Abstract Patient prognosis for individuals diagnosed with GBM remains incredibly poor with a median survival of only 15 months despite aggressive treatment with surgical tumor resection, chemotherapy, and radiation therapy. Therapeutic development to prolong survival has been hampered by a high degree of inter- and intra-tumoral heterogeneity. Contributing to tumor heterogeneity is a subset of highly tumorigenic cells, termed brain tumor initiating cells (BTICs), that can self-renew and have been shown to be highly invasive and resistant to therapy. BTICs can survive under nutrient poor conditions due to their increased expression of glucose transporter 3 (GLUT3). In a recently accepted study, we utilized structure based virtual screening (SBVS) using a GLUT3 homology model to identify two novel GLUT inhibitors. We are now generating a structure-activity relationship profile based on a highly efficacious compound (IC50= ~300 nM) with the same backbone structure identified in the initial screen and seek to improve the potency, selectivity and drug-like properties of the GLUT inhibitors. We have tested several novel analogs and identified four that have maintained efficacy against BTICs in vitro (IC50= 300–600 nM). Importantly, these analogs have displayed less toxicity to astrocytes than lead compounds in addition to improved stability in mouse liver microsomes. As proof of concept, we have begun to test the GLUT inhibitors alone and in combination with chemotherapy in vivo. Thus far, our work has demonstrated that targeting glucose transport is a promising therapeutic avenue to explore.