**11. PET/MRI**

PET/MRI is a relatively novel hybrid diagnostic imaging device that can simultaneously acquire PET and MR images of the brain and other body regions. PET images show the distribution of an intravenously injected radiotracer, whilst MRI depicts the local responses of atomic nuclei to high-frequency radio waves when placed in a strong magnetic field. PET/MRI represents an advance on hybrid PET/CT imaging systems that are currently used in routine clinical practice for the assessment of patients with cancer and other diseases.

The integration of PET with MRI rather than CT has several advantages:


Many of these advantages are particularly relevant to brain tumour imaging, and indeed, the first exploration offeasibility of hybrid PET/MRIin clinical oncology was in brain tumours [73].

PET/MRI could theoretically harness the advantages of PET imaging with various radiotrac‐ ers to accurately distinguish tumours from surrounding normal brain tissues—and the ability of advanced MRI techniques such as fMRI and diffusion tensor imaging to map the spatial relationship between tumours and adjacent functional brain tissues and white matter tracts at the same time.

To date, studies in PET/MRI have shown that it can diagnose, grade and evaluate treatment response in glioma patients [74, 75]. Other studies have also shown that PET/MRI can identify areas of greater cellular proliferation and vascularity in brain tumours using a combination of advanced MRI techniques and PET radiotracers for treatment targeting [76–78]. FMISO PET/MRI can quantify hypoxia in recurrent glioma for risk stratification prior to commence‐ ment of angiogenesis inhibitor therapy (such as bevacizumab) and assess response to treatment [79, 80].

PET/MRI may gradually replace PET/CT in paediatric oncology due to the radiation dose saving achieved with performing MRI in place of CT, and more specifically in the field of neurooncology, the superior characterisation of brain tumours afforded by MRI over CT. In this arena, PET/MRI with choline andFDOPAhave shown promise in the imaging of paediatric astrocytomas [81, 82].

The major disadvantages of PET/MRI currently relate to its high cost and consequent lack of access in many centres, need for optimisation of workflow and image acquisition parame‐ ters, and a greater body of evidence to evaluate its perceived superiority over existing tech‐ niques in neurooncology such as PET/CT, MRI or indeed PET/CT and MRI with software fusion of PET and MRI data. This would presumably require the results of large randomised controlled trials that should be a focus of future research efforts.
