Brain Network Patterns in Patients With Multiple System Atrophy: Spatial Independent Component Analysis Using FDG-PET Data.
Brain Network Patterns in Patients With Multiple System Atrophy: Spatial Independent Component Analysis Using FDG-PET Data.
Where did the research take place?
The study site has not been established. Author addresses may differ from where the research occurred.
Hangzhou, CN · Author affiliation
Department of Neurology, the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China.Location evidence
Shanghai, CN · Author affiliation
Human Phenome Institute, Fudan University, Shanghai, China.Location evidence
Huashan, CN · Author affiliation
Huashan Hospital, Fudan University, Shanghai, China.Location evidence
A plain-language reading has not been prepared for this paper yet.
Original abstract
BACKGROUND AND OBJECTIVES: Multiple system atrophy (MSA) is a progressive neurodegenerative disorder presenting significant clinical heterogeneity, posing challenges in diagnosis and treatment. This study uses spatial independent component analysis (ICA) of 18F-fluorodeoxyglucose (FDG) PET to deconstruct disease heterogeneity and elucidate the large-scale brain network mechanisms of MSA. METHODS: This cross-sectional study included patients with MSA and healthy controls (HCs) from the Second Affiliated Hospital, Zhejiang University School of Medicine. MSA diagnosis was based on the 2022 Movement Disorder Society MSA criteria. All participants underwent FDG-PET imaging. Clinical assessments and dopamine transporter (DAT) PET were performed in patients with MSA. Spatial ICA was applied to identify metabolic covariance networks. Moderation analysis and structural equation modeling (SEM) were used to explore underlying pathway mechanisms. RESULTS: Ninety-five patients with MSA and 102 HCs were included (mean age: 62.25 vs 61.84 years; 51% vs 43% female). Five MSA-related ICs were identified: cerebellar network, salience network, compensatory network, default mode network (DMN), and basal ganglia network (all q < 0.05, false discovery rate-adjusted, vs controls). The combination of ICs aligns with the MSA metabolic abnormalities. The cerebellar network was associated with cognitive impairment (Mini-Mental State Examination: β = 13.87, 95% CI 5.86-21.88, p = 9.04 × 10-4; Montreal Cognitive Assessment: β = 14.63, 95% CI 7.45-21.81, p = 1.13 × 10-4), cerebellar symptoms (β = -38.58, 95% CI -73.37 to -3.78, p = 0.03), and posterior putamen DAT (β = -72.79, 95% CI -120.35 to -25.23, p = 0.0031). The compensatory network showed increased metabolism associated with more severe parkinsonian symptoms (β = 3.66, 95% CI 1.51-5.81, p = 1.08 × 10-3). The basal ganglia network was associated with parkinsonian symptoms (β = -2.56, 95% CI -4.03 to -1.09, p = 8.37 × 10-4) and posterior putamen DAT (β = 48.24, 95% CI 15.99-80.50, p = 0.0038). DMN moderated the relationship between the cerebellar network and cognitive function. SEM demonstrated that posterior putamen DAT and basal ganglia network contributed to motor symptoms while the compensatory network acted as a compensatory mechanism. DISCUSSION: This study demonstrated that the metabolic abnormalities in MSA can be decomposed into 5 large-scale brain networks, providing a comprehensive understanding of the disease's heterogeneous mechanisms.