The availability of an accurate genetic test to identify Huntington’s Disease (HD) in the pre-symptomatic stage makes HD an important model to develop biomarkers for other neurodegenerative diseases, such as pre-clinical Alzheimer’s Disease. We reasoned that functional changes, measured by functional MRI (fMRI), would precede gray matter changes and that performing a task specifically affected by the disease would carry the clearest signature.Separate cohorts of HD gene mutations carriers and controls performed four different fMRI tasks, probing functions either primarly affected by the disease (i.e. motor control), higher cognitive functions (i.e. working memory and irritability), or basic sensory functions (i.e. auditory system). With the aim to compare fMRI and structural MRI biomarkers, all subjects underwent an additional high-resolution T1-weighted MRI. Best classification performance was achived from fMRI-based activations with motor sequence tapping and task-induced irritation. Classification performance based on gray matter probability maps was also significantly above chance and similar to that of fMRI. Both were sufficiently informative to separate gene mutation carriers that were on average 17 years before predicted disease onset from controls with up to 80% accuracy. Further analyses showed that classification accuracy was best in regions of interest with low within-group heterogeneity in relation to disease specific changes. Our study indicates that structural and some functional markers can accurately detect pre-clinical neurodegeneration. However, the lower variability and easier processing of the strucutral MRI data make latter the more useful tool for disease detection in a clinical setting.
Keywords: Automated disease diagnosis, pre-symptomatic neurodegeneration, early diagnosis, functional magnetic resonance imaging (fMRI), structural magnetic resonance imaging , support vector machine classification