Differential localization of glioblastoma subtype: implications on glioblastoma pathogenesis

Abstract

Introduction: The subventricular zone (SVZ) has been implicated in the pathogenesis of glioblastoma. Whether molecular subtypes of glioblastoma arise from unique niches of the brain relative to the SVZ remains largely unknown. Here, we tested whether these subtypes of glioblastoma occupy distinct regions of the cerebrum and examined glioblastoma localization in relation to the SVZ.

Methods: Pre-operative MR images from 217 glioblastoma patients from The Cancer Imaging Archive were segmented automatically into contrast enhancing (CE) tumor volumes using Iterative Probabilistic Voxel Labeling (IPVL). Probabilistic maps of tumor location were generated for each subtype and distances were calculated from the centroid of CE tumor volumes to the SVZ. Glioblastomas that arose in a Genetically Modified Murine Model (GEMM) model were also analyzed with regard to SVZ distance and molecular subtype.

Results: Classical and mesenchymal glioblastomas were more diffusely distributed and located farther from the SVZ. In contrast, proneural and neural glioblastomas were more likely to be located in closer proximity to the SVZ. Moreover, in a GFAP-CreER; PtenloxP/loxP; Trp53loxP/loxP; Rb1loxP/loxP; Rbl1-/- GEMM model of glioblastoma where tumor can spontaneously arise in different regions of the cerebrum, tumors that arose near the SVZ were more likely to be of proneural subtype (p < 0.0001).

Conclusions: Glioblastoma subtypes occupy different regions of the brain and vary in proximity to the SVZ. These findings harbor implications pertaining to the pathogenesis of glioblastoma subtypes.

Keywords: MR imaging; automatic tumor segmentation; glioblastoma; subtypes; subventricular zone.

Figure 1. Workflow for generation of total CE and centroid density maps

Preprocessed images A. were registered to the Montreal Neurological Institute (MNI) template and segmented according to the IPVL pipeline B. CE volumes were filled C. and centroid of each were calculated D. E. Filled CE volumes from each subject (first three panels) were summed (right most panel) and then converted to tumor density map.

Figure 2. Glioblastoma density map

Total CE probability map revealed that glioblastoma, as a whole, exhibit a strong predilection occurrence in proximity to the SVZ. Red indicates the highest frequency of overlap and light-blue indicating the lowest frequency of overlap.

Figure 3. Glioblastoma subtype density maps and regions of statistically significant subtype localization

A. Subtype-specific density maps were generated using total CE volume. Red indicates the highest frequency of overlap and light-blue indicating the lowest frequency of overlap. Proneural and neural tumors tend to occur in the temporal and frontal lobe, having higher densities in the left temporal region relative to the right. In contrast, the classical and mesenchymal subtypes were more diffusely distributed in the cerebrum, with significantly lower probabilities of overlap (p < 0.001). B. Axial (first row) and sagittal (second row) of statistically significant clusters (p < 0.05) by subtype. Statistical comparisons were carried out using voxel-wise Fisher's exact tests.

Figure 4. Association between SVZ distance and glioblastoma subtype

A. Graphical illustration of the SVZ distance measurement. SVZ distance is color coded so that blue indicates shorter SVZ distances while red indicates high SVZ distances. B. Median SVZ distances for the glioblastoma subtypes. * indicates statistical significance based on student's t-test when comparing the SVZ distance of the proneural subtype to those of the classical or the mesenchymal subtype. The same comparisons were performed for the neural subtype.

Figure 5. Analysis of glioblastoma location in a GEMM by subtype

The location and Olig2 staining characteristics of early glioblastomas were examined in GEMM model of glioblastoma [16]. A. Tumors were identified based on H&E staining (left column) and an adjacent slide stained for Olig2 (right column). Representative images demonstrating Olig2 staining intensity from 0 to 3 are shown. Scale bar in the upper right hand image is 50 μm and applies to all images. B. Olig2 staining of cortical tumors was compared to tumors of the subventricular zone (SVZ), overlying corpus callosum (CC), rostral migratory stream (RMS) and olfactory bulb (OB) in this scatter plot. Lines represent the mean and standard deviation. ****p < 0.0001.