Protein network map uncovers previously unknown cancer drivers

Reviewed by Emily Henderson, B.Sc.Sep 30 2021

In three studies from the Cancer Cell Map Initiative, researchers uncovered previously unknown interactions between proteins that drive cancer and combined this new data to generate a map of protein pathways informing cancer outcomes.

Their approach offers a framework that could improve scientists' understanding of cancer progression and aid in identifying therapeutic targets. For many cancers, which are genetic diseases, there is an extensive catalog of genetic mutations. But a consolidated map that organizes these mutations into pathways that drive tumor growth is missing. "A clearer picture would emerge if mechanisms critical for tumor growth were better consolidated into specific pathways," write Ran Cheng and Peter K. Jackson, in a related Perspective. "Identifying and consolidating these pathways and identifying how combinations of pathways drive cancer will simplify our search for effective cancer therapies."

Protein-protein interaction (PPI) networks are important tools in this effort because they extend beyond gene lists to define the protein biochemistry of tumor pathways and druggable targets. Danielle Swaney et al. studied protein-protein interaction data in head and neck squamous cells. They report that these cancer lines showed hundreds of distinct interactions compared with both noncancerous lines and with each other. Interactions with the PI3K pathway – commonly mutated in tumors – were predictive of drug response, Swaney and colleagues reported.

Minkyu Kim et al. focused on breast cancer and identified two proteins connected to the tumor suppressor gene BRCA1, as well as two proteins that regulate PIK3CA. Fan Zheng et al. combined the new PPI data from Swaney et al. and Kim et al. with existing public data to generate a map of protein pathways that they used to reveal previously hard-to-detect mutations potentially important in tumor metastasis. The studies provide a resource that will be helpful in interpreting cancer genomic data.