Identification and immuo-optic targeting of tumor stem cells in patients with glioblastoma

Central nervous system malignancies accounts for 1% of all adult cancers. The most common primary brain neoplasm is a glioma, which accounts for approximately half of all intracranial neoplasms. In patients with glioblastoma multiforme, the most malignant of gliomas, median survival has remained 9 to 12 months for decades.

The treatment of gliomas is based on surgical resection, but this is inadequate. The reason for this shortcoming has been hypothesized to be a result of tumour infiltration into healthy tissue, thus making it non resectable. Adjuvant radio- and chemotherapy may prolong the patients life, but eventually most patients relapse.

For nearly one hundred years, glia cells were thought to be the only dividing cells in the postnatal brain, thus, making them the candidates susceptible to malignant transformation. As the concept of neuronal stem cells emerges, the cellular antecedent of gliomas is being once again reviewed in the light of this new evidence. As stem cells already have the machinery necessary for self-renewal, fewer mutagenic events may be necessary for tumourigenic transformation than in differentiated cells. Neural stem cells also exhibit properties similar to tumour cells, i.e. high motility, diversity of progeny, robust proliferative potential, association with blood vessels and white-matter tracts and an immature expression profile.

Prospective isolation of tumour initiating cells is possible using tumour specific antigens. We are currently working on the isolation and characterization of such tumour initiating cells using primary cell cultures, both as floating neurosphere assays and adherent cultures, and combining this with FACS and MACS to prospectively isolate the putative tumour initiating cells.