A group of international researchers has devised a reliable method for subtyping breast tumors into the IntClust subtypes based on gene expression patterns, further demonstrating the clinical and biological validity of the IntClust classification.

IntClust is a software package for clustering gene expression data with repeated measurements based on interval data analysis. This can classify breast cancer into ten subtypes based on molecular drivers identified through the integration of genomic and transcriptomic data from 1,000 breast tumors and validated in a further 1,000.

The classification of breast tumors based on morphology and two key markers, oEGFRestrogen receptor and human epidermal growth factor receptor 2 are still the support of current clinical practice.

However, this approach lacks the underlying molecular changes, which ultimately constitute a tumor’s oncogenic drive,with recent genomic studies starting to reveal the complexity of somatic alterations in breast cancer at the levels of mutations and copy number aberrations.

In this study entitled “Genome-driven integrated classification of breast cancer validated in over 7,500 samples” and published in the Genome Biology journal, the team designed a gene expression-based approach for classifying breast tumors into the ten IntClust subtypes. They applied this method in 983 independent samples for which gene expression IntClust classification was available and gathered a significant external dataset of 7,544 additional breast tumors.
The researchers could observe that the ten subtypes are observable in most studies at comparable frequencies and are significantly associated with relapse-free survival.

Importantly, different IntClust subtypes reveal large differences in chemosensitivity, an important determinant of the relative utility of a disease classification scheme. The team also investigated how IntClust classification preforms when compared to PAM50 or SCMGENE, two other breast cancer classification methods, and concluded that patterns of genomic aberration observed in breast cancer are better explained by IntClust.

Based on the resulting data, the authors conclude that IntClust subtypes are reproducible entities, have clinical validity and can become increasingly relevant as more targeted biological therapies become available.