Secondary drug failure to adalimumab remains a significant challenge in managing Juvenile Idiopatihc Arthritis (JIA), with anti-adalimumab antibodies (ADAₐb) implicated as a potential mechanism. This study examines the relationship between ADAₐb formation and loss of response (LOR). In this retrospective, longitudinal study, a combination of conventional statistical methods and machine-learning algorithms were employed to investigate the association of adalimumab drug levels (ADAdrug) and ADAₐb, alongside clinical variables, on LOR among JIA patients treated between December 2016 and May 2024. LOR was defined as active disease. Among 184 patients, ADAₐb and ADAdrug emerged as the strongest predictors of LOR in random-forest analysis. In multiple logistic regression, higher ADAₐb levels were independently associated with increased risk of LOR (OR 1. 18, 95% CI 1. 08–1. 29, p = 0. 0003), while higher ADAdrug levels were associated with reduced risk (OR 0. 81, 95% CI 0. 68–0. 95, p = 0. 0103). An ADAₐb threshold of 69 AU/mL was associated with an eightfold increase in LOR (OR 8. 04, 95% CI 2. 8–26. 1, p = 0. 0002). Patients with high ADAₐb and undetectable ADAdrug had the highest risk of LOR (OR 28. 24, 95% CI 7. 11–111. 63, p = 0. 0001). Kaplan–Meier analysis demonstrated that 71% of patients with elevated ADAₐb experienced LOR within 12 months High ADAₐb levels are a significant predictor of LOR in JIA patients treated with adalimumab, particularly when coupled with low drug levels. We also report clinically meaningful one-year LOR rates. These findings support the use of therapeutic drug monitoring, which may help optimise treatment and improve long-term disease control.
Carrim et al. (2026) studied this question.