Object: Diffusely infiltrating astrocytomas are the most common primary brain tumors. As a group, they demonstrate an inherent tendency toward malignant progression. Histological grading using the guidelines of the World Health Organization (WHO) remains the gold standard for predicting the biological behavior of these tumors. Although useful, this grading system is often limited due to small sample sizes and the subjectivity in interpretation. Given the important roles for EGFR and PTEN in the malignant progression of astrocytomas, the authors hypothesized that the fraction of tumor cells with aberrations in these genetic loci would correlate with the histological grade.
Methods: The authors evaluated 217 consecutive diffusely infiltrating astrocytomas that were graded using the WHO guidelines, including 16 diffuse astrocytomas (WHO Grade II), 72 anaplastic astrocytomas ([AAs] WHO Grade III), and 129 glioblastomas multiforme ([GBMs] WHO Grade IV). Cases were evaluated quantitatively using dual-color fluorescence in situ hybridization with probes for the EGFR and PTEN loci and the centromeres of chromosomes 7 and 10.
Results: The population of tumor cells with polysomy of chromosome 7 and the EGFR locus and monosomy of chromosome 10 and the PTEN locus correlated significantly with histological grade. In particular, high-grade astrocytomas (that is, AAs and GBMs) had elevated fractions of tumor cells with polysomy of chromosome 7 and the EGFR locus and monosomy of chromosome 10 and the PTEN locus. Using these findings, the authors generated a mathematical model capable of subcategorizing high-grade astrocytomas. The successful model incorporated only the percentage of tumor cells with polysomy of EGFR and monosomy of PTEN, as well as patient age. The predictions of this model correlated with survival in a manner similar to histopathological grading.
Conclusions: The findings presented in this study emphasize the utility of combining histological interpretation and molecular testing in the evaluation of infiltrating astrocytomas. These results underscore the utility of building a grading framework that combines histopathological and molecular analysis.