Agent-based modeling reveals impacts of cell adhesion and matrix remodeling on cancer collective cell migration phenotypes

Abstract

Understanding the phenotypic transitions of cancer cells is crucial for elucidating tumor progression mechanisms, particularly the transition from a non-invasive spheroid phenotype to an invasive network phenotype. We developed an agent-based model (ABM) using Compucell3D, an open-source biological simulation software, to investigate how varying biophysical and biochemical parameters influence emerging properties of cellular communities, including cell growth, division, and migration. Our focus was on cell-cell contact adhesion and matrix remodeling effects on cancer cell migration. We simplified enzymatic remodeling of the extracellular matrix and the subsequent enhancements to cellular chemotaxis or durotaxis as a combined effect of localized cellular secretion of a chemoattractant. By varying the chemoattractant secretion rate and contact adhesion energy, we simulated their effects on cellular behavior and driving the transition from a spheroid phenotype to a network phenotype. The model serves as a digital twin for 3D cancer cell culture, simulating cancer cell growth, division, and invasion over 1 week, validated against published data. The simulations track the emergent morphological and collective phenotype changes using key metrics such as cell circularity and invasion. Our findings indicate that increased chemoattractant secretion enhances the invasiveness of the collective cells, promoting the transition to a network phenotype. Additionally, changing cell-cell contact energy from a strong cell-cell adhesion to a weak cell-cell adhesion affects the compactness of the spheroids, resulting in lower circularity and increased collective cell invasion. Our work advances the understanding of tumor progression by providing insights into the biophysical mechanisms behind invasive cancer cell phenotypic transitions.