Metabolic brain network in the Chinese patients with Parkinson's disease based on 18F-FDG PET imaging

The objective of this study was to validate Parkinson's disease-related pattern (PDRP) as a measure of network biomarker of Parkinson's disease (PD) in Chinese population by using 18F-fluorodeoxyglucose (FDG) and Positron Emission Tomography (PET). Resting-state brain FDG PET imaging was performed in a cohort of 33 PD patients and 33 age/gender-matched healthy controls to identify a PDRP. PDRP expression was then computed in a new cohort of 30 PD patients and 30 healthy controls using a voxel-based network quantification algorithm. Differences in PDRP expression were compared across groups and correlations with severities of PD were investigated. As a result, we identified a PDRP characterized by relative increases in pallidothalamic, pontine, and cerebellar metabolism, associated with concurrent metabolic decreases in the premotor and posterior parietal areas. PDRP expression in each of the two PD groups was significantly elevated relative to that of the healthy controls (P < 0.001). Receiver operating characteristic (ROC) analysis revealed that the PDRP-based discrimination for PD patients and controls had high sensitivity and specificity (both = 93.9%) in the derivation cohort, which declined slightly in the validation cohort (both = 90.0%) at the same diagnostic threshold. Moreover, PDRP scores correlated positively with Hoehn and Yahr scores (r ≥ 0.590, P ≤ 0.001) and Unified Parkinson's Disease Rating Scale motor scores (r ≥ 0.646, P < 0.001) in both patient groups. In conclusion, PDRP is highly reproducible in Chinese cohorts based on FDG PET imaging. Network activity of PDRP can differentiate PD patients from healthy controls and correlates with the severities of the disease.