Breast cancer subtyping from plasma proteins

Area of Science:

• Proteomics
• Cancer Biology
• Bioinformatics

Background:

• Early breast cancer detection is challenging due to tumor heterogeneity.
• Understanding blood-based heterogeneity is key for minimally invasive biomarkers.
• Major breast cancer subtypes (luminal A, luminal B, HER2+, basal-like) are known, but their blood-based profiles are less understood.

Purpose of the Study:

• To investigate breast cancer subtype heterogeneity in plasma proteome.
• To develop and apply methods for clustering and classifying breast cancer subtypes using proteomic data.
• To explore the relationship between different breast cancer subtypes based on pathway activation.

Main Methods:

• Integrative pathway-assisted clustering analysis of plasma proteome samples from 80 breast cancer patients and 80 healthy controls.
• Development and application of four distance metrics: Protein Intensity, Q-Value, Pathway Profile, and Distance Score Function.
• Permutation testing to identify significant pathway-level protein changes and Pathway Association Network analysis.

Main Results:

• Breast cancer subtypes were clustered into luminal (A and B) and basal (basal-like and HER2+) groups.
• Luminal A and B subtypes showed closer similarity than basal-like and HER2+ subtypes.
• Pathway-level alterations were more profound and less variable than molecular-level changes across subtypes.

Conclusions:

• Pathway activation patterns provide insights into the intrinsic relationships between breast cancer subtypes.
• Integrative pathway-assisted proteomics is a promising model for reliable cancer subtype classification.
• This approach can aid in the development of minimally invasive protein biomarkers for breast cancer detection, classification, and monitoring.