The brain is arguably the most important organ in the body. Unlike all other blood vessels in the body, the brain vasculature system has evolved a highly specialized protective system, called the blood-brain barrier (BBB) to selectively transport essential biomolecules needed to support the functions of the central nervous system (CNS) while preventing the entry of most other foreign agents that might be potentially harmful to the brain. Unfortunately, the presence of the BBB also poses a significant challenge for therapeutic delivery to treat diseases within the CNS; and indeed, this problem is highly relevant to brain cancer therapy.
Despite increasing development and approval of multiple new therapies, the survival rate of patients with malignant brain tumors (such as glioblastoma) has not improved (Drug Resist Updat 2015;19:1-12, Am Health Drug Benefits 2014:7;140-149). Patients still face abysmal prognosis even after surgical excision and aggressive post-operative chemo-radiotherapy (N Engl J Med 2005;352:987-996) and, to date, brain cancer is the leading cause of death in children and adolescents (CA Cancer J Clin 2016;66:7-30).
Healthy BBB forms a protective “wall” along the brain capillaries and is made up of predominantly brain capillary endothelial cells that are surrounded by closely associating pericytes and astrocytic endfeet (Figure) (Nat Rev Neurosci 2006;7:41-53). Paracellular flux at the BBB is restricted by the presence of tight junctions that connect adjacent endothelial cells (Trends Neurosci 2001;24:719-725).
Essential nutrients required for normal brain function (such as glucose and amino acids) are shuttled into the brain through specific transporters, while other large molecules or complexes (such as insulin and transferrin) are actively transported from the blood into the brain through receptor-mediated transcytosis (Trends Neurosci 2001;24:719-725). Additionally, efflux transporters (such as P-glycoprotein) on the surface of endothelial cells play an important role in actively pumping potentially toxic foreign agents back into the bloodstream (Adv Drug Deliv Rev 1999;36:179-194).
Solid tumors recruit new blood vessels through angiogenesis for a continuous supply of oxygen, nutrients, and necessary growth factors for rapid propagation. Angiogenic blood vessels have defective architecture and exhibit enhanced vascular permeability or “leakiness.” This phenomenon, which is universal in all solid tumors, is known as the enhanced permeability and retention (EPR) effect, and facilitates delivery of chemotherapeutics to tumor tissues (Adv Drug Deliv Rev 2011;63:136-151). In contrast to the BBB, abnormal and dysfunctional blood vessels in brain tumors form the blood-tumor barrier (BTB). Blood vessels in brain tumors are heterogenous, consisting of capillary populations that are:
- continuous and non-fenestrated (such as those of normal brain);
- continuous but fenestrated (exhibit enhanced permeability): and
- non-continuous (with interendothelial gaps) (Neuro-Oncol 2000;2:45-59).
In areas where fenestration or endothelial discontinuity is observed, drug delivery level in the tumor is substantially higher. This phenomenon is mostly found at or near the tumor core, but even then, due to the heterogeneity of the BTB, the distribution of drug permeability in the tumor is highly variable (Clin Cancer Res 2010;16:5664-5678). Invasive cells at the leading edge of the brain tumor border continue to thrive under the protection of the BBB (Figure) (Drug Metab Dispos 2013;41:33-39).
Failure to deliver therapeutics to the invasive cancer cells behind a functional BBB is one of the major causes for disease recurrence, even after primary tumor debulking by surgical means. Additionally, even though most tumors exhibit some increased BTB permeability, a high percentage of these lesions do not respond to cytotoxic drugs (such as paclitaxel and doxorubicin), highlighting the critical role of active efflux pumps in both the BTB and brain tumor cells in conferring chemoresistance (Cancer Res 2005;65:11419-11428).
