miR-363-3p Mediates Maintenance and Tumorigenicity of Breast Cancer Stem Cells and Provides A Potential Diagnostic and Prognostic Sera-Exosome Biomarker of Early Breast Cancer

Author(s): Stéphanie Tissot-Renaud, Athina Stravodimou, Fabrizio Benedetti, Valentina Scabia, Valérian Dormoy, Jacqueline Masternak, Cathrin Brisken, Jean-Yves Meuwly, Laetitia Rossier, Maryse Fiche, Marie Galmiche, Khalil Zaman, and Nicolas Mermod

Purpose: Despite great advances in breast cancer (BC) treatment, 20–30% of patients develop metastatic disease, eventually leading to death. Cancer stem cells (CSC) are key players in disease persistence and recurrence. However, methods for their detection and specific targeting are still lacking. MicroRNAs (miRNAs) are involved in the expression of many genes and biological processes. In this study, we aimed to identify miRNAs specific to BC stem cells (BCSC) that are resistant to treatment. Methods: By treating BCSC-enriched cells in vitro in mammospheres with chemotherapeutic agents, we selected miRNAs specific to resistant BCSCs. The most promising miRNA was then assessed in a mouse model and in clinical material from healthy women and patients treated with neoadjuvant chemotherapy. Results: In vitro enrichment of chemoresistant BCSC identified five-miRNA. miR-363-3p exhibited the highest expression in various BC cell lines and derived BCSC-enriched populations compared with non-tumorigenic cells. The inhibition of miR-363-3p decreased BCSC maintenance and tumorigenicity in vitro, whereas its overexpression increased mammosphere formation. Consistently, miR-363-3p downregulation decreased the growth and metastasis of human BC cells transplanted in mice. In human samples, miRNA-363-3p was overexpressed in cancer tissue before chemotherapy, whereas higher levels of miR363-3p in serum exosomes were correlated with better survival. Furthermore, a 7-exosomal miRNA (exomiRNA) signature, including miR-363-3p, enabled the discrimination of patients with BC from healthy women. Conclusion: In vitro, mouse and clinical models suggest that miR-363-3p and miRNA signatures act as biomarkers of the BCSC phenotype, providing a promising approach for BC detection, follow-up, and BCSC targeting.