Molecular genetic analysis of deep-seated glioblastomas

Abstract

Glioblastoma can be divided into genetic subsets. The most prominent criterion for dividing glioblastomas into subsets is the dichotomy between TP53 mutation and EGFR amplification, two genetic alterations that almost never coincide in the same tumor. Approximately one third of glioblastomas have TP53 mutations, one third have EGFR amplification, and one third have neither. When viewed in terms of tumor progression, secondary glioblastomas have a much higher incidence of TP53 mutations than do primary glioblastomas. When viewed in terms of the age of tumor onset, glioblastomas in young adults are likely to have TP53 mutations. However, no correlations have yet been found between the tumor locations and the genetic subsets. In this study, we evaluated the associations between the glioblastoma sites and the genetic subsets defined by the presence of the TP53 mutation or EGFR amplification in nine deep-seated glioblastomas of the thalamus and basal ganglia. All nine tumors were clinically defined as primary glioblastomas. Our investigation revealed that all tumors had TP53 mutations and none had EGFR amplifications. These findings suggest that glioblastomas deep-seated in the thalamus and basal ganglia can be grouped into a subset of glioblastomas with TP53 mutations, akin to the subsets of secondary and younger adult glioblastomas. The locations where the glioblastomas originate may be associated with the genetic features.

Introduction

Glioblastoma, the most malignant type of brain tumor, usually arises in the cerebral hemisphere and rarely in deep-seated structures such as the thalamus and basal ganglia [1], [2], [3]. With recent advances in molecular genetic analyses, glioblastomas can be subdivided into genetic subsets. In an unselected series of glioblastomas, approximately one third of glioblastomas, primarily those in older adults, have EGFR amplifications, while another third, primarily those in younger adults, have TP53 mutations. Of these alterations, the EGFR amplification and TP53 mutation occur in a mutually exclusive manner [3], [4], [5]. “Primary” (de novo) glioblastomas are likely to have EGFR amplifications, while “secondary” glioblastomas occurring after previous diffuse astrocytomas tend to have TP53 mutations [6].

Pediatric glioblastomas of the type frequently involving thalamic and brainstem regions have different genetic-clinical correlations distinct form those seen in adult glioblastomas. Primary glioblastomas in children tend to have the p53 pathway inactivation and no EGFR amplifications [7]. The distinct genetic features in primary glioblastomas of adults and children prompt speculation that there may be several progenitor (tumor-initiating) cells related to the periods of tumor initiation during the patients' lifetimes. Given the difference of preferential tumor location between primary glioblastomas in adults and primary glioblastomas in children, however, it may be fair to speculate that different types of progenitor cells reside in each location. For example, brainstem glioblastomas, a type likely to occur in children as primary tumors, tend to have TP53 mutations but no EGFR amplification [8], [9]. However, the genetic features of supratentorial deep-seated glioblastomas of the thalamus and basal ganglia, another subgroup likely to occur in children, are still unclear and relatively unexamined.

To evaluate whether the locations of glioblastomas are related to the specific genetic subsets, we examined a series of nine deep-seated glioblastomas of the thalamus and basal ganglia to identify mutations in the TP53 gene and amplifications of the EGFR gene.