Abstract PD05-06: Next-generation RNA-sequencing of triple negative breast cancer compared to donated microdissected normal epithelium and adjacent normal tissues

Abstract

Abstract Background: There currently exists an inadequate understanding of the transcriptional dysregulation that differentiates Triple-Negative Breast Cancer (TNBC) from normal breast epithelium. Currently, a popular comparator in gene expression studies is to use normal tissue adjacent to the primary tumor. Using RNA-seq data of TNBCs and donated microdissected normal breast epithelium from our institution combined with TNBCs and adjacent normal tissues from the The Cancer Genome Atlas (TCGA), we sought to determine the differential regulation of genes between TNBC compared to normal breast epithelium and adjacent normal tissues. Methods: Normal breast tissues from healthy pre-menopausal volunteers with no history of disease were procured from the Susan G. Komen for the Cure® Tissue Bank at the IU Simon Cancer Center. To eliminate bias from stromal tissue, normal tissues were laser capture microdissected for ductal epithelium. RNA from 20 microdissected normal breast tissues and 10 TNBCs from our institution were sequenced on a Life Technologies SOLiD sequencer. This data was combined with RNA-seq data from an additional 77 TNBCs and 10 adjacent normal tissues from the TCGA Data Portal. The merged gene expression values (RPKM) were quantile normalized, batch effect corrected, and submitted for differential gene expression analysis using Partek Genomics Suite. Differentially expressed genes were statistically determined with cutoffs of FDR<5% and Fold Change>±2. Pathway and network analyses utilized Ingenuity Pathway Analysis. Results: Principal components analysis (PCA) of all expressed protein coding genes revealed a close association of adjacent normal tissues to TNBCs whereas microdissected normal breast tissues were starkly separated. In comparing these 3 tissue types, 3841 genes were differentially expressed between TNBC vs. donated microdissected normal of which 1627 genes were also differentially expressed between adjacent normal vs. donated microdissected normal. The PCA along with the differential expression support the notion that adjacent normal gene expression is strongly influenced by the primary tumor, and in a significant set of genes, adjacent normal gene expression mimics the expression of primary tumors. The differential gene expression analysis also revealed 1240 genes that were restricted solely to the TNBC vs. donated microdissected normal that were not detected when TNBCs were compared to adjacent normal tissues. Examination of these genes reveals several potential kinase drug targets, genes involved in notch signaling, metabolism, cell cycle among other key cancer pathways. Preliminary analysis of the non-coding RNA component also reveals the same pattern of dysregulation in cancer associated microRNAs and lincRNAs. Conclusions: Gene expression within adjacent normal tissues is significantly impacted by the primary tumor and donated microdissected normal breast epithelium from healthy volunteers may be an optimal comparator for use in next-generation RNA sequencing breast cancer studies. Use of this comparator may identify therapeutic targets that would not otherwise be detected when using adjacent normal tissue as controls. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr PD05-06.