Abstract 6625: LinkPrep™ - a comprehensive assay for cancer genome characterization - enabling structural variant detection and discovery of enhancer hijacking

Abstract Cancer arises from genetic and/or epigenetic disruptions. Currently, no single technology can capture the full range of variation that drives tumorigenesis with high sensitivity. Furthermore, epigenetic changes resulting from somatic variation remain poorly understood. The LinkPrep™ technology, a multi-purpose NGS assay, generates unbiased linked-read libraries through DNA proximity ligation that enable the discovery of structural variants (SVs) while simultaneously capturing 3D chromosome conformation linked to the tumor’s epigenetic state. Importantly, LinkPrep data provides uniform coverage, allowing for the detection of single nucleotide variants (SNVs), insertions/deletions (InDels), and copy number variants (CNVs) with similar performance to whole genome shotgun sequencing (WGS). Libraries are generated from a single-day workflow and are compatible with Illumina sequencers. In a comparison study, a fresh-frozen, clinical ovarian serous adenocarcinoma sample was profiled with LinkPrep (30X), WGS (30X), and RNA-seq (100M reads). Additionally, the same sample stored in FFPE was profiled with Dovetail-FFPE (30X), a linked-read technology designed to extract long-range genomic interactions from FFPE tissue. LinkPrep & Dovetail-FFPE both identified three translocations (tlocs) at high SV-VAF (>30%) between chromosomes 1, 3, and 12. These tlocs were supported by strong evidence (LinkPrep: 589 - 3, 282 read pairs, Dovetail-FFPE: 417 - 5, 802 read pairs) and confirmed by copy number changes. The tlocs were further validated through PCR and Sanger sequencing. By comparison, RNA-Seq detected only one fusion consistent with tloc(1;12), and WGS identified two tlocs with low specificity (ranked 140 and 374 of 404 putative tlocs). Furthermore, LinkPrep detected a subclonal event present at SV-VAF of 4.5% at tloc (2;15). LinkPrep and WGS produced concordant variant calls that showed >95% overlap with the set of known variants. Similarly, copy number profiles between LinkPrep and WGS showed >98% similarity. To profile the promoter-enhancer genomic interactions at high coverage (200X), LinkPrep libraries were paired with the Dovetail® Pan Promoter Enrichment Panel. This revealed SV-driven neo-interactions, with interaction strength independent of SV VAF, suggesting biological mechanisms such as enhancer hijacking are at play. In summary, LinkPrep outperforms WGS and RNA-seq in SV discovery with superior sensitivity and specificity, while maintaining comparable performance to WGS in the detection of SNVs, InDels, and CNVs. Additionally, LinkPrep captures epigenetic information in the same library, providing insights into epigenetic changes and drivers. Taken together, LinkPrep is a powerful tool for comprehensive profiling of the cancer genome - capturing the whole range of genetic variants and the 3D genome in a single dataset. Citation Format: Ericca Stamper, Jon Torchia, Daniel Hwang, Chu Yuan Wu, Ek Danwanichakul, Cory Padilla, J. Zachary Sanborn, Lisa Munding. LinkPrep™ - a comprehensive assay for cancer genome characterization - enabling structural variant detection and discovery of enhancer hijacking [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 6625.