Predicting MS progression with 11 proteins

In MS, the immune system mistakenly attacks the protective myelin coating nerve axons in the brain and spinal cord. This damages the nerves and hampers the efficiency of signal transmission.

The physical symptoms of MS, such as fatigue, difficulty walking, imbalance and weakness, can be severe and limiting. As well as reducing an individual’s functional ability, living with MS can also cause emotional and mental health challenges. There is currently no cure for MS, but early treatment and intervention can help control the condition and relieve some symptoms.

Now, in a recent study led by Mika Gustafsson, a professor of bioinformatics at Linköping University, researchers have identified a combination of just 11 proteins capable of predicting long-term disability outcomes in MS. These proteins offer a potential tool for tailoring treatments based on the anticipated severity of the disease for individual patients.

The team measured and analyzed a large amount of proteins from 143 people with early-stage MS and 43 healthy controls. A combination of approaches, including proximity extension assays, next-generation sequencing and machine learning models, were used to identify potential MS biomarkers and to predict short- and long-term disease progression. The results were also later validated using samples from a separate group of 51 MS patients from Karolinska University Hospital (Stockholm, Sweden).

The significance of this discovery lies in its potential to guide early and more effective interventions. Gustafsson notes the important applications in balancing treatment benefits against potential side effects and costs. Since the analysis tool can identify which patients require more invasive and effective treatments, it can help avoid unnecessary side effects and expenses for patients who do not require these treatments.

“Having a panel consisting of only 11 proteins makes it easy should anyone want to develop analysis for this. It won’t be as costly as measuring 1,500 proteins, so we’ve really narrowed it down to make it useful for others wanting to take this further,” explained Sara Hojjati, a doctoral student at Linköping University.

Notably, the research team identified a specific protein, neurofilament light chain (NfL), leaking from damaged nerve axons, as a reliable biomarker for short-term disease activity. NfL has previously been recognized for its ability to identify nerve damage, and the current study reinforces its potential as an indicator of disease activity.

This breakthrough not only advances our understanding of MS but also opens avenues for targeted and efficient interventions, marking a significant stride in the application of machine learning for predicting the progression of chronic long-term conditions.