Genetic analysis of human astrocytic gliomas : xenografts as a research tool

Abstract

Cancer is a genetic disorder of somatic cells. An accumulation of genetic changes is believed to result in neoplastic transformation and malignant progression. In humans, astrocytic gliomas are the most common primary tumors of brain tissue. They are malignancy graded according to World Health Organization criteria. In adults, the tumors include astrocytoma, (grade II), anaplastic astrocytoma, (grade III) and the most malignant and common form glioblastoma (grade IV). A significant proportion of astrocytomas (A) and anaplastic astrocytomas (AA) progress into tumors of higher malignancy grade. Glioblastomas (GB) arise by progression, but may also develop de novo. Several genes and chromosomal regions are commonly found amplified, deleted or mutated in these tumors, suggesting a role for these abnormalities in the development and progression of astrocytic gliomas. An accumulation of genetic lesions is seen with increasing malignancy grade.

The aim of the studies described in this thesis was to document genetic abnormalities in human astrocytic gliomas and to establish human astrocytic gliomas as xenografts permitting the development of material for extensive and detailed studies of genetic abnormalities and their cell biological consequences. In addition, xenografts provide tumour tissue with well-documented patterns of genetic abnormalities, in an in vivo environment, for preliminary testing of novel treatment modalities. Other aims included the determination of whether the xenografts are representative of the patient's tumour, and to study whether further genetic abnormalities arise in xenografts and, if so, to identify these.

A series of 198 astrocytic gliomas was examined for loss of genetic material on chromosome 10. The microsatellite analysis revealed two commonly deleted regions on 10p, detected in all malignancy grades. Deletions on 10q were large and found mainly in the glioblastomas. The findings suggest that a number of loci on chromosome 10 may harbor tumour suppressor genes relevant to the development and progression of these tumors. Loci on 10p may be involved in the development of astrocytomas, and loci on 10q important in the progression of such astrocytomas to glioblastoma. The tumor suppressor gene PTEN is located at 10q23, and was studied in 207 astrocytic gliomas from patients, 23 glioblastoma xenografts and 13 glioma cell lines. Homozygous deletions were found in 7% of the glioblastomas and 40% of glioblastomas showed mutation of a single retained allele. The mutations mainly affected structurally conserved regions. PTEN loss was selected for in the glioblastoma xenografts. Only a few anaplastic astrocytomas harbored mutations and no alterations were found in astrocytomas. This suggests that PTEN abnormalities are associated with astrocytic tumour progression.

A glioblastoma xenograft model was established. Of 47 directly and subcutaneously transplanted patient tumors, 24 grew on serial passaging. To preserve the system and to increase accessibility, a cryopreservation protocol for long term storage of viable tumour tissue from any passage was developed and demonstrated to work successfully for the majority of the xenografts. A number of genes previously known to show aberrations in human glioblastomas were found to become mutated in the xenografts, as studied by comparing the genetic analysis of patient and xenograft tumour tissues. Acquisition of additional aberrations during passage as xenografts was detected. For example, one tumour with an amplified EGFR gene developed a rearrangement of the gene during xenograft cultivation. This demonstrates the significance of this aberrant variant reported earlier in clinical material. Furthermore, selection for tumour cells harboring genetic abnormalities deregulating the G1/S-phase transition control and p53 pathways was found in the xenografts. Commonly, abnormalities affecting both G1/S-phase transition control and p53 pathways developed simultaneously.

In summary, the patient tumors were characterized for genetic abnormalities. Some of the glioblastomas were successfully cultivated as xenografts. Cryopreservation was developed, and the xenografts characterized for the acquisition of and selection for genetic mutations. The characterized xenografts now offer the possibility of analyzing in detail the development of genetic changes, their cell biological consequences and the selection process, as well as offering the possibility of testing novel treatment modalities.