Abstract 6064: Patient-partnered research enables germline characterization of angiosarcoma predisposition genes

Abstract

Abstract Background: Angiosarcoma (ASC) represents 1-2% of soft tissue sarcomas. Due to its rarity, to date, no large-scale studies have been done to systematically identify ASC-predisposition genes and the broader landscape of pathogenic germline variants in patients with ASC remains unclear. Methods: Through the Angiosarcoma Project, an ongoing patient-partnered research approach that allows patients living anywhere in the U.S. or Canada with angiosarcoma to participate through a website, blood and saliva samples were collected from more than 300 patients and whole-exome-sequencing was performed. Machine-learning frameworks were used to identify germline short variants (SNPs, indels) and rare germline copy number variants (CNVs). The pathogenicity of variants in 107 curated cancer-predisposition and sarcoma-related genes were evaluated according to the ACMG guidelines. A short-variant-based gene-burden analysis was then performed using 223 unrelated ASC patients and 4557 ancestry-matched unrelated cancer-free controls. Results: In total, 38 patients (17.04%) carried at least one pathogenic short variant, including 37 carrying germline heterozygous pathogenic variants in POT1 (9, 4.04%), BRCA2 (3, 1.35%), CHEK2 (3, 1.35%), MUTYH (3, 1.35%), BRCA1 (2, 0.90%), PRF1 (2, 0.90%), ERCC3 (1, 0.45%), TP53 (2, 0.90%), ATM (1, 0.45%), TSHR (1, 0.45%), TSC2 (1, 0.45%), SDHA (1, 0.45%), SBDS (1, 0.45%), MSH6 (1, 0.45%), MITF (1, 0.45%), MET (1, 0.45%), LZTR1 (1, 0.45%), FH (1, 0.45%), ERCC2 (1, 0.45%), XPC (1, 0.45%). In addition, one patient harbored pathogenic short variants in both BRCA1 and ERCC3. In the 8 unique CNVs detected in 10 (4.55%) patients, 6 were classified as pathogenic including 3 heterozygous duplications overlapping MSH2, LZTR1, SMARCB1, and 3 heterozygous deletions overlapping NF1, MSH2, and WRN. Two unique heterozygous duplications met VUS criteria including one overlapping ERCC2 and one overlapping SMARCB1. In a comparison between ASC patients and cancer-free controls of predominantly European ancestry, we found ASC patients tobe 40 times more likely to carry a pathogenic variant in POT1 (OR: 42.91, CI: [15.40-174.08], q < 0.001, p < 0.001), while TP53 was nominally significant (OR: 7.37, CI: [1.19 - 41.13], p = 0.037). Conclusion: We confirmed prior observations of POT1 being responsible for ASC and called for attention to investigate the effect of germline CNVs in ASC. Overall, the elucidation of various germline elements associated with ASC risk highlighted the value and importance of patient-partnered research, particularly for rare cancers, and the analysis support using germline testing of patients for more precise clinical management. Citation Format: Hoyin Chu, Marissa Hollyer, Seunghun Han, Sabrina Y. Camp, Riaz Gillani, Eliezer Van Allen, Nikhil Wagle, Corrie A. Painter, Saud H. AlDubayan. Patient-partnered research enables germline characterization of angiosarcoma predisposition genes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6064.