Prediction of Lupus Outcome by Gene Expression Patterns

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Support is sought for performing mechanistic studies to define the gene expression profiles of glomeruli isolated from biopsy sections of lupus glomerulonephritis with the overall goals of determining whether the glomerular gene expression phenotype will predict outcome and efficacy in an ongoing parent trial. The parent trial compares administration of CellCept versus IV Cytoxan for initiating control of biopsy-proven lupus nephritis. We have demonstrated the feasibility of studying gene expression profiles by microarray analysis in glomeruli isolated from frozen biopsy sections by laser microdissection t6 characterize the molecular pathologic mechanisms leading to lupus nephritis. This work revealed considerable heterogeneity in gene expression patterns in samples classified as proliferative glomerulonephritis, suggesting the feasibility of lupus renal biopsy subclassification by gene expression criteria. The expressed genes formed 8 main clusters and the presence or absence of genes comprising these clusters in a given sample divided the biopsies into 3 distinct types. The hypothesis underlying the proposed studies is that differences in molecular pathologic mechanisms revealed by the various transcriptional phenotypes will predict the heterogeneous natural history and therapeutic outcome of lupus. The first aim of the proposed mechanistic studies is to cluster glomerular gene expression patterns in diagnostic renal biopsies performed in the current trial and use them to extend understanding of pathways involved in the molecular pathogenesis of lupus glomerulitis. Parallel quantitative PCR for selected genes found through the microarray assessment will be performed to validate the patterns found and determine if their differential expression can be used as a surrogate. The second aim will correlate the clusters and pathways with conventional pathologic features to identify the molecular basis of the pathologic findings. The third aim will determine whether particular outcomes of the trial could be predicted by the gene expression phenotype of the initial renal biopsy. In particular we will address, first, whether one gene expression type, characterized by apoptosis, TNF signaling and fibrosis, is highly correlated with poor outcome and thus predict the subset of non-responders to Cytoxan or CellCept. Second, whether CellCept will prove to be efficacious in a different gene expression subset, suggesting it could be used in these cases as a less toxic alternative to Cytoxan.