Background/Purpose:  Although proximal muscle weakness and skin rash are the typical features of juvenile dermatomyositis (JDM), little is known about disease pathogenesis, why other features develop, or why patients have differential responses to therapy. In complex diseases like JDM, large scale genomics and systems biology approaches may enable insight into dysregulated pathways involved in pathogenesis, and could lead to a more rational therapeutic approach.

Methods:  Peripheral blood samples were obtained from patients enrolled in the Juvenile Dermatomyositis Cohort and Biomarker Study (JDCBS) at pre-treatment (n=13) and 11.8 [11.3-13.2] months post-treatment (n=13), including n=10 matched samples. CD4, CD8, CD14 and CD19 cells were sorted from PBMC by flow cytometry and RNA prepared for RNA-seq. Samples were sequenced with 17.0 [14.5-18.5] million reads. Raw sequence reads were aligned to hg38 with TopHat and aligned reads processed using Picard tools and summarized by featureCounts. Differential expression analysis was performed using the edgeR package in R. Gene set enrichment analysis was performed using GSEA and the hallmark collection of gene sets. Over-represented gene sets are reported with the absolute value of the normalized enrichment score and Benjamini-Hochberg-adjusted q-value.

Results:  Across all subsets, when pre-treatment samples were compared to post-treatment samples, GSEA uncovered over-representation of gene sets involved in “INTERFERON_ALPHA_RESPONSE” (2.8, p<0.001 for CD4; 2.8, p<0.001 for CD8; 3.2, p<0.001 for CD19; 3.3, p<0.001 for CD14) and “INTERFERON_GAMMA_RESPONSE” (2.6, p<0.001 for CD4; 2.7, p<0.001 for CD8; 2.6, p<0.001 for CD19; 3.1, p<0.001 for CD14). When post-treatment samples were compared to pre-treatment samples, gene sets over-represented across all subsets included “TNFA_SIGNALING_VIA_NFKB” (2.4, p<0.001 for CD4; 2.2, p<0.001 for CD8; 3.0, p<0.001 for CD19; 2.3, p<0.001 for CD14), and “HYPOXIA” (2.0, p=0.004 for CD4; 1.95, p=0.002 for CD8; 2.3, p<0.001 for CD19; 1.7, p=0.016 for CD14). Additionally, some pathways differed between cell types in over-representation either pre- or post-treatment. “IL6_JAK_STAT3_SIGNALING” was over-represented at pre-treatment in CD4 (2.1, p<0.001) and CD8 cells (2.1, p<0.001), but over-represented at post-treatment in CD19 cells. Similarly, the “COMPLEMENT” and “INFLAMMATORY RESPONSE” gene sets were over-represented pre-treatment in CD4 (1.7, p=0.005; and 2.0, p=0.005, respectively), CD8 (1.8, p=0.005; and 1.8, p=0.005, respectively) and CD14 cells (1.3, p=0.03; and 1.4, p=0.02, respectively), but over-represented at post-treatment in CD19 cells (1.4, p=0.03; and 1.9, p<0.001, respectively).

Conclusion:  This study has identified several pathways that are altered during the first year of treatment of JDM. In addition to confirming involvement of the type 1 interferon pathway in active JDM disease, the purified subset approach enables identification of pathways that differ between immune subsets. Ongoing analyses are comparing pre- and post-treatment samples to age- and sex-matched healthy controls and investigating pathways associated with different clinical courses.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/transcriptomic-analysis-of-immune-subsets-in-juvenile-dermatomyositis-before-and-after-treatment-identifies-novel-pathways-involved-in-pathogenesis/