Epithelial-to-mesenchymal transition (EMT) and cancer metastasis: the status quo of methods and experimental models 2025

Abstract

Epithelial-to-mesenchymal transition (EMT) is a crucial cellular process for embryogenesis, wound healing, and cancer progression. It involves a shift in cell interactions, leading to the detachment of epithelial cells and activation of gene programs promoting a mesenchymal state. EMT plays a significant role in cancer metastasis triggering tumor initiation and stemness, and activates metastatic cascades resulting in resistance to therapy. Moreover, reversal of EMT contributes to the formation of metastatic lesions. Metastasis still needs to be better understood functionally in its major but complex steps of migration, invasion, intravasation, dissemination, which contributes to the establishment of minimal residual disease (MRD), extravasation, and successful seeding and growth of metastatic lesions at microenvironmentally heterogeneous sites. Therefore, the current review article intends to present, and discuss comprehensively, the status quo of experimental models able to investigate EMT and metastasis in vitro and in vivo, for researchers planning to enter the field. We emphasize various methods to understand EMT function and the major steps of metastasis, including diverse migration, invasion and matrix degradation assays, microfluidics, 3D co-culture models, spheroids, organoids, or latest spatial and imaging methods to analyze complex compartments. In vivo models such as the chorionallantoic membrane (CAM) assay, cell line-derived and patient-derived xenografts, syngeneic, genetically modified, and humanized mice, are presented as a promising arsenal of tools to analyze intravasation, site specific metastasis, and treatment response. Furthermore, we give a brief overview on methods detecting dissemination and MRD in carcinomas, highlighting its significance in tracking the course of disease and response to treatment. Enhanced lineage tracking tools, dynamic in vivo imaging, and therapeutically useful in vivo models as powerful preclinical tools may still better reveal functional interdependencies between metastasis and EMT. Future directions are discussed in light of emerging views on the biology, diagnosis, and treatment of EMT and metastasis.

Fig. 1

A schematic illustration depicting epithelial-to-mesenchymal transition (EMT) and the metastatic cascade. The figure shows tumor cells undergoing EMT, characterized by the reduction of epithelial markers and the acquisition of mesenchymal properties, promoting migration and invasion. Cancer cells then enter the vasculature, disseminate as disseminating tumor cells (DTCs), and extravasate to distant sites

Fig. 2

Depiction of various 2D and 3D in vitro models to study EMT in Cancer. In 2D culture, cells grow under adherent conditions, which confine them to microenvironment. These models are easy to implement, however, they only support to study cell–cell interactions. In contrast, 3D culture models mimic the tumor microenvironment by enabling both cell–cell and cell–matrix interactions within a three-dimensional structure. Moreover, 3D models allow for the co-culture of multiple cell types, effectively mimicking intra-tumor heterogeneity

Fig. 3

Schematic representation of the chorionallantoic membrane (CAM) model of the bred chicken egg and possible applications for cancer research. Schematic overview of the developing egg. Possible material to be used for the CAM model and potential applications (materials or compounds) are shown at the top level of the graph. Major compartments of the egg model are shown, including the lower and upper CAMs which are physiologically defined with respect to the window cut for placing the cells/material to be investigated (dotted line). Many implications of the model for cancer research are demonstrated in the lower part of the figure. Especially, after inoculation of the cells or tissues onto the CAM, invasion into the upper CAM can be measured, as well as tumor growth and angiogenesis. The lower CAM can be harvested specifically to evaluate the cells that have intravasated into blood vessels. PDX models are also possible. The model is excellent for rapid in vivo drug testing, amongst all of the other applications shown

Fig. 4

Schematic illustration of experimental mouse models to study metastasis. Syngeneic mouse models involve spontaneous development of tumors or transplantation of tumors within the same strain. Genetically engineered mouse models have targeted gene mutations which enable tumor development to be monitored from the initial stages. Immunocompromised mice are used for xenotransplantation of human tissues or cells. In the CDX model, a tumor is developed by injecting cells from established tumor cell lines into mice, whereas in the PDX model, the tumor is derived from human tumor tissue, mimicking the human tumor microenvironment. These tumors can be expanded within the same strain, facilitating metastatic studies. In humanized mouse models, the mice are inoculated with human cells (e.g., immune or stromal cells) to study tumor interactions with human normal cells. The PDXO model involves implanting human tumor tissue into mice to create 3D organoid cultures, which mimic the human tumor environment

Fig. 5

Comprehensive overview of the metastatic cascade and techniques that can be applied at each stage. Schematic overview of key in vitro and in vitro applications and animal models to investigate individual steps of the cascade of metastasis