Abstract:

The management of patients with triple-negative breast cancer (TNBC) remains problematic due to the absence, in most cases, of specific therapies, creating clinical obstacles due to its heterogeneity and lack of targeted therapies. This thesis addresses the need to improve therapeutic approaches using a unique mouse model of TNBC (MMTV-R26Met). This mouse model shows mild overexpression of the wild-type MET receptor tyrosine kinase (RTK) in the mammary gland, leading to the spontaneous formation of tumors that faithfully reproduce the main features of TNBC: histological and molecular heterogeneity, infiltration of immune cells, resistance to conventional chemotherapy and targeted therapies. The main aim of this thesis is to exploit this mouse model to explore tumor cell biology and the immune microenvironment, in order to develop and evaluate the efficacy of combined therapies, such as anti-cancer treatments and immunotherapy. The initial phase of my project was dedicated to characterizing the MMTV-R26Met model and identifying effective combinations of compounds. We discovered a new therapeutic combination targeting the anti-apoptotic protein BCL-XL and the cell cycle regulator WEE1, showing promising efficacy against TNBC cells. We then investigated the specificity of targeting cell cycle regulators in combination with BCL-XL inhibition. Our results revealed that high levels of BCL-XL and specific cell cycle regulators are associated with a poor prognosis in TNBC patients. Thus, targeting BCL-XL makes cells particularly vulnerable to agents targeting cyclin-dependent kinases (CDKs). This combination of drugs, highly effective in vitro and in vivo, deregulates the survival signals triggered by RTKs, leading to accumulated DNA damage and death by apoptosis. In line with our aim to elucidate the signaling networks underlying TNBC and response to treatment, we have developed a mathematical model integrating data from the literature and experimental proteomics. This model can serve as a valuable tool for assessing drug mechanisms of action and identifying vulnerabilities that can be targeted by treatments. Secondly, we exploited the immunocompetence of MMTV-R26Met mice, a unique model for exploring the interaction between immune cells and cancer cells. This question is particularly relevant in the current context of emerging immune therapies showing promising clinical results. By characterizing tumor cells and the tumor immune microenvironment (TIM) in the MMTV-R26Met model, using proteomics, spectral cytometry and single-cell RNA sequencing, we demonstrated a high degree of TIM heterogeneity among spontaneous TNBC tumors, reflecting the diversity of patients' immune profiles. Furthermore, using a syngeneic transplant model we have established, we aim to document in vivo the longitudinal evolution of the immune signature during tumor regression mediated by clinically used anticancer agents. In conclusion, this thesis contributes to advancing our understanding of TNBC biology and proposing innovative therapeutic strategies by integrating molecular, cellular and bioinformatics approaches. The results open up new prospects for exploring targeted treatment approaches that take into account the interactions between tumor cells and immune cells, and offer the hope of improving the prognosis of patients with TNBC.