Abstract 3479: Evaluation of subclonal deconvolution pipelines using reference cell-lines and patient plasma samples

Abstract Background: Intratumor heterogeneity plays a pivotal role in treatment resistance, emphasizing the need for a comprehensive understanding of tumor clonal architecture. While various methods exist for extracting this information from bulk tumor sequencing data, few studies have evaluated the entire workflow, from sample processing to subclone calls. Methods: To address this gap, we utilized cell lines with known copy number changes and somatic mutations, constructing 12 ground truth sample sets with varying cellular proportions and tumor purities. These cell lines underwent NGS pipeline using PredicineWES+ with increased sequencing depth across ~600 cancer-related genes. Subclonal analysis workflows, including FACETS, PyClone, and ECLIPSE, were implemented to to assess variant assignment and cellular fractions. The subclone analysis was extended to 91 plasma samples from cancer patients. Results: Our findings indicate that subclonal decomposition effectively segregates mutations from distinct cell lines. Notably, analyzing multiple samples with different proportions of the same subclones outperformed deconvolution from single mixture samples. In plasma samples from cancer patients, nearly one-third exhibited subclonal clinically actionable variants. Conclusions: The established subclonal deconvolution pipeline enhances biomarker analysis from assays utilizing WES with boosted sequencing depth for selected genes. Citation Format: Xiaoxi Dong, Yong Huang, Tom Zhang, Shidong Jia, Pan Du. Evaluation of subclonal deconvolution pipelines using reference cell-lines and patient plasma samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3479.