Background: Despite the expansion in the number of medications available in the treatment of juvenile idiopathic arthritis (JIA), inter-individual variation in therapeutic response and drug-associated toxicities continue to be a major concern and has driven efforts towards individualised therapy. Recent advances in paediatric treatment strategies and guidelines, as well as innovative approaches to identify a priori predictors of drug response, hold the promise for an individualised approach to therapy that will yield the highest efficacy and safety potential for each JIA patient.

Objectives: To determine whether baseline demographic, clinical, articular, laboratory variables and adherence to therapy may act as predictors of good response to DMARDs therapy in JIA.

Methods: Patients with JIA treated with DMARDs (methotrexate (MTX), leflunomide (LEF) or combination of MTX and LEF) were recruited for this study. Juvenile arthritis disease activity score (JADAS-27) was calculated at 3, 6 and 12 months. Multivariate logistic regression analysis was used to identify predictors of good response according to JADAS-27 score, development of deformities and extra-articular manifestations (uveitis).

Results: A total of 114 children were included in this study. The majority of patients were females 91 (79.8%). Mean age was 11.97±3.26 years; mean age at disease onset was 8.19±3.46 years whereas mean disease duration was 3.68±2.89 years. The most common ILAR subtype was polyarthritis RF negative (50.9%), Polyarticular RF positive (21%), oligoarticular extended (14.9%), oligoarticular persistent (7.9%) and systemic type (5.3%). Over half of the patients were on MTX (55.3%), and (34.2%) were using combined MTX and LEF while (10.5%) of patients were on LEF monotherapy. Mean MTX dose was 12.9±3.4 mg/week, and mean LEF dose was 16.67±4.78 mg/day. Prevalence of positive RF was 24/114 (21.05%). Anti-nuclear antibody (ANA) was detected in 20/114 (17.5%) of patients. 10.5% of patients had chronic anterior uveitis. At baseline, joint deformities were present in (13.1%) of patients. After one-year follow up (45.6%) of patients achieved remission while (27.1%) reached a state of low disease activity. Multivariate logistic regression analysis revealed that the most important predictors of JADAS-27 remission/low disease activity status were: Short disease duration <6.1 ±0.4 months (OR 1.93), polyarticular disease subtype, Childhood Health Assessment Questionnaire (CHAQ) disability index <2.125 (OR 1.75), absence of joint damage/deformities (OR 1.92). A short duration of NSAIDs therapy (<3 months) before moving to MTX therapy as well as from MTX to combination of MTX and LEF therapy (<4 months) were associated with the targeted disease activity status (OR 1.76 and 1.95 respectively). Adherence to therapy was also a predictor of good response (OR 1.86). 15.4% of patients did not achieve the disease activity target and received biologic therapy.

Conclusions: The subgroup of JIA patients with polyarticular disease onset, shorter disease duration, rapid optimisation/escalation of DMARDs therapy as well as those who were adherent to therapy were significantly associated with a good response to DMARDs therapy.

Disclosure of Interest: None declared

DOI: 10.1136/annrheumdis-2018-eular.4156