(815h) Agent Based Models of Heterogeneous Tumor Cell Populations

Despite significant improvements in treatment and understanding, cancer continues to grow at an alarming pace, accounting for 1 in 8 deaths worldwide. It is important to identify the mechanisms that control cancer growth, understand how these mechanisms interact, and discern how to manipulate these mechanisms in order to develop treatments. Models act as an invaluable tool for this, given the expensive and time-consuming nature of in vivo/in vitro experimentation.

A significant focus of researchers has been on theoretical, mathematical based models. These strongly deterministic models are fundamental in understanding the basic, physical mechanisms of cancer. These theoretical models, however, fail to  point to a single, conceptual framework in which data can be fitted. Given the rapid improvement in "wet lab" tools, it is critical to develop a model that can make use of the large amounts of this high-resolution data.

Cellular automata (CA) has been used to describe the growth of tumor cells using the concept of emergence, which suggest that complex behavior can develop from a simple set of fundamental rules. CA is a more realistic model, where changes in the state of a grid cell depend on a basic set of rules and the states of neighboring cells. The biggest drawback of cellular automata is its static nature. CA cells have a fixed position on the grid environment and only change state, not location, over time.

We propose to apply agent based models (ABM) to study emergent behavior of heterogeneous tumor cell populations. ABMs incorporate dynamic agents that can change both state and location over time based on a set of rules. We intend to use ABMs to model tumor growth based on rules suggested by two hypotheses: the cancer stem cell hypothesis and the clonal evolution model. The cancer stem cell hypothesis suggests that there are certain cancer stem cells that can replicate indefinitely and differentiate into all the cell types in a tumor. It is the migration, and subsequent colonization, of these cancer stem cells leads to metastasis. The clonal evolution model suggests that over time, cancer cells acquire various combinations of mutation. Tumor progression is driven by the natural selection of the most aggressive cancer cells. Metastasis results when the tumor cells have "acquired" the mutations needed to become invasive.

We hope that studying the emergent behavior of these two models and their alignment with experimental data will help clarify the mechanisms by which cancer develops and metastasizes.