Low pretreatment levels of myeloid-related protein-8/14 and C-reactive protein predict poor adherence to treatment with tumor necrosis factor inhibitors in juvenile idiopathic arthritis.

Abstract

Two thirds of patients with juvenile idiopathic arthritis (JIA) treated with tumor necrosis factor (TNF)-alpha inhibitors respond initially, but only about one third of patients achieve clinical remission at follow-up. We evaluated the 1-year response and long-term treatment adherence to TNF inhibitor treatment in JIA patients naive to biologics and investigated if baseline myeloid-related protein (MRP)-8/14 and C-reactive protein (CRP) were predictive of treatment response. One hundred fifty-two patients were included in a unicenter observational, prospective study from 2002 to 2015, excluding patients with systemic-onset JIA. One-year treatment response was evaluated by American College of Rheumatology-pediatric (ACR-ped) and by the number of patients achieving inactive disease (ID). Medical charts were reviewed for reasons of treatment withdrawal. After one year of treatment ACR-ped 30, 50, 70, and 90 were achieved by 61, 55, 38, and 10 % of the patients, and 23 % achieved a status of ID. Treatment adherence: 51 % withdrew from treatment due to lack of clinical effect, while 32 % continued treatment or withdrew due to disease remission. Increased MRP-8/14 concentrations at treatment initiation was associated with ID after 1 year (OR 1.55, CI 1.06-2.25, p = 0.02). Treatment withdrawal due to lack of effect was associated with low baseline levels of both MRP-8/14 (685 vs. 1235ng/ml, p<0.001) and CRP (0.75 vs. 2.73mg/l, p<0.001), verified by multivariable logistic regression analysis (OR 0.51, CI 0.34-0.77/OR 0.63, CI 0.48-0.83). In conclusion, an association was found between ID after 1 year of treatment and increased baseline levels of MRP-8/14. Furthermore, low baseline MRP-8/14 and CRP concentrations were associated with treatment withdrawal due to lack of clinical effect.