Modeling metastatic colorectal cancer subtypes to improve therapeutic strategies

Translational cancer research aims to optimize personalized treatment strategies and improve survival rates. However, developing innovative treatment approaches for colorectal cancer (CRC) faces challenges due to tumor subtype heterogeneity and the absence of suitable preclinical models that align with these subtypes. CRC can be categorized into specific tumor subtypes based on gene expression profiles and the status of the mismatch repair (MMR) system, characterized as either proficient (pMMR) or deficient (dMMR). Establishing preclinical models representing these CRC subtypes could rationalize therapeutic strategies. The objectives of the studies in this thesis were to uncover molecular tumor-intrinsic and microenvironmental factors contributing to aggressive behavior and drug resistance. Additionally, we aimed to develop a novel humanized mouse model for studying determinants of immunotherapy response. Furthermore, we investigated the mechanisms behind metastasis formation and its clonal relationship with primary tumors. We utilized patient-derived tumor organoids, animal models, CRC tissue specimens, and clinical data from patients at various stages of the disease. Our findings revealed that the upregulation of a tyrosine kinase receptor enhances tumor cell regenerative capacity and induces a pro-fibrotic tumor microenvironment. However, in vitro experiments showed that dMMR and pMMR tumor organoids had similar sensitivity to chemotherapy and targeted treatments, suggesting that microenvironmental factors may influence drug response and metastatic potential. We observed that metastasis formation in dMMR tumors is associated with the presence of inactive cytotoxic T cells in the pro-fibrotic microenvironment. Furthermore, we introduced a novel humanized mouse model that faithfully replicates the clinically observed benefits of immune checkpoint blockade therapy depending on the organ site, facilitating the evaluation of new immune-oncology approaches. Moreover, by modeling the formation of metastases, we demonstrated that fundamentally distinct processes drive the development of liver, lung, and peritoneal metastases. In summary, these findings contribute to advancing treatment strategies for CRC patients.