Abstract LB-122: High-throughput tumor genomic profiling by massively parallel sequencing

Abstract

Abstract The use of tumor genetic and molecular information to accurately predict a patient's response to therapy is critical to personalized cancer medicine. Knowledge of somatic genetic alterations in tumors - including mutations, copy number alterations, and pharmacogenomic polymorphisms - should ultimately facilitate individualized approaches to cancer treatment. However, the development and real-world implementation of comprehensive genomic profiling has not yet been achieved. Here, we describe the development of a high-throughput, massively parallel sequencing platform to detect tumor genomic alterations that may provide an efficient and cost-effective means to address this challenge. Our approach combines several recent innovations - solution phase hybrid selection, DNA barcoding and pooling, and massively parallel sequencing - to achieve a high-throughput genomic profiling platform. We designed and synthesized ∼7000 biotinylated RNA baits corresponding to the coding sequence of the top 150 “druggable” or potentially “actionable” genes known to undergo somatic genomic alterations in cancer. We obtained genomic DNA from cell lines with known genetic alterations and used this DNA to generate barcoded sequencing libraries. Following quantification of libraries, equimolar pools were generated consisting of up to 12 barcoded tumor DNAs and normal diploid control DNAs. These DNA pools were subjected to solution-phase hybrid capture with the biotinylated RNA baits followed by massively parallel sequencing. The sequencing data were deconvoluted to match all high-quality reads with the corresponding tumor samples and call base mutations and copy number alterations. In all samples tested, sequence reads were highly specific to the targeted exonic regions. In addition to detecting mutations, exon capture detected high-level amplifications and deletions, two types of “actionable” genomic alterations that are poorly detectable by mass spectrometric approaches to mutation profiling. Comparison with copy number data previously obtained using high-density SNP array data demonstrated a robust correlation. As a proof-of-principle, we performed the entire profiling approach on clinical tumor specimens from patients with refractory, aggressive cancers to assess the use of genomic profiling in a real-world situation where knowledge of “driver” genetic alterations might prove valuable. Altogether, our data endorse this sequencing-based approach as a promising method to detect critical “actionable” genetic alterations in a large panel of cancer genes. The platform is scalable, refractory to contaminating stromal DNA or hyperploidy, and completed at a fraction of the cost of comparable sequencing approaches. By providing a rapid, sensitive, cost-effective means to sequence >100 cancer genes deeply and simultaneously, this approach may ultimately empower the rational selection of specific drugs targeting the genetic alterations in each patient's tumor - clearly an unmet need in oncology. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr LB-122.