The cancer stem cell hypothesis represents a paradigm-shifting concept in tumor therapy resistance. This small population of tumor initiating cells is relatively quiescent. This makes them resistant to conventional cytoreductive cancer therapies such as radiation or chemotherapy which seek to damage the DNA of dividing cells. It is these cancer stem cells that may provide a major mechanism for cancer relapse and therapeutic resistance. In the case of brain tumors, either glioblastoma or metastatic breast cancer in the brain, the options for successful and complete surgical tumor removal are very limited. The prognosis for many brain tumors is very poor; the median overall survival for glioblastoma is roughly 12-15 months. Specifically targeting cancer stem cells may result in greater anti-tumor efficacy. Natural killer (NK) cells are terminally differentiated cytotoxic lymphocytes, which comprise the innate immune system and play a major role in eliminating transformed cells. NK cells not only attack blood-borne cancers and metastatic solid tumor cells in the blood, but also appear to have the unique ability to spontaneously recognize and reject allogeneic hematopoietic stem cells. It has been shown that NK cells preferentially attack cancer stem cells over other parts of the tumor. Before addition of NK cells, the tumor cells will first be primed for destruction by the addition of bortezomib, a proteasome inhibitor. Cells treated with bortezomib have been shown to upregulate tumor necrosis factor apoptosis-inducing ligand (TRAIL) which sensitizes them to NK-cell mediated apoptosis. Ideally the patient would first have their tumor debulked through conventional methods, i.e. radiation, so that the majority of the remaining tumor cells would be cancer stem cells. Then these cancer stem cells would be treated with bortezomib and exposed to NK cells