Final Report Summary - DUNHARROW (Evolutionary dynamics of growth factor production in cancer cell populations)
The production of diffusible factors by cancer cells is a form of cooperation, as diffusible molecules like growth factors are shared among neighbouring cells, affecting proliferation, resistance to apoptosis and immune system reaction, neo-angiogenesis and the Warburg effect — all major determinants of tumour development and resistance to therapies.
I developed mathematical tools (based on the properties of Bernstein polynomials) to study cooperation in large well-mixed populations of cells (to model the dynamics of liquid tumors) and computer simulations (on Voronoi graphs) to study cooperation in spatially structured populations (to model monolayers of cells in vitro). I also developed experimental systems to test the theory, using CRISPR to knock out genes coding for growth factors in different types of cancer cells, and developing experiments in which I manipulated different parameters of the models, to test the predictions of the theory.
Cooperation is studied in biology using evolutionary game theory, and studying cancer in the framework of evolutionary game theory is useful for two reasons: we can test open issues in the theory of cooperation using cancer cells; and we can use concepts from game theory to understand open issues in cancer biology.
1. Using cancer cells to test evolutionary theory. I used different cancer cell lines in which a particular growth factor has been knocked out, and their parental wild type cell line (that does produce the growth factor), to study cooperation between cancer cells. I manipulated the parameters of the models using different combinations of nutrients, to test open issues in evolutionary game theory that cannot be tested in natural populations or in bacteria, such as the effect of different types of benefits (whether cooperation affect birth rates or survival), the impact of spatial structure and the impact of non-linear benefits on the evolution of cooperation. The results confirm the predictions of evolutionary game theory, in particular, the importance of considering non-linear effects in the study of biological public goods.
2. Using evolutionary game theory to understand cancer. I developed theory and experimental tests to explain the evolution of intra-tumor heterogeneity, a feature that has long puzzled cancer biologists, but which can be explained using the logic of game theory developed during this project: the non-linear effect of growth factors on proliferation enables different types of cells to coexist in a stable mixed equilibrium. The results also help understand why therapies that target growth factors lead to a temporary reduction in tumor growth followed by relapse, as perturbations of this mixed equilibrium lead to a temporary decline in population fitness (corresponding to the short-term benefit of targeted therapies) followed by an adjustment to a new stable equilibrium (corresponding to the long-term evolution of resistance to therapies)
This work has implications for the study of cancer from a new point of view, that of evolutionary biology. Tumors are populations of cells that evolve over the course of an individual’s lifetime, and principles from evolutionary biology can be applied to the study of cancer as they are applied to the study of natural populations. This project aimed at introducing specific tools and concepts from evolutionary game theory in the study of cooperation between cancer cells for the production of diffusible molecules – one of the most important features of cell-cell interactions in tumors. Further work in evolutionary game theory of cancer could clarify the evolution of resistance to therapies, and could lead to the development of cell therapies based on knockout cells (like the ones used in these experiments) to drive intra-tumor cooperation to collapse.
I developed mathematical tools (based on the properties of Bernstein polynomials) to study cooperation in large well-mixed populations of cells (to model the dynamics of liquid tumors) and computer simulations (on Voronoi graphs) to study cooperation in spatially structured populations (to model monolayers of cells in vitro). I also developed experimental systems to test the theory, using CRISPR to knock out genes coding for growth factors in different types of cancer cells, and developing experiments in which I manipulated different parameters of the models, to test the predictions of the theory.
Cooperation is studied in biology using evolutionary game theory, and studying cancer in the framework of evolutionary game theory is useful for two reasons: we can test open issues in the theory of cooperation using cancer cells; and we can use concepts from game theory to understand open issues in cancer biology.
1. Using cancer cells to test evolutionary theory. I used different cancer cell lines in which a particular growth factor has been knocked out, and their parental wild type cell line (that does produce the growth factor), to study cooperation between cancer cells. I manipulated the parameters of the models using different combinations of nutrients, to test open issues in evolutionary game theory that cannot be tested in natural populations or in bacteria, such as the effect of different types of benefits (whether cooperation affect birth rates or survival), the impact of spatial structure and the impact of non-linear benefits on the evolution of cooperation. The results confirm the predictions of evolutionary game theory, in particular, the importance of considering non-linear effects in the study of biological public goods.
2. Using evolutionary game theory to understand cancer. I developed theory and experimental tests to explain the evolution of intra-tumor heterogeneity, a feature that has long puzzled cancer biologists, but which can be explained using the logic of game theory developed during this project: the non-linear effect of growth factors on proliferation enables different types of cells to coexist in a stable mixed equilibrium. The results also help understand why therapies that target growth factors lead to a temporary reduction in tumor growth followed by relapse, as perturbations of this mixed equilibrium lead to a temporary decline in population fitness (corresponding to the short-term benefit of targeted therapies) followed by an adjustment to a new stable equilibrium (corresponding to the long-term evolution of resistance to therapies)
This work has implications for the study of cancer from a new point of view, that of evolutionary biology. Tumors are populations of cells that evolve over the course of an individual’s lifetime, and principles from evolutionary biology can be applied to the study of cancer as they are applied to the study of natural populations. This project aimed at introducing specific tools and concepts from evolutionary game theory in the study of cooperation between cancer cells for the production of diffusible molecules – one of the most important features of cell-cell interactions in tumors. Further work in evolutionary game theory of cancer could clarify the evolution of resistance to therapies, and could lead to the development of cell therapies based on knockout cells (like the ones used in these experiments) to drive intra-tumor cooperation to collapse.