Investigation of local brain invasion by cerebral metastases and implications for clinical management

In the Western world, cancer affects more than a quarter of the population and many of those will die of this disease. The majority of deaths from the commonly occurring cancers arise not as a result of the growth of the primary tumour, but from its spread, often to distant sites in the body. One of the major sites of secondary spread of the commonly occurring cancers like breast, lung and melanoma is to the brain. In Liverpool, patients with secondary brain tumours or "metastases" are treated with either radiation to kill the tumour cells, surgery or a combination of these; such treatments can increase the average life expectancy by several months, particularly if the original cancer is under good control with drugs. The number of people with brain metastases, however, is rising and the treatments are becoming more and more complex, therefore markers are urgently needed to identify those patients who are most likely to benefit from treatment. Currently there are no such markers for brain metastases. In addition, we do not fully understand how deeply into the normal brain these metastases penetrate and whether this is different for different primary cancers. This would be useful as it would tell us how aggressive the tumour is and help decide what treatments we should give. We are therefore looking in more detail at this boundary between the tumour and the normal brain in patients with brain metastases.

We are first looking at advanced types of MRI scans called diffusion scans, which look at how easily water moves through the brain and therefore how dense the tumour border is. We will try to predict how the tumour will behave by taking readings at the boundary between the tumour and the brain from these scans. Next, we want to understand what is actually happening in the tumour and surrounding brain tissue that is being measured on those scans, so we will need to look at samples of brain metastases removed at surgery. A number of tissue or biological markers of invasion have shown promise in primary brain tumours and in other solid cancers therefore it is crucial to confirm whether they are also related to invasion and survival in brain metastases. The research is being carried out by the neuro-oncology team at The Walton Centre in cooperation with the Institute of Integrative Biology at Liverpool University. It involves looking at both historical patients who have died of brain metastases in the last few years, analysing their samples (which have been stored in our regional "tissue bank" with their consent) and MRI scans as well as investigating newly diagnosed patients who are being treated with surgery. We are not giving any new or experimental treatments, simply using the "spare" tumour tissue removed at surgery with their consent to perform these additional tests and taking readings from their MRI brain scans.

This research has a number of important results for how we treat patients with these tumours. For the patients themselves, their families and their doctors, it means we may be able to give a better estimate of their survival and possibly give treatments to patients who were previously thought to have a very poor prognosis. On the other hand, we may spare some patients undue harm or discomfort by not giving treatments that may not increase their life expectancy. Because the work focuses on the interface between the brain and these tumours, it will give an idea of how invasive brain metastases from different cancer types are and this might lead us to treat much wider areas around these tumours than we previously thought we had to. This would be very important for patients being treated with surgery or with radiation and could lead to a clinical trial, in which we give different combinations of treatment (such as localised versus whole brain radiation therapy) to different patients to try and find out which gives the best outcome.

Aims:
1. Identify MRI markers of brain metastasis invasion
2. Histopathology assessment of the tumour/brain boundary
3. Identify biological markers of brain metastasis invasion and prognosis

Resource: I will use a historical cohort of 121 patients with paraffin embedded tissue from operated brain metastases in the Walton Research Tissue Bank (WRTB) (approval granted - see Ethical Information) as well as obtaining tissue from new patients undergoing image guided surgical resection of brain metastases at Walton and being entered into the bank.

Methods:
1. Preoperative diffusion MRI scans will be used to record a number of measures both from within the tumour and across the brain-tumour interface. Pilot data has already confirmed the validity of this method (presented and submitted for publication) and is suggestive of relationships between these measures and patient outcomes. Suitable types of scans identified so far in 76 cases.

2. & 3. Using histology and immunocytochemistry prospectively compare cellularity, extraceullar matrix density, vascularity and expression of the known metastasis inducing proteins AGR2, S100A4, S100P, osteopontin & FANCD2 in tissue samples taken at the tumour boundary during routine image-guided surgery with MRI measures in (1) from co-localised preoperative advanced MRI scans. Method developed and 6 cases performed in this manner in 2 month pilot, we operate on c. 70 metastases per year and will recruit at least 40 cases in 18 months.

Outcomes: Actuarial survival analyses stratified for established oncology predictors (age, number of metastases, whole brain radiotherapy) in addition to the radiological and biological markers identified. Correlation of radiological and biological markers of invasion with different primary cancers.

Conclusion: The proposed research will increase understanding of how invasive brain metastases are and what the markers of this are, informing future clinical treatment.

Translational clinical impact
The proposed research will have an immediate translational benefit with use for oncologists, primary care physicians, neurosurgeons and neuro-oncologists who are treating patients with secondary brain tumours or metastases. With the rising incidence of this condition these represent an increasingly large group of clinicians. The proposed research will increase understanding of how much metastases invade normal brain, allowing better predictions of survival tailored to the particular cancer being treated. Further, knowledge about the degree of invasion seen on MRI scans for example, will potentially change the approach to how particular brain metastases are treated. We would hope this would lead to more individual treatment planning and therapy.

Patients and carers
For the patients suffering with newly diagnosed brain metastases and their families, the proposed research would increase the understanding of how the tumour affects the brain and therefore what type of treatments they undergo. This would apply for surgical treatment - someone with a more invasive tumour might need more aggressive surgery for example - as well as oncology or non-surgical treatments - if you had a less invasive metastasis you might not need a killing dose of radiation to the whole brain or if you were undergoing stereotactic radiosurgery such as gamma knife treatment, you might need a larger or smaller margin of normal brain around the tumour. Finally, the proposed research should lead to better estimates of survival for patients with these tumours depending, for example, on MRI scan findings and such information is of considerable emotional and practical value.

Government and public policy
At a governmental and policy level, the proposed research will have implications for how patients with brain metastases are allocated treatments. Better predictions of invasion and survival, from MRI scans non-invasively or after surgical biopsies would be expected. If this prevents unnecessary treatments, as well as preventing patient harm and distress it would be expected to allow savings to the health service, the government and the economy. Furthermore, in selecting patients who may have previously been felt to have a poor prognosis for life prolonging treatment (because of a more favourable finding at MRI scan for example), it could improve quality of life and reduce the extended burden on support services such as palliative care facilities. The proposed research would also allow public health physicians and those at a commissioning level to better predict the service needs in the future and allocate appropriate resources, again benefiting the health service and the economy as a whole.

Commercial impact
At a commercial level, the proposed research has already involved working with representatives from IT, radiology and medical instrument companies. The techniques and methods used would be expected to lead to wider use of post processing image software and image guided surgical tools in NHS practice, with economic benefit to these companies as well as the UK economy. We will continue working with them to trouble shoot these practical issues as the project develops and work with them to develop solutions which may have commercial applications.

Individual research fellow
Finally, the proposed research Fellowship will have a significant impact on my career as I aim to develop into a practising academic neurosurgeon. The skills I particularly hope to acquire during the Fellowship include the ability to develop a research question, plan laboratory and clinical studies and conduct research in the context of the NHS as background to an anticipated post-doctoral position as a clinical scientist.