Abstract 7177: Deep genomic analysis of undifferentiated sarcoma

Abstract Introduction: Undifferentiated sarcomas (USrc) are aggressive mesenchymal tumors that do not show characteristic features and therefore remain a diagnosis of exclusion. USrc typically have a low tumor mutational burden but are genomically complex, carrying many large structural variants (SVs) and copy number aberrations (CNAs). Improving diagnostic practice by adding novel whole genome-technologies based on high molecular weight DNA could increase the sensitivity for detecting genetic alterations in USrc, allowing identification of recurrent aberrations or patterns with clinical implications. Methods: Snap-frozen tissue from 41 USrc in 38 patients were analyzed by both conventional karyotyping and Optical Genome Mapping (OGM). Chromothripsis was determined based on OGM data and defined as more than eight copy number segments with oscillating copy number in co-occurrence with more than eight chromosomal rearrangements within 50 Mb, or more than 20 rearrangements within 50 Mb. These rearrangements should be interleaved and approximately half of them should be inversion-type aberrations. In addition, we identify known cancer genes (COSMIC Cancer Gene Census) which are amplified/lost or show evidence of disruptive SVs. DNA methylation profiling using EPIC arrays was done on 15 USrc followed by sarcoma classification with the Deutsches Krebsforschungszentrum (DKFZ) sarcoma classifier (v12.3). Results: 32 out of 41 USrc karyotypes showed at least subclonal whole genome-doubling with complex patterns of SVs and CNAs, including double minutes, ring chromosomes and giant markers. OGM was performed on all USrc biopsies yielding an average coverage of 352X. 25 out of 38 patient samples presented with chromothripsis with an average of 4.3 chromosomes affected (1-13 chromosomes). 70% of amplifications (CN greater than or equal to 5) and 83% of high, focal amplifications (CN greater than or equal to 10) occurred in regions with chromothripsis. Aberrations in 590 COSMIC cancer genes were detected. The most frequently altered cancer genes were CDKN2A (42%), RB1 (34%), PTPRD (29%), MGMT (26%), and ARHGEF12 (24%). For 3 patients, biopsies from two different time points were profiled, showing a high proportion of shared aberrations. DNA methylation patterns classified 8 out of 15 analyzed samples as USrc, and the remaining 7 showed no match to any reference sarcoma group. Copy number profiles derived from OGM and DNA methylation arrays were near identical. MGMT promotor hypermethylation was observed in 5 out of 16 cases. Conclusion: Implementation of OGM improves the detection of SVs and CNAs over standard-of-care. This is demonstrated by the detection of chromothripsis regions and novel recurrent aberrations in USrc. Further investigation to confirm these findings and to identify single nucleotide variants and allele-specific copy numbers are currently ongoing using long-read HiFi sequencing (PacBio). Citation Format: Jef Baelen, Raphael Sciot, Ruben Cools, Jonas Demeulemeester, Isabelle Vanden Bempt. Deep genomic analysis of undifferentiated sarcoma [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 7177.