Discontinuation rate to biologic and targeted synthetic DMARDs in rheumatoid arthritis patients: Systematic review and meta-analysis

Background: The 1-year discontinuation rate to tumor necrosis factor (TNF) and non-TNF inhibitors in rheumatoid arthritis (RA) is extremely variable (2.1%1–82%2). Objectives: To explore the impact of the type of exposure on discontinuation rate to biologics and targeted synthetic disease-modifying anti-rheumatic drugs (b/tsDMARDs) in prevalent and incident users with RA. Methods: This is part of the systematic review and meta-analysis performed according to the PRISMA and MOOSE guidelines (PROSPERO, CRD420191387765). We searched in Medline and Scopus for retrospective population-based studies evaluating discontinuation to b/tsDMARDs in RA until June 11th, 2019. We included original studies performed on healthcare administrative databases with results of discontinuation rate to b/tsDMARDs in adult RA available and full-length papers published in English. We conducted a meta-analysis of proportions using random-effects models. Results were presented as proportions (%) with a 95% confidence interval (CI). Heterogeneity within the group and between groups was evaluated using the Cochran Q test. A p < 0.05 was considered statistically significant. Results: We included 20 studies (65 groups). Incident users were 41,531 (12 studies, 29 groups, mean age 52.20 ±3.61, females 64.35%), prevalent ones 20,491 (9 studies, 24 groups, mean age 50.80 ±3.33, females 80.46%) and mixed ones 20,322 (2 studies, 12 groups, mean age 48.91 ±1.60, females 61.06%). The discontinuation rates in the user categories were 43.56% (95%CI: 33.61–54.05), 27.63% (18.55–39.01%), and 12.70% (7.99–19.58), respectively. Heterogeneity was substantial for all estimates (I2>99%). Among incident users, significant differences emerged between types of exposure considered (p < 0.0001): discontinuation rate ranged between 76.05% (16.11–0.9813) for TNF inhibitors (19,447 patients, 3 groups) and 34.75% (24.59–46.52) for infliximab (2,963 patients, 6 groups). In prevalent users, discontinuation rates ranged between 36.61% (17.77–60.67) for TNF inhibitors (9,493 patients, 5 groups) and 20.36% (4.51–58.07) for adalimumab (1,184 patients, groups) (p = 0.9251). Discontinuation rate in mixed users ranged between 12.50% (3.44–36.41) for adalimumab (6,926 patients, 2 studies, 2 groups) and 11.72% (3.72–31.31) for etanercept (9,637 patients, 2 groups) (p < 0.0001). Conclusions: The study showed that in RA the incident users of b/tsDMARDs had a moderate discontinuation rate, while prevalent a low one. The type of exposure affects the discontinuation measurement. Clinicians should consider the discontinuation rate in the evaluation of the benefit-risk profile of drugs for choosing the most appropriate biologic for RA patients.