Functional Neural Networks Stratify the Parkinson’s Disease Patients Across the Spectrum of Cognitive Impairment

Cognitive Impairment (CI) in Parkinson’s disease (PD) is one of the important non-motor symptoms that can begin even before the motor symptoms manifest. As the disease progresses into advance stages, however, virtually all patients suffer from cognitive decline. PD Patients hypothetically progress across PD with no CI (PD-NC), Mild Cognitive Impairment (PD-MCI), and PD dementia (PDD). The CI symptoms in PD are linked to different brain regions including dysfunctional subcortical regions and poorly elucidated neural pathways. However, it is still unknown how functional dysregulation in some brain regions correlates to CI progression in PD. Recently, rsfMRI has been shown to be a promising neuroimaging technique that can enable discovery of CI biomarkers in PD. Here, we investigated the differences in the clinical measures and the resting-state Functional Connectivity (FC) of three CI subtypes of PD. We included a total of 114 participants, (26 PD-NC, 32 PD-MCI, 31 PDD, and 26 Healthy Controls (HC), and performed intra- and inter-network FC analysis together with comprehensive clinical cognitive assessment. Our results showed the importance of several neural networks including Default Mode Network (DMN), Frontoparietal Network (FPN), Dorsal Attention Network (DAN), and Visual Network (VN) and their inter-intra network FC distinguishing between PD-MCI and PDD. Additionally, our results showed the importance of Sensory Motor Network (SMN), VN, DMN, and Salience Network (SN) in the discriminating PD-NC from PDD. Finally, in comparison to HC, we found DMN, FPN, VN, and SN as important networks for further differential diagnosis of CI subtypes of PD. We propose that resting state networks can be used in stratifying the CI subtypes of PD patients in the clinic.