Online Program

Return to main conference page

All Times ET

Thursday, June 3
Practice and Applications
Data-Driven Healthcare
Thu, Jun 3, 1:10 PM - 2:45 PM
TBD
 

Creation of Breast Cancer Subtypes Through a Consensus-Based Network Approach (309812)

*Christina Horr, University of Notre Dame 

Keywords: clustering, breast cancer genomics, network analysis, biostatistics, bioinformatics

Breast cancer is a heterogeneous disease composed of multiple subtypes, each with its own distinct biological characteristics, therapy, and clinical outcomes. There are five well-known subtypes, referred to as PAM50. These subtypes were introduced in 2001 using unsupervised clustering on whole-genome expression data. Recently, studies have suggested that these subtypes are unstable, resulting in large numbers of patients whose classifications are uncertain. Currently, there are no other subtyping systems that challenge PAM50’s use. Hence, there is a need to address classification inconsistencies and uncertainties in these subtypes.

An existing subtyping method is applied to breast cancer gene expression data in which different clustering approaches on a common set of samples are coalesced into a network resulting in consensus-based classifications of samples. This method produces subtypes based on multiple clustering strategies that are integrated together, and therefore, will not be biased towards one unsupervised clustering algorithm, as in PAM50.

A single-sample classifier for the subtype system is applied to expression data from different technologies, resulting in clusters that have the same biological features across numerous cohorts. This suggests that these clusters are biologically driven as their characteristics are conserved in the clustering, regardless of the type of dataset. Results identify a cluster that was not included in the PAM50 subtypes. This cluster, C5, exhibits characteristics associated with epithelial mesenchymal transition (EMT), along with increased abundance of endothelial cells and fibroblasts. The introduction of this subtype could have important effects on the future of breast cancer, as well as promoting different types of treatments for patients whose tumors belong to this new subtype. To gain deeper insights of this potentially novel cluster, more biological analyses focusing on the C5 subtype will be needed.