Chunyi Wang ¹ Jiajia Hu ² Puyu Li ¹ Ming Zhang ³ Liche Zhou ¹ Ningdi Luo ¹ Xue Zhu ¹ Qianyi Yin ¹ Min Zhong ¹ Xinyi Zhou ¹ Hongjiang Wei ³ Yuan Y. Li ¹ Biao Li ² Liu Jun ^1*

• ¹ Department of Neurology, Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China
• ² Department of Nuclear Medicine, Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China
• ³ School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, Shanghai Municipality, China

Background: Brain network dysfunction has been characterized by resting-state electroencephalography (EEG) and magnetic resonance imaging (MRI) in the prodromal stage.This study aimed to identify multi-modal electrophysiological and neuroimaging biomarkers for differential diagnosis in synucleinopathies and phenoconversion in isolated rapid eye movement sleep behavior disorder (iRBD).We enrolled 35 patients with multiple system atrophy (MSA), 32 with Parkinson's disease (PD), 30 with iRBD and 30 matched healthy controls (HC). Power spectral density (PSD) was calculated in different frequency bands. EEG functional connectivity (FC) was calculated using the weighted Phase Lag Index (wPLI) after source localization. Significant network disruptions were further confirmed by MRI FC analysis.Results: Quantitative EEG analysis demonstrated that delta and theta power spectral density significantly differed among MSA, PD and HC. The increased PSD was correlated with cognitive decline and olfactory dysfunction in PD. Band-specific FC profiles were observed in theta, alpha, and gamma bands. The hypoconnected alpha network significantly correlated with motor dysfunction, while the gamma FC distinguished PD from MSA. By integrating EEG and MRI network analyses, we found that FC between the olfactory cortex and dorsolateral prefrontal cortex was significantly different between MSA and PD. A multimodal discriminative model for MSA and PD, integrating spectral and FC attributes of EEG and MRI, yielded an area under the receiver operating characteristic curve of 0.900. Simultaneously, we found the FC abnormalities were more prominent than spectral features in iRBD indicating prodromal dysfunction. The decreased FC between the angular gyrus and striatum was identified in α-synucleinopathies. This hypoconnectivity was associated with dopaminergic degeneration in iRBD examined by dopamine transporter imaging.Discussion: Our study demonstrated EEG spectral and functional profiles in prodromal and clinical-defined synucleinopathies. Multimodal EEG and MRI provided a novel approach to discriminate MSA and PD, and monitor neurodegenerative progression in the preclinical phase.