Six-Minute Activity-95th Centile, a Novel Wearable-Derived Clinical Outcome Assessment for Duchenne Muscular Dystrophy

Nicholas Joy, Jonathan H. Soslow, W. Bryan Burnette, Andrew Liu, Christine C. Guo, Rakesh Pilkar, James C. Slaughter, Xu Meng, Kimberly Crum, Karry Su, Christopher F. Spurney, Nazia Husain, Katheryn Gambetta, Brian D. Soriano, Frank J. Raucci, Kan N. Hor, Larry W. Markham, Jaclyn Tamaroff, Jonathan Dayan, Jay A. Campbell, Jennifer S. Li, Beth D. Kaufman, Carol Wittlieb‐Weber, Weiqing WangPediatric 2025 · DOI

TL;DR

This paper introduces a novel wearable-derived metric, the "Six-Minute Activity-95th Centile" (6M95c), to assess skeletal muscle progression in Duchenne Muscular Dystrophy (DMD). 6M95c effectively differentiates between ambulatory and nonambulatory individuals and shows declines over time, mirroring disease progression, making it a promising effort-independent outcome measure for clinical trials.

What Problem Does This Paper Address?

Duchenne Muscular Dystrophy (DMD) is a progressive genetic disorder leading to severe muscle weakness and eventual cardiopulmonary failure. Accurately tracking disease progression is crucial for developing and evaluating new therapies. Current standard assessments, like the six-minute walk test (6MWT) and quantitative muscle testing (QMT), have limitations, particularly for nonambulatory patients, as they can be effort-dependent and require clinical settings. There's a need for an objective, effort-independent, and easily deployable measure that can capture skeletal muscle decline across the spectrum of DMD, including for those who have lost the ability to walk.

Methods

The researchers developed a novel metric called the "Six-Minute Activity-95th Centile" (6M95c). This metric is derived from data collected by triaxial accelerometers worn on the dominant wrist of participants for 7 days. The process involves:

  1. Splitting daily activity data into 6-minute segments.
  2. Calculating the total activity counts (Vector Magnitudes - VMs) for each segment.
  3. For each day, identifying the 95th percentile of these 6-minute activity counts. This captures the peak activity within a 6-minute window for that day.
  4. Excluding the first and last days of the wear period to ensure data quality.
  5. Averaging the daily 95th centiles over the remaining days to obtain the individual's 6M95c score.

This 6M95c metric was then analyzed to see if it could differentiate between ambulatory and nonambulatory individuals, track longitudinal declines, and correlate with existing measures like QMT.

Datasets

The study utilized data from a larger, ongoing longitudinal natural history study of individuals with DMD.

  • Participants: 139 individuals with DMD aged 8 years or older were initially enrolled.
  • Analysis Cohort: 102 male participants were included in the final analysis, with a median age of 12.0 years.
  • Wear Periods: A total of 184 valid wear periods were analyzed from these participants across three annual study visits.
  • Exclusions: 37 participants were excluded due to insufficient valid accelerometry data (less than 3 valid wear days).

No new benchmarks were introduced; the study aimed to validate a new metric against existing ones.

Key Insights and Ideas

  • Effort-Independent Measurement: The core innovation is creating a metric that reflects peak physical activity without requiring conscious effort from the participant, overcoming a major limitation of traditional tests like the 6MWT.
  • Bridging the Gap: 6M95c has the potential to serve as a continuous outcome measure, bridging the gap between ambulatory and nonambulatory stages of DMD, providing a consistent way to track progression.
  • Modeling Disease Trajectory: The observed declines in 6M95c over time and its correlation with time since loss of ambulation suggest it effectively models the natural progression of skeletal myopathy in DMD.
  • Convergent and Discriminant Validity: The moderate correlations with QMT (a measure of muscle strength) support 6M95c's convergent validity, while the lack of correlation with cardiac measures supports its discriminant validity as a measure of skeletal muscle function.

Key Results

  • Ambulatory vs. Nonambulatory: Significant differences were found in all 6-minute activity centiles (25th, 50th, 75th, and 95th) between ambulatory and nonambulatory wears, with nonambulatory individuals showing lower activity levels.
  • Longitudinal Decline: Over a 2-year period, significant declines were observed in the 50th, 75th, and 95th centiles of 6-minute activity, indicating disease progression.
  • Loss of Ambulation: While no significant differences were detected immediately before and after loss of ambulation in this small sample, regression analyses showed that 6M95c declined with increasing years since loss of ambulation.
  • Correlation with QMT: Moderately strong correlations were observed between 6M95c and indexed QMT (quantitative muscle testing) scores, particularly for total arm and total leg strength (rs = 0.65, P < 0.001).
  • Predictive Power: The regression models indicated that time since loss of ambulation explained a significant proportion of the variance in 6M95c, suggesting its utility in predicting disease progression in nonambulatory individuals.

Limitations and Future Work

  • Small Sample Size for Loss of Ambulation: The analysis of participants who experienced loss of ambulation was limited by a small sample size (N=11), which may have weakened statistical power and limited the ability to detect significant changes immediately before and after this transition. Future longitudinal studies are needed to better assess this.
  • Participant Compliance: A significant number of participants (37 out of 139) were excluded due to insufficient valid accelerometry data, highlighting challenges in ensuring participant compliance with wearing the device. Strategies to improve compliance, such as reminder communications and financial incentives, are suggested.
  • External Noise: The accelerometer's ability to account for noise from external movement sources, such as powered wheelchairs, is limited. Further study is needed to determine the extent of this bias and develop adjustment factors.
  • Validation Against Standard Measures: Future work should focus on correlating 6M95c with standard clinical measures like the North Star Ambulatory Assessment (NSAA) and 6MWT in ambulatory populations to assess its convergent validity.
  • Independent Cohorts: Validation in independent cohorts is necessary to confirm these findings.
  • Clinical Trial Applications: The authors suggest that 6M95c could serve as a baseline for future clinical trials, providing an effort-independent, at-home outcome measure for therapy recipients.

Summary updated

Abstract

Ambulatory and nonambulatory patients with Duchenne muscular dystrophy were differentiated by 6-min activity and declined over time, modeling progression of skeletal myopathy. Six-minute activity-95<sup>th</sup> centile has potential for future use as an effort-independent outcome of skeletal muscle progression and functional decline for both ambulatory and nonambulatory individuals.
BibTeX
@inproceedings{joy2025six,
  author = {Nicholas Joy and Jonathan H. Soslow and W. Bryan Burnette and Andrew Liu and Christine C. Guo and Rakesh Pilkar and James C. Slaughter and Xu Meng and Kimberly Crum and Karry Su and Christopher F. Spurney and Nazia Husain and Katheryn Gambetta and Brian D. Soriano and Frank J. Raucci and Kan N. Hor and Larry W. Markham and Jaclyn Tamaroff and Jonathan Dayan and Jay A. Campbell and Jennifer S. Li and Beth D. Kaufman and Carol Wittlieb‐Weber and Weiqing Wang},
  title = {{Six-Minute Activity-95th Centile, a Novel Wearable-Derived Clinical Outcome Assessment for Duchenne Muscular Dystrophy}},
  booktitle = {Pediatric},
  year = {2025},
  doi = {10.1016/j.pediatrneurol.2025.11.017},
}
Figure