Dr Tracey Warr Research

Radiological and Molecular markers of subsequent behaviour in adult low-grade glioma

Principal Investigator: Dr.Tracey Warr, Lecturer in Molecular Tumour Genetics, Department of Molecular Neuroscience, Institute of Neurology, National Hospital for Neurology & Neurosurgery, Queen Square, London.


In this pilot study, we aim to determine whether there are genetic correlates of tumour behaviour in a unique cohort of adult well-characterised low-grade gliomas (LGGs) who have been studied over the preceding six years in a serial multimodality imaging project (sponsored by the SDBTT Astro Brain Tumour Fund). This imaging study has defined a number of MRI parameters which are associated with early progression, usually due to malignant transformation. The proposed molecular study will characterise tumour samples from 40 patients at the genomic and expression level in an attempt to determine whether information from biopsy samples taken at diagnosis correlates with the results from the imaging data and subsequent tumour behaviour. Specifically, we wish to identify molecular markers which predict the rate of malignant transformation of these low grade tumours.


Gliomas account for more than 70% of all brain tumours, but current knowledge on the survival of adult glioma patients and on factors that are predictive of outcome is based largely on clinical trials on patients with malignant gliomas (anaplastic astrocytoma and glioblastoma multiforme). LGGs grow slowly when compared with their malignant counterparts but undergo malignant transformation to high-grade gliomas at a variable and unpredictable point in their natural history.  Most studies that have attempted to define prognostic variables have been retrospective – only four prospective randomised studies have been conducted and three of these have shown that increasing patient age and tumour volume correlate negatively with survival.

Similarly, a limited number of molecular genetic or immunohistochemical studies to determine prognostic markers have been carried out in LGGs and the majority of these have examined overall survival rather than time to malignant transformation. There is some evidence that abnormalities of the p53 tumour suppressor pathway may be of prognostic value.  In some studies, overexpression of the p53 protein, presence of p53 mutation and p14ARF hypermethylation have all been shown to be independent marker of progression-free survival (PFS) in grade II astrocytoma.  However, in a third series of grade II astrocytoma and oligoastrocytoma, there was no prognostic impact of p53 status on survival or time to malignancy.

We have recently  examined p53 mutation status in a panel of 38 patients with low grade astrocytoma that subsequently progressed to higher malignancy for which we had tumour biopsy samples at the time of diagnosis and at recurrence. In these patients, p53 mutations were significantly associated with rapid recurrence. Other molecular markers which have been demonstrated to be significantly associated with shorter PFS and overall survival in LGG include hypermethylation of the MGMT gene and overexpression of  PDGRF and EGFR. Allelic loss on chromosome arms 1p and 19q is an established marker of chemotherapeutic response and long-term survival in patients with anaplastic oligodendroglioma and 1p loss may also identify other treatment-sensitive malignant gliomas.  Tumours with preserved 1p can be divided into two subgroups those with p53 mutations which may also respond to chemotherapy but recur quickly and those without p53 mutations, which are poorly responsive, aggressive tumours that are clinically and genotypically similar to GBM.  However, the influence of 1p/19q status on time to malignant transformation has not been well documented in low-grade oligodendroglioma, oligoastrocytoma or astrocytoma.

There have also been few global genomic analyses of LGG to identify genetic prognostic markers and all but one of these have been limited to conventional metaphase comparative genomic hybridisation (CGH) studies.  This is in contrast to malignant glioma where CGH using array technology has been used to identify novel genetic aberrations associated with prognosis including amplification of SKP2 (associated with poor outcome) and gains of CDKN1C and INS and loss of TBR1 (significantly correlated with longer survival.

We are currently studying a large (> 60 patients) cohort of adult subjects with LGG and have quantified various imaging parameters, which correlate with clinical and radiological behaviour in individual tumours.

We now wish to extend these observations by analysing biopsy material from these patients both retrospectively and prospectively (on new subjects recruited to the imaging study) using a combination of immunohistochemical and genomic analyses. In particular we plan to ascertain the status of molecular markers including p53p14ARFMGMT, PDGFR, EGFR and 1p/19q deletions and to carry out array CGH to identify new regions of amplification or loss  and to correlate the molecular findings with the radiological status  of these LGGs. Ethics approval has been obtained from the National Hospital Local Ethics Committee to carry out this study.


Our understanding of how adult low-grade gliomas behave is hampered by the obvious problems of sampling the tumour at different times in the patient’s life. In order to circumvent this problem, we have chosen to study LGGs using state of the art imaging called Magnetic Resonance Imaging (MRI) and to develop a range of computerised methods to analyse the scans and to look at different aspects of the tumours behaviour. We are now hoping to extend the study by analysing samples from the patients’ tumours using a variety of laboratory methods which tell us about the genes and the proteins in individual tumours.

In this way we hope to improve our understanding of what changes occur in these tumours before they turn malignant both at a level visible by scanning as well as at the level of individual cells and molecules. By doing this we would then be able to identify patients who are at risk of malignant transformation before they come to any harm through rapid tumour growth. In this way we hope to see whether it is possible to identify more aggressive tumours early on and thereby treat the patient at an earlier stage than we would otherwise have done. In so doing we would hope to be able to reduce neurological disability, improve quality of life and even possibly overall survival.

The funding requested for this project will cover the cost of the research staff, laboratory materials and scan costs.

Grant Funding Required: £99,592 over 18 months from May 2007