News

A collaborative research team funded by the International Progressive MS Alliance has published results that advance the goal of finding a way to shorten the length of clinical trials and reduce the number of participants needed to test therapies for progressive MS. This work is part of the Alliance’s global research strategy to prioritize and coordinate efforts needed to find more and better treatments and improve quality of life for people living with progressive MS. The National MS Society plays a key leadership and supporting role in the Alliance.

• Background: Most clinical trials gather results about the effectiveness of a therapy by combining the results of all participants on the test therapy and calculating the average response. This may dilute the effects of a therapy that works in some, but not all, of the participants.
• An additional issue is that progression is not easily measured and usually happens over long periods of time, making it hard to quickly detect whether a therapy is impacting the course of disease. Developing markers such as blood tests or MRI characteristics that could facilitate the detection of the benefits of therapies more quickly would speed the development of more treatments for people with progressive MS.
• This Study: An international team led by Professor Douglas Arnold, M.D., of McGill University used advanced machine learning to predict progression using various MRI, disease, and demographic characteristics of participants at the beginning of previously conducted multi-year clinical trials in MS. They were able to identify characteristics that relate to treatment response over the short time intervals used in phase 2 trials.
• By testing their findings against results from an additional clinical trial, the team verified the ability of their machine learning model to predict who would be more likely to respond to immune-modifying therapies. They believe this model may be employed to “enrich” recruitment for phase 2 clinical trials with individuals who are most likely to respond to an experimental therapy, enabling shorter trials involving fewer participants.
• Once a therapy is quickly tested in likely responders, it could then be tested more thoroughly in a broader, more inclusive group of participants.
• This study is one facet of this collaborative network’s ongoing efforts to identify personalized markers of MS progression to facilitate testing of new therapies for progressive MS. The model that has been developed will be made available to the pharmaceutical industry and to the scientific community.

 
“Estimating individual treatment effect on disability progression in multiple sclerosis using deep learning,” by Jean-Pierre R. Falet, Joshua Durso-Finley, Brennan Nichyporuk, Julien Schroeter, Francesca Bovis, Maria-Pia Sormani, Doina Precup, Tal Arbel & Douglas Lorne Arnold, was published online on September 26, 2022 in Nature Communications (Volume 13, Article number: 5645). This open-access paper may be read by anyone without the need of a subscription.