Quantitative MRI Reveals Brain Microstructure Changes Across Huntington’s Disease Stages

Using multiparametric magnetic resonance imaging to assess tissue biophysical properties in living participants, researchers have demonstrated widespread microstructural alterations across cortical and subcortical brain regions in patients with Huntington’s disease. The quantitative mapping protocol simultaneously evaluated four distinct biophysical parameters across the brain, identifying widespread reductions in tissue integrity that were present across premanifest and manifesting stages of the inherited condition. By comparing voxel-level microstructural maps with volumetric anatomical analyses, the investigation established that these quantitative microstructural changes spatially overlap with macroscopic volume loss. The mapped parameters tracked clinical disease severity and cognitive performance scores, providing in vivo evidence of progressive demyelination, iron redistribution, and cellular loss. The work highlights the diagnostic capability of quantitative imaging for detecting cellular-level neuropathology before overt macroscopic degeneration becomes established.

Huntington’s disease is an autosomal-dominant neurodegenerative disorder caused by an expanded cytosine-adenine-guanine trinucleotide repeat in the huntingtin gene, where longer repeat lengths correlate inversely with an earlier age of motor onset and more severe disease progression. Although the genetic etiology has been well characterized, no disease-modifying therapies are approved, leaving clinicians reliant on purely symptomatic interventions. Affected individuals present with heterogeneous motor, cognitive, and psychiatric symptoms that worsen over time, with patient subgroups showing disproportionately rapid deterioration in specific functional domains. Structural magnetic resonance imaging serves as the standard clinical neuroimaging tool to detect basal ganglia atrophy, yet conventional macroscopic scans cannot probe underlying molecular or biophysical pathological processes. While post-mortem histological examinations provide cellular detail, they evaluate end-stage tissue, which limits their clinical utility for prospective tracking and patient stratification.

To overcome the resolution limits of standard clinical scans without invasive procedures, the research team deployed quantitative multiparametric mapping, an advanced imaging framework that derives quantitative physical units from multi-contrast gradient-echo acquisitions. The multiparametric mapping pipeline computes four specific tissue parameters: the longitudinal relaxation rate, the effective transverse relaxation rate, proton density, and magnetization transfer saturation. Biophysically, the longitudinal relaxation rate and magnetization transfer saturation are sensitive to local myelin content and macromolecular density, whereas the effective transverse relaxation rate reflects tissue iron concentration and demyelination. Proton density provides a semiquantitative index of the mobile proton pool sensitive to tissue water and macromolecular composition, though researchers caution that proton density decreases cannot be interpreted as direct evidence of reduced water content alone. Together, these quantitative parameters offer direct insight into myelin loss, iron alteration, and cellular architectural breakdown.

The clinical study evaluated twenty-six participants carrying genetically confirmed Huntington’s disease with fully penetrant huntingtin mutations of forty or more cytosine-adenine-guanine repeats, alongside forty-two healthy control volunteers who had no history of neurological illness, substance abuse, or structural brain lesions. The patient group comprised sixteen premanifest individuals who exhibited no unequivocal motor signs and ten manifesting individuals displaying unequivocal motor manifestations. Premanifest individuals had a mean age of 39.3 years and an average repeat length of 42 repeats, whereas manifesting patients had a mean age of 59.8 years, a disease duration of 7.7 years, and an average repeat length of 41 repeats. Healthy controls were stratified into matching sub-cohorts that aligned in age, sex distribution, and sample size with each patient group, ensuring unbiased comparative statistical evaluations across clinical stages.

All neuroimaging sessions were completed on a 3-Tesla scanner utilizing a sixty-four-channel head coil to collect coronal three-dimensional fast low-angle shot acquisitions across T1-weighted, proton-density-weighted, and magnetization-transfer-weighted contrasts. The quantitative voxel-based quantification analysis was carried out using the hMRI toolbox within the Statistical Parametric Mapping framework in MATLAB, applying specialized unified segmentation and sensitivity bias corrections to reconstruct normalized parameter maps at one-millimeter isotropic resolution. In parallel, the investigators performed voxel-based morphometry with the Computational Anatomy Toolbox to calculate local gray-matter and white-matter volume differences. Group-level statistical models applied two-sample t-tests incorporating age and sex as explicit covariates, using an initial uncorrected cluster-forming voxel threshold of p below 0.001 with family-wise error rate corrections applied at a significance threshold of p below 0.05.

