P096 Pathway directionality across different immune cell lineages contributes to methotrexate response in juvenile idiopathic arthritis

Abstract

Abstract Background/Aims Juvenile idiopathic arthritis (JIA) is the most common autoimmune rheumatic disease in children with methotrexate (MTX) as the first line treatment. However, about 50% of JIA patients will not respond well to MTX yet still experience drug side effects. Early prediction to MTX treatment response would be beneficial for patients and families to avoid incurring unnecessary MTX side effects and ongoing uncontrolled inflammation: such validated tools or biomarkers are currently not available. Methods Transcriptional analysis was performed on peripheral blood mononuclear cells (PBMC) from pre-MTX-treated JIA patients (n = 97) of all ILAR JIA subtypes, excluding systemic JIA. RNAseq was performed on total PBMC and sorted immune cell populations: CD4+ T cells, CD8+ T cells, CD19+ B cells, and CD14+ monocytes. Clinical data collected at the time of sampling (baseline) and at the follow-up time (between 3-12 months) were used to define outcomes, measured as change in active joint count (AJC), change in Physician VAS (PhysVAS), and change in cJADAS10. After batch normalisation with ComBat-seq, differential gene expression (DGE) analysis was performed using limma-voom with age, sex, ethnicity, and steroid status included as covariates. Log2 fold changes were utilised to rank genes to implement gene set enrichment analysis (GSEA) by fgsea. Results DGE analysis showed minimal significant differentially expressed (DE) genes that passed 5% false discovery rate (FDR). The greatest number of significant DE genes were observed in CD14+ monocytes, where baseline expression of 13 genes were significantly associated with change in PhysVAS. As alterations of gene expression for a heterogeneous disease such as JIA can be subtle and correlated between genes, GSEA was performed to investigate expression changes at pathway level. Using MSigDb Hallmark pathways, GSEA showed interferon-related and tumor-necrosis-factor pathways as significantly associated with MTX response in all cell types (5% FDR). However, the directionality (up/down regulation) differed between cell lineages, suggesting pathways divergence between T cell and non-T cell lineages. For example, in interferon gamma pathway, up-regulation associated with poor treatment response in T-cells and down-regulation in non-T cells. Preliminary analysis of the leading edge genes of the interferon gamma pathway in the non-T cell lineages showed that many of the genes are driven by interferon alpha, whilst within CD4+ and CD8+ T cells the majority of the leading edge genes are specific to interferon gamma. Conclusion Different directionality of pathways that might be relevant to JIA response to treatment with MTX is observed in different mononuclear cell lineages. This could potentially explain the difficulties of finding biomarkers which correlate with response to treatment from whole blood or PBMC. Shared genes across different interferon-related pathways also suggest that interaction between interferon pathways might be driving the different direction of interferon pathway expression in different cell lineages. Disclosure M. Kartawinata: None. W. Lin: None. B. Jebson: None. K. O'Brien: None. E. Ralph: None. R. Restuadi: None. G.T. Hall: None. S. Castellano: None. C. Wallace: None. L.R. Wedderburn: Grants/research support; LRW Declares in kind contributions to CLUSTER by AbbVie, GSK, UCB, Sobi and Pfizer inc and non renumerated collaborations with Lilly and Novartis.