In this paper, we have used an agent-based stochastic tumor growth model and presented a mathematical and theoretical perspective to cancer therapy. This perspective can be used to theoretical study of precision medicine and combination therapy in individuals. We have conducted a series ofin silicocombination therapy experiments. Based on cancer drugs and new findings of cancer biology, we hypothesize relationships between model parameters which in some cases represent individual genome characteristics and cancer drugs, i.e., in our approach, therapy players are delegated by biologically reasonable parameters.In silicoexperiments showed that combined therapies are more effective when players affect tumor via different mechanisms and have different physical dimensions. This research presents for the first time an algorithm as a theoretical viewpoint for the prediction of effectiveness and classification of therapy sets.