Figures and data
Demographic and clinical information of participants
Volumetric differences in the basal ganglia and thalamus between HD and healthy control (HC) groups.
Regions-of-interest (ROIs) were segmented using FreeSurfer v6. All ROIs, except the left thalamus, showed significantly smaller volumes in the HD cohort after FDR correction for multiple comparisons. Colours indicate the strength of rank-biserial correlations (rrb) from Mann-Whitney U tests: Red = strong effect (rrb ≥ 0.5), Yellow = medium effect (0.3 ≤ rrb < 0.5), White = small effect (rrb < 0.3). Raincloud plots show the distribution of the volumetric measures in each ROI per group with orange for HD and green for HC participants. * p < 0.05; *** p < 0.001
Descriptive and Mann-Whitney Statistics for Intracranial Volume-Normalised Regions of Interest
Microstructural differences in ROIs between HD and healthy control (HC) groups.
Median values of each microstructural measure were extracted per ROI. A) HD individuals show reduced soma density (fis) in the basal ganglia (BG). B) Soma radius (rs) is elevated in the caudate and putamen but reduced in the pallidum. C) Extracellular water fraction (fec) is increased in BG regions in the HD group. D) Extracellular diffusivity (De) is higher in the putamen and pallidum. E) Fractional anisotropy (FA) is elevated in the BG, and F) mean diffusivity (MD; expressed in x10-4mm²/s) is increased in the striatum. Colours indicate the strength of rank-biserial correlations (rrb) from Mann-Whitney U tests: Red = strong effect (rrb ≥ 0.5), Yellow = medium effect (0.3 ≤ rrb < 0.5), White = small effect (rrb < 0.3). Raincloud plots show the distribution of the volumetric measures in each ROI per group with orange for HD and green for HC participants. * p < 0.05; ** p < 0.01; *** p < 0.001
Descriptive and Mann-Whitney Statistics for Microstructural Measures in Regions of Interest
A) Correlation matrix and B-G) selected scatter plots illustrating Spearman’s rho correlations between SANDI measures and the Q-Motor principal component. A) Each cell represents the Spearman’s rho correlation strength, with pink indicating negative and green positive correlations. B-G) Each plot includes a best-fit least squares linear regression line with standard error indicated by the grey shaded area, along with the Spearman’s rho (ρ) and the corresponding FDR-p value. Regression lines are included for visualisation only and do not reflect variance explained (R²) or imply linear model fit. Scatter dot colours represent participants’ HD-ISS stage. Unclassified refers to those participants who could not be classified due to having CAG 36-40 or incomplete clinical data. Abbreviations: De: Extracellular diffusivity; fec: Extracellular signal fraction; fin: Neurite density signal fraction; fis: Soma density signal fraction; PC: Principal component; rs: Soma radius; TFC: Total functional capacity.
A) Correlation matrix and B-D) selected scatter plots illustrating Spearman’s rho correlations between DTI, volumetric measures and the Q-Motor principal component. A) Each cell represents the Spearman’s rho correlation strength, with pink indicating negative and green positive correlations. B-D) Each plot includes a best-fit least squares linear regression line with standard error indicated by the grey shaded area, along with the Spearman’s rho (ρ) and the corresponding FDR-p value. Regression lines are included for visualisation only and do not reflect variance explained (R²) or imply linear model fit. Scatter dot colours represent participants’ HD-ISS stage. Unclassified refers to those participants who could not be classified due to having CAG 36-40 or incomplete clinical data. Abbreviations: FA: Fractional anisotropy; MD: Mean diffusivity.
Standardised beta coefficients of SANDI microstructural metrics predicting volume (normalised for intracranial volume) in regions of interest in the healthy control group.
Data were modelled by firstly accounting for age, followed by the step-wise inclusion of all SANDI indices. The figure displays the predictor variables included in the final regression models for each region of interest. Abbreviations: De: Extracellular diffusivity; fis: Soma density signal fraction; rs: Soma radius; TFC: Total functional capacity. * p < 0.05; ** p < 0.01; *** p < 0.001.
Hierarchical Linear Regression Predicting Normalised Volumes from SANDI Microstructural Metrics, Controlling for Age in the Healthy Control Participants
Standardised beta coefficients of SANDI microstructural metrics predicting volume (normalised for intracranial volume) in A) left and B) right hemisphere regions of interest in individuals with Huntington’s disease.
Data were modelled by firstly accounting for age and Total Functional Capacity scores simultaneously, followed by the step-wise inclusion of all SANDI indices. The figure displays the predictor variables included in the final regression models for each region of interest. Abbreviations: De: Extracellular diffusivity; fec: Extracellular signal fraction; fis: Soma density signal fraction; rs: Soma radius. * p < 0.05; ** p < 0.01; *** p < 0.001.
Hierarchical Linear Regression Models Predicting Normalised Volumes in each Region of Interest from SANDI Microstructural Metrics, Controlling for Age and TFC in HD Participants
A) Correlation matrix and B-J) selected scatter plots illustrating Spearman’s rho correlations between SANDI, DTI, and volumetric measures with CAP100. Each scatter plot includes a best-fit least squares linear regression line with standard error indicated by the grey shaded area, along with the Spearman’s rho (ρ) and the corresponding FDR-p value. Regression lines are included for visualisation only and do not reflect variance explained (R²) or imply linear model fit. Scatter dot colours represent participants’ HD-ISS stage and those who were not classified due to having CAG<40 or incomplete clinical data. Abbreviations: De: Extracellular diffusivity; FA: Fractional anisotropy; fec: Extracellular signal fraction; fin: Neurite density signal fraction; fis: Soma density signal fraction; MD: Mean diffusivity; rs: Soma radius; vol: Normalised volume.
Scatterplot showing positive relationship between the Q-Motor principal component and the disease burden measure (CAP100) with the Spearman’s rho (ρ) test.
Scatter dot colours represent participants’ HD-ISS stage and those who were not classified due to having CAG<40 or incomplete clinical data.
Bar plot showing the effect sizes and 95% confidence intervals for exploratory pairwise comparisons between HD-ISS 0-1 (n = 13), HD-ISS 2-3 (n = 17), and healthy controls (HC).
Significant comparisons are marked with * (p < 0.05), ** (p < 0.01), and *** (p < 0.001).
Correlation heatmaps showing the cross-correlation of SANDI, DTI, and volumetric (normalised for intracranial volume) measures, averaged across the caudate, putamen, pallidum, and thalamus.
Heatmaps are shown separately for the full sample, healthy controls (HC), and Huntington’s disease (HD) participants. Correlations are expressed as Spearman’s rho coefficients.
Demographic and clinical information per HD-ISS stage
Descriptive statistics for motor outcome measures
Rotated Component Loadings on the Q-motor Outcome Measures
