Glioblastoma is the most common malignant brain tumour in adults. Categorised by The World Health Organization as a grade IV glioma, it is the most aggressive, invasive form of glioma with a median survival time of 14.6 months. The incidence in the United States is approximately is 2 to 3 per 100,000. Glioblastoma is about 40% more common in males than in females and the average age at diagnosis is 64 years. Despite aggressive medical intervention including surgical resection followed by radiotherapy and chemotherapy, treatment increases survival by a median of only 2.5 months. Tumour recurrence occurs in the brain, frequently adjacent to the site of the original tumour. New therapies are needed to combat metastatic disease and tumour regeneration after therapy.
Therapeutic resistance is the major cause of death in patients diagnosed with malignant tumours. Tumour heterogeneity has emerged as a major player in therapeutic resistance. In addition to this, treatments frequently contribute to the evolution of therapy resistance by introducing mutations which provide the cell with mechanisms to survive further rounds of treatment. The main aim of my thesis is to identify potential targets within glioblastoma that could be exploited to reduce intratumoural heterogeneity and prolong progression free survival.
We used three independent processes to interrogate phenotypic and genetic heterogeneity within populations of primary human glioblastoma cell isolates. In this first experimental chapter I have investigated whether the cancer stem cell hypothesis may explain the heterogeneity observed within our isolates of human glioblastoma. Cancer stem cells have been a major focus of research since they were implicated in tumour initiation in acute myeloid leukaemia in the 1990s. We tested the cancer stem cell hypothesis using glioblastoma cells. We found that glioblastoma contained “slow” and “fast” cycling populations of malignant cells. Within the slow cycling stem-like population of glioblastoma cells exist a population of cells with superior tumour forming ability. Some of the content of this chapter has been published (Brain, 2011, 134, 1331-1343).
The second chapter, explored chromosomal heterogeneity using fluorescence activated cell sorting and Flo-based assays to demonstrate that the “ploidy” of the malignant cells within primary glioblastomas was highly variable and was associated with highly diverse behaviour with respect to in vitro growth and proliferation as well as drug resistance. I also showed that hyperdiploid glioblastoma cells were capable of tumour formation, more metabolically active and sensitive to glycolysis than their euploid counterparts. Some of the content of this chapter has been published (Molecular Biosystems, 2014. 10, 741-758)
In the final experimental chapter we used another platform to interrogate the heterogeneity of primary glioblastoma cells. Cell surface proteins have been identified as prognostic markers and predictive of response to therapy in a range of human tumours. We analysed the expression of 118 cell surface proteins on primary human glioblastoma cells and identified a small cohort of proteins which in concert provide a molecular signature of GBM. This data highlights some potential targets for use in the design of antitumour specific therapeutics.
GBM is an incurable heterogeneous tumour with a poor response to treatment. The work presented within my thesis identifies a number of potential therapeutic targets which may be used to inform novel treatment strategies and targets. In particular, I provide new data that may be used to reduce the impact of the highly heterogeneous polyploid populations of cells that arise in tumour evolution. Moreover, the targeting of tumour heterogeneity may be used to design more effective therapies to reduce side effects, avoid the induction of further mutational events caused by current therapies and hopefully extend the progression free survival time of patients by slowing tumour growth.