1Reta Lila Weston Institute, UCL Queen Square Institute of Neurology, London WC1N 1PJ, UK.
2Dementia Research Centre, UCL Queen Square Institute of Neurology, London WC1N 3AR, UK.
3MRC Unit for Lifelong Health and Ageing, University College London, London WC1E 7HB, UK.
4Centre for Medical Image Computing, University College London, London WC1V 6LJ, UK.
5Biomedical Engineering and Imaging Sciences, King's College London, London WC2R 2LS, UK.
6Tenoke Ltd, Cambridge CB2 0AH, UK.
7Department of Clinical Biochemistry, Southampton General Hospital, Southampton SO16 6YD, UK.
8Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, London WC1N 3BG, UK.
9Wellcome Centre for Human Neuroimaging, UCL Queen Square Institute of Neurology, London WC1N 3AR, UK.
10Department of Clinical Chemistry, Northern General Hospital, Sheffield S5 7AU, UK.
11UCL Institute of Liver and Digestive Health and Royal Free Hospital, London NW3 2PF, UK.
12Sheffield Institute of Translational Neuroscience, Sheffield S10 2HQ, UK.
Wilson's disease is an autosomal-recessive disorder of copper metabolism with neurological and hepatic presentations. Chelation therapy is used to 'de-copper' patients but neurological outcomes remain unpredictable. A range of neuroimaging abnormalities have been described and may provide insights into disease mechanisms, in addition to prognostic and monitoring biomarkers. Previous quantitative MRI analyses have focused on specific sequences or regions of interest, often stratifying chronically treated patients according to persisting symptoms as opposed to initial presentation. In this cross-sectional study, we performed a combination of unbiased, whole-brain analyses on T1-weighted, fluid-attenuated inversion recovery, diffusion-weighted and susceptibility-weighted imaging data from 40 prospectively recruited patients with Wilson's disease (age range 16-68). We compared patients with neurological (n = 23) and hepatic (n = 17) presentations to determine the neuroradiological sequelae of the initial brain injury. We also subcategorized patients according to recent neurological status, classifying those with neurological presentations or deterioration in the preceding 6 months as having 'active' disease. This allowed us to compare patients with active (n = 5) and stable (n = 35) disease and identify imaging correlates for persistent neurological deficits and copper indices in chronically treated, stable patients. Using a combination of voxel-based morphometry and region-of-interest volumetric analyses, we demonstrate that grey matter volumes are lower in the basal ganglia, thalamus, brainstem, cerebellum, anterior insula and orbitofrontal cortex when comparing patients with neurological and hepatic presentations. In chronically treated, stable patients, the severity of neurological deficits correlated with grey matter volumes in similar, predominantly subcortical regions. In contrast, the severity of neurological deficits did not correlate with the volume of white matter hyperintensities, calculated using an automated lesion segmentation algorithm. Using tract-based spatial statistics, increasing neurological severity in chronically treated patients was associated with decreasing axial diffusivity in white matter tracts whereas increasing serum non-caeruloplasmin-bound ('free') copper and active disease were associated with distinct patterns of increasing mean, axial and radial diffusivity. Whole-brain quantitative susceptibility mapping identified increased iron deposition in the putamen, cingulate and medial frontal cortices of patients with neurological presentations relative to those with hepatic presentations and neurological severity was associated with iron deposition in widespread cortical regions in chronically treated patients. Our data indicate that composite measures of subcortical atrophy provide useful prognostic biomarkers, whereas abnormal mean, axial and radial diffusivity are promising monitoring biomarkers. Finally, deposition of brain iron in response to copper accumulation may directly contribute to neurodegeneration in Wilson's disease.