Background/Purpose: Biological therapy has dramatically improved the treatment of patients with JIA. However, there is still a group of patients that shows a lack of clinical response to this treatment. Furthermore, when patients do respond to treatment with etanercept, it is unclear when they can discontinue this treatment. The use of robust predictive markers of response to identify individuals who are likely to respond to anti-TNF treatment (etanercept or adalimumab) and who are able to stop etanercept may provide guidance in optimizing treatment strategies. Our objective was to test the ability of baseline MRP8/14 serum levels to differentiate between responders and non-responders to anti-TNF treatment and to correlate longitudinal follow-up of this marker in response to treatment. A second objective was to test the ability of MRP8/14 to distinguish between patients who are likely to flare after discontinuing etanercept.

Methods: Samples were collected from 89 JIA patients (13 polyarthritis rheumatoid factor (RF) positive, 33 polyarthritis RF negative, 24 extended oligoarthritis, 5 persistent oligoarthritis, 4 enthesitis related arthritis, 10 psoriatic arthritis) included in the Dutch Arthritis and Biologics in Children Register, the German Registry for Biologics in Paediatric Rheumatology and Great Ormond Street Hospital for Children London treated with TNF blockers. The patients were categorized into responders (ACRpedi≥50 and/or inactive disease according to the Wallace) and non-responders (ACRpedi<50). Serum concentrations of MRP8/14 complexes were measured by ELISA at start of biologic treatment and if available also within 6 months after start of treatment. A flare was defined as having at least three of the following: VAS physician or patient ≥ 20 mm, ≥ 2 active joints, any worsening on the CHAQ and ≥30% worsening on ESR or limited joints. Non-parametric tests were used for analyses.

Results: Before initiation of etanercept treatment, responders (n=71) showed significantly higher levels of MRP8/14 serum complexes compared to non-responders (n=18) (p=0.004, median in responders: 1490 ng/ml (IQR 1020-3323 ng/ml), median in non-responders: 788 ng/ml (IQR 442-1233 ng/ml)). No significant correlation was found between baseline MRP8/14 and JADAS10 disease activity. In non-responders MRP8/14 levels did not significantly change after initiation of treatment whereas levels decreased in responders (p<0.001). Change in JADAS10  disease activity was significantly correlated to change in MRP8/14 levels (Spearman’s rho: 0.4, p=0.03). Samples were available from 28 patients at the time of discontinuation of etanercept. Patients who flared within 6 months after the discontinuation of etanercept had higher MRP levels at discontinuation than patients who did not flare (p= 0.031, median 1025 ng/ml (IQR 588-1288 ng/ml) vs. 505 ng/ml (IQR 346-778 ng/ml)).

Conclusion: High levels of baseline serum MRP8/14 have prognostic value in predicting a subgroup of JIA patients who will respond well to anti-TNF treatment. Decrease of MRP8/14 after initiation of treatment is associated with response to treatment. High MRP8/14 serum levels at time of discontinuation of etanercept are associated with a higher chance to flare.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/mrp814-serum-level-as-predictor-of-response-to-starting-and-stopping-anti-tnf-treatment-in-non-systemic-juvenile-idiopathic-arthritis/