Signaling Network Assessment of Mutations and Copy Number Variations Predicts Breast Cancer Subtype-specific Drug Targets

TL;DR

This study integrates genomic and functional data onto signaling networks to identify subtype-specific mechanisms and predict targeted therapies in breast cancer, validated through experiments.

Contribution

It introduces a network-based approach combining sequencing and RNAi data to uncover subtype-specific signaling mechanisms and predict drug targets in breast cancer.

Findings

Identified two subtype-specific signaling networks in breast cancer.

Network genes can classify tumors into subtypes based on genomic alterations.

Predicted drug targets were experimentally validated.

Abstract

Individual cancer cells carry a bewildering number of distinct genomic alterations i.e., copy number variations and mutations, making it a challenge to uncover genomic-driven mechanisms governing tumorigenesis. Here we performed exome-sequencing on several breast cancer cell lines which represent two subtypes, luminal and basal. We integrated this sequencing data, and functional RNAi screening data (i.e., for identifying genes which are essential for cell proliferation and survival), onto a human signaling network. Two subtype-specific networks were identified, which potentially represent core-signaling mechanisms underlying tumorigenesis. Within both networks, we found that genes were differentially affected in different cell lines; i.e., in some cell lines a gene was identified through RNAi screening whereas in others it was genomically altered. Interestingly, we found that highly connected network genes could be used to correctly classify breast tumors into subtypes based on genomic alterations. Further, the networks effectively predicted subtype-specific drug targets, which were experimentally validated.