(P 244) MUTATIONAL BURDEN OF COPY NUMBER VARIANTS VS STRUCTURAL VARIANTS IN EXTRASKELETAL MYXOID CHONDROSARCOMA: A CASE STUDY

Objective: Extraskeletal myxoid chondrosarcoma (EMC) is rare sarcoma subtype that typically presents in the proximal extremity soft tissue in middle-aged males. Though it has an indolent course, it has high rates of recurrence and metastasis to the lungs. EMC is generally unresponsive to chemo- or radiation therapy with the standard of care being wide surgical resection. It is in most cases characterized by the translocation of the NR4A3 gene on chromosome 9 to the EWSR1 N-terminal transactivation domain on chromosome 22 (70% of cases), but NR4A3 can also fuse with TAF15 (27%); FUS, TCF12, and TGF ( < 5%). Our lab presents an extremely rare primary EMC sample collected from a patient with matching lung metastasis and peripheral blood samples. We performed Whole Genome Sequencing to identify the NR4A3-EWSR1 gene fusion in the tumor samples. We analyzed copy number variants (CNVs) and structural variants (SVs) to identify other genomic alterations that may either lend the tumor pro-cancer properties or may be targeted for therapy.
Our genomic characterization of a wide spectrum of SVs from single nucleotide polymorphisms to tandem duplications, inversions, translocations, and other complex events will add to the currently extremely limited literature of genomic variation present in EMC. We hypothesize that the lung metastasis (LM) will have a higher mutational burden than the primary tumor (PT), particularly in the larger scale SVs.

Methods: DNA was extracted from the primary EMC, lung metastasis, and peripheral blood mononuclear cells utilizing the QIAamp Fast DNA Tissue Kit (Qiagen, Cat#51404). DNA concentrations were determined with the Qubit 3.0 Fluorometer and DNA Broad Range Kit (ThermoFisher, Cat#Q32580). Samples were then submitted to our institution for further preparation, sequencing, and analysis. Libraries were prepared via the Illumina Nextera DNA Flex Library kit (Illumina, Cat# 20018704) and samples were sequenced on the Illumina NovaSeq 6000. For sequencing, 30X coverage was used to obtain greater than 1.6 billion, 150 bp paired-end reads for each sample. For analysis, the DRAGEN genome pipeline (Edico Genome) was used to map reads to the GRCh38 genome assembly. CNVkit was used to identify CNVs. Delly 0.7.7 was used to identify SVs. Sequencing data from patient PBMCs served as a control to filter out normal somatic and germline variants in the tumors. Data was visualized using SVCircos.

Results: CNVs in chromosomes 1 and 8 were increased in the primary tumor and lung metastasis compared with the blood (BN). CNVs in chromosomes 6 and 10 were decreased in the primary tumor and lung metastasis. Delly identified the NR4A3-EWSR1 EMC-defining translocation in both samples. Both samples had a high mutational burden of translocations. Compared with the primary tumor and normalizing by chromosome length, the lung metastasis had a higher proportion of inversions particularly in chromosomes 2, 8, 9, 12, and X. The lung metastasis also had a higher burden of deletions particularly in chromosomes 4, 10, and 19.

Conclusion: While the primary tumor and the lung metastasis had similar changes in CNVs across chromosomes, their SV mutational burden and composition was different. In particular, the lung metastasis had a greater proportion of inversion and deletion SVs compared with the primary tumor. Future directions will examine SVs conserved between the tumors, as well as SVs present in only the lung met, as these SVs may be advantageous for identifying anti-metastatic treatment strategies. If expanded beyond this case report, these SVs may be used to predict which patients presenting with EMCs are likely to metastasize and require a more aggressive approach.

LM vs BN Delly.pdf

PT vs BN Delly.pdf