Background/Purpose: More than 40% of children with polyarticular forms of Juvenile Idiopathic Arthritis (JIA) experience clinical inactive disease on medication (CR). No clinical variable or biomarker has been identified to predict which patients in CR can effectively stop anti-TNF biologic therapy.

Methods: In a 16 center, prospective protocol driven clinical trial, 106 JIA pts. (polyarticular RF+/RF- and extended oligoarticular) in CR while on anti-TNF treatment stopped anti-TNF therapy (all other meds remained stable) and were followed prospectively for up to 10 mos. Flare was determined by prespecified criteria. Serum drawn at the time of stopping anti-TNF therapy were analyzed for 12 individual biomarkers included in the multi-biomarker disease activity score (MBDA; Vectra DA) using multiplexed immunoassay methods previously described ^1,2: vascular cell adhesion molecule-1 [VCAM-1], epidermal growth factor [EGF], vascular endothelial growth factor A [VEGF-A], interleukin-6 [IL-6], tumor necrosis factor receptor 1 [TNF-R1], matrix metalloproteinase-1 [MMP-1], matrix metalloproteinase-3 [MMP-3], human cartilage glycoprotein-39 [YKL-40], leptin, resistin, serum amyloid A [SAA], and CRP (1,2). The Vectra DA score (range 1 to 100) was calculated with a validated algorithm. Six additional biomarkers were also assayed: interleukin-8 [IL-8], interleukin-1B [IL-1B], macrophage-derived chemokine [MDC], interleukin-6 receptor [IL-6R], intercellular adhesion molecule 1 [ICAM-1] with a custom multiplex assay (MSD, Bethesda, MD). S100A12 levels were determined as previously described ^3,4. Univariate analysis with students t-test for significance and multivariate discriminant analysis was used to calculate relationship of clinical and biomarker variables with risk for flare.

Results:

39/106 (37%) of the pts. Flared, mean/median/SEM for time to flare was 212/250/9.8 days. Univariate analysis of correlations with flare/no flare status shown in Table 1. Multivariate discriminant analysis identified a model that includes 7 variables (MMP-3, JIA duration, JIA Dx age, VCAM-1, CR duration, VEGF, and resistin) with an AUC 0.80, correct prediction in 79% of cases, sensitivity 79%, specificity 79%, PPV 91%, NPV 79%.

Table 1.

Conclusion:

37% of the pts. flared ≤ 10 mos of stopping anti-TNF. The MBDA did not discriminate significantly flare/no flare. A combination of clinical and lab biomarkers predicted not flaring accurately in 79% of cases.

References.

1. Eastman et al. Journal of Pharmaceutical and Biomedical Analysis 2012;70: 415-24.

2. Curtis et al. Arthritis Care & Research 2012;64:1794-1803.

3. Foell D, et al. Gut. 2003;52:847-53.

4. Frosch M, et al. Arthritis Rheum. 2000;43:628-37.

« Back to ACR/ARHP Annual Meeting 2017

ACR Meeting Abstracts - https://acrabstracts.org/abstract/validation-of-biomarkers-to-predict-flare-in-polyarticular-jia-upon-stopping-anti-tnf-therapy/