FRI0462 PREDICTIVE FACTORS OF RELAPSE AFTER METHOTREXATE DISCONTINUATION IN JIA PATIENTS WITH INACTIVE DISEASE

Abstract

Background:Methotrexate (MTX) is the most widely used conventional synthetic disease-modifying antirheumatic drug (csDMARD) in the treatment of juvenile idiopathic arthritis (JIA).1,2When remission is achieved, questions remain about discontinuing MTX. There is some evidence that a longer period of inactive disease before MTX withdrawal is associated with lower likelihood of relapse, while both rheumatoid factor (RF) positive polyarthritis and extended oligoarthritis categories are associated with higher probability of disease relapse.2,3Objectives:To identify predictive factors of relapse after discontinuation of MTX in JIA patients with inactive disease.Methods:Prospective multicentre cohort study in patients diagnosed with JIA, according to the ILAR classification, using real world data from the Portuguese national register database, Reuma.pt (Fig 1).4We evaluated patients who have reached JADAS27 inactive disease (≤1 and no active extra-articular manifestations) and discontinued MTX before the age of 18 years-old.5Relapse was defined as recurrence (>1 or extra-articular manifestations) or restarting a DMARD.5To identify differences of relapse risk, univariate analyses were performed. Persistence in remission was estimated using the Kaplan-Meier method. Subsequently, Cox regression analyses were performed to identify predictors of relapse.Results:119 JIA patients discontinued MTX due to inactive disease (Fig 1). 69.7% were females and 60.6% had oligoarticular JIA. Sociodemographic and clinical characteristics are shown in Table 1. Relapse has occurred in 32.8%. Table 2 shows the disease characteristics at MTX initiation and discontinuation and at relapse or last visit.In univariate analysis, relapse was associated with the use of NSAIDs at the time of MTX discontinuation (p=.027) and with a period of less than two years in inactive disease before MTX suspension (p=.040). We found no association with gender, race, immunology (RF, antinuclear and cyclic citrullinated peptide antibodies), MTX dose, discontinuation modality (tapering and spacing the doses or just tapering the dose), extra-articular manifestations, previous corticotherapy, family history, body mass index, JADAS, CHAQ index, inflammatory parameters, tender and swollen joint counts at MTX initiation or discontinuation nor with age at remission or at MTX suspension. Median persistence in inactive disease was significantly higher in patients with more than two years in remission before MTX discontinuation (p=.034) and in those who did not use NSAIDs at time of MTX discontinuation (p=.026) (Fig 2).After adjustment for age at diagnosis, MTX tapering and JIA category, use of NSAIDs at the time of discontinuation (HR, 1.98 95%CI 1.03-3.82) and less than two years in remission (HR, 3.12 95%CI 1.35-7.13) remained associated with relapse.Conclusion:In this large cohort we found that the use of NSAIDs at the time of MTX discontinuation was associated with two times the likelihood of relapse. Like in other studies we also showed that the time in remission before MTX discontinuation is the main predictor of relapse. We found no association between the JIA category and the risk of relapse.