Abstract 4843: A next generation sequencing (NGS) genome wide copy number variation (CNV) assay for comparison of circulating tumor cell (CTC) heterogeneity

Abstract

Abstract Background: Analysis of CTCs from patient blood represents a non-invasive method for real time monitoring of disease status. Often, targeted therapies require prior knowledge of a patient's somatic mutational status and overall tumor heterogeneity at therapeutic decision points for optimum efficacy. CTCs are a product of tumor metastasis and their heterogeneity is reflective of tumor clonal evolution in response to treatment decisions and/or sequence. We sought to develop a method to characterize genomic heterogeneity of CTCs by NGS mediated CNV analysis. Typically, CTC detection platforms enrich CTCs by epithelial cell surface markers or physical properties in pools of CTCs. The loss of cells without typical CTC characteristics and the lack of single cell resolution may bias clonal evolution monitoring. The Epic Sciences CTC platform is a non-selective method for the detection and characterization of CTCs with single cell resolution, enabling individual phenotype to genotype comparisons. Characterization of CTC sub-populations at this level, prior to therapeutic decisions and in response to targeted therapy may represent a paradigm shift in the treatment of cancer. Methods: 2 healthy donor blood samples were spiked with 1 of 3 prostate cancer cell lines (n = 6). Spiked samples were processed using the Epic Sciences platform: ∼3M nucleated cells were deposited on replicate glass slides, Ab detection of CD45/CK/AR expression in conjunction with DAPI nuclear staining identified individual tumor cells. Cells were isolated, lysed and whole genome amplified. Barcoded shotgun libraries were constructed, sequenced (>4M reads/cell) in replicate runs (n = 2), reads were aligned to hg19, parsed into 0.1M bp windows, normalized for read count and germline CNV. Correlation of each cell's CNV vs. replicate cells determined the reproducibility of the method. Accuracy was confirmed by the detection of known genomic alterations in each cell line. Results: All cells processed yielded successful libraries with mean read lengths >150 bp, alignment of >94% and AQ20>80%. Cells from duplicate samples and sequencing runs demonstrated good correlation in CNV profile (r = 0.90-0.95) in comparison to their correlation with other cell lines (r = 0.62-0.88). LNCaP cells showed the lowest correlation to both VCaP (r = 0.62) and PC3 (r = 0.63), while higher correlation was observed between VCaP and PC3 (r = 0.88). Known AR amplification (VCaP), PTEN deletion (VCaP,LNCaP,PC3) and Y chr (PC3) null status were confirmed in all samples tested. Conclusions: The high level of correlation observed between replicate cells sequenced in separate runs and the detection of known PCa cell line CNV in all samples tested indicate the developed method is both reproducible and accurate. Further, the results demonstrate the utility of single CTC analysis for the purposes of disease monitoring. Citation Format: Stephanie Green, Mark Landers, Jessica Louw, Adam Jendrisak, Ryan Dittamore, Dena Marrinucci. A next generation sequencing (NGS) genome wide copy number variation (CNV) assay for comparison of circulating tumor cell (CTC) heterogeneity. [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 4843. doi:10.1158/1538-7445.AM2015-4843