Wir überwinden Grenzen

(20.08.2025) New model describes multiple sclerosis as a disease continuum with dynamic stages instead of subtypes. Inflammation is seen as a key determinant of disease progression. The approach has relevance far beyond MS. Publication in Nature Medicine.

Multiple sclerosis (MS) was previously regarded as a disease with various subtypes such as "relapsing" or "progressive". An international study published in Nature Medicine on August 20, 2025 under the leadership of the University Medical Center Freiburg and the University of Oxford radically questions this dogmatic model after analyzing the NO.MS cohort (study data from Novartis). Instead of fixed disease phenotypes, an AI-supported model identifies four central condition dimensions that map the course of MS much better: physical disability, brain damage, clinical relapses and silent inflammatory activity. The findings could fundamentally change the diagnosis and treatment of MS patients and also be important for other diseases.

Continuous disease process with definable state transitions

"Our data clearly show that MS cannot be characterized by different subtypes such as relapsing or progressive MS, but is a continuous disease process with definable state transitions," says Prof. Dr. Heinz Wiendl, Medical Director of the Department of Neurology and Neurophysiology at the Freiburg University Medical Center.

The results are based on the analysis of over 8,000 patients and more than 35,000 MRI images from various studies (NO.MS Cohort, Roche Ocrelizumab Cohort, MS PATHS Cohort).

Disease as a dynamic system: a new view of MS

The probabilistic model describes MS as a sequence of states with specific transition probabilities. Earlier, mild states usually transition via inflammatory intermediate phases into advanced, irreversible stages of the disease. Remarkably, a direct transition to the severe stages without prior inflammatory activity is practically impossible - silent, symptom-free inflammation or relapses are key drivers of deterioration.

Implications for diagnostics, therapy and approvals

In many cases, the previous classification system makes access to effective medication more difficult, as approvals are based on rigid subtype definitions. The new model allows an individualized risk assessment - regardless of the diagnosed subtype.

"Instead of categorizing patients, we should quantify their condition and track it dynamically," says Wiendl. Patients with active but clinically silent inflammatory activity in particular require early treatment decisions, as the model impressively demonstrates.

A model with a broad impact - beyond MS

State-based modeling using artificial intelligence methods is not just a scientific breakthrough in MS research. "The principle is fundamental and groundbreaking - and it can also be applied to many other diseases, both in neurology and beyond," says Prof. Dr. Lutz Hein, Dean of the Faculty of Medicine at the University of Freiburg. The key is to move away from rigid, fixed disease categories and instead focus on data-based, flexible disease states within the disease.

Next steps: translation into clinical practice and research

"It is now important to translate these possibilities of individualized risk assessment into clinical practice and to collect prospective data for this purpose," emphasizes Prof. Dr. Peter Berlit, Secretary General of the German Neurological Society. The model has already been successfully tested within the study using external clinical and real-world data sets. The next step is now to transfer it to everyday clinical practice, for example for therapy decisions or better patient education. In the future, dynamic classification could also fundamentally change the approval logic of future therapies.

Original title of the publication: AI-driven reclassification of multiple sclerosis progression
DOI: 10.1038/s41591-024-03001-z
Link to the study: AI-driven reclassification of multiple sclerosis progression | Nature Medicine

More interesting articles