Engineered microvascular brain-on-a-chip model for the study of tumor progression

Abstract

Cancers of the brain tend to be among the most fatal, due to their rapid rates of growth and the difficulty in transporting therapeutics across the blood-brain barrier (BBB), one of the tightest vascular barrier in humans¹. High-grade glioma, the most common type of primary brain cancer, has one of the worst prognoses of all cancers with a five-year survival rate of ~2.4%²⁻⁴. In addition, an estimated 20% of all cancer patients develop metastatic tumors in the brain⁵. Highly lethal, these stem from circulating tumor cells in brain capillaries that transmigrate into the parenchyma despite the presence of highly-regulated transport mechanisms at the BBB⁶⁻⁸. The lack of physiologically relevant in vitro human BBB models as well as the challenges involved in translating results from animal experiments to the clinic have significantly hindered progress in improving patient prognoses⁹⁻¹¹.

A better understanding of the mechanisms of tumor progression at the brain in a microvascular human brain-on-a-chip model that allows for high spatio-temporal resolution imaging is critical to developing new therapeutic strategies that address tumor extravasation across the BBB and glioma-BBB interactions. In this thesis, we develop an in vitro microvascular model of the human BBB in a microfluidic chip to assess the cellular and molecular interactions between cancer cells and brain stromal cells. The selfassembled BBB vascular networks are generated with induced pluripotent stem cell-derived endothelial cells, primary brain pericytes, and astrocytes. The addition of brain stromal cells resulted in improved barrier function and decreased vessel permeabilities, comparable to in vivo measurements.

In addition, the engineered model has the capability to recapitulate the early steps of the metastatic cascade at the brain and primary tumor progression and interaction with the BBB in real-time via confocal microscopy. The BBB microvascular assay is then employed to obtain biological insights into the roles of brain stromal cells in the extravasation of cancer cells from various primary sites. Particularly, astrocytes are identified to play a major role in tumor transmigration through their secretion of CCL2. This chemokine is internalized by CCR2-expressing tumor cells and promotes their extravasation via both chemotaxis and chemokinesis. The translational strength of our in vitro BBB model was validated in vivo in mouse brains. We uncovered that CCR2 knock-down on tumor cells significantly reduces transmigration and can thus be harnessed as a potential therapeutic strategy to mitigate the early steps of the metastatic cascade at the brain.

Furthermore, we expand upon the current BBB assay to recapitulate the complex tumor-stroma interactions with the incorporation of a high-grade glioma spheroid in the in vitro brain vasculature. Specifically, we explore the mechanisms of drug delivery, across the BBB and into the brain tumor, of layered nanoparticles that bind to endothelial receptors. With this novel platform and in vivo validation in glioma tumor-bearing mice, we demonstrate that transport occurs via transcytosis and is improved with LRP1-binding nanoparticles compared to control carriers, particularly across the vasculature near the glioma tumor. Keywords: cancer, extravasation, blood-brain barrier, microfluidics, organ-on-a-chip, glioma, nanoparticle