Abstract 5693: Low coverage genome-wide sequencing of cell-free DNA enables detection, therapeutic monitoring, and characterization of copy number alterations in patients with solid tumors

Abstract

Abstract Circulating tumor DNA (ctDNA) has been detected in the blood of patients with many cancer types and is thought to reflect the combined genetic profile of the primary tumor and metastases. The majority of studies into “liquid biopsies” have focused on detecting focal events including single nucleotide variants, fusions, and specific copy number alterations (CNAs); however, these assays are limited to mutations occurring in targeted regions and require deep sequencing. The objective of this study was to use low coverage, genome-wide sequencing of cell-free DNA (cfDNA) to evaluate the frequency, distribution, and persistence of genome-wide CNAs in the plasma of patients with solid tumors. cfDNA was isolated from a total of 203 plasma aliquots from 72 patients diagnosed with a variety of solid tumors. For almost half of the patients, multiple collections were available (range: 2-13) to enable longitudinal monitoring of treatment response. Greater than 25% of baseline cfDNA samples showed evidence of one or more CNAs and greater than 40% of patients had a CNA detected in at least one sample. Among patients with longitudinal monitoring, greater than 35% of individuals showed evidence of CNAs at baseline and approximately 25% of samples showed similar evidence in subsequent sampling. Collectively, CNAs were detected on all autosomes, suggesting that genome-wide profiling is necessary to fully assess CNA distributions. Additionally, we devised a metric which quantified the level of genomic instability. This metric, termed the genomic instability number (GIN), was associated with disease classification. In patients capable of longitudinal monitoring, GIN values were observed concomitant with therapeutic response (including, interestingly, response to immunotherapy) or disease progression and may provide information earlier than standard imaging-based methods. In addition, total cfDNA amount was evaluated as a predictor. When analysis was performed across all patients, there was a poor correlation between total cfDNA amount and the GIN; however, total cfDNA amount was strongly correlated with the cumulative abundance of CNA events within individual patients, suggesting that the CNAs detected in the cfDNA were derived from the additional genetic material contributed by the tumor. These data demonstrate the feasibility of detecting ctDNA in a significant portion of cancer patients solely on the basis of low coverage CNA profiling. Further, the nature of CNA events observed varied both between patients and across longitudinal samples from individual patients, suggesting potential for monitoring therapy response or disease progression. This study suggests that genome-wide CNA profiling may be useful alone or as a complement to standard variant detection strategies in the development of liquid biopsy technologies. Citation Format: Christopher K. Ellison, Aaron M. Goodman, Sabine Riethdorf, Amin R. Mazloom, Lisa Tran, Prachi Nakashe, Erin McCarthy, Tobias Gorges, Alexander Stein, Julia Quidde, Daniel S. Grosu, Mathias Ehrich, Taylor J. Jensen, Klaus Pantel, Razelle Kurzrock. Low coverage genome-wide sequencing of cell-free DNA enables detection, therapeutic monitoring, and characterization of copy number alterations in patients with solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5693. doi:10.1158/1538-7445.AM2017-5693