Rheumatoid arthritis (RA), a chronic autoimmune disease characterized by circulating autoantibodies, involves many cytokine-mediated signaling pathways in multiple immune cell subsets. Most studies of immune cells in RA have limitations, such as analysis of a small number of cell subsets or pathways, and limited longitudinal data on patient phenotypes. In this study, we used an innovative systems immunology approach to simultaneously quantify up to 882 signaling nodes (Jak/STAT signaling readouts modulated by cytokines and other stimuli) in 21 immune cell subsets. We studied 194 RA patients and 41 controls, including 146 well-characterized RA patients prior to, and 6 months after, initiation of methotrexate or biologic agents from the Treatment Efficacy and Toxicity in RA Database and Repository (TETRAD). There was strikingly attenuated signaling capacity in RA patients in IFNα stimulation followed by measurement of phosphorylated STAT1 [IFNα→p-STAT1] in six immune cell subsets. Multiple nodes showed negative association with disease activity, including IFNα→STAT5 signaling in naive and memory B cells. In contrast, IL-6-induced STAT1 and STAT3 activation in central memory CD4-negative T cells showed a positive association with disease activity. Multiple nodes were associated with treatment response, including IFNα→STAT1 in monocytes and IL-6→STAT3 in CD4+ naive T cells. Decision tree analysis identified a model combining these two nodes as a high-performing classifier of treatment response to TNF inhibitors. Our study provides novel information on RA disease mechanisms and serves as a framework for the discovery and validation of biomarkers of treatment response in RA.
ASJC Scopus subject areas
- Biochemistry, Genetics and Molecular Biology(all)
- Agricultural and Biological Sciences(all)
- Immunology and Microbiology(all)
- Pharmacology, Toxicology and Pharmaceutics(all)