Molecular markers of gliomas: a clinical approach.

Over the last decade, the knowledge on the molecular genetic background of gliomas has dramatically increased. This information provides the basis for the molecular target therapies and molecular tests serve to complement the subjective nature of histopathologic criteria and add useful data regarding response to treatments and prognosis. In particular, the use of loss of heterozygosity (LOH) and methylation specific polymerase chain reaction (PCR) (MSP) based testing of gliomas is already in place and used clinically in several centers. This paper provides a brief overview of these molecular genetic aberrations and discusses the clinical utility, as well as the advantages and disadvantages of such approach. Newly developed molecular techniques, such as LOH testing, fluorescence in situ hybridization (FISH), DNA sequencing and MSP, are currently being employed in assessment of gliomas in some laboratories. However, the clinical use of some markers and the context in which the information obtained should be used are still not entirely understood. Therefore, this paper will focus on validation and implementation of molecular testing in gliomas, with emphasis on LOH on chromosomes 1p, 19q, 17p and 10q and O(6)-methylguanine-DNA methyltransferase (MGMT) methylation status.

Reclassification of oligoastrocytomas by loss of heterozygosity studies.

Oligoastrocytomas (OAs) are WHO grade II or III tumors composed of a mixture of 2 neoplastic cell types morphologically resembling the cells in oligodendrogliomas and diffuse astrocytomas. Investigations on the genetic profile of OAs may yield important information for their classification and help for their clinical management. We have studied, in 94 OAs (46 WHO grade II and 48 WHO grade III), the patterns of loss of heterozygosity (LOH) of 4 genomic regions: 1p, 19q, 17p and 10q. Results were as follows: LOH 1p was present in 46% of the tumors; LOH 19q in 45%; LOH 17p in 22%; LOH 10q in 16%. LOH 1p and 19q were associated in 32%, other LOH associations were rare (<3%). Patients had a median follow-up of 30 months. Patients without LOH on 1p had shorter progression free survival than patients with LOH on 1p: 30 vs. 132 months, p < 0.0001. MRI indicated that tumors without LOH on 1p were often temporal (p < 0.02), and showed signal inhomogeneity on T1 and T2 images (p < 0.02) and contrast enhancement (p < 0.04). Thus, LOH on 1p identifies two subgroups of OAs. OAs without LOH on 1p behave like WHO grade II or III diffuse astrocytomas: they have shorter survival, MRI characteristics implying malignancy and genetic alterations associated with tumor progression. OAs with LOH on 1p, on the other hand, behave like WHO grade II or III oligodendrogliomas with 1p loss: they are associated with longer survival and do not have MRI or genetic alterations associated with malignancy. These findings suggest that the definition of OAs or mixed gliomas could be reshaped in agreement with the genetic information.