Multiple Sclerosis Treatment Is Being Rethought After Two Hidden Forms Are Found
What if Multiple Sclerosis were never just a deadly disease? For ages, people living with MS have faced the same diagnosis and labels, but their experiences could be different. Some progress rapidly, whereas others undergo slower, more unpredictable changes. Until now, medicine has struggled to explain why. Treatments are often selected based on visible symptoms, even when these reflect only part of the story.
Using AI (Artificial Intelligence) in conjunction with MRI scans and blood tests, Researchers have identified two previously unknown biological subtypes of Multiple Sclerosis. It is an extraordinary innovation that experts say could reshape how the disease is treated, diagnosed, as well as monitored.
The innovation provides a clearer Biological understanding of MS and raises the possibility of more personalised therapies that target the underlying disease processes rather than symptoms alone.
Multiple Sclerosis affects millions of people and causes damage to the spinal cord & brain, across the world. While several treatments are available, most are prescribed based on clinical presentation, laboratory findings, and reports. This symptom-based approach can overlook Biological differences among patients, thereby limiting the effectiveness of therapeutics in preventing long-term damage or even slowing disease progression.
Combining AI, Blood Tests, and Brain Imaging
Researchers at University College London and Queen Square Analytics led the novel Research. The Research team analysed data from over 600 people living with Multiple Sclerosis. They used a combination of MRI brain scans and blood tests to investigate disease activity at the biological level.
Researchers focused on sNfL (Serum Neurofilament Light Chain). It is a protein found in the blood that increases when nerve cells are damaged in the body. sNfL levels are increasingly used as a marker of disease activity in Multiple Sclerosis.
To interpret the complex datasets, Scientists employed an ML (Machine Learning) model known as SuStaIn (Subtype and Stage Inference). By analysing sNfL levels alongside MRI scans, the advanced AI systems were able to identify patterns of disease progression that aren’t visible through Clinical symptoms alone. The results were published in the medical journal “Brain.”
Two Distinct MS Subtypes Revealed, Redefining Multiple Sclerosis Treatment
The analysis identified two clear and distinct Biological subtypes of Multiple Sclerosis, termed early sNfL and late sNfL, based on when nerve damage markers appeared and how brain changes unfolded.
In the early sNfL subtype, patients exhibited high sNfL levels early in disease progression. MRI (Magnetic resonance imaging) scans revealed early damage to the Corpus Callosum. This structure connects the left and right hemispheres of the brain and is associated with the rapid development of brain lesions. Researchers described this subtype as more aggressive and biologically active.
By contrast, patients with late-stage sNfL MS exhibited brain volume loss in regions such as the limbic cortex and deep grey matter before sNfL levels increased. In this group, overt nerve damage appeared later, suggesting a slower disease trajectory.
Implications for Personalised Multiple Sclerosis Treatment
Researchers state that identifying these Biological patterns would help Clinicians and Medical Professionals better predict disease risk and customize treatment strategies for Multiple Sclerosis patients.
The lead author of the study and a Researcher at UCL, Dr Arman Eshaghi, stated that current MS classifications are limited. “MS is not one disease, and current subtypes fail to describe the underlying tissue changes, which we need to know to treat it,” he said.
“By using an AI model combined with a highly available blood marker with MRI, we have been able to show two clear Biological patterns of MS for the first time. This will help Clinicians understand where a person sits on the disease pathway and who may need closer monitoring or earlier, targeted treatment.”
Eshaghi added that, in the future, patients identified as having early-stage sNfL MS could be offered higher-efficacy treatments earlier and monitored more closely. Individuals with late-stage sNfL MS may benefit from various approaches, including therapies designed to protect brain cells and neurons.
Expert Views on the Discovery and Its Impact on MS Treatment
The MS Society welcomed the findings. Caitlin Astbury, Senior Research Communications Manager at the charity, described the study as “an exciting development in our understanding of MS.”
She explained that while Clinical symptoms currently categorise Multiple Sclerosis, these labels often fail to reflect the Biological processes driving the disease. “MS is complex, and these categories often don’t accurately reflect what is going on in the body, which can make it difficult to treat effectively,” she said.
With around 20 treatment options available for relapsing MS and limited choices for progressive forms, Researchers believe this work supports a shift toward Biology-based definitions of the disease.
By uncovering two distinct Biological subtypes of MS using Artificial Intelligence, this Research highlights the limitations of symptom-based classifications and points toward a future of more precise, personalised care. The breakthrough could help ensure patients receive the proper treatment at the right time, potentially improving long-term outcomes for people living with Multiple Sclerosis.
More importantly, the Research findings point toward a better and advanced future where treatments will be guided by what is actually happening inside the body, rather than by symptoms alone. For patients, this could mean closer monitoring when the disease is more active and more carefully chosen treatments at the right time. While more work is needed, the study offers a clearer, more promising direction, one in which care for Multiple Sclerosis becomes more precise, personalized, and effective.
