Amyotrophic Lateral Sclerosis (ALS) is a devastating, progressive neuromuscular disease that has seen very little progress in therapeutic interventions in the last 30 years. This isn’t because there isn’t a drug pipeline; over 110 Phase 2 and Phase 3 industry-sponsored trials were started over the last decade (2013 – 2023 on clinicaltrials.gov). Sadly, this trend continued with two recent failed clinical trials, including the Phase 3 trial for Relyvrio and the Phase 2 trial for SAR443820/DNL788. In both instances, the therapies did not demonstrate a reduction in the ALS Functional Rating Scale-Revised (ALSFRS-R), which was used as the primary endpoint.

Do these trials fail because the drugs aren’t having any effect, or because the endpoints aren’t sensitive enough to detect a change in patient function? 

The ALSFRS-R clinical rating scale is a staff-administered questionnaire that can provide non-invasive and diverse insight into how people with ALS are functioning. However, this rating requires in-clinic visits and can suffer from recall bias and inter-rater variability. These types of point-in-time measurements may not be sensitive enough to capture the day-to-day variabilities in a progressive disease such as ALS. Wearable, sensor-based digital health technologies (DHTs) can continuously collect data for an objective measure of two main functions that are important to assess in patients with ALS – gross motor function and fine motor skills.  

In the first installment of our Digital Health Monthly scientific webinar series, we learned about new data collected with DHTs that can collect objective, continuous, and sensitive measures to track function in patients with ALS. 

Real-World Measures of Gait in ALS using DHTs 

Rakesh Pilkar, Senior Data Scientist at ActiGraph, shared data from a natural history research study, conducted in collaboration with Frederik Steyn of The University of Queensland, where ALS patients wore an ActiGraph GT9X DHT device on their wrist. The analysis of data from this project began with the question: What is meaningful to patients with ALS when it comes to function in their daily life? According to the Voice of the Patient ALS Patient-Focused Drug Development Survey Report published by the ALS Association, walking was the most common response when asked what they wished they could still do, indicating that functional mobility is meaningful to patients.  

Gait characteristics provide more detailed information about functional mobility than step count alone, but obtaining measures of gait from wrist-worn accelerometry data has historically been a challenge. The wrist is a dynamic joint compared to lumbar or ankle locations, which have traditionally been used as wear locations when deriving gait measures. Since asking patients to wear an actigraphy device on the wrist can increase adherence and reduce burden, this data analysis focused on three questions -  

• Can wrist-derived DHT measures of gait distinguish between ALS and the controls? 
• Can wrist-derived DHT measures of gait track ALS progression? 
• How well do the wrist-derived DHT measures of gait correlate with the clinical measure? 

Rakesh and the science team applied a novel algorithm that classified raw accelerometry data from the wrist into non-rhythmic or rhythmic data, which they used to derive the gait characteristics of steps and cadence. Gait characteristics of speed and distance were estimated by combining anthropometric data (age, height, weight, gender) with rhythmic data.  

Using this algorithm and the subsequent gait metrics, Rakesh and the team found that all four DHT-derived measures of gait distinguished between ALS patients and controls and that DHT-derived measures of gait significantly decreased as measures were obtained over 9-12 months. This finding suggests these measures were sufficient to differentiate between functional states and track changes over time. Further, each of the gait measures also correlated to some degree with the ALSFRS scale, supporting the clinical validity of these digital measures of gait.  

This analysis is a promising sign that wrist-derived DHT measures could be a sensitive, valid, and low-burden way to track disease progression and treatment effect in ALS. Next steps for this project include determining how much walking data is necessary to collect reliable measures, especially in populations where walking volume is low and declines over time. Rakesh and the team are also working on determining how the wearable data should be aggregated to determine the most clinically meaningful measures. 

Tracking Upper Limb Movement in ALS with DHTs 

Marcin Straczkiewicz, PhD, Researcher, DHT Consultant and Start-Up Advisor, shared data from his recent paper in The Lancet, Upper limb movements as digital biomarkers in people with ALS. The authors of the paper spent time with clinicians to better understand what the main challenges were for patients living with ALS and learned that progressive disability in the upper limbs significantly impacted patients’ quality of life. The authors had access to a data set that included actigraphy data from ALS patients who wore a DHT device (ActiGraph CentrePoint Insight Watch) on their wrists continuously for 6 months. However, using raw accelerometry data to derive meaningful measures of upper limb function is difficult because of the heterogeneous nature of upper limb movements, which has been a historical challenge of creating meaningful digital measures of upper limb function. 

The publication provides an innovative use of accelerometry data to measure the movement of the wrist, using the spherical coordinate system of the sensor (pitch and roll) to derive a new set of measures:  

• four forearm movement types: flexion, extension, pronation, and supination  
• the count and duration of each movement type 
• the angle threshold in degrees 

The authors found that these upper limb movement metrics showed a statistically significant change over time, and many of the measures correlated with ALSFRS-RSE scores, supporting the clinical validity that these measures changed with disease progression. These digital measures of upper limb function can provide additional insights into patients’ function in everyday life and the specific movements associated with daily tasks. For example, DHT-based measures of duration of upper limb flexion and extension are associated with strength required for everyday activities such as dressing and rotating in bed, but not with detailed activities that require precision like handwriting or using utensils.  

Marcin concluded with real data from an ALS patient collected over 20 weeks, one that shows the stark progression of the disease. While heartbreaking, this data provides hope that DHT-based measures like those discussed here are a part of building more effective clinical tools that can help drug development programs better assess disease progression and the impact of therapy in patients with ALS.  

We were excited to see the publication of another paper in The Lancet, Remote monitoring of amyotrophic lateral sclerosis using wearable sensors detects differences in disease progression and survival: a prospective cohort study, soon after our live webinar. This research showed a novel measure derived from a hip-worn accelerometer correlated with and was predictive of ALS disease progression in 97 patients over a multi-year observation time period. This work further develops evidence that DHT-based measures can be used to assess functions meaningful to patients with ALS, and that they can sensitively capture changes over time that may help the drug development community advance treatments for patients with this devastating disease. 

Click here to watch the on-demand recording of this first installment of our Digital Health Monthly scientific webinar series, Advancing ALS Research with DHTs: Where Are We Now?