Predictors of relapse after withdrawing biotherapies in children with inactive juvenile idiopathic arthritis: a retrospective cohort study of the JIR cohort

Objectives In patients with juvenile idiopathic arthritis (JIA), the approach to discontinuing biological treatments after a period of remission remains exploratory. We aimed to identify one or more predictors of relapse after the discontinuation of biologic/targeted synthetic disease-modifying anti-rheumatic drugs. Methods We reviewed JIA patients followed in two French tertiary centers and included in the JIR cohort who had discontinued their biologic treatments from 2000 to 2023. Remission was defined according to the Wallace criteria. The primary outcome was relapse within 1 year after treatment withdrawal. Multivariate regression was used to analyze predictors of relapse. Results A total of 697 JIA patients received treatment in the two centers during the study period. We analyzed 130 instances of treatment discontinuation in 115 patients (16.5%) (70.8% girls [92/130], median age 5.29 years at diagnosis [interquartile range 2.3–10.4]). Patients discontinued biological treatments after a median of 2.58 years, and 56.2% (73/130) of patients experienced relapse within 1 year after treatment withdrawal. The psoriasis subgroup had the highest relapse rate (89.5% [17/19]) and the systemic subgroup the lowest (33.3% [7/21]). The probably of relapse was increased with antinuclear antibody positivity at diagnosis (p < 0.05) and reduced with longer duration of clinically inactive disease before treatment withdrawal. Conclusion Relapses are common after discontinuing biologic therapies in JIA, particularly in patients with antinuclear antibody positivity and psoriatic arthritis and those who stop treatment shortly after reaching inactive disease status.