Objective: To identify sensor-based digital measures of motor function that exhibit reliability, association with MDS-UPDRS Part 3, and longitudinal progression in early-stage Parkinson’s disease (PD), to support development of a composite digital biomarker of MDS-UPDRS 3 progression for use in PD drug development.

Background: Digital health technologies can improve precision in assessing early PD progression compared with traditional clinical assessments. Prior work has shown that composite digital measures can exhibit larger progression effect sizes as compared to MDS-UPDRS 3¹. This work extends earlier efforts by advancing a multidomain, clinically interpretable composite digital measure of motor function that can serve as an indicator of MDS-UPDRS 3 progression.

Method: Using data from the 12-month observational WATCH-PD study (N=82 early PD, N=50 healthy controls) digital measures of motor function were derived from research grade inertial sensors worn during standardized in-clinic movement tasks. Individual measures were evaluated for: (1) interpretability and reliability; (2) association with MDS-UPDRS 3; (3) 12-month progression in early PD. A redundancy-reduction step was then performed for highly correlated measures that captured similar motor function concepts.  Composite digital measures were then constructed from the resulting list of digital measures and explored for potential future utility in assessing disease-modifying drug efficacy.

Results: Thirteen digital measures were selected, each meeting predefined reliability, validity, and progression criteria, spanning multiple motor domains, including bradykinesia, gait, turning, and upper-limb movement. Composite digital measures constructed from these measures closely tracked MDS-UPDRS  3 progression and demonstrated larger progression effect sizes.

Conclusion: Sensor-based digital measures of motor function progression across multiple domains were identified from in-clinic, task-based assessments in early PD. Composite digital measures constructed from these measures demonstrated larger progression effect sizes than MDS-UPDRS 3, indicating potential value for assessing disease progression in future trials. Ongoing work will further evaluate their suitability as digital biomarkers for PD drug development trials.

References: 1. S. Zhai, et al. Developing Composite Digital Measures for Tracking Parkinson’s Disease (PD) Progression using a Comprehensive Machine Learning-based Framework [abstract]. Mov Disord. 2024; 39 (suppl 1).

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/identification-of-digital-measures-of-motor-progression-from-in-clinic-assessments-in-early-parkinsons-disease-for-composite-digital-biomarkers-results-from-watch-pd/