Brain radiomics-based network tracks distinct subtypes in prodromal Parkinson's disease

193 

Abstract Submission 

Lin Hua¹, Canpeng Huang¹, Fei Gao², Zhen Yuan¹

¹University of Macau, Taipa, Macau, ²Fudan University, Shanghai, Shanghai

Lin Hua  
University of Macau
Taipa, Macau

Canpeng Huang  
University of Macau
Taipa, Macau

Fei Gao  
Fudan University
Shanghai, Shanghai

Zhen Yuan  
University of Macau
Taipa, Macau

Individuals in the prodromal phase of Parkinson's disease (PD) exhibit significant heterogeneity and can be divided into distinct subtypes based on clinical symptoms, pathological mechanisms, and brain network patterns. However, little has been done regarding the valid subtyping of prodromal PD, which hinders the early diagnosis of PD. In this study, we aimed to identify the subtypes of prodromal PD using the brain radiomics-based network and examine the unique patterns linked to the clinical presentations of each subtype.

Individualized brain radiomics-based network was constructed for normal controls (NC; N=110), prodromal PD patients (N=262), and PD patients (N=108). Data-driven clustering approach using the radiomics-based network was carried out to cluster prodromal PD patients into higher-/lower-risk subtypes. Then, the dissociated patterns of clinical manifestations, anatomical structure alterations, and gene expression between these two subtypes were evaluated. Finally, to ensure the consistency of the prodromal PD subtypes identified through brain radiomics-based network, reproducibility was used to access their robustness across various brain atlases or parcellation schemes.

Compared with NC, widespread radiomics-based connections were statistically significant changes in PD. Furthermore, 50 key connections that contributed to separating NC and PD were mainly involved in the thalamus, precentral gyrus, and inferior temporal gyrus. Clustering findings based on key connections indicated that one prodromal PD subtype closely resembled the pattern of NCs (N-P; N=159), while the other was similar to the pattern of PD (P-P; N=103). Additionally, significant differences (p<0.05) were observed between two prodromal PD subtypes in terms of multiple clinical measurements, neuroimaging for morphological changes, and gene enrichment for synaptic transmission. Finally, the prodromal PD subtypes were able to reproduce among 13 brain atlases or parcellation schemes.

The present study confirmed that patients in the prodromal phase of PD manifest heterogeneous clinical presentations, and that variation across individuals cannot be attributed solely to a single impairment. Furthermore, prodromal PD subtypes exhibited unique neuroanatomic patterns and clinical symptoms. Notably, the morphological alterations observed between prodromal PD subtypes are meaningfully associated with gene expression, which provided a more stable alternative to the symptom-based definitions of subtypes. Therefore, our work could significantly advance our understanding into the heterogeneity in the biological mechanisms that underlie prodromal PD and facilitates the accurate prediction of disease profiles for individuals. Ultimately, the findings can further inform precise and personalized intervention in PD during its early stages.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ¹

Aging ²

Aging

Degenerative Disease

Machine Learning

MRI