Improving the diagnosis and treatment of childhood cancer through functional imaging

Overall Vision and Aim: To translate functional imaging into routine clinical use for the diagnosis and treatment of childhood cancer and facilitate its incorporation into the investigation of children with other diseases. Key Objectives: 1) To improve non-invasive diagnosis and characterisation of children’s brain tumours through the use of functional imaging. 2) To Integrate functional imaging into the standard clinical management and treatment protocols for childhood cancer across Europe. 3) To ensure a structured introduction of functional imaging for children on 3T Magnetic Resonance Imaging scanners, capitalising on the improvements they make to improve information whilst reducing scan times. Background and Rationale The focus of Dr Peet's research has been to develop functional imaging for the diagnosis, treatment and understanding of childhood brain tumours and more recently to expand this to children with other diseases. This programme aims to translate these advances into improvements in routine clinical care. Cancer is the most common cause of death from disease in childhood and brain tumours are the most common solid tumours in this age group. Magnetic resonance imaging (MRI) is the main investigation used in children with brain tumours and is increasingly used in children with tumours elsewhere in the body. MRI provides images of exquisite detail, enabling the site of the tumour and its relationship with surrounding structures to be visualised. However, it does not answer many questions of clinical importance, for example, the tumour type, how aggressive it is and whether it will respond to treatment. Attention has turned towards developing imaging methods which can give information on tissue properties, such as their chemical makeup, blood flow and cellular density(1). These methods, commonly termed functional imaging, provide this information non-invasively and there is increasing evidence that it can answer many of these questions of clinical importance. Whilst Functional Imaging methods span many different technologies, Dr Peet's work has focussed on those which can be performed on standard MRI scanners facilitating their incorporation into clinical use. Applying functional imaging techniques to children is particularly challenging but the non-invasive nature of the methods means that they are often the patients who stand to gain most. The relative rarity of childhood diseases means that multi-centre studies must be undertaken and this causes an extra level of complexity. Despite this, there is sufficient evidence now available to recommend the clinical use of functional imaging in many clinical situations(1). The challenge is shifting from one of development and early evaluation to translation and evaluation within well defined clinical settings. This programme will aim to make functional imaging a routine clinical tool for investigating childhood cancer in major centres across Europe and further its use for other diseases of childhood. Techniques and Facilities The main techniques which will be used in the programme are all advanced MRI techniques. Collaboration with experts in other functional imaging techniques such as positron emission tomography will occur through national and international groups but these techniques will not be the focus of Dr Peet's work. The main focus will be on magnetic resonance spectroscopy (MRS) which provides a chemical profile of tissue, diffusion weighted imaging (DWI) which provides a measure of cell viability and cellularity, diffusion tensor imaging (DTI) which measures tissue structure, and perfusion MRI which measures tissue blood vessel properties. All of these methods are commercially available and commonly used in many major centres. However, their routine implementation in clinical practice is hampered by a lack of agreed quality control measures, acquisition protocols and analysis techniques for specific clinical scenarios. An additional challenge is that technical advances in this field take place frequently and need to be taken account of. Protocols which are adaptable enough to allow for this are essential. Over the next five years there is likely to be a widespread introduction of higher field (3T) scanners into hospitals. These scanners have major advantages over 1.5T scanners for functional imaging and will make an important contribution. Ensuring that this is achieved in an optimum fashion will be an important component of the programme and will be greatly facilitated by the NIHR 3T MRI Centre at Birmingham Children’s Hospital (BCH) which is located next to its Wellcome Clinical Research Facility. Diagnosis of Childhood Brain Tumours The functional imaging techniques MRS and DWI can be used as diagnostic aids with high accuracy for childhood brain tumours (2,3). Diagnostic decision support tools based on pattern recognition applied to these data types (2,3) will be tested prospectively on multi-centre data from the Children’s centres in the National Cancer Research Network (NCRN). Increased accuracy from shorted scanning times are likely to be achieved due to both increased sensitivity and better metabolite resolution (4) from new higher field scanners such as the one available in the NIHR 3T MRI Centre at BCH and other