Rania Khzouz
Research Mentor: Dawn Morario
Mentor Department: Not Available, Not Available
Author(s): Rania Khzouz, Mohamad Orabi, Ali Dabaja, Mark Slayton, Nathan Merril, Sofia Merajver
Session: Session 3 (11:00 AM – 11:50 AM)
Presentation Type: Poster 79
Abstract
Despite extensive research into brain metastasis, targeting metastasized breast cancer cells remains a significant challenge in oncology. While in vivo studies offer physiological relevance, they often lack the real time imaging and precise quantification necessary to understand cellular extravasation from blood vessels into the brain parenchyma. In this study, we developed a microfluidic device using soft-lithography to mimic the blood brain barrier (BBB) on-chip. The platform consists of an upper channel simulating vasculature, a lower chamber mimicking the brain parenchyma, compromised of normal human astrocytes in a 10% w/v collagen matrix, and a 20µm polycarbonate membrane with 5µm pores. To establish the BBB, the membrane was coated with 2% w/v Matrigel before seeding human cerebrum microvasculature endothelial cells. Following endothelial adhesion, breast cancer cells were introduced into the upper channel. We evaluated the efficacy of two drugs, Gemcitabine and Gilteritinib, using a custom AI model to quantify cancer cell extravasation from 3D confocal images. Our findings revealed distinct dosing requirements: Gemcitabine exhibited high toxicity toward astrocytes and endothelial cells, necessitating an off-chip pre-treatment approach for cancer cells. Conversely, Gilteritinib showed no significant toxicity to healthy cells, allowing for direct on-chip dosing. This study demonstrates the advantages of a tailored on-chip platform for BBB research, providing a novel methodology that integrates AI-driven quantification with sophisticated cell seeding techniques for drug discovery.
