The theme of the #ACTRIMS Forum 2019 was “Precision Medicine Approaches for MS: Scientific Principles to Clinical Application.” Many talks and presentations were focused on precision medicine which is a model in which healthcare is customized to an individual through decisions, practices, and treatments. Check out what we heard about precision medicine in MS.
Current State of Precision Medicine in MS
Dr. Gavin Giovannoni, Queen Mary University of London, , London, United Kingdom, discussed some prognostic factors that can be used to select treatments for individual patients, and de-risking therapies. Firstly, Dr. Giovannoni discussed using a risk calculator to categorize people with MS into three prognostic groups: good, intermediate and poor prognosis. This calculator could include collecting information on age, magnetic resonance imaging (MRI) measures, system changes (motor, cognition etc), recovery from relapses, vitamin D levels, comorbidities and more. Using this calculator, people with MS would be assigned to a prognostic group and have a list of treatment that would best suit the them. In another part of his talk, Dr. Giovannoni discussed de-risking MS therapies. One example of this could be prescribing an antibody against an adverse effect of a therapy. There are many steps associated with de-risking immunosuppression which include performing a comprehensive baseline analysis examining leukocytes, platelets, serology tests, and MRI. Furthermore, individuals should be followed by close monitoring of infusion reactions and antibiotics with treatments and regular monitoring via blood tests, urine analysis, MRI, infections and disease activity. All these factors will allow for personalized decision making for the individual with MS.
Selecting the right patient for the right treatment
In his talk, titled “Selecting the right patient for the right treatment,” MS-Society funded neurologist, Dr. Mark Freedman, University of Ottawa and the Ottawa Hospital Research Institute, discussed factors to consider when personalizing therapy for people diagnosed with MS. Three profiles are key when determining treatment options: profile of the disease (disease activity and prognostic factors), profile of the therapy (mechanism of action, tolerability, safety, monitoring frequency), and profile of the individual (adherence, comorbidities, personal factors). When profiling MS, it is important to determine the risk of imminent disease progression. This can be done by looking at many factors including the patient age, sex, ethnicity, relapse severity, relapse recovery, and disability level. Furthermore, similar to Dr. Giovannoni, Dr. Freedman mentioned that it is important to look at prognostic factors such as MRI which includes measures such as the number of lesions, the location of the lesions, and level of atrophy. When profiling the therapy, there are many mechanisms of action including therapies that modulate the immune system, those that prevent harmful immune cells from entering the central nervous system and those that deplete harmful immune cells. The mechanism of action along with other factors such as adverse effects and mode of treatment is key when selecting therapies for MS. Ultimately, the individual should choose the right therapy for them using the disease profile, drug profiles, and personal considerations such as occupation, lifestyle, travel, monitoring availability, and pregnancy.
From visible to invisible- using biosensors to monitor individuals
In the Algorithms informing precision medicine management of MS session, Dr. Dina Katabi from Massachusetts of Technology spoke about using biosensors for measuring disease activity. Dr. Katabi developed a new technology, called Emerald, which monitors an individual’s breath, sleeping patterns, heart rate, and gait. What is novel about this technology is that it can monitor an individual using electromagnetic waves which means that the individual is not wearing any device on their body. Emerald is a box, similar to a wifi box, that sits in one part of the home and can obtain multiple measures on the individual. For example, the gait measurement is an important measure for MS, Parkinson’s, Huntington’s and other neurological conditions. While gait is measured in the clinic, it may not always represent how an individual is walking on a day-to-day basis. Emerald monitors the individual’s gait in their own home and over many days. Beyond gait, Emerald is developed to record activities of daily living, such as toileting and dressing, and also sleep patterns. In fact, Dr. Katabi’s research team is using Emerald to measure different stages of sleep from awake, rapid eye movement, light sleep, and deep sleep. In one study, Dr. Katabi measured the accuracy of the Emerald technology by monitoring 25 individuals for 100 nights. These individuals were connected to sleeping devices to monitor sleeping patterns. The outcomes from the sleeping devices were compared to the Emerald device which was kept in the corner of the same room. The Emerald device had 80% accuracy which is comparable to individuals that go to sleep labs for study monitoring. Currently, Dr. Katabi has deployed the technology in 200 homes of individuals with depression, Parkinson’s, pulmonary disease, Alzheimer’s, and MS.
Machine learning using MRI to predict disease progression
Finally, in a talk from Dr. Tal Arbel from McGill University, we heard about the machine learning methods used to predict lesion activity and progression in MS using MRI. Machine learning in medical imaging has the potential to inform diagnosis, disease development, patient outcomes, speed up clinical trials and promote a more personalized approach to healthcare. Specifically, Dr. Arbel presents on deep learning methods used to predict the development of new lesions and future disability progression in MS. To develop deep learning methods, the research team obtained data from large MS clinical trials through industry partners and analyzed thousands of MR images. Using machine learning methods, the team automatically detected lesions in images and are working on further developing models that can predict future lesion activity and clinical progression from baseline images. Machine learning methods have a potential to impact healthcare by accelerating precision medicine and drug discovery. Dr. Arbel and her research team will continue to develop deep learning to model the evolution of the disease with a specific focus on progression to progressive forms of the disease as well as trying to developing a way to identify individuals that are likely to respond to a treatment compared to those that are not.
What interests you most about precision medicine in MS? Leave a comment below.
Expressability