Unraveling Parkinson's disease heterogeneity using subtypes based on multimodal data

Area of Science:

• Neuroscience
• Neurology
• Medical Imaging

Background:

• Parkinson's disease (PD) is a neurodegenerative disorder with significant clinical and neuroanatomical heterogeneity.
• Previous subtype identification in PD has lacked multimodal data integration.
• Multimodal data, including clinical, motor, cognitive, and neuroimaging assessments, can offer complementary insights into PD subtypes.

Purpose of the Study:

• To identify distinct subtypes of Parkinson's disease using a comprehensive multimodal dataset.
• To leverage clinical, cognitive, motor, and neuroimaging variables for PD subtyping.
• To enhance understanding of PD heterogeneity for potential personalized interventions.

Main Methods:

• Utilized cross-sectional data from 95 participants with mild-to-moderate Parkinson's disease.
• Employed random forest clustering for subtyping based on clinical, motor, cognitive, and MRI data (structural and resting-state).
• Analyzed neuroimaging profiles and descriptive statistics to characterize identified subtypes.

Main Results:

• Identified three distinct Parkinson's disease subtypes: motor-cognitive, cognitive dominant, and motor dominant.
• The motor-cognitive subtype exhibited widespread brain alterations and motor/cognitive impairments.
• The cognitive dominant subtype showed frontoparietal changes and cognitive deficits, while the motor dominant subtype had motor impairments without detectable brain alterations.

Conclusions:

• Three distinct PD subtypes were successfully identified using a multimodal approach.
• Motor variables and sensorimotor structural MRI were the most influential factors in subtyping.
• These findings can advance the understanding of PD heterogeneity, paving the way for tailored treatments and rehabilitation strategies.