Mathematical Modelling Reveals Selective Dynamics Invisible to Imaging

Tumors arise and progress according to the laws of Darwinian evolution. i.e., an interplay between mutational diversification and environmental selection. While the mutational component of somatic evolution is relatively well understood, the nature of the selection forces acting on populations of tumor cells, and how these forces are impacted by interactions between subclonal populations as they cooperate and compete for limited resources, remains poorly understood. As space is the key growth-limiting resource, both the ability to create this resource through engineering, achieved by degrading extracellular matrix, and the ability to utilize this resource effectively through maximizing proliferation, are likely to be under strong selection. While the ability to proliferate is a cell-intrinsic property, degradation of ECM is achieved through secretion of matrix-degrading enzymes, representing public good, which can benefit both producers and non-producers. Assuming the well-established principle of evolutionary tradeoff, we used agent-based modeling to explore the interplay between engineering and consumption strategies in evolving populations of tumor cells, as well as the impact of this interplay on tumor heterogeneity and clonal composition in space and time. We found that that this interplay results in ecological succession, enabling generation of large, heterogenous and highly proliferative populations. Surprisingly, even though in our simulations both engineering and consumption strategies were under strong positive selection, their interplay led to sub-clonal architecture that could be interpreted as neutral evolution with sampling strategies commonly used in tumor genome analyses. Our results warrant more careful interpretation of inferences from sequencing cancer genomes and highlight the importance of consideration of ecological aspects of somatic evolution.