Abstract 4563: Tumor fraction estimation and tissue copy number inference using copy number signal from a liquid biopsy assay

Purpose: Estimating tumor fraction and inferring copy number alterations (CNAs) from cell-free DNA (cfDNA) liquid biopsy is challenging due to SNV heterogeneity and occasional lack of SNV signal. About 90% of solid tumors exhibit chromosome arm level gains or losses [1-2], and this effect on copy number could potentially be used as a tumor load biomarker. This study aims to use aneuploidy signal in cfDNA to estimate tumor fraction and infer tissue copy number, addressing a limitation of variant-based cfDNA assays and providing a robust method for benchmarking against gold-standard techniques like FISH. Methods: We employed Northstar Select for copy number quantification, BillionToOne’s treatment-selection liquid biopsy assay, which is optimized for accurate plasma copy number measurement of genes and is resistant to technical and biological noise. A Gaussian mixture model (GMM) was applied to quantify liquid aneuploidy signal on measured genes. The periodicity pattern was used to estimate tumor fraction independently of SNV signal. From a databank of 3180 clinical samples, we analyzed 92 clinical samples (from 20 patients) for concordance with SNV-based methods over multiple timepoints, and 136 clinical samples (from 128 patients) with homozygous copy number loss calls for concordance with the tissue copy number estimate. Results: Aneuploidy-based tumor fraction estimates showed robustness above 2% tumor fraction. GMM tumor fraction estimates demonstrated concordance with samples exhibiting clear SNV alterations (Pearson r = 0.65 across multiple accessions and patients), and the correlation can be as good as r = 0.96 in samples from patients with longitudinally trackable SNVs. In the retrospective analysis, 141 homozygous copy number losses were called by Northstar Select assay, and among those, aneuploidy-based tissue copy number estimation identified homozygous losses in 126 samples (89%). The algorithm also identified additional homozygous losses (both inside of and outside of the reportable gene set of Northstar Select). Analysis of serially tested clinical plasma samples revealed that—despite large tumor fraction fluctuations (≤5% to ≥50%)—tissue copy number estimates for the same patient remained stable. Conclusions: Our aneuploidy-based approach provides a reliable alternative to SNV-based methods for tumor fraction estimation and tissue copy number inference from liquid biopsy data. These findings suggest potential improvements in the utility of liquid biopsy for cancer diagnostics and monitoring, particularly in cases with low/mixed SNV signals. The ability to accurately infer tissue copy number is crucial for predicting amplification levels, identifying driver mutations, and anticipating therapeutic responses. References: 1. Ben-David, U., Amon, A. Nat Rev Genet 21, 44-62 (2020). 2. Sansregret, L., Bakhoum, S.F., et al. Genome Medicine, 13(1), 93 (2021). Citation Format: Patrick D. Cherry, David Tsao, Wen Zhou. Tumor fraction estimation and tissue copy number inference using copy number signal from a liquid biopsy assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4563.