Abstract LB338: Genome-wide haplotype-specific copy number analysis from clinical targeted tumor sequencing data

Abstract Background: Integrating clinical genomic sequencing into the molecular profiling of patient tumors has transformed cancer research and precision oncology. Routine use of targeted sequencing panels, such as MSK-IMPACT and UW-OncoPlex, has enabled the detection of actionable somatic alterations across hundreds of cancer-associated genes. However, these assays often discard off-target sequencing reads representing untapped genomic information. Expanding the utility of these off-target reads to analyze germline haplotypes and allele-specific copy number alterations (CNAs) offers an opportunity to explore previously inaccessible genomic features, including non-coding regulatory regions. Current methods for detecting genome-wide CNAs, loss of heterozygosity (LOH), and homologous recombination deficiency (HRD) from targeted sequencing data are limited in their ability to leverage off-target information. In this study, we developed a haplotype-specific copy number caller optimized for targeted sequencing panels, utilizing both on- and off-target reads. Methods: We implemented a computational pipeline integrating both on-target and off-target sequencing reads from clinical genomic panels. Using CNVkit, read depths were calculated, normalized, and GC-corrected separately for on- and off-target regions. Haplotypes were phased and reconstructed using the statistical phasing algorithm, GLIMPSE, which utilizes population-level reference datasets, enabling the identification of haplotype blocks across the genome. We expanded the TitanCNA framework to analyze haplotype-specific copy number alterations (CNAs) by integrating read depth and haplotype information. Results: We applied our pipeline to analyze prostate and ovarian tumors sequenced using clinical targeted panels from the UW-OncoPlex platform. The pipeline’s performance was validated using matched 30x whole genome sequencing (WGS) data for 24 UW-oncoplex samples to ensure sensitivity, specificity, and accuracy in CNA and LOH detection. Loss of heterozygosity (LOH) events were identified through phased haplotype analysis, and genome-wide copy number signatures, such as HRD, were computed based on patterns of CNAs and LOH events. We also found that the pipeline could reliably determine CN signatures concordant with the matched WGS data and published reports. Impact: This comprehensive approach addresses current limitations in leveraging off-target reads and provides a robust framework for analyzing genome-wide alterations in targeted sequencing panels. With this approach, we aim to enhance the detection of genomic alterations, characterize germline haplotypes, and integrate these data with real-world clinical outcomes from diverse patient populations. Citation Format: Eden Cruikshank, Pooja Chandra, David Ma, Tomin Perea-Chamblee, Patricia Galipeau, Jian Carrot-Zhang, Gavin Ha. Genome-wide haplotype-specific copy number analysis from clinical targeted tumor sequencing data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB338.