Cerebral cortex structure in prodromal Huntington disease

Abstract

Neuroimaging studies of subjects who are gene-expanded for Huntington Disease, but not yet diagnosed (termed prodromal HD), report that the cortex is “spared,” despite the decrement in striatal and cerebral white-matter volume. Measurement of whole-cortex volume can mask more subtle, but potentially clinically relevant regional changes in volume, thinning, or surface area. The current study addressed this limitation by evaluating cortical morphology of 523 prodromal HD subjects. Participants included 693 individuals enrolled in the PREDICT-HD protocol. Of these participants, 523 carried the HD gene mutation (prodromal HD group); the remaining 170 were non gene-expanded and served as the comparison group. Based on age and CAG repeat length, gene-expanded subjects were categorized as “Far from onset,” “Midway to onset,” “Near onset,” and “already diagnosed.” MRI scans were processed using FreeSurfer. Cortical volume, thickness, and surface area were not significantly different between the Far from onset group and controls. However, beginning in the Midway to onset group, the cortex showed significant volume decrement, affecting most the posterior and superior cerebral regions. This pattern progressed when evaluating the groups further into the disease process. Areas that remained mostly unaffected included ventral and medial regions of the frontal and temporal cortex. Morphologic changes were mostly in thinning as surface area did not substantially change in most regions. Early in the course of HD, the cortex shows changes that are manifest as cortical thinning and are most robust in the posterior and superior regions of the cerebrum.

The Corticospinal Tract in Huntington's Disease

“…As our results show some changes (higher iron level) in the CST of Pre-HD subjects, but not significant motor cortex thickness reduction in Pre-HD, we suggest that WM damage may precede GM changes in the motor cortex. This hypothesis is in agreement with previous studies pointing in this direction in HD patients (Bartzokis et al 2007) and Pre-HD subjects (Nopoulos et al 2010). In any case, a future longitudinal investigation is needed to resolve this question.…”

Brain Structure in Preclinical Huntington's Disease: A Multi-Method Approach

“…It is also possible that clinical characteristics, such as the estimated time to motor onset, could have contributed to negative findings. For instance, although in preHD less than 15 years from the estimated motor onset decreased striatal and cortical volume has been robustly detected using semiautomated analysis techniques [5,9,11,42,43], several studies have suggested that in preHD far from the estimated motor onset there may be no such structural changes when studied cross-sectionally [7,8,9] nor substantial rates of change over time [14,42]. We added to these reports in employing two additional techniques for structural analysis, cortical surface modeling and subcortical segmentation.…”

“…Indeed, cross-sectional and longitudinal MRI studies have suggested that structural brain changes in carriers of the HD mutation may serve as biomarkers and potentially as surrogate endpoints in clinical trials [2,3,4,5,6]. However, intact brain structure in preHD has also been reported [7,8,9,10]. It is unlikely that these discrepancies can be solely explained by differing sample sizes, since clinical characteristics, such as the proximity to the onset of motor signs [5,11,12], and different analysis procedures [13] may also account for data heterogeneity.…”

“…So far, several methods have been applied to the analysis of high-resolution MRI data; these include a priori regions-of-interest techniques [2], unbiased whole-brain-based methods such as voxel-based morphometry (VBM) [7,8,14] and cortical thickness analyses [9,15]. Each of these approaches may offer particular methodological advantages; however, given the increasing number of imaging postprocessing software using different methodological algorithms, a considerable number of heterogeneous findings may also be attributed to the methods used to analyze the data.…”

“…To this end, we employed three well-established automated methods to analyze structural data obtained from healthy controls and far-from-onset preHD: (1) the VBM Diffeomorphic Anatomic Registration Through Exponentiated Lie Algebra algorithm (DARTEL); (2) the Oxford Centre for Functional MRI of the Brain's (FMRIB's) Integrated Registration and Segmentation Tool (FIRST), and (3) the FreeSurfer software package. We went on to examine the relationship between subcortical and cortical volume with genetic and clinical variables, such as CAG repeat length and the estimated time to motor onset considering the increasing subclinical and neurobiological changes which occur with closer proximity to the motor onset of the disease [5,9,11,17,18]. We finally administered a complementary neuropsychological test battery to relate structural changes to cognitive performance.…”