Abstract
The goal of the studies described in this thesis was to identify genetic markers with prognostic or therapeutic implications for the less common glial tumors, the oligodendroglial tumors (OTs), including both pure oligodendrogliomas (Os) and mixed oligo-astrocytomas (OAs), and the ependymal tumors (Es). We screened 29 Os and 39 OAs for chromosomal imbalances using comparative genomic hybridization (CGH). A loss of 1p (-1p) and -19q were most frequently detected and 4 different genetic subtypes were identified. Comparing histopathology, genetic subtype and clinical data, we showed that genetic subtyping correlates with survival, independent of histopathological diagnosis. Furthermore, we confirm the previous observation of Cairncross et al. that -1p is predictive for chemosensitivity. Unequivocal criteria for the delineation of OAs from Os and pure astrocytic tumors (As) are lacking, and application of a more 'strict' as opposed to a more 'relaxed' histopathological approach, resulted in a higher frequency of '-1p/-19q' tumors among the hg-OAs. However, some tumors containing -1p (predicting chemosensitivity) are not classified as OAs and therefore these patients will not be included in the chemotherapy protocols for OTs, underlining the clinical value of genetic analysis. Performing CGH on 20 Es showed that -22 was the most frequently detected imbalance. Combining our results with those of others (58 tumors), we showed that some aberrations occur predominantly in a specific category suggesting age and/or localization related genetic subtypes. Overall, our results show that different genetic subtypes can be identified among OTs and Es which can provide information about the tumor type, grade of malignancy, and chemosensitivity. Genetic analysis appears to be a valuable diagnostic tool, on the basis of which a more accurate estimation of prognosis can be provided and a more optimal and 'tailor made' therapeutic approach can be chosen for the individual patient.
