Abstract Prostate cancer metastasis to the bone is predominantly lethal and results from the ability of successful metastatic prostate cancer cells to co-opt microenvironmental cells and processes involved in bone remodelling. Understanding how the interactions between tumour and stromal cells determine successful metastases and how metastatic tumours respond to treatment is an emergent process that is hard to asses biologically and thus can benefit from mathematical models. In this work we describe a mathematical model of bone remodelling and the establishment of a prostate cancer metastasis in the bone using evolutionary game theory. We have mathematically recapitulated the current paradigm of a vicious cycle driving the tumor growth and we have used this tool to investigate the key interactions between the tumour and the bone stroma. Crucially, the model sheds light on the role that the interactions of heterogeneous tumor cells with the bone microenvironment have in the treatment of cancer. Our results show that resistant populations naturally become dominant in the metastases under a number treatment schemes and that schedules designed by an evolutionary game theory approach could be used to better control the tumour and the associated bone growth than the current standard of care.