Tumor profiling from whole-genome and whole transcriptome sequencing to uncover gene fusions and structural variations in clinically relevant cancer genes.

Abstract

e23118 Background: Current targeted cancer therapies rely on the identification of clinically relevant somatic alterations in the tumor. Hotspot gene-panels and exome sequencing are designed to quickly assess somatic variations in frequently mutated regions and/or the coding regions of relevant genes, but they have limited ability to detect complex genomic rearrangements or novel structural variations. Here, we describe an integrative and comprehensive approach to fully characterize the genomic complexity of solid tumors using high throughput whole genome sequencing (WGS) and whole transcriptome sequencing (RNA Seq). Methods: We performed WGS and high-depth sequencing of known cancer genes in 14 paired tumor-normal samples of a variety of tumor types. Tumor-specific somatic alteration assessments included protein-coding mutations, copy number variations, gene fusions and structural variants. In addition, RNA Seq data was analyzed to identify expressed somatic alterations. Results: We identified 2 novel fusion genes as well as important structural alterations which could have clinical and therapeutic implications. We described a novel BRAF fusion gene in a cholangiocarcinoma devoid of other known driver mutations. BRAF fusions have not been described previously in cholangiocarcinoma; this fusion may represent an alternative mechanism for MAPK activation and could be a useful drug target. We also identified a novel NTRK3 fusion partner in a glioblastoma tumor. This fusion may imply a novel mechanism for NTRK3 activation. Finally, we identified numerous tandem duplications in an ovarian cancer. Recent advances describe tandem duplication hotspots in ovarian cancer as a potential driver mechanism characterizing a specific mutational signature. Conclusions: Comprehensive genomics assessment of paired tumor-normal samples through whole-genome and transcriptome sequencing can yield additional clinically actionable genomic characteristics that may not be detected in whole-exome or hotspot gene-panel sequencing. These findings have the potential to aid in clinical decision making.