Computational modeling of breast cancer cell invasion into adipose tissue

Cancer cell invasion into adipose tissue plays a key role in breast cancer progression and metastasis. The process of cancer cell invasion from a primary tumor into healthy tissue can exhibit different phenotypes (e.g. individualistic versus collective invasion), depending on the properties of the cancer cells, as well as the surrounding tissue. Understanding the important variables that control invasion phenotypes is crucial for our understanding of cancer invasion. We developed novel discrete element method (DEM) simulations of tumor cell invasion into adipose tissue in three dimensions. We model the adipocytes as deformable particles that can explicitly change their shape in response to external forces and the cancer cells as active, sticky spheres. We first show that cohesion between cancer cells hinders tumor invasion, while increased persistence in the cancer cell migration direction promotes invasion. We identify a characteristic dimensionless action A_c (that incorporates compression, cohesion, and persistence) above which cancer cell invasion occurs. We then show that cancer cells must be cohesive and highly persistent to exhibit collective invasion. Finally, we find that the distribution of the local packing fraction of the adipose tissue and alignment of the extracellular matrix play important roles in cancer invasion.