How Digital Health Technologies are Addressing Unmet Measurement Needs in Parkinson’s Disease

Feature
Article

In a rare disease space that faces challenges in measurement quality, these technologies can enable the use of real-world data and improve study timelines.

© Pixel-Shot - © Pixel-Shot - stock.adobe.com

Image Credit: © Pixel-Shot - stock.adobe.com

Clinical development in Parkinson’s disease (PD) faces numerous challenges and pitfalls. There is an unmet need to improve measurement quality in PD, and digital health technologies (DHTs) are helping to address these gaps. Wearable devices can provide accurate and sensitive data to complement traditional assessments with known limitations. However, the use of DHTs in clinical development must be applied thoughtfully and include considerations for the maturity of products, regulatory implications, and operational feasibility. Here, we look at the essential aspects of DHT use in PD drug development and give examples of how digital measures can improve trial design.

The progression of PD

When we think of PD, most focus on the cardinal symptoms such as tremors, walking difficulties, and balance problems. However, non-motor symptoms such as cognitive impairment, mood disorders, and sleep disturbances are also common aspects of the disease and significantly impact daily living and quality of life.

The progression of PD is poorly understood, with variability in how symptoms appear among individuals. Non-motor symptoms like sleep disorders, depression, and loss of smell can emerge up to 20 years before the onset of motor symptoms or clinical diagnosis, making early intervention crucial but challenging. Interventions typically begin only after motor symptoms develop, meaning many people with early non-motor symptoms remain undiagnosed and ineligible for clinical trials. Trials often enroll individuals at a mid-stage of the disease course, complicating efforts to achieve meaningful long-term outcomes.

The variability and unpredictability of PD are evident not only in how the disease progresses across different individuals but also in how symptoms can fluctuate within the same person over a day. In the early motor stages, a person may feel fine in the morning with no signs of slowed movements, known as bradykinesia. However, as the day goes on and dopamine levels that are required for movement are depleted, bradykinesia emerges. These late-day symptoms of slowness and fatigue are frequently mistaken for the effects of a long, productive day. As the disease advances and dopamine levels drop further, these symptoms begin to appear earlier in the day. When they become disabling, medications that replenish dopamine may be necessary, allowing individuals to regain their earlier level of function temporarily.

Progression in PD isn't just a gradual worsening of existing symptoms. New symptoms may develop, and existing symptoms can change. Worsening bradykinesia, for example, can lead to "freezing," where a person cannot initiate walking or turning. Some symptoms, such as tremors, may develop or disappear over time. Traditional clinical outcome measures of PD often fall short because they cannot identify the intra- and inter-day variability of symptoms due to their need for certified raters.1

Measurement and biological variability

Some have viewed the sensitivity of PD outcome measures as an issue that could be addressed by digitizing existing rating scales. However, this approach has not proven successful because precision is only part of the problem; the more significant challenge lies in the disease's biological variability. As PD progresses, this variability increases, making it even more difficult to quantify the severity of various symptoms.

DHTs are the most effective way to better characterize biological variability. These technologies allow for continuous or near-continuous at-home assessments, where individuals are monitored during their daily activities over time, providing a clearer view of their functional abilities and symptom variability.2 This measurement precision is essential in disease-modifying therapy trials, where the challenge is identifying differences between rates of progression in the treatment and placebo groups and involves detecting the treatment signal amid the noise inherent in biological variability.

We focus in this article on DHTs that enable passive data collection of patient function (i.e. sensor-based wearables), but active digital assessments (i.e. app-based assessments) in a patient’s daily environment also provide an opportunity for more frequent assessments to reduce measurement noise. The benefits of incorporating active digital measures to complement standard clinical assessments for evaluating the progression of PD and the response to treatment in clinical studies, particularly those targeting the early stages of the condition, have been demonstrated in the work of the Critical Path for Parkinson's Consortium 3DT Initiative3 and the WATCH-PD study.4

The path to mature digital endpoints

Currently, no regulatory-endorsed endpoints exist for PD, but the Digital Medicine Society's V3+ Framework5 has provided valuable guidance for validating these tools and offering a systematic roadmap. This lengthy, complex, and costly roadmap makes it a daunting challenge for many to take on alone, but partnering with an experienced solutions provider can supplement this expertise.

Integrating a digital measure in the development pathway requires careful strategic planning. First, the current phase of a clinical development program must align with the maturity of the digital outcome assessment. This involves clearly defining the goal of integrating DHTs: is the aim to identify signals, guide go/no-go decisions, or pursue a registrational endpoint strategy as the molecule advances? The answers will determine the necessary resources and time commitment.

Additionally, companies should consider the long-term implications. How will this integration affect future interactions with regulators and payers? Will it offer competitive advantages? Could it benefit or streamline adjacent programs? Addressing these questions is essential for successful integration.

Implementing PD digital endpoints in clinical trials

When considering the different phases of drug development, whether early phases, feasibility, proof of concept, safety and efficacy testing, or later stages such as large registrational trials or post-market registries, there are implications for which types of digital endpoints to consider.

