Objective: This study evaluates the feasibility, reliability, and validity of smartphone app-based testing in patients with Parkinson’s disease (PD) in the D1PAM (D1 positive allosteric modulator) phase 1B clinical trial.

Background: Currently, the lack of well-validated biomarkers to diagnose and monitor PD is a significant hurdle in the discovery of disease-modifying therapeutics. Smartphones, equipped with sensitive touchscreen and sensors, promise an unbiased approach to assess disease state and treatment response.

Method: An iPhone trial app, modeled after the public mPower study [1], was custom designed using the ResearchKit software framework incorporating tests to assess (1) sustained phonation, (2) finger tapping, (3) walking and resting, and (4) cognition in PD patients. The iPhones were locked-down with only the trial app and accompanying training materials accessible.During the two-week at-home pre-treatment, 6-week in-clinic D1PAM intervention and 2-week at-home post-treatment, 24 participants were instructed to complete the four active tests daily at 8 am, 10 am, 2 pm and 8 pm. During the 2-week intervention, each participant received daily doses of LY3154207 or placebo [2]. Compliance was quantified as completion of required daily tests during pre-treatment, intervention, and post-treatment. Test-retest reliability was quantified as the intra-class correlation coefficient (ICC) using features derived from pre-treatment sensor data. Clinical validity was quantified as the correlation between features and MDS-UPDRS clinical severity rating.

Results: Compliance: compliance was superior during D1PAM treatment averaging above 75% and about 50% during pre-treatment and post-treatment. Test-retest reliability: multiple features from tapping, walking and rest, and cognition tests had good reliability (ICC > 0.75). Clinical validity: among the most reliable features, some of tapping and walking features are moderately yet significantly correlated with MDS-UPDRS total and subscores (Spearman r> 0.4).

Conclusion: The results demonstrated the applicability of incorporating smartphone app-based testing in PD clinical study. These tests are amenable to regular at-home or in-clinic use and have the potential to enable digital biomarker development to objectively detect a subtle disease state change in response to treatment [3].

References: [1] Bot, B. M. et al. The mPower Study, Parkinson Disease Mobile Data Collected Using ResearchKit. Sci. Data 3:160011 (2016) [2] A Study of LY3154207 in Healthy Participants and Participants with Parkinson’s Disease. www.clinicaltrial.gov NCT02562768 [3] Li, Y. et al. Use digital sensors and deep learning to evaluate motor performance in the D1PAM (LY3154207) phase 1B Parkinson’s disease clinical trial. MDS (2019)

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/treatment-monitoring-using-objective-and-frequent-digital-testing-in-the-d1pam-ly3154207-phase-1b-parkinsons-disease-clinical-trial/