FRI0330

DEVELOPMENT OF CUT-OFF VALUES FOR HIGH DISEASE ACTIVITY IN JUVENILE IDIOPATHIC ARTHRITIS BASED ON THE JUVENILE ARTHRITIS DISEASE ACTIVITY SCORE (JADAS)

A. Consolaro ^{1 2,*}G. Bracciolini ²A. Frisina ²N. Ruperto ³S. Magni-Manzoni ⁴C. Malattia ^{1 2}S. Pederzoli ²N. Solari ²S. Davì ²A. Martini ^{1 2}A. Ravelli ^{1 2}

¹Pediatrics, Univeristy of Genova, ²Pediatria II, ³PRINTO, IRCCS G. Gaslini, Genova, ⁴Rheumatology, IRCCS Ospedale Pediatrico Bambino Gesù, Roma, Italy

Background: In the last decade, there have been major advances in the management of juvenile idiopathic arthritis (JIA). A reliable documentation of these progresses creates the need for validated criteria that describe precisely patient states. We recently, developed the cut-off values for remission, minimal disease activity and acceptable symptom states in JIA based on the JADAS (1).

Objectives: To determine the JADAS cut-offs for high disease activity (HDA) in JIA.

Methods: For the selection of cut-offs, data from a clinical database including 618 children with JIA were used. Patients were defined as having HDA when the physician made one of the following therapeutic interventions: 1) start of methotrexate (MTX); 2) intra-articular corticosteroid injection; 3) start of a biologic medication; 4) start of systemic corticosteroids. Patients were defined as having low disease activity (LDA) when they were receiving no therapy or had therapy discontinued, tapered or left unchanged for > 1 year. For each patient, 1 visit in HDA and 1 visit in LDA was retained for the analyses. ‘Optimal’ JADAS cut-offs were determined by calculating the 25^th percentile of cumulative score distribution in patients with HDA and by assessing their ability to discriminate between HDA and LDA through ROC curve analysis (including calculation of Youden index and fixed 90% specificity). Cross-validation of cut-offs was performed in 490 JIA patients enrolled in the PRINTO MTX trial (2) and was based on assessment of discriminant validity.

Results: The cut-offs were calculated separately for patients with oligoarticular and polyarticular course of joint disease (irrespective of ILAR category) owing to the different severity of these 2 JIA phenotypes. The table shows the cut-offs for HDA for JADAS-71 version, defined according to the different statistical methods. The cut-offs that showed the best trade-off between sensitivity and specificity were selected for cross-validation analyses. These analyses showed that at baseline visit of MTX trial 94.7% of patients had a JADAS-71 higher than the HDA cut-off for polyarthritis. At 6-month visit, the percentage of patients with a JADAS-71 higher than the cut-off was 85.6% among nonresponders and 23.8% among responders.

Table. JADAS-71 cut-off values for high disease activity in oligoarthritis and polyarthritis. Sensitivity and specificity are reported in parenthesis.

Conclusions: We developed the JADAS cut-offs for HDA in JIA. The cut-offs revealed strong discriminant ability in a clinical trial and are, therefore, potentially applicable in clinical practice, observational investigations, and therapeutic studies.

References: 1) Consolaro A et al. A&R, in press. 2) Ruperto et al. A&R 2004;50:2191-201

Disclosure of Interest: None Declared