Optimizing T-cell-mediated cancer killing: An agent-based model

Abstract

Immunotherapies for cancer aim to improve immune cell function by helping immune cells recognize and kill tumors. A group of immunotherapies target T-cells, aiming to improve the patients prognosis. However, these therapies are not always successful, and the underlying mechanisms of T-cell elimination of tumors remain poorly understood. We developed a three-dimensional agent-based model to explore this behavior that probabilistically models cell growth, migration and interaction parameters. Through this project, we explore the influence of cytokines like IL-2 and CCL2 on T-cell behavior and optimize the interactions between T-cells and cancer cells to most efficiently kill tumors using modeling. How individual parameters such as T-cell killing rate, proliferation, infiltration and cytotoxicity is still largely unexplored, and our study aims to fill that gap by stochastically simulating T-cells as they attack a cancer spheroid in order to identify new potential strategies to enhance T-cell killing of tumors. Our study found that, from greatest to least effect, increasing the killing, proliferation, and infiltration rate of T-cells increased the clearance rate of the tumor. A 17% increase in killing rate was enough to control and shrink the tumor, while a 15% increase in their proliferation rate was sufficient to control the tumor without regressing it; a 30% increase in their proliferation rate was sufficient to shrink the tumor. Increasing the infiltration was only able to control the tumor, not cause it to regress.