Measuring and Mapping Cognitive Decline After Brain Radiosurgery

Overview

Background Stereotactic Radiosurgery (SRS) is a localised radiotherapy treatment for patients with brain metastases or other benign tumours in the brain, like meningiomas. The Investigators do not currently know if, or how much, SRS affects brain function. Patients with brain tumours do not get tested routinely for their brain function. Understanding short- and long-term side-effects is important for SRS. Brain metastases patients have short life expectancies (6-months to 1-year). However, meningioma patients can live 10 years or more. SRS is used to treat both. The Montreal Cognitive Assessment will be used to test the participants' brain function. Quality-of-life questionnaires QLQ-C30 and BN20 will also be used to assess the participants' physical and mental wellbeing . These are specific for patients with brain cancer. Why is it important This study aims to identify areas in the brain that relate to changes in brain function after SRS. These areas can then have the radiation dose reduced to them in future patients, hoping to minimise side-effects. Research Question Which regions of the brain contribute to a decline in brain function following SRS. Study Design This is a single centre observational study with prospective and retrospective collection of data. This study will look at two groups of patients: Group1: Patients will complete the MoCA and two quality-of-life questionnaires before the treatment and every 3 months for a year. Group2: Patients will complete the MoCA and two quality-of-life questionnaires once. The investigators will use these tests, MRI scans and the SRS treatment plan to identify areas of the brain that are responsible for any problems with the participants' brain function. The participants for Group 1 will be recruited from the SRS Clinics, at City Campus, Nottingham University Hospitals NHS Trust. The participants for Group 2 will be identified through the Mosaiq Oncology Information System. This pilot study is funded by the Midlands Mental Health and Neurosciences Network.

Brain metastases affect roughly 20-40% of all cancer patients with a primary extracranial disease. Lamba et at estimated that patients' median survival without treatment is approximately one month. This number increases from one month to between three and 12 months if whole brain radiotherapy (WBRT) is used. WBRT is a radiotherapy treatment that is delivered in five to ten sessions and where the whole brain is irradiated. With the advent of new systemic treatments, prognosis for patients is constantly improving . The widespread use of MRI imaging has also increased the number of patients being diagnosed by improving the detection of very small tumours (one or two millimetres in diameter). Consequently, more people are referred for treatment with a curative intent for their brain metastases. Patients with brain metastases can be treated with surgery, WBRT, Stereotactic Radiosurgery (SRS) or a combination of the above. SRS is the treatment of choice for single or multiple metastases, unless there is a large mass (more than 3 cm in diameter) when surgery would be the preferred option. This is due to the invasiveness of the surgery, SRS is non or minimally invasive, and the associated risks of having a surgical resection. SRS is a highly focused treatment that makes use of stereotactic localisation techniques to deliver a high radiation dose to an outlined area of the brain. The treatment is usually delivered in a single radiotherapy treatment. In prospective randomised controlled trials, it was shown that Stereotactic Radiosurgery is as effective as whole brain radiotherapy, in terms of local tumour control, but the burden of cognitive decline for the patients is less. The primary tumours can vary, but the majority of the patients have a lung, breast, melanoma, kidney or colon primary. Radiation damages the brain, with different areas being more or less susceptible to radiation. The main organs used to optimise radiotherapy treatment plans by reducing the dose delivered to these, include the eyes, optic nerves, chiasma, brainstem, cochleas, lenses and lacrimal glands. These are organs that affect physical functions. Currently, during radiotherapy treatment planning, no consideration is given to areas of the brain that affect cognitive function, such as the hippocampus, basal ganglia, insula and corpus callosum to name a few. The effects of radiotherapy in the brain are well documented but this is mostly for WBRT. The effects of focal radiotherapy, such as SRS, on cognition have not been studied in detail. This pilot study will be using voxel-based lesion symptom mapping (VLSM) to correlate the treatment planning dose maps with clinical outcomes on neurocognitive function. Voxel-based lesion symptom mapping is a technique where the relationship between damage to certain locations of the brain can be corelated with symptoms that affect cognitive function. This is done on a voxel-by-voxel basis. A voxel is a volume in a three-dimensional structure. VLSM has been used extensively in the past in stroke patients, but the literature is scarce for patients with brain metastases. Individual MR images of the brain will be used along with the dose maps from each patient's SRS treatment plan. The dose maps will be overlayed and correlated with results from neurocognitive testing. The identified areas of the brain will then be outlined and the dose level of significance (the level of dose at which cognitive change becomes quantifiable) will be assessed. Benefits The identification of areas in the brain and associated doses with cognitive changes will help advise future guidelines on volumetric dose constraints for those areas in the brain. As lesion location will be correlated with clinical outcomes from neurocognitive testing, a better understanding of the side-effects of the radiotherapy treatment may be gained. This will allow the optimisation of future radiotherapy treatments, to spare these areas thereby minimising the side-effects of the radiotherapy treatment. Finally, with the acquired knowledge, consultants will be better placed to inform patients about the effects that their treatment will have. Patients will therefore be able to make informed decisions about their treatment. This is especially important in this population, as the majority of patients treated with SRS do not have a very long life-expectancy. Methods Magnetic Resonance Imaging or MRI is a medical imaging technique that uses magnetic fields and radio waves to generate detailed images of the inside of the body. MRI is the preferred imaging method used to diagnose tumours in the brain, both benign and malignant, due to its ability to provide high contrast for these tissues. Tumours as small as 2 mm can be easily visualised with the aid of an MRI scanner. The methodology that will be used to analyse the data, VLSM, is well established and has been used extensively in patients with stroke lesions. It uses a regression model that is univariate and can associate the cognitive performance of patients to whether or not a lesion is present in each voxel. In this pilot study instead of using lesions the investigators will be using the dose maps of the radiotherapy treatment plans. The structural MRI scans of the patients will be spatially normalised to a common coordinate system, the Montreal Neurological Institute (MNI) space. The transformations will then be used to fit the dose maps to the same coordinate system. The data will then be smoothed. The voxel-based lesion symptom mapping analysis will involve the use of a Matlab statistical parametric mapping toolbox. For each voxel a group comparison between patients will be undertaken with the use of the dose maps from the radiotherapy treatment plans. The t statistic will be computed based on the results from the neurocognitive testing. A corresponding t-map of the brain will be created showing all the voxels related to the relevant cognitive deficits. A correction for multiple comparisons needs to be carried out to ensure that the rate of false positives will be minimised.