Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement
Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.
Gene Expression Profiling in Peripheral Blood Mononuclear Cells Predicts Disease Flares in Children with Polyarticular Juvenile Idiopathic Arthritis Who Have Achieved Clinical Remission on Medication.
Jarvis1, James N., Osban2, Jeanette, Chen1, Yanmin, Jiang1, Kaiyu, Armstrong1, Nicholas E., Frank2, Mark Barton, Knowlton2, Nicholas
Children with polyarticular juvenile idiopathic arthritis (JIA) patients pursue variable clinical courses, even after disease control has been achieved. Developing biomarkers to identify which children can come off therapy and which children need continued or more aggressive therapy has been elusive.
To better understand the underlying biology of disease flares in polyarticular, RF-negative juvenile idiopathic arthritis (JIA), and to determine the feasibility of developing gene array-based prognostic biomarkers to guide therapy in this disease.
We performed gene expression profiling and in silico pathway analysis on peripheral blood mononuclear cells (PBMC) of 17 children with RF-negative polyarticular JIA at the time they achieved inactive disease (ID) status. Of these children, 8 subsequently experienced a disease flare within 11 months of achieving ID status, while the remaining 9 experienced sustained disease control over observation periods ranging from 18 months to 6 years. Computational analyses were undertaken in an attempt to distinguish the 2 groups.
Gene expression profiling identified 103 genes that distinguished the Flare and No-flare groups. Gene expression profiling comparing children in ID/CRM who were fated to flare with those who were not ("baseline specimens") showed distinct differences between the two groups. The group fated to flare was already distinguishable from the group that remained stable at the time of baseline sampling. One hundred forty-four genes were differentially expressed between the two groups, all of which were over-expressed in children fated to flare. In silico modeling demonstrated that the differentially expressed genes could be incorporated into 4 large, linked networks. These networks incorporated genes associated with MAP kinase signaling, type 1 interferon production, and genes regulating the cell cycle. Furthermore, the gene expression profile remained largely stable in each group over the 515 month period over which the patients were followed. That is, comparison of the "flare" group at the time they achieved ID/CRM with the same patients at the time of their flare revealed only a single gene whose expression differed from the baseline sample. Similarly, children who remained stable also demonstrated a stable gene expression profile. In addition, network analysis suggests a role hepatocyte nuclear factor 4a (HNF4a), a transcription factor not previously identified in leukocytes, in maintaining treatment response in JIA.
Although the number of patients studied here is small, this work demonstrates the feasibility of developing prognostic biomarkers in JIA. These findings also cast light on the nature of the inflammatory networks that sustain the disease process in polyarticular JIA even when the illness clinically appears to be inactive.
To cite this abstract, please use the following information:
Jarvis, James N., Osban, Jeanette, Chen, Yanmin, Jiang, Kaiyu, Armstrong, Nicholas E., Frank, Mark Barton, et al; Gene Expression Profiling in Peripheral Blood Mononuclear Cells Predicts Disease Flares in Children with Polyarticular Juvenile Idiopathic Arthritis Who Have Achieved Clinical Remission on Medication. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :655