The molecular signature that defines tumor microvasculature will likely provide clues as to how vascular-dependent tumor proliferation is regulated. Using purified endothelial cells, we generated a database of gene expression changes accompanying vascular proliferation in invasive breast cancer. In contrast to normal mammary vasculature, invasive breast cancer vasculature expresses extracellular matrix and surface proteins characteristic of proliferating and migrating endothelial cells. We define and validate the up-regulated expression of VE-cadherin and osteonectin in breast tumor vasculature. In contrast to other tumor types, invasive breast cancer vasculature induced a high expression level of specific transcription factors, including SNAIL1 and HEYL, that may drive gene expression changes necessary for breast tumor neovascularization. We demonstrate the expression of HEYL in tumor endothelial cells and additionally establish the ability of HEYL to both induce proliferation and attenuate programmed cell death of primary endothelial cells in vitro. We also establish that an additional intracellular protein and previously defined metastasis-associated gene, PRL3, appears to be expressed predominately in the vasculature of invasive breast cancers and is able to enhance the migration of endothelial cells in vitro. Together, our results provide unique insights into vascular regulation in breast tumors and suggest specific roles for genes in driving tumor angiogenesis.

Detection of hallmark genomic aberrations in acute myeloid leukemia (AML) is essential for diagnostic subtyping, prognosis and patient management. However, cytogenetic/cytogenomic techniques used to identify those aberrations, such as karyotyping, fluorescence in situ hybridization (FISH) or chromosomal microarray analysis (CMA), are limited by the need for skilled personnel as well as significant time, cost and labor. Optical genome mapping (OGM) provides in a single, cost-effective assay significantly higher resolution than karyotyping with comprehensive genome-wide analysis comparable to CMA and the added unique ability to detect balanced structural variants (SVs). Here, we report in a real-world setting the performance of OGM in a cohort of 100 AML cases, which were previously characterized by karyotype alone or karyotype and FISH or CMA. OGM identified all clinically relevant SVs and copy number variants (CNVs) reported by these standard cytogenetic methods when representative clones were present in >5% allelic fraction. Importantly, OGM identified clinically relevant information in 13% of cases that had been missed by the routine methods. Three cases reported with normal karyotypes were shown to have cryptic translocations involving gene fusions. In 4% of cases, OGM findings would have altered recommended clinical management and in an additional 8%, OGM would have rendered the cases potentially eligible for clinical trials. The results from this multi-institutional study indicate that OGM effectively recovers clinically relevant SVs and CNVs found by standard of care methods and reveals additional SVs not reported. Furthermore, OGM minimizes the need for labor-intensive multiple cytogenetic tests while concomitantly maximizing diagnostic detection through a standardized workflow.

Detection of hallmark genomic aberrations in acute myeloid leukemia (AML) is essential for prognosis and patient management. Clinical practice guidelines for identifying such structural variants (SVs), established by the World Health Organization (WHO), European Leukemia Net (ELN) and National Comprehensive Cancer Network (NCCN), rely substantially on cytogenetic/cytogenomic techniques such as karyotyping, fluorescence in situ hybridization (FISH) or chromosomal microarray analysis (CMA). However, these techniques are limited by the need for skilled personnel as well as significant time and labor, making them cost-prohibitive for some patients. Optical genome mapping (OGM) addresses these limitations and allows for the accurate identification of clinically significant SVs using a novel, high throughput, inexpensive methodology. In a single assay, OGM offers a significantly higher resolution than karyotyping with comprehensive genome-wide analysis comparable to CMA and the added unique ability to detect balanced SVs that are missed by microarray. Here, we report the performance of OGM in a cohort of 100 AML cases, which were previously characterized by karyotype alone or karyotype and FISH. CMA was performed as an additional test in some cases. OGM identified all the clinically relevant SVs and CNVs reported by these standard cytogenetic methods. Moreover, OGM identified clinically relevant SVs in 11% of cases that had been missed by the routine methods. In 24% of cases, OGM refined the underlying genomic structure reported by traditional cytogenomic testing (13%), identified additional clinically relevant variants (7%) or both (4%). Three of 48 (6.25%) cases reported with normal karyotypes were shown to have cryptic translocations involving gene fusions. Two of these cases included fusion between NSD1-NUP98. Based on the comprehensive genomic profiling of the AML patients in this multi-institutional study, we recommend that OGM be considered as a first-line test for detection and identification of clinically relevant SVs.

Genomic structural variants comprise a significant fraction of somatic mutations driving cancer onset and progression. However, such variants are not readily revealed by standard next-generation sequencing. Optical genome mapping (OGM) surpasses short-read sequencing in detecting large (>500 bp) and complex structural variants (SVs) but requires isolation of ultra-high-molecular-weight DNA from the tissue of interest. We have successfully applied a protocol involving a paramagnetic nanobind disc to a wide range of solid tumors. Using as little as 6.5 mg of input tumor tissue, we show successful extraction of high-molecular-weight genomic DNA that provides a high genomic map rate and effective coverage by optical mapping. We demonstrate the system’s utility in identifying somatic SVs affecting functional and cancer-related genes for each sample. Duplicate/triplicate analysis of select samples shows intra-sample reliability but also intra-sample heterogeneity. We also demonstrate that simply filtering SVs based on a GRCh38 human control database provides high positive and negative predictive values for true somatic variants. Our results indicate that the solid tissue DNA extraction protocol, OGM and SV analysis can be applied to a wide variety of solid tumors to capture SVs across the entire genome with functional importance in cancer prognosis and treatment.