Influence of active efflux transport on the distribution of targeted agents to brain tumors

Abstract

Glioblastoma multiforme (GBM) is a lethal disease of the whole brain. Despite complete surgical resection of the tumor, recurrence is inevitable and leads to patient death. Several molecularly-targeted agents have shown promising results preclinically, although clinical results have been disappointing. One plausible explanation for clinical failure of drugs is their inability to effectively target the invasive glioma cells that reside in areas away from the tumor core. These regions of the brain have an intact blood brain barrier (BBB), which through a combination of endothelial tight junctions and active efflux transporters restricts brain penetration of several drugs. The objective of this thesis was to investigate the influence of active efflux transporters, P-glycoprotein (P-gp) and breast cancer resistance protein (Bcrp) on the brain distribution of molecularly targeted agents. To enhance our understanding of brain distribution with statistical certainty, we proposed a population-based analysis method to estimate variability around the brain partition coefficient (Kp). We showed that the brain distribution of sunitinib (a tyrosine-kinase inhibitor) is limited by active efflux mediated by P-gp and Bcrp at the BBB. We further demonstrated that brain distribution could be enhanced by administration of a dual P-gp/Bcrp inhibitor. To statistically ascertain the variability associated with Kp in a serial sacrifice design, we developed a pharmacokinetic model to simultaneously describe plasma and brain concentration time profile data. We further evaluated the influence of study design features such as between subject variability (BSV) and sample size at each time point on bias and precision of estimation of Kp. Our results show that bias is unaffected by the assumptions regarding the magnitude of BSV and sample size, however, precision improves with sample size.We further examined the influence of BBB on delivery and efficacy of dual PI3K/mTOR inhibitors, GNE-317 (higher BBB permeability) and GDC-0980 (restricted BBB penetration) using three glioma mouse models, GL261-luc-GFP model, GBM10 and U87. Disruption of the BBB in the tumor core resulted in higher drug concentrations for GDC-0980. However, intact BBB in the areas adjacent to the core restricted the brain concentrations of GDC-0980. Contrary to this, the brain concentrations of GNE-317 were similar in all the three regions of the brain. Furthermore, results from survival studies indicated that although both GNE-317 and GDC-0980 significantly improved survival compared to placebo, the treatment groups were not significantly different from each other. Therefore, it is important to consider that effective treatment of GBM relies not only on effective delivery across the BBB to invasive glioma cells but also on an effective drug.