Background/Purpose: IL-1 inhibitors are both widely used and highly effective as first-line therapy for children with systemic juvenile idiopathic arthritis (sJIA); however, the mechanisms underlying response or non-response to these treatments are not well understood. We sought to characterize inflammatory cytokine and protein levels in responders and non-responders to IL-1 inhibitors in a real-world cohort of new-onset sJIA patients in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry.

Methods: We identified all patients in the FROST study who were started on IL-1 inhibitors and had available serum or plasma. IL-1 responders were classified as those patients who fulfilled the Wallace criteria of clinically inactive disease (CID) or modified CID at 6 months, with no change of therapy. If all other criteria for CID were met but inflammatory markers were not obtained, patients were classified as modified CID. Children who did not fulfill these requirements were characterized as non-responders. IL-18 and CXCL9 levels were measured using an Ella assay, and other cytokine and inflammatory protein levels were obtained using two different custom Luminex panels. Expression levels were compared using a student t-test or paired t-test.

Results: We identified 27 patients with existing baseline biosamples who were started on IL-1 inhibitors. Of these, 16 were responders and 11 were non-responders. At baseline, median IL-18 levels measured by Ella were higher in non-responders (41,744 pg/mL; range 4,612-157,713) vs responders (22,400; range 1,485 – 150,479), but this difference was not significant. Baseline median CXCL9 values measured by Ella were also higher in non-responders (1,887 pg/mL; range 80 – 3,757) when compared to responders (435 pg/mL; range 281 – 4,238), but the difference was not statistically significant.  The IL-18/CXCL9 ratio at baseline was similar between non-responders (30.74) and responders (35.18). Luminex data revealed decreased baseline CCL25 (p=0.049) and increased baseline CD25 (p=0.028) levels in non-responders when compared to responders. Many cytokine levels were significantly decreased after 6 months of treatment in responders, including CD163, IFNγ, IL-17a, CD25, IL-18, GDF15, IL-23, CXCL11, TNFRI, and TNFRII. Among non-responders, fewer changes after 6 months of treatment were observed, but levels of IL-18, CD25, IL12p70, and TNFRI were significantly decreased at 6 months when compared to baseline.

Conclusion: Non-responders to IL-1 inhibitors had overall higher IL-18 and CXCL9 levels and significantly decreased CCL25 and increased CD25 levels compared to responders at baseline. Responders to IL-1 inhibitors demonstrated significant reductions in numerous inflammatory protein levels at 6 months. These findings support the hypothesis that subgroups of sJIA patients with different immunologic phenotypes exist, and that immune phenotyping could assist in predicting IL-1 inhibitor non-response.

*This study utilized data (and biospecimens) collected in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry. The views expressed are the authors’ and do not necessarily represent the view of CARRA.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/analysis-of-protein-biomarkers-for-the-prediction-of-il-1-inhibitor-treatment-response-in-the-carra-first-line-options-for-systemic-juvenile-idiopathic-arthritis-treatment-frost-study/