Background/Purpose: Up to one-third of the patients who receive TNFi lose responsiveness over time. Options available to these patients include treatment with an alternative TNFi (cycling strategy) or switching to another therapy with a different mode of action (swapping strategy). We conducted a systematic review and meta-analysis of RCTs to compare the efficacy of various approaches.

Methods: Five electronic databases were searched (MEDLINE, EMBASE, Cochrane Library, and Web of Sciences). Sources of gray literature (unpublished records) were searched through clinicaltrials.gov and other websites. The search was broad in scope to capture all available evidence. The selection process was performed by two independent reviewers. We included studies evaluating the efficacy and safety of targeted therapies in RA that included only adult patients with failure to respond to at least one TNFi. We excluded non randomized studies, RCTs with no separate data for TNFi failures, evaluating retreatments (e.g. rituximab or tocilizumab vs. placebo after one or two cycles of rituximab or tocilizumab), with insufficient data to evaluate the outcomes of interest, or comparing a brand-name product with a biosimilar (with no control group). Outcomes included: 50% improvement according to the American College of Rheumatology criteria (ACR50), serious adverse events, and withdrawals.

Results: Out of 44,651 citations we found 13 studies (with multiple publications n=68). The mean age of patients ranged between 45.1 and 58.2 years. The majority of the patients were females (82%) with a disease duration ranging between 8.6 and 13.2 years and a baseline disability score ranging from 1.1 to 1.9. Cycling strategies included: golimumab, certolizumab, adalimumab, etanercept or infliximab in patients failing any other TNFi. Swapping strategies included rituximab, tocilizumab, abatacept, or tofacitinib. Four studies were head-to-head comparisons, one compared two different doses of tocilizumab combined with methotrexate with tocilizumab monotherapy, one compared nonTNFi plus TNFi versus TNFi only and the remaining studies compared a targeted agent combined with a disease modifying anti-rheumatic drug (DMARD) versus DMARD monotherapy (with or without placebo). Three studies (n=898) for the cycling strategy and four studies (n=1,774) for the swapping strategy were suitable for direct meta-analysis. When an alternative TNFi was used in combination with DMARD, ACR50 response rates were statistically significantly improved at 12 to 24 weeks compared to the DMARD alone group (pooled for all TNFi = RR 2.6 (95% CI 1.6 to 4.3). Improvements were also observed with each swapping strategy (pooled RR including rituximab, abatacept, tocilizumab, or tofacitinib was 6.1 (95% CI 4.1 to 8.9)). No differences were observed in the rates of withdrawals due to adverse events or serious adverse events between groups for any of the strategies.

Conclusion: Current evidence from RCTs shows that cycling to a different TNFi after failure can be effective, and can be considered before switching to a different targeted therapy, suggesting that many patients fail a specific agent and not TNFi as a class.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/therapy-in-patients-with-rheumatoid-arthritis-ra-with-inadequate-response-to-tumor-necrosis-factor-alpha-inhibitors-tnfi-a-systematic-review-and-meta-analysis-of-randomized-controlled-trials-rct/