Mapping molecular subtype specific alterations in breast cancer brain metastases identifies clinically relevant vulnerabilities

Nicola Cosgrove, Damir Varešlija, Stephen Keelan, Ashuvinee Elangovan, Jennifer M. Atkinson, Sinéad Cocchiglia, Fiona T. Bane, Vikrant Singh, Simon Furney, Chunling Hu, Jodi M. Carter, Steven N. Hart, Siddhartha Yadav, Matthew P. Goetz, Arnold D.K. Hill, Steffi Oesterreich, Adrian V. Lee, Fergus J. Couch, Leonie S. Young

Research output: Contribution to journalArticlepeer-review

Abstract

The molecular events and transcriptional plasticity driving brain metastasis in clinically relevant breast tumor subtypes has not been determined. Here we comprehensively dissect genomic, transcriptomic and clinical data in patient-matched longitudinal tumor samples, and unravel distinct transcriptional programs enriched in brain metastasis. We report on subtype specific hub genes and functional processes, central to disease-affected networks in brain metastasis. Importantly, in luminal brain metastases we identify homologous recombination deficiency operative in transcriptomic and genomic data with recurrent breast mutational signatures A, F and K, associated with mismatch repair defects, TP53 mutations and homologous recombination deficiency (HRD) respectively. Utilizing PARP inhibition in patient-derived brain metastatic tumor explants we functionally validate HRD as a key vulnerability. Here, we demonstrate a functionally relevant HRD evident at genomic and transcriptomic levels pointing to genomic instability in breast cancer brain metastasis which is of potential translational significance.

Original languageEnglish (US)
Article number514
JournalNature communications
Volume13
Issue number1
DOIs
StatePublished - Dec 2022

ASJC Scopus subject areas

  • Chemistry(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Physics and Astronomy(all)

Fingerprint

Dive into the research topics of 'Mapping molecular subtype specific alterations in breast cancer brain metastases identifies clinically relevant vulnerabilities'. Together they form a unique fingerprint.

Cite this