Putative cancer-initiating stem cells in cell culture models for molecular subtypes of clinical breast cancer.

Abstract

Cancer-initiating stem cells (CISC) represent a minor subpopulation of heterogeneous breast cancer. CISC are responsible for the acquired resistance to conventional chemoendocrine therapy and eventual relapse observed in patients with breast cancer. Certain molecular subtypes of clinical breast cancer that exhibit differential expression of genes coding for hormone and growth factor receptors differ in their response to conventional chemoendocrine therapy and targeted therapeutic inhibitors. Thus, the development of reliable cell culture models for CISC may provide a valuable experimental approach for the study of stem cell-targeted therapy for the treatment of breast cancer. The present study utilized optimized cell culture systems as experimental models for different molecular subtypes of clinical breast cancer, including luminal A, human epidermal growth factor receptor (HER)-2-enriched and triple negative breast cancer. Biomarker end points, including control of homeostatic growth, cancer risk and drug resistance, were quantitatively analyzed in the selected models. The results of the analyses indicated that, compared with the non-tumorigenic controls, the cell models representing the aforementioned molecular subtypes of clinical breast cancer exhibited aberrant cell cycle progression, downregulated cellular apoptosis and loss of control of homeostatic growth, as evidenced by hyperproliferation. Additionally, these models displayed persistent cancer risk, as indicated by their high incidence and frequency of anchorage-independent (AI) colony formation in vitro and their tumor development capacity in vivo. Furthermore, in the presence of maximum cytostatic drug concentrations, the drug-resistant phenotypes isolated from the parental drug-sensitive cell lines representing luminal A, HER-2-enriched and triple negative breast cancer exhibited an 11.5, 5.0 and 6.2 fold increase in cell growth, and a 5.6, 5.4 and 4.4 fold increase in the number of AI colonies, respectively, compared with the drug-sensitive controls. Collectively, the data of the present study demonstrated the presence of putative CISC in these breast cancer models.