White matter hyperintensities in Down syndrome: associations with demographic, genetic, fluid and imaging biomarkers of Alzheimer's disease

Abstract Background Strong associations between age, Alzheimer’s disease (AD), and vascular risk factors create challenges to delineate the specific relationships between white matter hyperintensities (WMH) and AD pathophysiology. Down syndrome (DS), a genetically determined form of AD, is not typically complicated by systematic vascular risk factors like hypertension and atherosclerosis and can therefore provide critical information in this regard. We aim to further characterize topography of WMH in DS along the AD continuum and define associations with demographic, genetic, fluid and other imaging biomarkers. Method In this cross‐sectional study, adults with DS from Down‐Alzheimer Barcelona Neuroimaging Initiative (DABNI) underwent a 3T‐MRI protocol, including 3D‐T1 and 3D‐FLAIR acquisitions. Gray matter (GM) volumes were computed using the CAT12 Toolbox. WMH were segmented using the lesion prediction algorithm implemented in the Lesion Segmentation Toolbox for SPM12. Segmented maps were binarized (threshold of 0.5) and WMH volumes were extracted in each brain lobe, subregions of the corpus callosum, and periventricular versus deep regions (using the standard 10mm limit from the lateral ventricles). Non‐parametric statistical tests were used to assess effects of disease severity on WMH and define associations between WMH and demographic and genetic data, CSF AD biomarkers and GM volumes. Results We included 116 individuals with DS (Table 1). WMH volume was higher in prodromal and demented groups for all brain regions (Fig 1). Moreover, WMH increased with disease severity such that the demented group demonstrated larger WMH volume than the prodromal group in periventricular regions, frontal and temporal lobes, and body and splenium subregions of the corpus callosum. Total WMH volume was associated with age but not sex, intellectual disability or APOE haplotype (Fig 2). Total WMH volume was negatively related to CSF Aβ 42/40 ratio and positively related to CSF t‐tau and p‐tau‐181 concentrations (Fig 3). Associations with GM volumes were widespread, but strongest in parieto‐occipital‐temporal regions. Conclusion WMH increases with AD pathology and the emergence of clinical symptoms in DS, a population with lower vascular risk factors. It also relates to neurodegeneration in typical AD regions at both asymptomatic and symptomatic stages. Altogether, this suggests that WMH might be intrinsically related to AD pathophysiological processes.