Becker, Chani2016-09-192016-09-192016-05https://hdl.handle.net/11299/182319University of Minnesota Ph.D. dissertation. May 2016. Major: Neuroscience. Advisors: David Largaespada, William Elmquist. 1 computer file (PDF); xxiii, 226 pages.Glioblastoma multiforme (GBM) is a lethal cancer. Without treatment, patients diagnosed with this disease survive nine months. With the best therapeutics science has to offer, including surgical resection, radiation therapy, and temozolomide, patients survive only five more months. Despite numerous clinical trials, the vast majority of tested drugs fail to provide therapeutic benefit to patients. It was the intent of this thesis to characterize the molecular mechanisms that prevent or limit the efficacy of targeted agents against malignant glioma. This work specifically explores how the internal characteristics of the tumor including its invasiveness and genetic heterogeneity as well as external attributes of therapeutic agents including brain penetrance contributes to the chemotherapeutic failure in GBM. By clarifying the biological processes that constrain treatment of this disease, scientists can strategize the development of better therapeutics with greater likelihoods for clinical success. We compared the brain distribution, molecular targeting efficiency, and survival benefit of GDC-0980 and GNE-317, two PI3K/mTOR inhibitor analogues. We showed that GDC-0980 is liable for efflux by P-glycoprotein (Pgp) and Breast Cancer Resistance Protein (BCRP) at the blood-brain barrier (BBB) while GNE-317 remains relatively resistant to efflux. Because GNE-317 is more brain penetrant than GDC-0980, it showed greater accumulation in the brain and stronger ability to impede the activation of PI3K/mTOR pathways in the GL261 mouse glioma model. Unexpectedly, neither drug affected survival, an effect that underscores the challenges presented by the genetic heterogeneity associated with cancer and the consequences of inadequate target selection. We also sought to determine the influence of anti-angiogenic therapy (AAT) on the delivery and efficacy of concurrently administered targeted agents. Again, we used GDC-0980 and GNE-317 to determine whether susceptibility to efflux impacted these parameters. We demonstrated that the vascular endothelial growth factor (VEGF) monoclonal antibody, bevacizumab (Avastin) could decrease the brain distribution of GDC-0980, although not significantly, but had no effect on the brain accumulation of GNE-317. We further showed that while bevacizumab alone provided a survival benefit in patient-derived glioma xenograft models, this therapeutic benefit could only be enhanced with co-treatment of a brain-penetrant drug like GNE-317. Collectively, these data suggest that AAT-induced BBB normalization is more likely to limit the delivery of targeted agents that are subject to active efflux. Finally, we examined the therapeutic potential of targeting cancer stem cells (CSCs) through experiments with parthenolide and LC-1 in the GL261 mouse glioma model. Effectively killing CSCs is an important goal in brain tumor research because this cell population is thought to responsible for tumor growth and recurrence, and is known to be particularly resistant to chemotherapies. In vitro studies of parthenolide and LC-1 in multiple glioma cell lines demonstrated that both drugs exhibited similar cytotoxicity profiles and were able to induce total cell death. LC-1 was also shown to be brain penetrant and non-toxic after prolonged exposure, and produced a demonstrable delay in tumor growth and a significant survival benefit. For these reasons, glioma stem cells remain a compelling therapeutic target for future clinical therapies.enBBBGlioblastomaGliomaThesis or Dissertation