Login / Signup

Genomic Landscape of Angiosarcoma: A Targeted and Immunotherapy Biomarker Analysis.

Andrea P Espejo-FreireAndrew ElliottAndrew RosenbergPhilippos Apolinario CostaPriscila Barreto-CoelhoEmily JonczakGina D'AmatoTy SubhawongJunaid ArshadJulio A Diaz-PerezWilliam M KornMatthew J OberleyDaniel MageeDon DizonMargaret von MehrenMoh'd M KhushmanAtif Mahmoud HusseinKirsten LeuJonathan C Trent
Published in: Cancers (2021)
We performed a retrospective analysis of angiosarcoma (AS) genomic biomarkers and their associations with the site of origin in a cohort of 143 cases. Primary sites were head and neck (31%), breast (22%), extremity (11%), viscera (20%), skin at other locations (8%), and unknown (9%). All cases had Next Generation Sequencing (NGS) data with a 592 gene panel, and 53 cases had Whole Exome Sequencing (WES) data, which we used to study the microenvironment phenotype. The immunotherapy (IO) response biomarkers Tumor Mutation Burden (TMB), Microsatellite Instability (MSI), and PD-L1 status were the most frequently encountered alteration, present in 36.4% of the cohort and 65% of head and neck AS (H/N-AS) (p < 0.0001). In H/N-AS, TMB-High was seen in 63.4% of cases (p < 0.0001) and PDL-1 positivity in 33% of cases. The most common genetic alterations were TP53 (29%), MYC amplification (23%), ARID1A (17%), POT1 (16%), and ATRX (13%). H/N-AS cases had predominantly mutations in TP53 (50.0%, p = 0.0004), POT1 (40.5%, p < 0.0001), and ARID1A (33.3%, p = 0.5875). In breast AS, leading alterations were MYC amplification (63.3%, p < 0.0001), HRAS (16.1%, p = 0.0377), and PIK3CA (16.1%, p = 0.2352). At other sites, conclusions are difficult to generate due to the small number of cases. A microenvironment with a high immune signature, previously associated with IO response, was evenly distributed in 13% of the cases at different primary sites. Our findings can facilitate the design and optimization of therapeutic strategies for AS.
Keyphrases
  • copy number
  • stem cells
  • gene expression
  • machine learning
  • nucleic acid
  • genome wide
  • cancer therapy
  • deep learning
  • soft tissue
  • big data
  • protein kinase
  • artificial intelligence
  • data analysis
  • cell free