Abstract 1183: Targeted sequencing of cell-free DNA data enables comprehensive profiling of tumor copy number landscape from blood

Abstract

Abstract Background: The evolution of cancer genomes within a single tumor creates mixed cell populations with divergent somatic mutational architectures. Inference of tumor subpopulations has been disproportionately focused on the assessment of somatic point mutations, whereas computational methods targeting copy number alterations (CNA) in targeted sequencing data remain underdeveloped. Methods: Using a large database of clinical cell-free DNA (cfDNA) sequencing data (Guardant Health, CA), we developed a coverage-based probabilistic model to simultaneously normalize molecular coverage, segment the genome, predict copy number alterations, and estimate the tumor content in cfDNA samples, while accounting for mixtures of cell populations. This model was technically validated using tissue fluorescence in situ hybridization (FISH) data and then applied to a unique set of cfDNA sequencing data from >5,000 normal and clinical samples spanning multiple cancer types, where model predictions were compared to the observed allelic frequencies of somatic driver mutations and heterozygous germline SNPs. Results: Technical validation against FISH-derived tissue copy number estimates demonstrated high concordance with model estimates. Analysis of clinical samples demonstrated a wide range of copy number architectures, including prevalent copy number neutral loss-of-heterozygosity, large chromosomal deletions, and high focal amplifications, all of which are not easily detected and/or differentiated with standard sequencing analysis approaches on highly-fragmented cfDNA. Conclusion: Our results show that probabilistic modeling of coverage data generated from targeted cfDNA sequencing can detect and differentiate heterogeneous tumor populations with diverse somatic variations, CNA, and LOH landscape. This method may enable improvements in CNA detection accuracy, sensitivity, and specificity and provides more precise interrogation of tumor fraction and clonal architecture. Citation Format: Catalin Barbacioru, Eric Collisson, Darya Chudova, Justin Odegaard, Richard Lanman, AmirAli Talasaz. Targeted sequencing of cell-free DNA data enables comprehensive profiling of tumor copy number landscape from blood [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1183.