Author summary Tumor therapy with replicating oncolytic viruses is based on the premise that if the tumor specific virus infects and is amplified by the tumor population and spreads sufficiently within the tumor, it will lead to eradication of the cancer. The outcome of this approach is an exercise in population dynamics, and, as in ecology, depends on the detailed interactions between the various players involved. Mathematical models have been used to capture these dynamics, but space is often explicitly excluded from these models. We combine in vitro experiments studying tumor growth and virus spread in two and three dimensions to inform the development of a spatially explicit computational model of tumor virotherapy and compare the outcome with non-spatial, mean-field models. Viruses generally spread from cell to cell, and therefore the number of nearest neighbors close to an infected cell is important. Experimental data show that in three dimensions, the median number of nearest neighbors is 16 compared to 6 in the 2D plane. However, while simulations in 3D reach equilibrium faster than in 2D, tumor eradication is much less common in 3D than in 2D. Three dimensional space plays a critical role in the outcome of tumor virotherapy and this additional spatial dimension cannot be ignored in modeling.
展开▼
