Integration of multi-omics data reveals a novel hybrid breast cancer subtype and its biomarkers

Abstract

Tumor heterogeneity in breast cancer hinders proper diagnosis and treatment, and the identification of molecular subtypes may help enhance the understanding of its heterogeneity. Therefore, we proposed a novel integrated multi-omics approach for breast cancer typing, which led to the identification of a hybrid subtype (Mix_Sub subtype) with a poor survival prognosis. This subtype is characterized by lower levels of the inflammatory response, lower tumor malignancy, lower immune cell infiltration, and higher T-cell dysfunction. Moreover, we found that cell-cell communication mediated by NCAM1-FGFR1 ligand-receptor interaction and cellular functional states, such as cell cycle, DNA damage, and DNA repair, were significantly altered and upregulated in patients with this subtype, and that such patients displayed greater sensitivity to targeted therapies. Subsequently, using differential genes among subtypes as biomarkers, we constructed prognostic risk models and subtype classifiers for the Mix_Sub subtype and validated their generalization ability in external datasets obtained from the GEO database, indicating their potential therapeutic and prognostic significance. These biomarkers also showed significant spatially variable expression in malignant tumor cells. Collectively, the identification of the Mix_Sub breast cancer subtype and its biomarkers, based on the driving relationship between omics, has deepened our understanding of breast cancer heterogeneity and facilitated the development of breast cancer precision therapy.

Keywords: biomolecular markers; breast cancer; molecular subtypes; multi-omics; spatial variability.