The Etiology and Progression of Brain Tumors

Aims of the study:

• To collect adequate brain cancer tissue material for high throughput morphological, protein, RNA and DNA analyses.
• To combine tissue-related results of these analyses with clinical data to examine the potential of the new biomarkers to assess diagnosis, prognosis and heredity of brain tumors.

Materials and Methods

Paraffin-embedded tumor material

Paraffin blocks of more than 4,000 brain tumours have been collected into the Laboratory of Pathology at Tampere University Hospital (Department of Pathology in Fimlab Laboratory). They have been used primarily for purposes of clinical diagnosis, but once patient or authority consent has been obtained the surplus of the material can be used for research purposes. A neuropathologist has marked the most representative region of each tumour in a tissue slide. Using these markings, tumour tissue regions are then biopsied into the tissue microarray block (Micro-Array Technology, Beecher Instruments, Inc.). Up to 1,000 histological samples can be collected into one tissue microarray block, which can be cut into 200 tissue sections. These sections can be used in various kinds of analysis (immunohistochemistry, fluorescence and chromogene in situ hybridisation and other histological standings). Among the advantages of the method are its high capacity, potential for automation, limited damage to the original tissue block and optimised circumstances for molecular biological analyses. We produced the first Finnish TMA block in 1999. Until now up to 3000 different brain tumors have been sampled to our TMA:s of which 2000 gliomas and meningiomas are used for this study. The following provides examples of projects in which the method is used. Similar strategies are used in the present study:

• We have studied the relationship between carbonic anhydrases and brain tumours. Carbonic anhydrase IX (CA IX) is a hypoxia-induced enzyme that is associated with tumorigenesis. CA IX immunopositivity was found in 80% of 362 astrocytic gliomas collected in tissue microarray blocks. According to multifactorial survival analysis, CA IX intensity was a significant and independent prognostic factor. CA IX is a possible candidate for targeted therapy.
• By using the cDNA-microarray method it is possible to analyse the expression of several cancer genes in one hybridisation. The cDNA-CGH microarray method makes simultaneous use of the cDNA-microarray method and comparative genomic hybridisation. This method allows us to study the relationship between gene copy number and the expression of the particular amplified gene, and also to define, at high resolution, the genes in cell lines that are amplified and over- expressed. The analysis gives a more focused picture of the relationship between changes in gene copy number and their expression in different kinds of tumours. With these methods we have found several new cancer gene candidates that may be involved in the pathogenesis of neuroblastomas. To ensure the role of these genes we have built neuroblastoma tissue micro array blocks and performed FISH and IHC analyses using these TMAs. We succeeded in characterising an amplification region which occurred in almost half of the tumour samples of neuroblastoma patients and which was significantly associated with patient survival.

Fresh tissue and frozen tissue microarrays

Our team at the Department of Pathology has been taking steps to prepare for molecular biological and genetic studies that require fresh tissue archives. To this end we have systematically collected and recorded brain tumour tissue for modern methods of cancer diagnosis and research. The processing, storage and archiving of brain tumour tissue from the neurosurgical operation theatres at

Tampere University Hospital are undertaken centrally at the Department of Pathology frozen tissue laboratory. This work has been ongoing since 1992 and by now more than 1,000 fresh tumour samples have been collected. The sample material is used for exact tumour diagnosis in the phase of primary diagnosis (e.g. genetic 1p19q LOH analysis of oligodendrogliomas). Our novel frozen tissue application serves several new molecular pathological methods. We have developed a completely new method based on the snap frozen technique and produced dozens of samples from different brain localisations (frozen brain array). The following provides one example of a project in which frozen tissue method was used. Similar principles are to be used in the present study:

• We have developed a new method which facilitates the differential diagnosis of brain tumours during operation. Using the Ultrarapid Ki-67 staining method, the proliferation marker Ki-67 can be analysed intraoperatively with the snap frozen technique and light microscopy within 10 - 15 minutes. We were able to ascertain the applicability of the method using frozen tissue sections of gliomas that had been previously collected into our tissue bank. On the basis of their proliferation indices, the gliomas could be divided into different malignancy grades and prognosis groups. This highly specific diagnostic method can be applied for example in situations where therapeutic drugs are placed into an intracranial operation field.

Extracted RNA and DNA

Our laboratory is well equipped and prepared for RNA and DNA studies. Following extraction, RNA and DNA are archived and kept at -700 C for later use. Example of projects that use this method:

• Linkage analysis offers a powerful tool for localising genes that predispose to familial diseases, provided that there is a sufficient number of families with the disease concerned. For linkage analysis, the polymorphic regions of the genome are investigated using markers of chromosomal regions that have been passed on from the family's founder parents to all members with gliomas in the family. We have collected blood and DNA samples from very rare glioma families (with a total of 183 members) for purposes of linkage analysis. On the basis of genome-wide linkage analysis, we found a new chromosome locus that was significantly associated with the familial glioma. Deep sequencing of the samples of familial glioma patients is the following step of our study.

Data collection

Our neuro-oncological material at the Tampere University Hospital Department of Pathology comprises 5,000 tissue samples (4,000 paraffin and 1,000 frozen tissue samples). This is too large a dataset to be managed by individual researchers in their own databases. Our aim therefore is to develop a new integrated research data system for the effective management of the tissue material that has been collected over the past 30 years, including a detailed register on all the samples. The register will also include digital photo material from TMA and other histological slides as well as virtual microscopy slides. The relevant clinical data of the patients (e.g. at least three year follow-up of glioma and meningioma patients operated during 1983 - 2009) is combined with the tissue data with the permission of Finnish authorities and the Tampere University Hospital.

The project observes the Helsinki Declaration, current Finnish legislation and the principles of data protection, laid down by Tampere University Hospital. This study is retrospective and purely observational. The assignment of the medical intervention is not at the discretion of the investigator. The collection of samples for research purposes required each individual patient's informed consent in the familial glioma study. The archiving of diagnostic material in a tissue bank involves no ethical problems. Only excess material from diagnostic samples is used for research purposes, either with the patient's informed consent or with the permission of the relevant Finnish authorities (the National Authority for Medicolegal Affairs of Finland). The Ethical Committee of Tampere University Hospital has given permission for our project (R07042). The familial glioma study is conducted under a separate permit (Ministry of Social Affairs and Health , Diary number 127/08/95). Research permits have also been obtained on the basis of the Finnish tissue law (Valvira: Diary number 7796/05.01.00.06/2011).

Inclusion criteria

all glioma and meningioma patients 1983 - 2009

Exclusion criteria

-