Background Perivascular spaces (PVS) are fluid-filled cavities surrounding small cerebral blood vessels. They form part of the glymphatic system as a conduit for fluid transport, facilitating waste clearance from the brain and contributing to the maintenance of brain health. There are limited reports of enlarged PVS in manifest Huntington’s disease (HD). This may compromise cerebral fluid and toxin clearance, potentially contributing to HD pathology. PVS enlargement may also affect the distribution and success of intraparenchymal and intrathecally-administered therapies via increased drainage, thus, a more in-depth understanding of PVS in HD may facilitate future clinical trial design or interpretation.

Aims To better characterise differences in PVS in premanifest and early-manifest HD subjects compared with controls.

Methods Thirty controls, 30 premanifest HD subjects and 30 early-manifest HD subjects were selected from the Vancouver site of the TRACK-HD study. Perivascular Space Semi-Automatic Segmentation (PVSSAS) was used to segment PVS in the white matter of T2W 3T MRI using a 2D Frangi-based detection algorithm. Manual refinement of PVS segmentations was performed on all datasets with the rater blinded to groups. PVSSAS reproducibility with manual intervention was evaluated on a sub-cohort including controls and HD subjects (n=12). PVS count, total PVS volume, long-axis-length, and short-axis-length were measured for each subject. Group differences were examined, controlling for nuisance variables.

Results Three subjects were removed due to motion artefact on T2W MRI. Four subjects were removed due to white matter mask failures. Image analysis is complete. Statistical analysis is ongoing and will be presented in due course.