British News Agency

Millions of people worldwide live with the disease; however, treatments are usually selected based on symptoms and may be ineffective because they do not target the patient’s underlying biology.

Now, scientists have identified two new biological types of MS using artificial intelligence, a simple blood test, and MRI scans. Experts said this “exciting” discovery could revolutionize the treatment of the disease globally.

In a study led by University College London (UCL) and Queen Square Analytics involving 600 patients, researchers examined blood levels of a specific protein called serum neurofilament light chain (sNfL). This protein can help indicate the extent of nerve cell damage and how active the disease is.

sNfL results and patients’ brain scans were interpreted by a machine-learning model called SuStaIn. The findings, published in the medical journal Brain, revealed two distinct types of MS: early sNfL and late sNfL.

In the first subtype, patients had high sNfL levels early in the disease, with visible damage occurring in a brain region known as the corpus callosum. Brain lesions also developed rapidly. Scientists said this type appears more aggressive and active.

In the second subtype, patients showed brain shrinkage in regions such as the limbic cortex and deep gray matter before sNfL levels rose. This type progresses more slowly, with marked damage appearing later.

Researchers say this breakthrough will enable doctors to more precisely understand which patients face higher risks of different complications and will open the door to more personalized care.

The study’s lead author, Dr. Arman Eshaghi of UCL, said: “MS is not a single disease, and current subtypes are insufficient to describe the underlying tissue changes we need to know about for treatment.

“By using MRI and an artificial intelligence model combined with a highly accessible blood biomarker, we were able for the first time to show two clear biological patterns of MS. This will help clinicians understand where a person is on the disease pathway and who may need closer monitoring or earlier, targeted treatment.”

Eshaghi added that in the future, when the AI tool suggests a patient has early sNfL MS, they may become eligible for higher-impact therapies and be monitored more closely.

By contrast, those with late sNfL may be offered different types of treatment, such as personalized therapies to protect brain cells or neurons. “Innovations will therefore be two-pronged: transforming clinical and neurological examinations that have changed little for centuries with the help of AI algorithms, and delivering personalized treatments based on disease profiles.”

Caitlin Astbury, senior research communications manager at the MS Society, said: “This is an exciting development in our understanding of MS.

This study used machine learning to analyze MRI and biomarker data from people with relapsing and secondary progressive MS. By combining these data, they were able to identify two new biological subtypes of MS.

In recent years, we have developed a better understanding of the biology of the disease. However, current definitions are based on the clinical symptoms a person experiences. MS is complex, and these categories often do not accurately reflect what is happening in the body, which can make it harder to treat effectively.”

Astbury noted that there are around 20 treatment options for people with relapsing MS, and some options are beginning to emerge for progressive MS, but many people still have no options. “The more we learn about the disease, the greater the likelihood of finding treatments that can stop its progression.”

“This research adds to the growing evidence supporting a shift away from existing MS descriptors (such as ‘relapsing’ and ‘progressive’) toward terms that reflect the underlying biology of the disease. This could help identify people at high risk of progression and offer more personalized treatment.”