Abstract 4859: Development of a cancer transcriptome analysis toolkit: identification of gene fusions in chronic lymphocytic leukemia

Abstract

Abstract RNA-Seq has become an invaluable tool for improving the understanding of the biological mechanisms at work in various cancers, with tantalizing prospects for application in the area of clinical diagnostics and for the design and monitoring of therapy. Transcriptome data uniquely provides information on gene expression, and the presence of somatic expressed splice variants, point mutations, CNVs and gene fusions. Although diverse algorithms have been developed to identify specific classes of somatic alterations, an approach for streamlined, comprehensive and large-scale processing and mining of RNA-Seq data remains a formidable challenge. To address this challenge, we have developed the Cancer Transcriptome Analysis Toolkit (CTAT).The CTAT is an amalgamation of both existing and novel tools for RNAseq data analysis, wrapped in Galaxy as a user-friendly bioinformatics pipeline. We have tested CTAT on a RNA-seq dataset generated from 138 patients with chronic lymphocytic leukemia (CLL), whose malignant cells have been already extensively characterized at the DNA level by analysis of whole-exome sequencing (WES) and SNP array data. In addition, we have also analyzed RNA-seq data from CD19+ B cells isolated from 19 healthy adult volunteers. As expected, CTAT could detect the established features of CLL. We could reliably detect at the RNA-level previously identified mutated CLL driver genes as per WES analysis. Overexpression of CLL-associated genes (i.e. ZAP70, and CD38) was also detected. CTAT also could detect novel features of CLL. Focusing on gene fusions, an important class of genetic alteration across blood malignancies but which have not been well-characterized in CLL, we detected a total of 509 gene fusion events across the trio of fusion detection algorithms used, and 25 were detected unanimously. From an arbitrary subset of 40 fusions, we validated 38 by RT-PCR (95% validation rate). 450 of 509 (88%) of gene fusions involved intrachromosomal partners. After adjusting for gene density, CTAT detected the highest incidence of high confidence gene fusions on chr13 (30%). GSEA revealed that gene fusions appeared to cluster to known driver pathways of CLL, such as the BCR, NOTCH, NFKB, RNA processing and Apoptosis pathways. Specific gene fusions in our CLL patient cohort appeared to be predominantly private events and with variable predicted oncogenic potential. However, several potential oncogenic drivers (NCOR2–UBC, PHF3–PTP4A1, CYTIP–ERMN, G3BP2–CCNG2) each appeared in 5-10% of patients, targeting cell cycle control, NOTCH signaling and cell adhesion pathways, and their biological significance can now be investigated. Citation Format: Nikolaus D. Obholzer, Brian J. Haas, Dan-Avi Landau, Nathalie Pochet, Aviv Regev, Catherine Wu. Development of a cancer transcriptome analysis toolkit: identification of gene fusions in chronic lymphocytic leukemia. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4859. doi:10.1158/1538-7445.AM2015-4859