Xinyi Li, Jiaying Lu, Qi Shen, Yi‐Xin Zhao, Kun‐Wang Chan, Wen Liu, Jingjie Ge, Jue Zhao, Yi‐Min Sun, Tzu‐Chen Yen, Jianjun Wu, Chuantao Zuo, Jian Wang, Fengtao Liu, the Progressive Supranuclear Palsy Neuroimage Initiative (PSPNI)
Abstract Background Corticobasal syndrome (CBS) may arise from heterogeneous neuropathological substrates, including corticobasal degeneration (CBD), progressive supranuclear palsy (PSP), and Alzheimer's disease (AD), and biomarker‐based stratification has therefore become central to its characterization in vivo . In a large CBS cohort, we examined whether 18 F‐fluorodeoxyglucose (FDG) positron emission tomography (PET) metabolic patterns reflect underlying amyloid‐ β deposition and tau topography, and how these patterns relate to the clinical phenotype. Methods We conducted a cross‐sectional study of consecutive patients with CBS and were enrolled at Huashan Hospital, Fudan University, between December 2019 and February 2025. Each patient underwent 18 F‐FDG PET for cerebral glucose metabolism, amyloid status assessment using amyloid PET or cerebrospinal fluid biomarkers, and florzolotau( 18 F) PET for tau topography. Results Among 113 patients, 18 F‐FDG PET identified five distinct metabolic subgroups: AD type (n = 20), CBD type I (n = 45), CBD type II (n = 18), PSP type (n = 26), and frontotemporal dementia type (n = 4). The AD metabolic pattern showed high specificity for amyloid‐ β positivity (98.5%) and for combined AD pathology (amyloid‐ β plus tau, 98.6%), with correspondingly lower sensitivity (54.2% and 72.2%). 18 F‐FDG PET patterns further showed high specificity for the PSP tau profile (96.2%) and high sensitivity for the CBD tau profile (97.7%). On spatial analysis, amyloid‐ β deposition significantly mediated the regional coupling between tau accumulation and cerebral glucose hypometabolism. Conclusions 18 F‐FDG PET identifies metabolic patterns that correspond to specific proteinopathies underlying CBS with high specificity but requires integration within multimodal biomarker frameworks for comprehensive patient stratification. © 2026 International Parkinson and Movement Disorder Society.