Association between basal forebrain and medial temporal lobe tau, and gray matter volume
Andrea T Shafer, Maiya F. MacAlpine, Cristina Bañuelos, Abhay Moghekar, Susan M. Resnick, Murat Bilgel- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Tau pathology is a hallmark of Alzheimer’s disease (AD). The basal forebrain (BF) is a significant cholinergic projection site with cholinergic groups 1‐3 (Ch1‐3) projecting to the hippocampus, hypothalamus, and olfactory bulb, and group 4 (Ch4) projecting to the cortical mantle and amygdala. Braak stage 1 includes BF tau and recent literature suggests BF degeneration precedes medial temporal lobe (MTL) degeneration, thus implicating early involvement of the BF in AD. This study examines BF 18F‐flortaucipir (FTP) retention and gray matter (GM) volume, and their associations with MTL FTP retention and GM volume.
Method
Data were acquired for 111 older adults (Table 1) in the Baltimore Longitudinal Study of Aging. Binary masks for BF subregions Ch1‐3 and Ch4 were generated using a probabilistic atlas (Figure 1). We computed mean standardized uptake value ratio (SUVR) using an inferior cerebellar GM reference region and mean GM volume using average voxel‐based morphometry values for each BF subregion. Colocalized relationships between BF tau and volume were assessed using Pearson correlations. Proximal relationships between BF tau/GM and MTL tau/GM were assessed at the voxel level controlling for age, sex, and intracranial volume and corrected for multiple comparisons using non‐parametric permutations.
Results
Within the BF, FTP‐SUVR in Ch1‐3 was strongly related to FTP‐SUVR in Ch4, and FTP‐SUVR was associated with larger volume in Ch1‐3, but not Ch4 (Figure 2). Voxelwise analyses showed strong BF‐MTL associations with higher Ch1‐3 and Ch4 FTP‐SUVR associated with higher FTP‐SUVR throughout the MTL. The strongest associations were observed anteriorly (amygdala, peri/entorhinal cortices) and medially along the longitudinal axis of the hippocampus/parahippocampal gyrus. Moreover, for both BF Ch groups, larger volume was associated with larger MTL volume. This association was strongest in the amygdala and hippocampal head and body.
Conclusion
As expected, BF FTP‐SUVR was positively associated with MTL FTP‐SUVR, and BF volume was positively associated with MTL GM volume, supporting the use of these BF measures. Further analysis is necessary to better understand the positive associations between BF FTP‐SUVR and GM volume and to assess the temporal ordering of FTP‐SUVR accumulation in the BF and MTL.