Background/Purpose: Interleukin (IL)-1 inhibitors are now regarded as the first line treatment option for systemic juvenile idiopathic arthritis (sJIA), however in up to half of cases there may be an incomplete response. Our objective was to evaluate the use of candidate biomarkers to predict responsiveness to IL-1 inhibition in a real-world cohort of new-onset sJIA patients who were enrolled in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry FROST study.

Methods: FROST patients who received IL-1 inhibitors and had available DNA, RNA or plasma were included. Response was defined as cJADAS10 inactive disease (≤ 2.5, without fever or steroids) at 6 months while on IL-1 monotherapy. Whole genome sequencing was used to determine class II HLA alleles and IL1RN single nucleotide polymorphisms (SNPs) associated with poor response to IL-1 therapy (PMID: 29609200). Expression of candidate genes associated with response and non-response (PMID: 33452871) and 28 interferon stimulated gene panel (ISG-28) were measured with custom NanoString assays. IL1RA autoantibodies were measured using Luciferase Immunoprecipitation System.

Results: 17/31 (54%) patients responded to IL-1 inhibition. We found no significant differences in expression of IL-1 response genes or in ISG-28 scores; however, STAB1, CD163 and VCAN expression were significantly higher at baseline in non-responders compared to responders. A composite geomean score of these three genes >545 yielded 78% sensitivity and 93% specificity for non-response to IL-1 inhibitors. Non-responders trended toward lower neutrophil counts at baseline (p=0.0944). 6/29 (20%) patients were positive for HLA-DRB*15. There was no difference in response rate based on homozygosity for high-expressing IL1RN SNPs, presence of HLA-DRB1*15 or HLA-DRB1*11. No patients presented with autoantibodies to IL1RA at disease onset, but 2 patients developed them at 6 months. Both patients had been exposed to anakinra and achieved clinical response, which was maintained through 18 months. 3 patients had adverse events reported which may be suggestive of a possible drug hypersensitivity reaction (2 severe injection site reactions, one hepatitis). None of these patients were positive for HLA-DRB1*15 and 2 were positive for HLA-DRB1*11. There was no significant difference in the median ISG-28 scores between baseline and follow up of patients with paired samples (329.5 vs 509.0, p = 0.42), suggesting stability of the signature.

Conclusion: In this real-world cohort of new onset sJIA patients, we confirmed that STAB1, CD163 and VCAN expression at baseline correlated with poor response to IL1 inhibition, suggesting a different immunologic phenotype that might benefit from alternative therapy. 2 patients who responded to IL-1 inhibitors developed anti-IL1RA antibodies at follow up. We identified that IL-1 inhibitors do not significantly increase type I interferon signatures in sJIA patients. IL1RN SNPs and HLA-DRB1*15 were not different between responders and non-responders in this cohort. It is possible that biomarkers have less predictive effect on treatment response in new onset patients, which might include monophasic patients who will improve independently of therapy.

« Back to ACR Convergence 2025

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