Histogram Analysis of the Microvasculature of Intracerebral Human and Murine Glioma Xenografts

The purpose of this study is to examine the usefulness of histogram analysis combined with vessel size index (VSI) magnetic resonance imaging for the specific characterization of brain tumor microvasculature in a panel of six volume-matched glioma xenografts. Using a simple descriptive histogram analysis, significant differences of the mean tumoral VSI (P = 0.0035 for 9L, P = 0.008 for glioma mix, P = 0.05 for C6), the 75th VSI percentile (P = 0.003-0.075) as well as the 25th and median blood volume (BV) percentiles were found in murine gliomas compared to their contralateral healthy brain. Using a segmented histogram analysis, dilatation of already existing vessels in murine gliomas and development of new small caliber vessels in human glioblastomas were suggested. Most gliomas showed a higher proportion of pixels with BV below 1% (glioma mix [21% vs 1%], Glioblastoma 2 (GBM2) [9% vs 3.7%]) and a smaller proportion of pixels with BV in the range 1.7-6.3% (65 vs 90% for glioma mix, 80 vs 85% in GBM2) relative to their contralateral part. In glioblastomas, VSI and BV distributions were similar to normal brain distributions and in agreement with immunohistochemical findings. The histogram analysis of VSI and BV heterogeneity in experimental brain tumors allowed detection of microregional differences in gliomas from different origins. Magn Reson Med 65:778-789, 2011. (c) 2010 Wiley-Liss, Inc.