A groundbreaking international study led by the University Medical Center Freiburg and the University of Oxford, published in Nature Medicine, fundamentally questions the previous understanding of multiple sclerosis (MS). Instead of the previous classification into subtypes such as relapsing or progressive, a new, AI-supported model describes MS as a dynamic disease continuum with four central condition dimensions: physical disability, brain damage, clinical relapses and silent inflammatory activity. These findings could decisively change the diagnosis and treatment of MS and have far-reaching implications for other diseases.

The study is based on the analysis of data from Novartis’ NO.MS cohort, complemented by other datasets such as the Roche ocrelizumab cohort and the MS PATHS cohort. In total, more than 8,000 patients and more than 35,000 MRI images were evaluated. The new model shows that MS is not characterized by rigid subtypes, but by dynamic state transitions. Inflammatory processes, whether in the form of flare-ups or clinically silent activity, are considered to be central drivers of disease deterioration. A direct transition to severe stages of the disease without prior inflammatory activity is almost impossible.

The previous subtype classification often led to restrictions on access to effective therapies, as approvals are based on these categories. The new model enables an individualized risk assessment that quantifies the current disease status regardless of defined subtypes. Patients with silent inflammatory activity in particular could benefit from early, targeted therapeutic approaches.

Beyond MS, the condition-based model offers potential for other medical areas. The data-driven approach, supported by artificial intelligence, could revolutionize the classification and treatment of numerous diseases in neurology and beyond by replacing rigid categories with flexible, dynamic state descriptions.

The next steps are aimed at integrating the model into clinical practice. Prospective studies are intended to validate individualized risk assessment and promote its application in therapy decisions and patient education. In the long term, the model could also change the approval logic of new therapies by creating the basis for more precise, patient-centered medicine.

Original Paper:

AI-driven reclassification of multiple sclerosis progression | Nature Medicine

Editor: X-Press Journalistenbüro GbR

Gender Notice. The personal designations used in this text always refer equally to female, male and diverse persons. Double/triple naming and gendered designations are used for better readability. ected.