For example, several digital endpoints in a Phase I study may be tested to discover signals and responses to a novel drug or device. This data can potentially power a more extensive efficacy study for regulatory registration. In Phase II or Phase III studies, where the goal is to prove the hypothesis from the feasibility study, the focus may be more on specific digital endpoints. While PD studies still use traditional clinician-rated scales such as the Movement Disorders Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) or patient diaries, digital endpoints can complement these measures to support efficacy objectives during the regulatory trial process.

In post-market studies, which involve larger sample sizes to assess the drug or device's safety and efficacy, digital endpoints can help improve the diversity of the trial. This allows for examining specific patient population subsets, providing additional information to better characterize the real-world patient population.

PD digital endpoints

Motor and non-motor digital endpoints in PD trials include measurements of bradykinesia and dyskinesia,6 tremor,7 immobility, fluctuations, nighttime sleep, gait, and walking scores, such as those developed by PKG Health. These endpoints help characterize disease progression over time by indicating changes in symptom stability, variability, response to PD medications, or the ability to move at an average pace.

These digital endpoints have been validated both analytically and clinically, showing robust reliability, responsiveness to dopaminergic or surgical therapies, and strong correlations with traditional PD scales, such as the MDS-UPDRS, the Abnormal Involuntary Movement Scale (AIMS) score for dyskinesia, and sleep scales like the Epworth Sleepiness Scale (ESS).

Comparing digital endpoints to traditional patient-reported and clinician-rated scales reveals key relationships: Bradykinesia and tremor scores align with the motor components of the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS-III), dyskinesia and fluctuations correspond to Part 4 of the UPDRS (abnormal movements and side effects), and immobility and nighttime sleep scores correlate with the Parkinson’s Disease Sleep Scale (PDSS-2) or the Epworth Sleep Scale (ESS). Gait and walking measurements are comparable to the 6-Minute Walk Test (6MWT). ‘Off’ time scores, when a person's symptoms return between medication doses or when the effects of their medication wear off, reflect the severity and duration of bradykinesia or dyskinesia symptoms and are comparable to diaries collected from participants.

These correlations are modest, but they are not expected to correspond exactly. There is a difference in sampling of DHTs, which provide continuous measures over a longer period, whereas Clinical Rating Scales are sampled at a point in time. The alignment of digital measures with these traditional rating scales supports their clinical relevance, but digital measures provide unique insights in addition to what can be learned from traditional scales, hence their value.

Digital endpoint use cases

In trials assessing symptomatic therapies, data from wearable DHTs can provide objective measures to determine whether symptoms have changed at follow-up compared to baseline. This involves assessing symptom severity from mild to advanced phases and evaluating fluctuations and variability in those symptoms.

For disease-modifying therapies, the focus is on a smaller subset of scores to demonstrate the slowing or modification of disease progression over a more extended period. This is typically more relevant in the early to moderate phases of the disease. These trials may use derivatives of the wearable-derived bradykinesia and gait scores, which have been shown to indicate changes in disease progression over time.

Digital endpoints can also serve as enrichment tools, whereby patients can be monitored for the presence and severity of bradykinesia and dyskinesia.This can help clinical trial sponsors identify and stratify patients as primarily bradykinetic or dyskinetic, with the ability to exclude patients with dyskinesia in early disease trials or, conversely, select individuals with dyskinesia in trials of antidyskinetic therapies.

Digital endpoints address the measurement needs of PD trials

DHTs offer advantages over traditional scales in PD trials by enhancing statistical sample size calculations and improving the identification and stratification of patient populations before enrollment. This approach minimizes missing data and ensures the inclusion of participants sensitive to novel therapies.

Digital endpoints can increase study power, offering higher feasible sampling frequencies that can increase signal-to-noise. It is, though, important to acknowledge that DHTs are likely to be noisier without these higher sampling frequencies than rating scales.

Using objective, continuous, real-world data collected from wearable DHTs can help study teams measure the impact of interventions on meaningful aspects of health, providing clinical insights into how patients function in their daily lives. This transition is crucial for improving the pace and success of clinical development programs in PD. Current DHTs facilitate signal finding in shorter, smaller trials and enhance go/no-go decision-making processes. Future DHTs will benefit from the ongoing validation efforts and data collection, aiming to elevate them from promising exploratory endpoints to secondary or even primary endpoints.

Christine Guo, PhD, chief scientific officer, ActiGraph; Fatta B. Nahab, MD, executive director, clinical and digital development, Neuron23, Inc.; and Karen Krygier, consultant, PKG Health

References

  1. https://pubmed.ncbi.nlm.nih.gov/15853525/
  2. https://pubmed.ncbi.nlm.nih.gov/24030855/
  3. https://c-path.org/program/critical-path-for-parkinsons/#early-motor-parkinsons-dat-clinical-trial-enrichment-tool
  4. https://www.nature.com/articles/s41531-023-00497-x.pdf
  5. https://datacc.dimesociety.org/v3/
  6. https://content.iospress.com/articles/journal-of-parkinsons-disease/jpd11071
  7. https://content.iospress.com/articles/journal-of-parkinsons-disease/jpd160898

Recent Videos
Related Content
© 2024 MJH Life Sciences

All rights reserved.