prescription of TNFi or a novel mode of action (MOA) DMARD (abatacept, anakinra, rituximab, tocilizumab, tofacitinib). Patients were divided into 2 cohorts: those who switched from previous TNFi to another TNFi (TNF cyclers), and those who switched to a novel MOA DMARD. Differences in baseline characteristics and reasons for discontinuing prior TNFi were analyzed using bivariate analysis. Multiple logistic regression was used to identify independent factors associated with switching.

Results: 519 adult RA patients were included in the study: mean age 56.5 years, 79.4% women, 86.3% Caucasian, 95.8% US West Coast resident, clinical disease activity index (CDAI) 22.8, mean prior TNFi treatment duration 29.4 months. Reasons for discontinuing prior TNFi therapy were primary lack of efficacy (38.9%), secondary loss of efficacy (20.0%), adverse events (12.1%), cost/insurance-related reasons (5.6%), and 23.3% other. After discontinuing index TNFi, more than half of patients switched to another TNFi (n = 278, 53.6%) rather than novel MOA DMARDs (n = 241, 46.4%). Patients switching to a novel MOA DMARD were older (57.8 vs. 55.4 years, P = 0.0332), less likely to be commercially insured (61.51% vs. 50.21%, P = 0.001), and more likely to be Medicare insured (46.06% vs. 35.61%, P = 0.0156) than patients cycling to another TNFi.

Multivariate analysis showed that patients were more likely to switch to a novel MOA DMARD than cycling to another TNFi if the first TNFi was administered intravenously (adjusted odds ratio [aOR] [95% CI] = 4.18 [2.50-6.99], P < 0.0001). Additionally, those with positive rheumatoid factor (aOR = 2.13 [1.28-3.56], P = 0.0038) and those with higher CDAI (aOR = 1.02 [1.01-1.04], P = 0.008) were more likely to switch to a novel MOA DMARD. Patients were less likely to switch to novel MOA DMARDs than to another TNFi if the reason for discontinuing the prior TNFi was insurance/cost-related (aOR = 0.30 [0.32-0.95], P = 0.0094) or loss of efficacy (aOR = 0.50 [0.32-0.95], P = 0.0321).

Conclusion: TNF cycling in RA treatment is common in clinical practice. The main reasons for discontinuation of prior TNFi therapy were lack or loss of efficacy. Increased understanding of switching patterns in RA patients may help improve patient care and foster further definition of drug development.

« Back to ACR/ARHP Annual Meeting 2015

ACR Meeting Abstracts - https://acrabstracts.org/abstract/factors-associated-with-tnf-switching-a-retrospective-real-world-study-of-patients-with-rheumatoid-arthritis/