Abstract Despite available treatments including surgical resection, radiation and chemotherapy, glioblastoma (GBM) is incurable with rapid recurrence and low median survival rate of just fourteen months. Development of more effective treatments is difficult due to the highly heterogeneous nature of GBM. One aspect of that heterogeneity involves brain tumor initiating cells (BTICs) that have a stem cell-like ability to self-renew. BTICs can readily alter their metabolism and survive in low nutrient environments due in part to increased GLUT3 expression. We believe that the higher expression of GLUT3 in cancer cells compared to non-tumor cells makes it a therapeutic target, although the potential for toxicity must be considered. In recently accepted studies by Libby et al., we reported on two novel GLUT inhibitors identified by structure based virtual screening (SBVS) using a GLUT3 homology model. We are creating a structure-activity relationship profile and seek to increase the potency, selectivity and stability of the GLUT inhibitors. In this study we have tested a number of novel analogs and identified three that have maintained efficacy against BTICs in vitro. Importantly, these compounds display minimal toxicity against human astrocytes. The novel derivatives have increased stability compared to the lead compounds and are efficacious in the nanomolar range. In the future, we intend to utilize our anti-GLUT compounds alone and in combination with radio- and chemotherapy with the hope of clinical translation.