Machine learning analysis of motor evoked potential time series to predict disability progression in multiple sclerosis.

BACKGROUND: Evoked potentials (EPs) are a measure of the conductivity of the central nervous system. They are used to monitor disease progression of multiple sclerosis patients. Previous studies only extracted a few variables from the EPs, which are often further condensed into a single variable: the EP score. We perform a machine learning analysis of motor EP that uses the whole time series, instead of a few variables, to predict disability progression after two years. Obtaining realistic performance estimates of this task has been difficult because of small data set sizes. We recently extracted a dataset of EPs from the Rehabiliation & MS Center in Overpelt, Belgium. Our data set is large enough to obtain, for the first time, a performance estimate on an independent test set containing different patients. METHODS: We extracted a large number of time series features from the motor EPs with the highly comparative time series analysis software package. Mutual information with the target and the Boruta method are used to find features which contain information not included in the features studied in the literature. We use random forests (RF) and logistic regression (LR) classifiers to predict disability progression after two years. Statistical significance of the performance increase when adding extra features is checked. RESULTS: Including extra time series features in motor EPs leads to a statistically significant improvement compared to using only the known features, although the effect is limited in magnitude (ΔAUC = 0.02 for RF and ΔAUC = 0.05 for LR). RF with extra time series features obtains the best performance (AUC = 0.75±0.07 (mean and standard deviation)), which is good considering the limited number of biomarkers in the model. RF (a nonlinear classifier) outperforms LR (a linear classifier). CONCLUSIONS: Using machine learning methods on EPs shows promising predictive performance. Using additional EP time series features beyond those already in use leads to a modest increase in performance. Larger datasets, preferably multi-center, are needed for further research. Given a large enough dataset, these models may be used to support clinicians in their decision making process regarding future treatment.

Best Practices for Long-Term Monitoring and Follow-Up of Alemtuzumab-Treated MS Patients in Real-World Clinical Settings.

Multiple sclerosis (MS) is a chronic autoimmune neurological disease that typically affects young adults, causing irreversible physical disability and cognitive impairment. Alemtuzumab, administered intravenously as 2 initial courses of 12 mg/day (5 consecutive days at baseline, and 3 consecutive days 12 months later), resulted in significantly greater improvements in clinical and MRI outcomes vs. subcutaneous interferon beta-1a over 2 years in patients with active relapsing-remitting MS (RRMS) who were either treatment-naive (CARE-MS I; NCT00530348) or had an inadequate response to prior therapy (CARE-MS II; NCT00548405). Efficacy with alemtuzumab was maintained over 7 years in subsequent extension studies (NCT00930553; NCT02255656), in the absence of continuous treatment and with a consistent safety profile. There is an increased incidence of autoimmune events in patients treated with alemtuzumab (mainly thyroid events, but also immune thrombocytopenia and nephropathy), which imparts a need for mandatory safety monitoring for 4 years following the last treatment. The risk management strategy for alemtuzumab-treated patients includes laboratory monitoring and a comprehensive patient education and support program that enables early detection and effective management of autoimmune events, yielding optimal outcomes for MS patients. Here we provide an overview of tools and techniques that have been implemented in real-world clinical settings to reduce the burden of monitoring for both patients and healthcare providers, including customized educational materials, the use of social media, and interactive online databases for managing healthcare data. Many practices are also enhancing patient outreach efforts through coordination with specialized nursing services and ancillary caregivers. The best practice recommendations for safety monitoring described in this article, based on experiences in real-world clinical settings, may enable early detection and management of autoimmune events, and help with implementation of monitoring requirements while maximizing the benefits of alemtuzumab treatment for MS patients.

Does including the full CVLT-II and BVMT-R improve BICAMS? Evidence from a Belgian (Dutch) validation study.

BACKGROUND: The Brief International Cognitive Assessment for Multiple Sclerosis (BICAMS) is a fast, easy-to-administer and already widely validated neuropsychological battery for cognition in multiple sclerosis. OBJECTIVE: The goals of our study were to validate the BICAMS in a Belgian Dutch-speaking population and to investigate to what extent including extensive versions of two of the three BICAMS subtests improved its psychometric qualities. METHODS: Ninety-seven persons with MS and ninety-seven healthy controls were included and group-matched on age, education level and gender. All participants performed the BICAMS with an extensive version of the CVLT-II and BVMT-R. RESULTS: The SDMT and BVMT-R were able to dissociate between the MS and healthy control group, while the CVLT-II was not. Distributions of CVLT-II scores suggest learning effects in the MS group, indicating the need for alternative word lists or the construction of an adapted version fitted for repeated administration. Including the full CVLT-II and BVMT-R did not markedly improve the psychometric qualities of the BICAMS. CONCLUSION: This study validates the BICAMS in a Belgian Dutch-speaking population and facilitates the use of it in clinical practice, while providing evidence that including full versions of the CVLT-II and BVMT-R does not increase its psychometric qualities markedly.