Abstract P5-04-17: From transcriptome meta-analysis to targeted therapies in triple negative breast cancer

Abstract

Abstract Triple-Negative Breast Cancer (TNBC) is a subtype of breast cancer that urgently requires the identification and approval of novel targeted therapies. Even for breast cancer subtypes that have approved targeted therapies such as tamoxifen in ER+ and herceptin in HER2+ patients, there are a proportion of patients that do not respond to these therapies or develop resistance and succumb to metastatic recurrence. Thus, there is a clinical need to identify patients who do not benefit from current standard therapies and developing new strategies for therapy for non-responsive patients across all breast cancer subtypes. We hypothesised a potential prognostic and drug discovery approach by meta-analysis of multiple global gene expression profiles of breast cancer studies to identify significantly over- and under-expressed genes that associate with clinical outcomes (metastatic and/or death events within five years). These genes were filtered through 3 methods to annotate their predicted functions (1). The third and most generalised method identified a 133-gene signature set that was prognostic in all subtypes of breast cancer. These signatures, particularly with the devised score calculated as the ratio of the average expression of the over-expressed genes to the under-expressed genes, were patented (2). Of the 133 genes, we selected a 21-gene list representing novel and new targets in breast cancer drug development. The 21 genes were selected through an unbiased ranking of least-published cancer and non-cancer studies in PubMed. This poster reports the characterisation of these 21 genes as drug targets in TNBC. A siRNA screen measuring proliferation and viability (xCELLigence RTCA) was performed (2 siRNAs per gene) on 3 TNBC cell lines (MDAMB231, SUM159PT and BT549) and MCF10A. The top gene hits that show a dependence on cancer proliferation and viability were followed up with clonogenic assays, cell cycle profiles and apoptosis assays. Parallel to these experiments, a negative-selection clustered regularly interspaced short palindromic repeats (CRISPR) screen involving the 21 genes (6 constructs per gene) is currently underway on MDAMB231 cell line in vitro and in vivo to identify cancer dependence on these genes for survival and metastasis.