In a landmark study that could revolutionise how multiple sclerosis (MS) is treated, scientists have harnessed artificial intelligence to uncover two entirely new biological subtypes of the condition.
AI and Blood Tests Reveal Hidden Patterns
The research, led by experts at University College London (UCL) and Queen Square Analytics, analysed data from 600 patients. The team focused on a specific protein in the blood called serum neurofilament light chain (sNfL), which acts as a marker for nerve cell damage and indicates how active the disease is.
By feeding sNfL data and MRI brain scans into a sophisticated machine learning model named SuStaIn, the researchers identified two clear biological patterns, detailed in the journal Brain. This moves beyond the traditional classification of MS based purely on a patient's symptoms.
The Two New Subtypes: Early and Late sNfL
The first subtype, termed 'early sNfL', is characterised by patients who show high levels of the protein early in their disease progression. These individuals also exhibited rapid development of brain lesions and visible damage in a critical area known as the corpus callosum. This form appears to be more aggressive.
The second subtype, 'late sNfL', follows a different path. Here, patients demonstrated brain shrinkage in regions like the limbic cortex and deep grey matter before their sNfL levels increased. This suggests a slower, more insidious disease course where overt damage becomes apparent later.
Paving the Way for Personalised MS Care
The lead author of the study, Dr Arman Eshaghi of UCL, emphasised the importance of the find. "MS is not one disease and current subtypes fail to describe the underlying tissue changes, which we need to know to treat it," he stated.
This breakthrough means doctors could soon predict which patients are at higher risk of severe complications. Those identified with the 'early sNfL' subtype could be fast-tracked for higher-efficacy treatments and closer monitoring. Conversely, patients with 'late sNfL' might benefit more from therapies designed to protect brain cells.
Caitlin Astbury, Senior Research Communications Manager at the MS Society, hailed the work as "an exciting development in our understanding of MS." She noted that moving towards definitions based on biology, rather than just symptoms, is crucial for finding effective, personalised treatments that can halt progression.
With around 20 treatment options available for relapsing MS but far fewer for progressive forms, this AI-driven discovery offers new hope for millions living with the condition worldwide, promising a future where care is tailored to the individual's unique disease biology.
