Background The use of TNF inhibitors (TNFi) for the treatment of rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), and juvenile idiopathic arthritis (JIA) is well established, but observational studies comparing long-term TNFi retention in adult and juvenile-onset inflammatory arthritides are still lacking.
Objectives To evaluate the 10-year drug survival and reasons for discontinuation of the first-line TNFi (infliximab [IFX], adalimumab [ADA], or etanercept [ETN]) in a real-life large cohort of RA, AS, PsA, and JIA patients.
Methods RA, AS, PsA, and JIA patients treated with their first TNFi between October 1999 and October 2014 were selected from a local registry. The analysis was limited to patients treated with IFX, ADA, or ETN. Drug survival up to 10-years follow-up was evaluated overall by the Kaplan-Meier method and compared according to diagnosis, TNFi agent, and reason for discontinuation by a stratified log-rank test.
Results The analysis included 360 JIA patients (71.9% female, mean age [±SD] 14.9 [±8.5] years, mean disease duration 7.2 [±8.1]), treated with IFX (n=89), ADA (n=66), or ETN (n=205); and 927 (583 RA, 188 AS, and 156 PsA) adult patients (66.5% female, mean age 54 [±12.4] years, mean disease duration 8.6 [±18.9] years), treated with IFX (n=447), ADA (n=262), or ETN (n=218). The median survival on treatment in the whole group was 60.9 months (66.3 and 51.3 months in adult and juvenile population, respectively). The overall 10-year retention rate was 30.3%, with no significant difference between adult and juvenile patients (31.2 and 27.5%, respectively; HR [95% confidence interval] 0.96 [0.81-1.13])) [Figure 1]. The risk of stopping TNFi because of adverse events (AEs) was significantly lower in juvenile than adult patients (HR 0.78 [0.61-0.99]), whereas no differences were observed in discontinuation due to inefficacy (HR 1.16 [0.92-1.46]). In the sub-analysis of AEs leading to discontinuation, infections (n=22 vs n=0) and malignancies (n=19 vs n=0) were more frequent in adult than juvenile patients, whereas neuropsychiatric (n=16 vs n=1), ocular (n=14 vs n=3), and gastrointestinal (n=11 vs n=0) complications were more frequent in juvenile than in adult-onset arthritis. The 10-year drug persistence was significantly higher for ETN than IFX and ADA in adult population (55.5 vs 27.5 and 23.3%, respectively; p<0.0001) and significantly higher for ADA than IFX and ETN in juvenile population (52.1 vs 11 and 32.2% respectively; p<0.0001).
Conclusions In a real-life setting, the 10-year retention rate of the first TNFi was similar in juvenile and adult-onset arthritides. The risk of stopping TNFi because of AEs was significantly lower in juvenile than adult patients. Neuropsychiatric, ocular, and gastrointestinal complications were more frequent in juvenile patients whereas infections and malignancies in adult population. ETN showed the higher retention rate in juvenile and ADA in adult-onset diseases.
Disclosure of Interest None declared