Quantitative hybrid modeling reveals predictable mechanical stress response of growing tumor spheroids

By a hybrid computational modeling strategy we demonstrate that the growth response of tumors under mechanical stress may be quantitatively predictable even under largely differing growth conditions. In our computational models, cells are represented by individual units parameterized by measurable biophysical and cell-biological parameters. We assume a simple functional relation how cell cycle progression is controlled by volumetric strain, the latter in our model being derived from a biomechanical relation between applied pressure and cell compressibility. After parameter calibration from experiments with CT26 cells growing against resistance of a thin elastic alginate capsule, the model adequately predicts both the growth curve in thick capsules and, after accounting for differences in the growth conditions, in a Dextran solution experiment, where the stress is generated by osmosis. Our modeling strategy is very general, and provides an important technical achievement for dealing with high compression in individual-based models that enjoy increasing interest.