Background/Purpose:

Canakinumab (CAN), a monoclonal anti-interleukin (IL)-1β antibody, is approved for the treatment of systemic juvenile idiopathic arthritis (SJIA). CAN-treated patients with SJIA showed a high treatment response rate in an open-label long-term extension study. However, little is known about the correlation of various serum biomarkers in patients with different treatment outcomes.

Methods:

Serum samples from 54 patients treated with CAN in the open-label long-term extension study were studied in a 14-plex bead array assay (Luminex) at different time points during the clinical trial, including days 1 (prior to first administration of CAN), 3 and 15, and months 1, 3, 6, 12, 18 and 24. Treatment outcomes included (1) a modified pediatric American College of Rheumatology (pACR) 90 response within 15 days of treatment with CAN and (2) an pACR100 response or clinical inactive disease within 15 days plus no disease flare or macrophage activation syndrome (MAS) during the study (=sustained complete response). Biomarker data, multiple clinical parameters and treatment outcomes were analysed using rank correlation, two-group comparisons via non-parametric testing (Mann-Whitney U test), receiver operating characteristic (ROC) analysis and hierarchical clustering.

Results:

Twenty-six of 54 patients (49%) reached a modified pACR90 response within 15 days. Within a median follow-up of 23 months (range 0.5-32 months), 12 of 54 (22%) patients had a sustained complete response and 5 (9%) had developed MAS. Several biomarkers were moderately to strongly associated with each other prior to initiation of CAN therapy (Spearman rank correlation coefficient highest for interferon (IFN)-γ and IL-18: 0.90). Biomarkers did not correlate significantly with age, duration of disease or active joint count. Several biomarkers (IL-1β, IL-18, IL-6 and S100A12) were markedly elevated when compared to healthy controls and decreased during CAN therapy. There were significant differences in day 1 CXCL9 levels between patients exhibiting an pACR90 response by day 15 (responders: median 296 pg/ml, non-responders: 709 pg/ml, p<0.01). Further, day 1 levels of CXCL9, IL-7 and IL-18 as well as the CXCL9:IFN-gamma ratio differed between sustained complete responders and those without sustained complete response (CXCL9 median: 278 pg/ml vs. 566 pg/ml; IL-7: 12.3 pg/ml vs. 7.4 pg/ml; IL-18: 1700 pg/ml vs. 556 pg/ml; CXCL9:IFN-γ ratio: 0.12 vs. 0.50). As determined via ROC analysis, these markers had moderate accuracy in predicting a sustained complete response (area under the curve: for CXCL9 0.72, CXCL9:IFN-γ ratio 0.79, IL-18 0.73, IL-7 0.83). Several clusters were apparent but not clearly related to clinical outcomes.

Conclusion:

Several serum biomarkers measured prior to and shortly after initiation of CAN therapy in patients with SJIA showed an association with treatment response. Lower CXCL9 serum concentrations, a lower CXCL9:IFN-γ ratio, and higher IL-18 and IL-7 serum concentrations measured before CAN therapy were moderately associated with a better treatment response.

« Back to ACR/ARHP Annual Meeting 2018

ACR Meeting Abstracts - https://acrabstracts.org/abstract/multiplex-serum-biomarker-analysis-before-and-during-therapy-with-canakinumab-in-patients-with-systemic-juvenile-idiopathic-arthritis/