Summary:

Pancreatic cancer is one of the most deadly cancers, with an extremely poor prognosis. In 2020, the 5-year survival rate remains very low (only 3 to 9 %) and the median survival is less than 6 months. Despite progress in patient management, current therapies are not as effective as hoped. This is due to the high level of chemoresistance observed in this cancer. The key factor in this resistance is the complex tumour microenvironment, consisting mainly of stroma and a dense extracellular matrix, which limits the access of therapies to the tumour. The limitations of current study models, particularly in terms of physiological relevance, are a major obstacle to understanding this chemoresistance. In response to this problem, researchers are turning to new approaches by developing new alternative study models to those already available (in vitro and in vivo). The objectives of this work were: (i) to develop a microfluidic 3D in vitro culture device that can reproduce the tumour microenvironment in biological and mechanical terms, and to model the flow and mass transport present in a pancreatic tumour, and (ii) to address the morphological changes of co-culture by studying epithelial-mesenchymal markers and to study the impact of FOLFIRINOX chemotherapy in this model. We first demonstrated numerically and experimentally the feasibility of such an in vitro model. The extracellular matrix chosen is a combination of collagen I and hyaluronic acid, creating a rigid structure close to in vivo conditions. It enables the culture to be maintained over the long term with little or no degradation under the effect of perfusion, as well as activating the pancreatic stellate cells. The chosen perfusion rate applied an interstitial flow equivalent to that observed in the in vivo microenvironment, inducing hydrostatic pressure and shear stress on the cells. We then demonstrated the biological contribution of this modelling by showing increased chemoresistance to the FOLFIRINOX protocol in tumour cells both in mono- and co-culture in the microfluidic device. We also show the establishment of a process with characteristics of epithelial-mesenchymal transition and a possible promotion of a dedifferentiated tumour cell phenotype by activated pancreatic stellate cells. In conclusion, in this thesis we present an original microfluidic study model for modelling a tumour (co-culture of epithelial and mesenchymal cells) and studying the kinetics of a complex multidrug chemotherapy. In the future, this device should enable us to study in greater depth the mechanisms of chemoresistance and tumour-stroma interactions in pancreatic cancer.

Abstract:

Pancreatic cancer is one of the most deadly cancers with an extremely poor prognosis. In 2020, the 5-year survival rate remains very low (only 3-9%) and the median survival is less than 6 months. Despite advances in patient management, current therapies are not as effective as hoped. This is due to the high chemoresistance observed in this cancer. The key factor of this resistance is the complex tumor microenvironment composed mainly of stroma and a dense extracellular matrix limiting the access of therapies to the tumor. The limitations of current study models, particularly in terms of physiological relevance, are a major obstacle in understanding this chemoresistance. In response to this problem, researchers are turning to new approaches by developing new alternative study models to those already available (in vitro and in vivo). The objectives of this work were: (i) to develop a 3D microfluidic in vitro culture device allowing to reproduce the tumor microenvironment on the biological and mechanical levels, as well as to model the flows and mass transports present in a pancreatic tumor, and (ii) to address the morphological changes of the co-culture by the study of epithelial-mesenchymal markers and to study the impact of the FOLFIRINOX chemotherapy in this model. We first demonstrated numerically and experimentally the feasibility of such an in vitro model. The chosen extracellular matrix is a combination of collagen I and hyaluronic acid creating a rigid structure close to in vivo conditions. It allows the maintenance of the culture in the long term by not degrading or only slightly degrading under the effect of the perfusion as well as the activation of the pancreatic stellate cells. The chosen perfusion rate allows to apply an interstitial flow equivalent to the one observed in the in vivo microenvironment, inducing a hydrostatic pressure and a shear stress on the cells. We then highlighted the biological contribution of this modeling by showing an increased chemoresistance to the FOLFIRINOX protocol of tumor cells both in mono- and co-culture in the microfluidic device. We also show the establishment of a process with characteristics of epithelial-mesenchymal transition and a possible promotion of a dedifferentiated phenotype of tumor cells by activated pancreatic stellate cells. In conclusion, we present in this thesis an original microfluidic study model allowing to model a tumor (epithelial and mesenchymal cells co-culture) and to study the kinetics of a complex multidrug chemotherapy. This device should allow us in the future to study more deeply the mechanisms of chemoresistance and tumor-stroma interactions in pancreatic cancer.