Abstract 4986: Kinome analysis of cancers by high-throughput transcriptome sequencing: Ideas in personalized medicine

Abstract

Abstract Targeted cancer therapy, which takes into account the intrinsic molecular heterogeneity of individual cases, is the ultimate goal of cancer genomics research. Despite intrinsic genetic complexity and heterogeneity, a few genes exert an overarching influence on the growth and sustenance of cancers, a phenomenon described as “oncogene addiction”. The most common class of addictive genes in cancers are protein kinases, and some of the most successful therapeutics include kinase inhibitors such as imatinib for BCR-ABL1 positive chronic myeloid leukemia, herceptin and lapatinib for ERBB2 positive breast cancers and gefitinib for lung cancers with EGFR mutations. Inhibitors of several other common driver kinases are under various stages of clinical use, trials or development. Unfortunately, targeting a primary driver almost always fails because, (1) Target patients are not well defined (2) cancers display de novo or acquired resistance to the inhibitors, and (3) there is an absence of additional/ alternate targets. We hypothesize that, in addition to the primary oncogenic driver, cancers harbor additional/ secondary drivers that are capable of providing escape from addiction to the primary driver following targeted therapy. Using high throughput sequencing, cancer samples can be interrogated individually to identify patient specific drivers/ therapeutic targets. We have used gene expression readout of transcriptome sequencing data from individual cancer samples to identify potential therapeutic targets showing outlier expression. In a proof of principle study focusing on kinase outliers, we nominated sample-specific, multiple outlier kinases in breast and pancreatic cancer and tested combinatorial targeting through siRNA knockdown experiments. In herceptin resistant, ERBB2 positive cancers, we observed an outlier expression (or mutation, or both) of other druggable kinases, such as FGFR4, RET, EGFR or MET, among others. Similarly, pancreatic cancer cell lines harboring oncogenic KRAS mutation were found to display outlier expression of AXL, MET, EPHA2, or MST1R and other potentially targetable kinases, in a sample specific expression profile. Based on these observations, we hypothesize that in addition to the primary aberration in every cancer sample multiple secondary drivers act through parallel/collaborating pathways. A combinatorial therapeutic strategy targeting the primary aberration together with sample specific secondary aberrations could preempt or overcome resistance to targeted therapy focused solely on the primary driver. Through an integrative analysis of high throughput transcriptome and exome sequencing of cancer samples, we could nominate personalized therapeutic candidates in real time. This could provide a novel paradigm for personalized medicine for cancers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4986. doi:10.1158/1538-7445.AM2011-4986