Prostate cancer cells depend on androgen for their survival. A standard treatment of metastatic prostate cancer (mPC) is androgen deprivation treatment (ADT). However, after a period of remission, some cancer cells changed into androgen-independent cells, and then treatment proceeds with a combination of ADT and chemotherapy. Senescent cells are cells that stop dividing but sustain viability. Senescence cancer cells are common in cancer, and they affect cancer treatment negatively by secreting inflammatory cytokines and pro-cancer VEGF. In this paper, we include the effect of senescence in a model of mPC. We consider combinations of ADT, chemotherapy, and senolytic drug, which eliminate senescent cells, in a spatio-temporal partial differential equations model, and demonstrate that simulations of the model are in agreement with experimental results. We evaluate the synergy between different doses of chemotherapy and senolytic drugs, at different fixed doses of ADT. We also consider optimal scheduling of the drugs, and the hypothesis that, in optimal schedule, a senolytic drug is to be administered immediately following the chemotherapy drug.