Abstract LB-409: Network analysis of RNA-seq data comparing triple negative breast cancer to normal breast tissues

Abstract

Abstract Introduction: Triple-negative breast cancer (TNBC) accounts for 15% of all breast cancers and results in disproportionally higher mortality compared to other breast cancers. Moreover, there is a paucity of therapies for this subtype of breast cancer resulting primarily from an inadequate understanding of the transcriptional differences that differentiate TNBC from normal breast. Using next-generation RNA-seq, we embarked on a study to compare the transcriptomes of TNBC and normal breast to comprehensively identify dysregulated networks, pathways, transcription factors, and high connectivity drug targets that can be exploited for therapeutic discovery. Methods: cDNA libraries from 10 normal breast tissues from the Susan G. Komen Tissue Bank at the IU Simon Cancer Center and 10 TNBC tumors were sequenced on an Applied Biosystems (ABI) SOLiD3 sequencer. Mapping of reads to the human genome (hg18) was performed using ABI BioScope 1.2; differential gene expression was analyzed using Partek Genomics Suite; and network, pathway, and transcription factor analysis was performed using Ingenuity Systems iReport and IPA 9.0. Results: Canonical pathway analysis revealed highly elevated and enriched expression in BRCA/DNA repair pathways, PI3K/AKT, and integrin signaling. Particularly, targeting BRCA/DNA repair pathways with PARP inhibitors has demonstrated some clinical success. Our pathway analysis reveals additional targets: ATM, CHK1/2, & PLK1 that are highly overexpressed and are integral to the BRCA pathway. To determine key regulators of our dataset, we employed IPA's transcription factor (TF) analysis that utilizes a literature knowledge base to identify TFs that explain a set of differential expression data, and classifies them as either activated or inhibited. We observed that TP53, RB1, & CDKN2A were significantly inhibited. This is congruent with current literature showing that TP53 & RB1 as the most predominantly mutated genes in TNBC. In terms of activation, TBX2 and c-MYC were significantly activated in our dataset, suggesting a highly pro-tumorigenic transcriptional programme. In addition, RNA-seq detects pre-miRNAs. When integrated into the analysis, the top ranked network reveals a large set of dysregulated miRNAs secondary to inhibited TP53. When looking at druggable kinases, we performed a connectivity analysis to determine major hub proteins. Our top three kinases were SRC, IKBKE, and AKT1. Of note AKT inhibitors have recently entered early phase clinical trials in breast cancer. Conclusions: Network analysis reveals a concerted programme of proteins and transcription factors that drive a highly pro-tumorigenic phenotype revolving around nonfunctional TP53, and upregulated PI3K/AKT & DNA damage repair pathways. The analysis reveals additional DNA damage repair proteins in addition to PARP and highly connected kinases as potential drug targets for TNBC. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-409. doi:1538-7445.AM2012-LB-409