Identification of passive wrist-worn accelerometry outcomes for improved disease monitoring and trial design in motor neuron disease.

Background: Motor neuron disease (MND) leads to progressive functional decline, making reliable measures of disease progression critical for patient care and clinical trials. Current clinical outcome measures lack the ability to continuously and objectively track functional decline in daily life of patients with MND. This study assessed and validated wrist-worn accelerometry outcome measures for continuous monitoring in MND, with the potential to refine clinical trial outcomes.

Methods: This longitudinal study included 95 patients with MND who wore an ActiGraph GT9X Link device on their non-dominant wrist for 8 days, with follow-up every 3-4 months. Accelerometer data were processed using ActiLife and GGIR. Joint models were used to simultaneously investigate the longitudinal change in ALS Functional Rating Scale-Revised (ALSFRS-R) scores and accelerometer-derived outcomes alongside their relationship with overall survival. Sample size estimates for clinical trials were generated using both accelerometer- and ALSFRS-R-based outcomes, and principal component analysis (PCA) explored outcome relationships.

Findings: Accelerometer outcomes showed a slower rate of decline (-0.03 to -0.07 SD/month) compared to ALSFRS-R (-0.10 SD/month) and had stronger correlations with ALSFRS-R motor subdomains (partial r: 0.60-0.73). PCA revealed that longitudinal measures of accelerometry were distinct from the ALSFRS-R, highlighting the complementary nature of these measures. Peak 6-min activity predicted smaller clinical trial sample sizes for studies over 12 months. Accelerometer-derived outcomes were not significantly associated with survival.

Interpretation: Wrist-worn accelerometry offers a practical solution for continuous monitoring in MND, complementing ALSFRS-R. Measures of peak performance, and specifically peak 6-min activity shows promise, potentially reducing sample sizes and improving disease tracking over longer duration studies. Further refinement and validation are needed to adopt actigraphy measures as clinical assessment outcomes.

Funding: This study was supported by Wesley Medical Research (2016-32), the Honda Foundation, Motor Neurone Disease Research Australia, and FightMND. CJH received a Higher Degree Research Scholarship from UQ. STN received support from the Scott Sullivan Fellowship (MND and Me Foundation/RBWH Foundation), a FightMND Mid-Career Fellowship, and the AIBN.