Comparing Lesion Volume Dynamics Between Multiple Sclerosis and Neuromyelitis Optica Spectrum Disorder During...

Area of Science:

• Neuroimmunology
• Neuroimaging
• Machine Learning in Medicine

Background:

• Multiple sclerosis (MS) and aquaporin-4-positive neuromyelitis optica spectrum disorder (AQP4+ NMOSD) are distinct inflammatory demyelinating diseases of the central nervous system.
• Subclinical lesion progression during remission is poorly understood in both MS and NMOSD.
• Machine learning (ML) offers advanced quantitative analysis of brain lesions on MRI.

Purpose of the Study:

• To compare subclinical lesion volume changes between MS and AQP4+ NMOSD patients during remission.
• To utilize ML-based lesion segmentation for precise volumetric assessment.
• To investigate differences in disease activity during inter-attack periods.

Main Methods:

• Retrospective analysis of 3D brain MRI scans from 31 MS and 30 AQP4+ NMOSD patients.
• Application of ML-based segmentation for lesion quantification.
• Statistical analysis of lesion volume changes during the remission period.

Main Results:

• MS patients showed a significant median lesion volume increase (3,493 mm³ to 4,430 mm³, p<0.001), indicating ongoing subclinical activity.
• NMOSD patients exhibited no significant change in median lesion volume (640 mm³ to 930 mm³, p=0.129), suggesting an attack-dependent course.
• Annual lesion volume increase was substantially higher in MS (193 mm³/year) compared to NMOSD (25 mm³/year, p=0.017).

Conclusions:

• MS and AQP4+ NMOSD demonstrate fundamentally different pathogenic processes.
• Distinct disease mechanisms necessitate specialized therapeutic and monitoring strategies for each condition.
• ML-based neuroimaging is crucial for understanding subclinical disease progression in demyelinating disorders.