Wearable Sensor-Derived Digital Outcomes for Quantifying Gait in ALS

Topic:

Clinical Trials

Poster Number: 190 M

Author(s):

Laurent Servais, MD PhD, University of Oxford, Katherine Burke, Neurological Clinical Research Institute and Sean M. Healey & AMG Center for ALS, Boston, MA, USA, Charlotte Rasser, SYSNAV, Vernon, France, Camila Gonzalez Barral, SYSNAV, Vernon, France, Narghes Calcagno, Department of Neurology and Laboratory of Neuroscience, IRCCS Istituto Auxologico Italiano, Milan, Kendall Carney, Neurological Clinical Research Institute and Sean M. Healey & AMG Center for ALS, Boston, MA, USA, Sydney Hall, Neurological Clinical Research Institute and Sean M. Healey & AMG Center for ALS, Boston, MA, USA, Sravan Mandepudi, Neurological Clinical Research Institute and Sean M. Healey & AMG Center for ALS, Boston, MA, USA, Timothy Royse, Neurological Clinical Research Institute and Sean M. Healey & AMG Center for ALS, Boston, MA, USA, Oihana Piquet, SYSNAV, Vernon, France, Charlotte Hermanne, University Department of Neurology, CHR Citadelle, Liège, Belgium, Laura Buscemi, CRMN, Department of Pediatrics, University Hospital Liege & University of Liege, Belgium, Laura Desire, SYSNAV, Vernon, France, Guillaume Parinello, Msc, SYSNAV, Vernon, France, Damien Eggenspieler, MSc, Sysnav, Stéphanie Delstanche, MD, CRMN, Department of Pediatrics, University Hospital Liege & University of Liege, Belgium, James Berry, MD, Massachusetts General Hospital, Margaux Poleur, MD, CRMN, Department of Pediatrics, University Hospital Liege & University of Liege, Belgium

Gait impairment in amyotrophic lateral sclerosis (ALS) can emerge early and contributes to loss of independence. Traditional gait assessments in ALS rely on a single ambulation item of the ALS Functional Rating Scale-Revised (ALSFRS-R) or infrequent in-person walking tasks. These measures are not quantitative, have limited sensitivity to change and lack ecological validity. Digital Health Technologies (DHTs) offer a promising alternative with continuous, remote monitoring of ambulation in real-world settings. Digital endpoints derived from these sensors, such as the stride velocity 95th centile (SV95C), provide objective measures of mobility. SV95C has been qualified by the European Medicines Agency as a primary endpoint in Duchenne Muscular Dystrophy. SV95C may similarly offer more sensitive quantification of ALS progression in clinical trials and, eventually, in clinical care.
We hypothesize that ankle-worn wearable sensor-derived digital endpoints can sensitively and reliably quantify ambulation in people with ALS.
Two longitudinal studies are being conducted to assess the feasibility, reliability, and construct validity of digital measures targeting key concepts of interest in ALS: number of strides per hour (NbStrides/h), SV95C, and peak walking distance (WD95C), quantifying respectively activity, top ambulation ability, and endurance. Participants undergo standard clinical evaluations (e.g. ALSFRS-R, 6MWT) every 3 months for 1 year. They wear sensors daily for 1 to 3 months after the baseline visit and for 1 month after subsequent visits. Reliability is assessed using the intra-class correlation coefficient (ICC) over two 15-day periods. Construct validity is evaluated via correlations with clinical anchors. Longitudinal changes are assessed using Wilcoxon signed-rank test and standardized response means (SRM).
To date, 47 ALS patients have been enrolled. Participants provided an average 281 hours of wear time during the first month. SV95C showed significant correlation (p<0.001) with 3 of 4 clinical anchors. It also demonstrated significant decline at month 3, 6 & 9, with SRM of -0.59, -0.81 & -0.71, respectively. These preliminary findings demonstrate the feasibility of using DHTs to capture real-world ambulation in ALS and suggest that sensor-derived endpoints may offer greater sensitivity to change than traditional clinic-based measures. Future analyses aim to further evaluate their potential utility for ALS research and clinical care.