Data-driven hypothesis discovery from disease trajectories in multiple sclerosis.

Area of Science:

• Neurology
• Biostatistics
• Data Science

Background:

• Multiple sclerosis (MS) is an incurable autoimmune disease with unpredictable progression.
• Lack of reliable biomarkers hinders accurate prognosis and personalized treatment for MS patients.

Purpose of the Study:

• To introduce a novel trajectory-based statistical approach for identifying patterns in multiple sclerosis patient histories.
• To uncover previously unrecognized progression patterns and generate new hypotheses for MS research.

Main Methods:

• Longitudinal clinical data from 1,025 MS patients were analyzed using a trajectory-based statistical approach.
• Two analyses were performed: one on a large dataset (n=985) and another on a smaller cohort (n=83) to assess robustness.

Main Results:

• The approach identified novel progression patterns in MS, suggesting potential effects of Alemtuzumab on bowel/bladder function and glatiramer acetate on relapse occurrence.
• Confirmed known associations, such as the link between relapse activity and brain lesions.

Conclusions:

• The trajectory-based method is robust across different dataset sizes and can reveal previously unseen relationships in MS.
• Findings support hypothesis generation for biomarker discovery and therapeutic optimization in multiple sclerosis care.