Survival Outcomes are Associated with Genomic Instability in Luminal Breast Cancers

Abstract

Breast cancer is the leading cause of cancer related death among women. Breast cancers are generally diagnosed and treated based on clinical and histopathological features, along with subtype classification determined by the Prosigna Breast Cancer Prognostic Gene Signature Assay (also known as PAM50). Currently the copy number alteration (CNA) landscape of the tumour is not considered. We set out to examine the role of genomic instability (GI) in breast cancer survival since CNAs reflect GI and correlate with survival in other cancers. We focussed on the 70% of breast cancers classified as luminal and carried out a comprehensive survival and association analysis using Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) data to determine whether CNA burden quartiles derived from absolute CNA counts are associated with survival. Luminal A and B patients were stratified by PAM50 subtype and tumour grade and then tested for association with CNA burden using multiple statistical tests. Analysis revealed that patients diagnosed with luminal A grade 3 breast cancer have a CNA landscape associated with disease specific survival, suggesting that these patients could be classified as at-risk. Furthermore, luminal A grade 3 cases largely occupy a region of stratification based on gene expression at the boundary where luminal A and luminal B cases overlap. We conclude that GI reflected by absolute CNA score is a statistically robust prognostic factor for survival in luminal A grade 3 breast cancer. Therefore, luminal A grade 3 breast cancer patients in CNA burden quartiles 3 or 4 may benefit from more aggressive therapy. This demonstrates how individual genomic landscapes can facilitate personalisation of therapeutic interventions to optimise survival outcomes.