centres in the network. Acquisition protocols will be optimised for 3T MRI and MRS and theminimum scan time required to achieve high diagnostic accuracy determined. There are approximately 350 new cases per year in the UK and 120 of these present to the main collaborating partners with 45 at BCH. The CCLGs Functional Imaging Database at the University of Birmingham already contains more than 300 cases and will provide the platform for data storage and analysis. Prognostic Markers for Childhood Brain Tumours Prognostic biomarkers, measured at diagnosis in children with brain tumours, have been discovered and evaluated on a retrospective cohort. They include total choline, mobile lipids, myoinositol, glycine, (5,6,7) and diffusion characteristics at the boundary of tumours (8). These biomarkers will be evaluated prospectively for the main specific diagnostic tumour groups through incorporation in national and international protocols thereby starting to identify their clinical value for treatment stratification The SIOP-E Brain Imaging Group, of which Dr Peet is the co-chair, provide the imaging protocols for international clinical trials in Europe providing a mechanism for incorporation of functional imaging in the trials. Two phase III trials in the major brain tumour groups will start over the next year providing an excellent opportunity for initiating this strategy. The UK Clinical Trials Centre for children's cancer is at the University of Birmingham and Dr Peet collaborates closely with the Director, Dr Kearns. Biomarkers for Monitoring Response of Tumours to Treatment Decreased cellular density, decreased perfusion and characteristic changes in key metabolites such as total choline, mobile lipids and myoinositol have all been identified as potential early markers of treatment response (1). Incorporating the measurement of these biomarkers into the imaging protocols used in the international treatment studies used for children with cancer will allow their systematic evaluation as early markers of treatment response for a range of tumour types. This is already underway for a phase II clinical trial in high grade gliomas and in an advanced stage of planning for a phase II trial in neuroblastoma and a phase III trial in Wilms tumour of the kidney. This strategy is particularly applicable to tumours for which resection is not the preferred initial treatment. This applies to most solid tumours outside the brain and to unresectable brain tumours. Underpinning Laboratory Studies for Research on Metabolism in Cancer An important although smaller component of Dr Peet's research is the laboratory application of MRS to cancer tissue and cells. This work allows validation of the in vivo findings (11) and has been crucial to both the discovery of biomarkers of prognosis (7) and the development of non-invasive diagnostic tools (12). The technique of Magic Angle Spinning MRS is particularly suited to this type of study since it is non-destructive, can be used directly on intact tissue obtained at surgery and closely mirrors the in vivo MRS. This research will continue during the programme where it supports the translation of MRS in vivo into clinical practice. In particular, the improved accuracy of metabolite level determination by 3T MRS compared with 1.5T MRS will be determined by comparison of in vivo values with those in tissue. In addition to this supportive role, Magic Angle Spinning MRS of tumour tissue has the potential to act as a rapid intra-operative diagnostic tool. Diagnostic aids based on this previous finding (12) will also be evaluated as part of the programme in comparison with current intra-operative diagnostic methods based on histopathology. Disorders associated with Neurodegeneration BCH is a major centre for conditions which lead to neurodegeneration and MRI plays a key role in diagnosing and monitoring these conditions. Children who survive a brain tumour often have neuro-cognitive dysfunction and this can worsen over time, particularly after radiotherapy. In addition many inherited metabolic disorders are associated with neurodegeneration and Dr Peet has collaborated in a functional imaging study of these patients. They have applied in vivo MRS to a cohort of 300 children with a broad range of neurocognitive function and over the whole age range to 16 years. They found that for specific inherited metabolic disorders, MRS can both aid diagnosis and provide biomarkers of neurocognitive function, such as N-acetyl aspartate, which could be used to monitor progress (9,10). This study will be continued, in particular concentrating on the improved metabolite determination which is available using the 3T MRI scanner. Collection of functional imaging on a comparator cohort of children across the age range will be important. Treatments such as gene replacement therapy have recently started at BCH and elsewhere for inherited metabolic disorders and non-invasive methods which can directly monitor the effects of these treatments on the brain are becoming increasingly important. These studies will be supported by active collaborations with the Aston Brain Centre and the Birmingham University Imaging Centre which have extensive research experience in psychology. Multi-centre studies will be managed through the Medicines for Children Network for which BCH is a lead. LaySummary>