Therapeutic Delivery to Tumors
Osmotic shrinkage using hypertonic solutions (such as mannitol) have been used to increase BBB permeability for enhancing drug delivery to the brain in the clinic. Various other therapeutic strategies are also employed to increase drug delivery to brain tumors, including injecting the chemotherapy agents directly into the CNS through intrathecal, intranasal, or intraventricular administration. Direct drug administration into the brain interstitial system using biodegradable wafers (such as carmustine) and catheter-based convection-enhanced delivery (CED) have also been applied in the clinics. Although they have been successful in improving patient survival, clinicians remain hesitant with the utility of these approaches due to morbidity associated with perioperative complications (Curr Pharm Des 2016;22:1177-1193, J Neurooncol 2012;107:373-378).
BBB disruption by radiation (Oncol Rep 2001;9:683-688) or imaging-guided focused ultrasound using intravenously administered microbubbles (Proc Natl Acad Sci USA 2006;103:11719-11723), where activation of the microbubbles in the vessels by an acoustic field produced by ultrasound treatment temporarily disrupts the tight junctions that connect the endothelial cells of the BBB have also been explored (J Acoust Soc Am 2011;130:3059-3067). While there are safety concerns relating to toxicity and long-term effects due to radiation, the use of focused ultrasound appears to be relatively safe when tested in monkeys (Cancer Res 2012;72:3652-3663). Additionally, due to the EPR effect, microbubbles accumulation in tumors upon administration followed by ultrasound treatment allows for further selectivity in BBB opening and enhancement of therapeutic delivery within the tumor(s).
In other areas of research, due to unknown long-term effects of BBB modulation, scientists have taken a different approach to overcome the BBB by developing “Trojan horses” that can hijack the transport systems of the BBB, including the insulin receptor, transferrin receptor, and low-density lipoprotein receptor to deliver chemotherapy into the brain (Curr Pharm Des 2016;22:1177-1193).
Drug formulations that have been designed to cross the BBB include attachment of monoclonal antibodies or peptides that recognize these receptors (and then become endocytosed) to targeted liposomes that contain toxic anti-tumor agents, or even directly to the drug itself (Clin Cancer Res 2007;13:1663-1674). Even with successful BBB penetration, drug targeting to specifically kill tumor cells is necessary to achieve efficacy without harming healthy brain tissues, which can lead to severe neurotoxicity.
BBB Modeling for Therapeutic Development
In vitro screening platforms have played crucial roles in facilitating the discovery of new brain therapeutics and optimization of existing drugs with enhanced BBB penetration. In 1983, Bowman and colleagues show that brain endothelial cells can be grown in tissue culture to model the BBB (Ann Neurol 1983;14:396-402). This finding has led to a wave of in vitro BBB mimetic model development, including the mid-throughput static transwell system, which involves the culture of brain endothelial cells within a transwell insert followed by culture of astrocytes (and/or pericytes) at the basal compartment of the transwell (J Cereb Blood Flow Metab 2016;36:862-890). Although the transwell model is highly versatile, simple, and currently the most widely used BBB model, it has been criticized for many well-known limitations (Eur J Nanomedicine 2014;6:185).
Efforts to improve BBB modeling through simulation of a more realistic representation of the BBB environment in a living brain have led to the development of the dynamic in vitro BBB model and other microfluidic BBB systems, which account for blood flow and shear stress (Brain Res 2006;1109:1-13, Lab Chip 2012;12:1784-1792, Lab Chip 2013;13:1093-1101).
More recently, multicellular BBB organoids spawned through the co-culture of human brain endothelial cells, pericytes, and astrocytes under low-adhesion condition have been shown to closely reproduce key BBB elements and functions (Nat Commun 2017;8:15623). The ease of culture and robustness, along with the offer of a high-throughput capacity for drug screening make this model particularly attractive compared to preceding models. Although culture-based models play a critical role in advancing neuroscience, they are not a substitute for in vivo models and should always be complemented with results gained from animal (or clinical) studies.
CHOI-FONG CHO, PHD, is Instructor in Neurosurgery at the Brigham and Women's Hospital and Harvard Medical School, Boston, and Research Fellow in the Department of Chemistry, Massachusetts Institute of Technology.
Can't Find a Past Issue?
Go online to http://bit.ly/2wCfymG where you can view all the articles from past issues of Oncology Times.