Treatment > Cerebral Edema

Meeting Abstract

Effect of infiltrating tumor cells on the blood-brain barrier in a neonatal rat glioma model

Lissa C. Baird, Peter Canoll, and Jeffrey Bruce

Department of Neurological Surgery, The Gabriele Bartoli Brain Tumor Research Laboratory [L.C.B., J.B.], and Department of Pathology [P.C.], Columbia University College of Physicians and Surgeons, New York, NY, USA

Introduction. Peritumoral edema is a serious complication of malignant gliomas. The exact mechanism behind the formation of this edema is not completely understood. Malignant gliomas are characterized by a diffuse infiltrative growth pattern within the brain parenchyma. Although vessels within the primary tumor mass are known to lack normal blood-brain barrier function, the effect on the parenchymal vasculature by infiltrating tumor cells has yet to be established. The neonatal rat glioma model demonstrates an infiltrative pattern of growth whereby tumor cells migrate along parenchymal host blood vessels. In this study we investigated the effect of infiltrating C6 glioma cells on the blood-brain barrier.

Methods. Neonatal rats were stereotactically injected with GFP-labeled C6 glioma cells at 3 days of age. Evans blue, an albumin-binding dye, was used to assess vascular permeability. Animals were injected intravenously with 2% Evans blue at multiple postinjection time points. The dye was allowed to circulate 60 minutes, after which animals were perfused with saline until colorless effluent was obtained. Animals were then perfused with 4% paraformaldehyde. Brains were sectioned into 100-micron-thick slices. Immunohistochemistry was done with a rat endothelial cell marker (RECA) and an astrocyte marker (GFAP). Brain sections were analyzed under fluorescent microscopy.

Results. C6 glioma cells were observed migrating away from the primary tumor mass along parenchymal blood vessels between endothelial cells and astrocyte end-feet. Red fluorescence was seen throughout the primary tumor mass, indicating Evans blue extravasation. Evans blue fluorescence was absent from the contralateral hemisphere and areas of the brain distant from the tumor site. At 10 days post tumor injection, Evans blue extravasation from parenchymal vessels ensheathed with tumor cells was absent in most vessels. At 15 days post–tumor injection, a significant amount of Evans blue extravasation from parenchymal vessels ensheathed with tumor cells was observed in many vessels. Conclusions. Blood-brain barrier breakdown within the primary tumor mass is well known and clearly demonstrated in this study. The infiltration of C6 glioma cells along parenchymal vessels appears to have a time-dependent effect on vascular permeability to Evans blue. The data suggest that dye time glioma cells alter the vasculature in a way that leads to the breakdown of the blood-brain barrier and an increasing leakiness to vascular contents. A possible contributing factor to this permeability could be the disruption of the astrocyte-endothelial cell interactions important for blood-brain barrier formation and maintenance. Likewise, it is possible that angiogenic factors known to be secreted by glioma cells are altering the blood-brain barrier of parenchymal vessels in contact with the infiltrating cells.

Future Directions. Studies are currently underway to characterize the effect of glioma cell migration on aquaporin-4 and blood-brain barrier-related proteins such as zo- and claudin-. The time-dependent effect on vascular permeability and blood-brain barrier breakdown caused by infiltrating tumor cells has clinical implications for edema control and chemotherapeutic efficacy in the clinical setting.

Copyright © 2003 by the Society for Neuro-Oncology