Tumor growth analysis using cellular automata based on the cancer hallmarks

Abstract

[Resumen]En esta tesis se ha realizado un modelado del crecimiento tumoral, considerando éste consecuencia emergente de las interacciones entre las células y su entorno. El modelado se ha considerado en el nivel de comportamiento celular, modelando los procesos de mitosis y muerte celular en función de la adquisición de una serie de rasgos característicos del cáncer (hallmarks) y del entorno inmediato de cada célula. Para el modelado hemos considerado la herramienta de Autómata Celular (AC). En la tesis se ha analizado la relevancia de los diferentes hallmarks en diferentes escenarios, las transiciones de comportamientos al aplicar un tratamiento, además de introducir la modelización de células madre de cáncer (CSCs). Al incorporar CSCs en el modelado se analizan además diferentes estrategias de tratamientos en el contexto de CSC, teniendo en cuenta la capacidad de recrecimiento del tumor debido a la presencia de CSCs. Finalmente, hemos aplicado optimización evolutiva para la obtención automática de los tratamientos que minimicen el efecto de la recidiva.

[Abstract] In this thesis we used computational models based on cellular automata and the abstract model of cancer hallmarks to analyze the emergent behavior of tumor growth at cellular level. Tumor growth is modeled with a cellular automaton which determines cell mitotic and apoptotic behaviors. These behaviors depend on the cancer hallmarks acquired in each cell as consequence of mutations. The presence of the cancer hallmarks defines cell states and cell mitotic behaviors. Additionally, these hallmarks are associated with a series of parameters, and depending on their values and the activation of the hallmarks in each of the cells, the system can evolve to different dynamics. With the simulation tool we performed an analysis of the first phases of cancer growth. Firstly, we studied the evolution of cancer cells and hallmarks in different representative situations regarding initial conditions and parameters, analyzing the relative importance of the hallmarks for tumor progression; Secondly, we focused on the analysis of the effect of killing cancer cells, inspecting the time evolution of the multicellular system under such conditions and the possible behavioral transitions between the predominance of cancer and healthy cells. Later, we analyzed the effect of treatment applications on cancer growth taking into account the presence of Cancer Stem Cells (CSCs) and their regrowth capacity. Finally, we used evolutionary computing to analyze the implications of treatment strategies in a CSC context. In this way, we determined the best strategies of treatment applications in terms of intensity, duration and periodicity considering the regrowth capacity of CSCs.