CHARACTERIZATION OF ASTROGLIAL VERSUS OLIGODENDROGLIAL PHENOTYPES IN GLIOBLASTOMAS BY MEANS OF QUANTITATIVE MORPHONUCLEAR VARIABLES GENERATED BY COMPUTER-ASSISTED MICROSCOPY

Citation
C. Decaestecker et al., CHARACTERIZATION OF ASTROGLIAL VERSUS OLIGODENDROGLIAL PHENOTYPES IN GLIOBLASTOMAS BY MEANS OF QUANTITATIVE MORPHONUCLEAR VARIABLES GENERATED BY COMPUTER-ASSISTED MICROSCOPY, Journal of neuropathology and experimental neurology, 57(8), 1998, pp. 791-802
Citations number
35
Categorie Soggetti
Pathology,Neurosciences,"Clinical Neurology
ISSN journal
00223069
Volume
57
Issue
8
Year of publication
1998
Pages
791 - 802
Database
ISI
SICI code
0022-3069(1998)57:8<791:COAVOP>2.0.ZU;2-G
Abstract
The current WHO classification places glioblastomas in the astrocytoma category. However, whether or not glioblastomas also show oligodendro glial differentiation remains a matter of controversy. This study inve stigates, at the morphonuclear level, the hypothesis that some gliobla stomas (GBMs) may also represent the ultimate level of malignancy in t he oligodendroglial lineage. Using a series of 164 GBMs, we sought to ascertain whether any of these GBMs exhibited phenotypical characteris tics that were more closely related to oligodendroglial lineages than astrocytic lineages. Phenotypical features were quantitatively determi ned by means of the computer-assisted microscope analysis of Feulgen-s tained nuclei, a process that made it possible to quantitatively descr ibe the]patterns of the cell nuclei (and, more specifically, of their chromatin) through 16 variables, and the distribution of the nuclear D NA content (DNA ploidy) through 8 variables. The phenotypical characte ristics typical of astrocytic and oligodendroglial tumors were analyze d by means of Discriminant Analysis, a statistical multivariate analys is, performed on a series of 65 astrocytic and oligodendroglial tumors . This series consisted of 14 WHO grade II and 19 grade III astrocytom as and 24 WHO grade II and 8 grade :[II oligodendrogliomas. This multi variate analysis enabled an accurate model to be produced that disting uished between astrocytomas and oligodendrogliomas on the basis of 5 c ytometry-generated variables. This model was used to characterize the phenotype of each of the 164 glioblastomas. The results show that of t hese 164 glioblastomas, 6 (about 3.5%) displayed phenotypes that were very similar to oligodendrogliomas, and 141 displayed phenotypes that were very similar to astrocytomas. The phenotypes of the 17 remaining GBMs were too ambiguous to be categorized as having a pure astrocytic or oligodendroglial lineage.