Specific breast cancer prognosis-subtype distinctions based on DNA methylation patterns

Area of Science:

• Oncology
• Epigenetics
• Genomics

Background:

• Tumour heterogeneity presents a significant challenge in breast cancer diagnosis and treatment.
• DNA methylation, a key epigenetic regulator of gene expression, offers a potential avenue for characterizing this heterogeneity.

Purpose of the Study:

• To identify and characterize distinct breast cancer prognosis-subtypes based on DNA methylation patterns.
• To compare the granularity of DNA methylation-based subtypes with existing gene expression-based classifications (e.g., PAM50).

Main Methods:

• Utilized consensus clustering on DNA methylation data (3869 CpGs) from 669 breast cancer samples in The Cancer Genome Atlas (TCGA) database.
• Analyzed associations between identified DNA methylation subtypes and clinical/demographic factors.
• Developed a Bayesian network-based prognosis model for classifying breast cancer subtypes.

Main Results:

• Identified nine distinct breast cancer subgroups based on DNA methylation profiles, correlating with race, age, tumour stage, receptor status, histology, metastasis, and prognosis.
• DNA methylation subtypes provided a more refined classification than PAM50 subtypes, further subdividing the Basal-like subtype.
• Discovered 1252 differentially methylated CpG sites (888 genes) specific to each subgroup.
• Validated the prognosis model's ability to accurately classify independent test sets.

Conclusions:

• DNA methylation-based subtyping offers a more elaborate and clinically informative classification of breast cancer heterogeneity.
• These findings can elucidate the heterogeneity within existing molecular subtypes and guide the development of personalized treatment strategies for newly defined subgroups.