Modeling and Pharmacological Control of Tumor Spheroid Growth in the Agent-Based Chaste Simulator

Systems and synthetic biology increasingly rely on computational simulation to study complex biological phenomena, such as tumour growth and response to therapy, while reducing the need for costly and time-consuming experiments. In these simulations, however, drug delivery is usually treated as a prescribed input rather than as a quantity regulated in response to tumour evolution. This thesis takes a first step in this direction by extending Chaste, an open-source multi-agent simulation framework in which each cell is modelled as an individual agent, with a closed-loop control module that regulates drug delivery to avascular tumour spheroids. To this end, the underlying cell-cycle model is extended so that cells can become quiescent and die in response to the drug, allowing the tumour to both grow and shrink. The framework is evaluated through simulation studies of increasing complexity. Under feedback control, an instability caused by the synchronous reactivation of quiescent cells emerges near the target population and is mitigated by limiting how fast the drug dose can change. Moreover, tighter regulation of the tumour does not necessarily imply lower drug consumption, but depends on how the drug acts on the cells. Finally, when the drug diffuses through the tissue rather than being applied uniformly, controller settings that ensure convergence under uniform delivery lead to persistent oscillations, identifying spatially resolved drug delivery as an open challenge.