When analyzing the combined patient cohort against healthy controls, the team identified pervasive reductions across all four multiparametric mapping parameters throughout multiple cerebral cortices and deep subcortical structures. Significant microstructural clusters showing decreased values emerged in the bilateral middle temporal gyri, superior temporal gyri, left posterior orbital gyrus, left fusiform gyrus, left supramarginal gyrus, and left postcentral gyrus. Within the basal ganglia mask, t-values were significantly lower across all four modalities in the bilateral caudate nuclei. Concurrent voxel-based morphometry revealed extensive volumetric gray-matter reductions in the left caudate nucleus and right middle frontal gyrus. Notably, these volumetric reductions directly overlapped with the quantitative microstructural abnormalities, confirming that cellular-level biophysical changes accompany macroscopic gray-matter tissue loss across the patient population.

Subgroup comparisons revealed contrasting spatial distributions between premanifest and manifesting stages of the condition. Premanifest participants demonstrated localized parameter reductions confined to specific frontal and temporal regions, including decreased longitudinal relaxation rates in the left superior frontal and middle temporal gyri, decreased transverse relaxation rates in the right middle frontal gyrus and superior parietal lobule, and reduced magnetization transfer saturation in the right superior frontal and posterior cingulate gyri. Voxel-based morphometry in premanifest participants detected localized volumetric loss in the right middle orbital gyrus and cerebellar vermis, but found no significant basal ganglia differences. In contrast, manifesting patients exhibited pronounced, widespread reductions across the bilateral caudate nuclei, putamen, and frontoparietal cortices, with much higher statistical effect sizes reflecting extensive, disease-stage-dependent cellular vulnerability and accelerated neurodegenerative processes.

Examination of white matter tracts revealed extensive microstructural disruption that paralleled the gray matter findings, particularly in participants with manifest disease. While premanifest individuals showed quantitative alterations in white matter maps without detectable volumetric loss, manifesting patients exhibited severe reductions in longitudinal relaxation rate, transverse relaxation rate, and magnetization transfer saturation within the splenium and genu of the corpus callosum. These microstructural abnormalities overlapped with significant white-matter volumetric atrophy identified through morphometry in the posterior corona radiata, right superior corona radiata, sagittal stratum, and left cingulum. The authors observed that although diffusion tensor imaging studies often report white matter changes preceding gray matter volume loss, their quantitative morphometric analysis detected no premanifest white-matter volume loss, suggesting that multiparametric biophysical alterations may capture early myelin remodeling before gross tract loss occurs.

Within regions exhibiting group-level abnormalities, cross-modal correlation analyses using subject-level cluster means demonstrated strong, statistically significant positive relationships among the measured parameters following false discovery rate corrections. Across the combined patient cohort, longitudinal relaxation rates correlated positively with effective transverse relaxation rates in the left fusiform gyrus, right thalamus, bilateral caudate nuclei, and left middle occipital gyrus, while also correlating positively with magnetization transfer saturation. Furthermore, subject-level clinical correlation analyses demonstrated that higher effective transverse relaxation rates in the right middle occipital gyrus and left precuneus were positively correlated with higher Montreal Cognitive Assessment scores. Additionally, cytosine-adenine-guanine repeat lengths correlated positively with regional relaxation rates and magnetization transfer saturation across several neocortical regions, including the left middle cingulate, supramarginal, right precentral, and left superior temporal gyri.

The study authors highlighted several methodological limitations that warrant consideration, including the cross-sectional design, which captures a single temporal snapshot rather than tracking longitudinal microstructural trajectories. Dividing the twenty-six patients into premanifest and manifesting cohorts necessarily limited statistical power to detect subtle basal ganglia changes in premanifest participants. The relatively narrow distribution of cytosine-adenine-guanine repeat lengths in the enrolled participants may also have constrained the detection of wider genetic associations, while standard bedside cognitive screenings provide general performance metrics rather than domain-specific cognitive profiles. Nonetheless, the findings confirm that quantitative multiparametric mapping can characterize microstructural brain changes in living patients with Huntington’s disease, providing a sensitive, non-invasive biomarker platform to track disease progression, support clinical trial stratification, and monitor therapeutic responses across distinct disease stages.

Subject of Research: Quantitative MRI assessment of brain microstructure across stages of Huntington's disease

Article Title: In vivo assessment of brain microstructure in patients with Huntington’s disease using quantitative MRI

Article References: Pokotylo, M. M., Aßmann, J.-L., Mertins, M. G. K., Henkel, J., Uter, J., van Well, L., Münchau, A., Loens, S., Göttlich, M., Brüggemann, N., & Prasuhn, J. (2026). In vivo assessment of brain microstructure in patients with Huntington’s disease using quantitative MRI. Journal of Neurology, 273(11), Article 707. https://doi.org/10.1007/s00415-026-14244-8

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14244-8

Keywords: Huntington's disease, multiparametric mapping, quantitative MRI, neuroimaging, brain microstructure, neurodegeneration, voxel-based quantification, biomarkers, vivo, assessment, brain, microstructure