Chimeric antigen receptor T cell (CAR-T) therapy is a novel approved treatment for hematological malignancies, still under development for solid tumors. Here, we use a rate equation-based mathematical model to discover regimens and schedules that maintain efficacy while potentially reducing toxicity by decreasing the amount of CAR-T infused. Tested on an in vivo murine model of spontaneous breast cancer, we show that our mathematical model accurately recapitulates in vivo tumor growth results achieved in the previous experiments. Moreover, we use the mathematical model to predict results of new therapy schedules and successfully prospectively validated these predictions in the in vivo. We conclude that using one tenth and even one percent of a full CAR-T dose used in preclinical trials can achieve efficacious results similar to full dose treatment.