Background Methotrexate (MTX) is the DMARD of first choice in patients with juvenile idiopathic arthritis (JIA). However, limited data is available on MTX survival, including reasons for stopping MTX, and predictors for stopping MTX.

Objectives The objectives of this study were to i) describe survival time and reasons for stopping MTX and ii) to identify possible predictors for stopping MTX.

Methods Consecutive children with JIA treated with MTX from the Childhood Arthritis Prospective Study (CAPS), a large prospective longitudinal inception cohort study, were included. At baseline, 6 months and at annual follow-up visits a clinical examination was performed including the physician’s global assessment (PGA) and number of active and limited joints. The CHAQ was completed by the parent or the child. Start and stop dates and reasons for stopping MTX were also collected at each visit. For the present study the following definitions were applied: 1) adverse event (AE), stopped due to AE and MTX not restarted for at least one month; 2) inefficacy, stopped MTX due to inefficacy or added a biologic; 3) efficacy, stopped MTX because of efficacy and MTX not restarted for at least one year; 4) other. Kaplan Meier survival curves were calculated to determine the survival probability at two years for overall, AE, inefficacy or efficacy survival. Patients were included until the date of stopping MTX or until the last follow-up date when MTX treatment was continued. Cox proportional hazards regression analyses were applied to assess the predictive ability of variables assessed at time of starting MTX treatment with AE or inefficacy. Since data on disease activity were not always collected at time of MTX therapy start, disease activity and CHAQ-score assessed up to maximum of three months prior to MTX start was used.

Results 501 children (340 (68%) females; median [IQR] age at MTX start 8.3 [4.0–12.2] yrs) received MTX after a median time since symptom onset of 7.1 [3.5–19.1] months. 244 (49%) stopped MTX, reasons: AE (n=58), inefficacy (n=121), efficacy (n=34) and other reasons (n=31). Overall median survival time was 2.4 [1.1–4.4] yrs. The estimated survival rates at two years were 0.87 (95%CI 0.83-0.90) for AE, 0.75 (95%CI 0.70-0.79) for inefficacy and 0.93 (95%CI 0.88-0.95) for efficacy. Older children were more likely to stop MTX medication because of AE (HR 1.08, 95%CI 1.01-1.14) or inefficacy (HR 1.06, 95%CI 1.01-1.10) and a high PGA score measured at MTX start (HR 1.21, 95%CI 1.04-1.42 (n=236)) was associated with MTX related AE survival. No other associations with AE or inefficacy survival were found, probably due to lack of power, respectively: female gender (HR (95%CI); 0.83 (0.48-1.43) and 0.98 (0.66-1.44)), active joint count (0.99 (0.95-1.03) and 1.02 (0.99-1.04) (n=323)), limited joint count (0.99 (0.95-1.04) and 1.02 (0.99-1.04) (n=321)), and CHAQ-score (0.71 (0.36-1.41) and 1.03 (0.64-1.66) (n=147)).

Conclusions In this cohort of patients with JIA starting MTX we found that children stayed on MTX therapy for a median of ∼2.5 years. Older age was a predictor for stopping MTX due to AE and inefficacy.

Disclosure of Interest None Declared