A new gene expression-based Breast Cancer Purity Score

In a new study recently published on NPJ Precision Oncology, researchers from Fondazione Michelangelo and Milan-Bicocca University developed and validated a new gene expression-based tumour purity score as a method specific for breast cancers. The 9-gene Breast Cancer Purity Score outperformed existing methods for estimating tumour content in patient-derived samples.

Tumour purity, defined as the relative abundance of cancer cells in a tumour, could represent a biologically relevant, intrinsic tumour feature but is also affected by extrinsic sampling bias. Data analysis and interpretation of molecular characterization of large clinical cohorts can be biased by tumour purity variability, thus a variety of strategies to estimate tumour purity has been proposed (such as gene expression, genomic or DNA-methylation profiles) but none is available specifically for breast cancer. Researchers analyzed over 6000 expression profiles from ten breast cancer datasets to generate and validate a tumour purity score; the resulting Breast Cancer Purity Score outperformed ESTIMATE, one of the commonly used transcriptomics-based methods developed on pan-cancer data without including any tumour related genes. The new Score can also capture treatment-induced changes carrying predictive and prognostic information: Breast Cancer Purity Score-estimated tumour purity improved prognostication in luminal breast cancer, correlated with pathologic complete response in on-treatment biopsies from triple-negative breast cancer patients undergoing neoadjuvant treatment and effectively stratified the risk of relapse in HER2+ residual disease post-neoadjuvant treatment. As authors conclude, «We developed and validated a straightforward tool to estimate tumour content from bulk transcriptomic breast cancer data, useful to explore the role of tumour purity, aid data interpretation and improve prognostication. The framework presented here could be successfully applied to other cancer types».