Premium
White matter hyperintensities and regional tau‐PET signal independently contribute to cognitive deficits in symptomatic patients on the Alzheimer’s disease continuum
Author(s) -
Edwards Lauren,
Iaccarino Leonardo,
Tanner Jeremy A,
Cobigo Yann,
Pham Julie Q.,
Rosen Howard J.,
SoleimaniMeigooni David N.,
Strom Amelia,
Wolf Amy,
Kramer Joel H,
Miller Bruce L.,
Rabinovici Gil D.,
La Joie Renaud
Publication year - 2021
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.055667
Subject(s) - hyperintensity , neuropsychology , white matter , psychology , cognition , standardized uptake value , nuclear medicine , fluid attenuated inversion recovery , dementia , medicine , audiology , magnetic resonance imaging , neuroscience , positron emission tomography , radiology , disease
Abstract Background We examined the associations of white matter hyperintensities (WMH), amyloid‐PET, and tau‐PET with multi‐domain cognitive performance in symptomatic patients on the Alzheimer’s disease (AD) continuum. Method 119 amyloid‐PET‐positive patients diagnosed with MCI or AD dementia underwent 3‐Tesla structural MRI, 18F‐flortaucipir (tau)‐PET,11C‐PIB (amyloid)‐PET, and neuropsychological assessment. Standardized uptake value ratio (SUVR) flortaucipir‐ and PIB‐PET images were created using tracer‐specific reference regions and warped to template space. Patients were split based on availability of FLAIR‐MRI (Figure 1). The exploratory sample (n=50; no FLAIR) was used to run voxelwise models to identify regions where flortaucipir‐ and PIB‐PET were associated with episodic memory, executive, language, semantic memory, and visuospatial composite scores. In the validation sample (n=69), WMH were segmented from FLAIR and T1‐MRIs using a two‐step supervised algorithm. Linear regression models assessed the independent contributions of global WMH volume (corrected for intracranial volume and log‐transformed) and cognitive domain‐specific regional PET‐SUVR (derived in the exploratory sample) to each cognitive domain. Result The two samples had comparable demographics and clinical characteristics, with relatively young age (mean ∼65.5) and moderate impairment (mean MMSE ∼ 22, Table 1). In the exploratory sample, cognitive scores were associated with regional flortaucipir‐SUVR patterns (Figure 2); PIB‐SUVR was not robustly associated with cognition. In the validation sample, global WMH volume was inversely correlated with global PIB‐SUVR ( r =‐.25, p =.04) but not global cortical flortaucipir‐SUVR ( r =‐.18, p =.15). Older age predicted lower cortical flortaucipir‐SUVR ( r =‐.70, p <.001) and greater WMH volume ( r =.46, p <.001). Female sex was specifically associated with greater cortical flortaucipir‐SUVR ( d =0.64, p =.01; psdspsConclusion In patients at the early clinical stages of AD, tau‐PET and global WMH volume were not associated with each other, but independently contributed to cognitive impairment. These findings emphasize the clinical relevance of cerebrovascular injury in the context of AD pathology, even in younger patients.