In this talk two multidimensional bladder cancer models with various types of therapy are examined. The first model is four dimensional, another model is nine dimensional. The main researching interest is twofold: finding ultimate densities of cells populations and tumor clearance conditions. We study ultimate dynamics of interactions between tumor and the immune system via finding ultimate upper and lower bounds for all variables involved into these models (densities of cells populations and concentration of treatments). Besides, the dissipativity property in the sense of Levinson is shown for both of models. Further, the global tumor clearance problem is considered as a control problem in which the treatment is chosen as a constant control satisfying some inequalities expressed in terms of model parameters. Our method is based on localization method of compact invariant sets and may be exploited for a prediction of the cells populations dynamics involved into cancer tumor models.