A BMI1/CHD7 signature in medulloblastoma with poor prognosis.

Brain tumours account for a large proportion of childhood tumours. Medulloblastomas are the most common brain cancers seen in young children. They are malignant tumours formed from poorly developed cells at a very early stage of their life. They develop in a region at the back of the head called the posterior fossa in the region of the brain called the cerebellum, but may spread to other parts of the brain. Very rarely, medulloblastomas may spread to other parts of the body. The main treatments available to children with medulloblastomas today are surgery, radiotherapy or chemotherapy. These treatments can be effective and kill the tumour cells in a proportion of patients; however they almost invariably also result in severe side effects which are particularly damaging in young children as the brain is still growing. These side effects mainly affect a child's physical and intellectual development with hearing and visual disturbances, growth and hormonal changes, reduced fertility, behavioural changes, learning problems, difficulties with coordination and secondary cancers. A report recently compiled by the New Philanthropy Capital, has identified brain tumours as one of the worst funded cancer, the cumulative research spend on brain tumours between 2002 and 2011 was less than 1% of all NCRI research spend.

We have recently employed powerful new genetic approaches to identify specific molecular changes that characterize a particularly aggressive subtype of medulloblastomas for which no specific treatment exists. We have established cells from these tumours that can be grown in the laboratory and we have the ability to produce mice that develop the same types of tumours. We will use state-of-the-art methods to study the mechanisms responsible for the formation and growth of these tumours. These studies will lead to the identification of new genes and pathways that can be targeted to stop or reduce tumour formation and growth. We will target these genes and pathways to identify the ones that show most promise as therapeutic targets, which will represent the first important steps towards developing effective, new generation treatments for one of the most aggressive subtypes of this devastating disease.

Medulloblastomas are the most common malignant brain tumours of childhood, accounting for a quarter of all cancer-related childhood deaths. Genome-wide expression, copy number analysis and whole genome sequencing have significantly advanced our understanding of the molecular pathogenesis of these tumours, identifying four distinct subgroups affecting prognosis and predicting response to therapy. Despite improvements in the clinical treatment of these tumours, the outlook for aggressive subgroups remains bleak, while survivors often suffer from pronounced neurocognitive delays resulting from the surgical resection of the tumour and subsequent chemotherapy/radiotherapy.

We have identified a subset of medulloblastomas that exhibit a novel BMI1highCHD7low molecular signature associated with reduced overall survival. Our initial analyses of primary medulloblastoma cells of this particular subgroup and CHD7-deficient primary mouse granule neuron progenitors have led to the identification of three potential mechanisms of BMI1-CHD7 convergence. We will use these cells and novel mouse models to test the biological relevance of these molecular mechanisms in medulloblastoma development. In addition, we will employ state-of-the-art, unbiased, genome-wide approaches (RNA-seq, ChIP-seq and ATAC-seq) to identify novel tumour-promoting pathways in medulloblastomas displaying this signature. The significance of these genes and pathways in tumour formation and growth will be assessed by over-expression and knock-down experiments in human medulloblastoma cells and xenografts.

Comparative analysis of the candidate genes identified in the experimental models with the human datasets followed by functional validation of the most promising candidates in human primary medulloblastoma cells will lay the basis for future high- throughput drug screening to identify new single-agents or combinatorial therapeutics with anti-tumour properties.

A report recently compiled by the New Philanthropy Capital, has identified brain tumours as one of the worst funded cancer, the cumulative research spend on brain tumours between 2002 and 2012 was less than 1% of all NCRI research spend. The severe lack of funding for brain tumour research has been repeatedly highlighted in the media. Investment in basic research with translational potential is urgently needed to tackle this devastating disease.

Brain tumours account for a large proportion of childhood tumours. Medulloblastoma is the most common brain cancer seen in young children. The main treatments available to children with medulloblastoma today are surgery, radiotherapy or chemotherapy. These treatments can be effective and kill the tumour cells in a proportion of patients; however they are almost invariably also causing severe side effects which are particularly damaging in young children as their brain is growing quickly. These side effects mainly affect a child's physical and intellectual development with hearing and visual disturbances, growth and hormonal changes, reduced fertility, behavioural changes, learning problems, difficulties with coordination and secondary cancers.

Importantly, recent experiments have revealed that medulloblastoma can be classified into different subgroups which predict their prognosis, including how they will respond to treatment. We have recently shown that a gene called BMI1 is strongly expressed in these tumours. We have observed that BMI1 is most strongly expressed in a subgroup of medulloblastoma that is associated with a particularly poor outcome.

We will take advantage of next generation sequencing techniques applied to a unique set of experimental model systems to explore the epigenetic regulation of medulloblastoma cells. Candidate genes and pathways identified in this screening will be robustly validated to assess their translational functional relevance (impact on the scientific community). As the selection criteria for identifying candidate genes from the genomic screening will be that they represent targetable pathways and/or carry prognostic value, their validation on human medulloblastoma datasets will directly lay the basis for high throughput drug screening to identify novel and more effective drugs against these currently lethal tumours (impact on patient health and pharma).

It is increasingly clear that interdisciplinary approaches are required to build upon our emerging understanding of brain cancer biology towards translation into new and improved treatments. There is currently a shortage of researchers with the requisite skills to cope with such an interdisciplinary approach to cancer research, hence there is a clear need to train young scientists to work and communicate effectively in this fast-changing, interdisciplinary research environment. This is a collaborative project between four research groups working on various aspects of cancer research, including experimental models of brain cancer, epigenetic regulation of gene expression, next generation genomics, advanced integrated bioinformatics platforms and molecular pathology. The postdoctoral fellow employed on this grant will have an excellent opportunity to train in these various aspects of brain cancer research and will develop essential skills to effectively interact with teams of scientists with diverse scientific background (impact on the scientific community and education).
The groups involved in this project meet regularly and the postdoctoral fellow will attend these meetings and actively present their data. Moreover, the post-doctoral fellow will attend and present at national and international conferences relating to brain tumour research (impact on the scientific community and education).
Within the Blizard Institute there are excellent opportunities to become involved in outreach activities, see public engagement section for details (impact on the general public).