DDC AADC overlap (Cieselska 2017 / 18F-L-DOPA / Why increase / Qs), Diagnosis of PD overlap, Stages of early PD overlap, Digital section opener (Lipsmeier 2018 PMID 29701258 / 6-month Phase 1b smartphone trial / APDM composite motor endpoint), Konectom Roche-Biogen partnership, Digital Trials APDM ML feature reduction (122 → 28 → 9 → Random Forest 25 features), WATCH-PD Takeda + Biogen + MJFF program, APDM walking-feature characteristic table opener (Ratio of peak power / spontaneous tremor detection)
4 min read
DDC AADC overlap (Cieselska 2017 / 18F-L-DOPA / Why increase / Qs), Diagnosis of PD overlap, Stages of early PD overlap, Digital section opener (Lipsmeier 2018 PMID 29701258 / 6-month Phase 1b smartphone trial / APDM composite motor endpoint), Konectom Roche-Biogen partnership, Digital Trials APDM ML feature reduction (122 → 28 → 9 → Random Forest 25 features), WATCH-PD Takeda + Biogen + MJFF program, APDM walking-feature characteristic table opener (Ratio of peak power / spontaneous tremor detection)
Digital
Lipsmeier et al. 2018, PMID 29701258 / Clinical P1
6-month, phase 1b clinical trial with 44 Parkinson participants, and an independent, 45-day study in 35 age-matched healthy controls; participants completed six daily motor active tests (sustained phonation, rest tremor, postural tremor, finger-tapping, balance, and gait), then carried the smartphone during the day (passive monitoring), enabling assessment of, for example, time spent walking and sit-to-stand transitions by gyroscopic and accelerometer data. analysis of finger tapping and memory tests collected remotely using smartphones.
Approach: Using a large cohort (312 PD subjects and 236 controls) of participants in the mPower study.
demonstration of a method for continuous, objective assessment of activity tremor and bradykinesia based on data from a single wrist-worn accelerometer
APDM smartwatch and smartphone
Devices are available
To build a composite motor endpoint analysis of WATCH-PD results (~Sep 2024) (will inform ePOC feasibility)
GOOD resource: https://www.aifoflorum.org/news/conference-coverage/de-smartphones-collect-better-clinical-data-paper-and-pencil-tests
goal
developing novel, objective, efficient (or comprehensive) digital assessments of individuals with early, untreated PD42)
Identify digital outcomes measures that best correlate with MDS-UPDRS
Identify digital outcomes measures that best correlate with DaTscan
Identify digital outcome measures in subjects with possibly untreated effect size → Reduce clinical trial sample size, length (longitudinal 가으 가)
subtype identification: Clustering individuals into groups with similar progression rates
eg generating phenotypes specific to genetic sub-types of PD (eg, due to LRRK2 or GBA mutation)
Identify rapid progressors
enhance our understanding of the future of the disease, including its expression, pathophysiology, and potential modulators
Validation required
analytic validate
whether algorithm accurately processes the data (i.e., the calculation of gait speed from accelerometry data true?)
how to reduce noise (idiosyncratic behaviors, environmental settings and changes, and effects of comorbid conditions)
clinical validate
correlation with clinical scales
correlation with Datscan
Digital Trials
Konectom
Device
Konectom
Indici and Biogen entered into an agreement to develop digital biomarkers for Parkinson's disease. Part of the agreement allows Biogen to license Konectom - a smartphone-based digital biomarker platform to Indici. Indici will also be responsible for the digital management of Konectom in the Phase 1B LUMA study that is enrolling B/IB12 in patients with early-stage Parkinson's. This drug is being developed in collaboration with Denali Therapeutics. Indici will ensure the continued deployment of Konectom, including ongoing service availability, development, maintenance, algorithm development and data processing to generate derived digital measures from the LUMA study. Additionally, Biogen and Indici plan to develop and validate novel digital endpoints in Parkinson's and explore additional disease areas for endpoint development leveraging Konectom.
only walking and postural sway
APDM ML
Total 122 features
Over 18 m
1
MDS-UPDRS-III-Motor exam guide tile sum of 13.10 Cad. 3.11 Freezing of Gait. 3.12 Postural Stability, 3.13 "Posture", "13 progression feature [target fig (?)"
No progression (and visit-to-visit fluctuation)
2
28 (?) two progression with lowest p-value : (i) (5) of terminal double support of contralateral, (ii) (5) of swing of contralateral)
28 features that showed statistically significant progression scores only, 19 reflect walking variability, the sum of 23 weakened
3
29 (?) nine most accurate estimate of MDS-UPDRS III progression : 가
Progression %35 points per Visit (i.e., per 3 months); from 71 in either baseline, due to by the induced variability of the modeled data, compared to MDS-UPDRS
4
Random forest validation : using 25 features (cross-tab, dynamics, baseline) showed good analytical validation and known-groups validity
(Lipsmeier 2022 #2626) / Roche PD Mobile Application v2 #2625
Start: March 2019 End: Sept 2021 50% enrollment March 2020, LPO July 2021, data read out Sept 2021
over the study duration of 12 months subjects may meet medication for treatment of PD
WATCH-PD (study)
Takeda is sponsoring with Biogen and MJFF, Univ. Rochester. Eliza (?) Custom Motor and Cognitive Assessments of Episode (a continuously evolving (motor, gestureorth) sensor (APDM, Inc.), Apple Watch, ankle and ankle accelerometer, gyroscope and magnetometer data https://www.apdmcom/2/