Gray Matter Structural Covariance Networks Patterns Associated with Autopsy-Confirmed LATE-NC Compared to Alzheimer's Disease Pathology.
Kaicheng Li,Xiao Luo,Qingze Zeng,Xiaocao Liu,Jixuan Li,Siyan Zhong,Xinyi Zhang,Xiaopei Xu,Shuyue Wang,Hui Hong,Yerfan Jiaerken,Zhirong Liu,Shuai Zhao,Peiyu Huang,Minming Zhang,Yanxing Chen
DOI: https://doi.org/10.1016/j.nbd.2023.106354
IF: 7.046
2023-01-01
Neurobiology of Disease
Abstract:Background: Cases with the limbic-predominant age-related TAR DNA-binding protein 43 (TDP-43) encephalopathy neuropathologic change (LATE-NC), Alzheimer's disease (AD), and mixed AD+TDP-43 pathology (AD+LATE-NC) share similar symptoms, which makes it a challenge for accurate diagnosis. Exploring the patterns of gray matter structural covariance networks (SCNs) in these three types may help to clarify the underlying mechanism and provide a basis for clinical interventions. Methods: We included ante-mortem MRI data of 10 LATE-NC, 39 AD, and 25 AD+LATE-NC from the ADNI autopsy sample. We used four regions of interest (left posterior cingulate cortex, right entorhinal cortex, frontoinsular and dorsolateral prefrontal cortex) to anchor the default mode network (DMN), salience network (SN), and executive control network (ECN). Finally, we assessed the SCN alternations using a multi-regression model-based linear-interaction analysis. Results: Cases with autopsy-confirmed LATE-NC and AD showed increased structural associations involving DMN, ECN, and SN. Cases with AD+LATE-NC showed increased structural association within DMN while decreased structural association between DMN and ECN. The volume of peak clusters showed significant associations with cognition and AD pathology. Conclusions: This study showed different SCN patterns in the cases with LATE-NC, AD, and AD+LATE-NC, and indicated the network disconnection mechanism underlying these three neuropathological progressions. Further, SCN may serve as an effective biomarker to distinguish between different types of dementia.