Systematic Literature Review and Meta-analysis of Tumor Necrosis Factor–Alpha Experienced Rheumatoid Arthritis

Abstract

PurposeThe goal of this study was to compile all available evidence regarding the efficacy of tumor necrosis factor–α (TNF) inhibitors, non-TNF biologics, and tofacitinib for TNF-experienced patients who have rheumatoid arthritis (RA). MethodsA systematic literature review of MEDLINE, EMBASE, and rheumatology conference abstracts was performed to identify observational studies and randomized controlled trials (RCTs) reporting American College of Rheumatology response rates (ACR 20/50/70) for adult patients with RA who switched from at least 1 TNF to another TNF or a non-TNF therapy. A direct random effects meta-analysis was performed to evaluate ACR 20/50/70 response rates for TNF and non-TNF therapies. Separate analyses were conducted among 3-, 6-, and 12-month observational studies and for 6-month RCTs. FindingsA total of 18 observational studies and 6 RCTs were selected. Among 3-month observational studies, the percentages of ACR20/50/70 responders switching to another TNF were similar to those switching to a non-TNF biologic (ACR20, 54.5% vs 58.6%; ACR50, 33.3% vs 33.3%; and ACR70, 13.0% vs 14.6%, respectively). Among 6-month observational studies, the percentages of TNF ACR20/50/70 responders were higher than those of non-TNF responders (ACR20, 67.7% vs 50.4%; ACR50, 50.4% vs 26.6%; and ACR70, 24.9% vs 11.6%). Among 6-month RCTs, the percentages of non-TNF biologic ACR20/50/70 responders were similar to those in the 6-month non-TNF observational studies (ACR20, 50.7% vs 50.4%; ACR50, 27.5% vs 26.6%; and ACR70, 11.9% vs 11.6%). For 12-month observational studies, TNF biologic ACR20/50/70 percentages were higher than those of non-TNF therapies (ACR20, 72.2% vs 57.0%; ACR50, 42.1% vs 28.9%; and ACR70, 22.9% vs 10.0%). ImplicationsFor TNF-experienced patients with RA, subsequent TNF therapy and non-TNF biologic therapy have comparable efficacy.