Premium
Reduced [ 18 F]flortaucipir retention in white matter hyperintensities compared to normal‐appearing white matter
Author(s) -
Moscoso Alexis,
Whitman A.J.,
Baker Suzanne L.,
La Joie Renaud,
Pascoal Tharick A.,
RosaNeto Pedro,
Rabinovici Gil D.,
Jagust William J.,
Grothe Michel,
Schöll Michael
Publication year - 2020
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.046610
Subject(s) - white matter , hyperintensity , nuclear medicine , voxel , neuroimaging , standardized uptake value , grey matter , fluid attenuated inversion recovery , partial volume , medicine , magnetic resonance imaging , psychology , radiology , neuroscience , positron emission tomography
Background Recent research has suggested that the use of white matter (WM) reference regions for longitudinal tau PET imaging might result in more stable estimates of tau accumulation. However, tau tracers display affinity for the β‐sheet structure formed by proteins such as the myelin basic protein, a major component of axons, and thus WM lesions might influence tau tracer retention in the WM. Here, we explored whether [ 18 F]flortaucipir shows reduced retention in WM lesions, assessed as WM hyperintensities (WMH), and further studied how [ 18 F]flortaucipir retention in WM changes over time. Method We included data from 678 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) who had undergone a baseline [ 18 F]flortaucipir‐PET as well as T1 and FLAIR MRI scanning. WMH were automatically delineated in the FLAIR images using the Lesion Segmentation Toolbox in SPM12. To avoid partial volume effects, we eroded the WM and WMH masks so that all voxels within a distance of 3 mm from any non‐WM voxel were removed. The inferior portion of the cerebellar gray matter was used as the reference region to derive Standardized Uptake Value Ratios (SUVR). SUVR in WMH and normal‐appearing WM (NAWM) were compared using paired t‐tests. Among participants with at least one‐follow up [ 18 F]flortaucipir scan (N=217, mean follow‐up time 1.5 years), longitudinal SUVR changes in the WM were estimated using Linear Mixed Models. Result Older age was associated with lower SUVR in both NAWM (r=‐0.10, p=0.007) and WMH (r=‐0.21, p<0.001) but not with [ 18 F]flortaucipir retention in the medial temporal cortex after adjusting for amyloid‐β (Aβ) status and diagnosis (r=‐0.04, p=0.36, Meltzer‐corrected for partial volume effects). Compared to NAWM, WMH areas displayed significantly reduced SUVR independently of cognitive impairment and Aβ status (p<0.001, Figure 1). Longitudinal analyses revealed that SUVR in the WM decreased over time (‐0.009 SUVR/year, p=0.003), while medial temporal SUVR increased (0.009 SUVR/year, p=0.03). Conclusion Low [ 18 F]flortaucipir retention in the WM is associated with advancing age and presence of WM lesions, supporting the hypothesis that [ 18 F]flortaucipir retention in the WM might reflect its myelin content. These findings do not support the use of WM reference regions for [ 18 F]flortaucipir PET imaging.