Researchers at University College London and Queen Square Analytics used artificial intelligence to analyze brain scans and blood markers from 634 multiple sclerosis patients, identifying two distinct biological subtypes termed "early-sNfL" and "late-sNfL" that reflect different disease trajectories.
The early-sNfL subtype shows elevated nerve injury markers and rapid brain lesion formation early in the disease with a 144% increased risk of new lesions, while the late-sNfL subtype exhibits brain volume loss before overt nerve damage and is more common in older patients.
The breakthrough could enable personalized treatment approaches beyond current symptom-based classifications, with early-sNfL patients potentially receiving more aggressive therapies and closer monitoring while late-sNfL patients may benefit from neuroprotective treatments.
Read detailed
