Parkinson’s Disease is a complex, progressive neurological condition which requires challenging, multi-modal endpoint assessment across both motor and non-motor symptoms. Trials typically involve a combination of neuroimaging-based approaches and clinical assessments conducted using the Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS), a revised version, released in 2008, of the original UPDRS scale already in use from the mid-80s.

The MDS-UPDRS provides a more comprehensive picture of the experiences of Parkinson’s patients, capturing a wide range of motor and non-motor symptoms. However, challenges remain with subjectivity and the episodic nature of the assessments, with MDS-UPDRS scores only representing a “snapshot” of the patient’s overall condition.

Fluctuations are a key feature of Parkinson’s Disease, caused by shifting levels of dopamine in the brain. These fluctuations often take the form of “on-off” or “wearing off” phenomena, where symptoms wane and return often depending on when the last dose of levodopa, which restores the brain’s dopamine levels, was taken. As a result, clinical assessments may not capture with maximum accuracy true disease state or treatment effects, and this can be particularly challenging in shorter trials with less assessments.

Enter wearable devices and digital biomarkers

Wearable devices have been gaining increasing attention in neurological clinical trials, and their use within protocols is on the rise. Wearables now provide continuous motor monitoring using sensors such as accelerometers, gyroscopes, and magnetometers to record posture, motion, and gestures in real-time and real-world conditions, with machine learning algorithms analysing the data to detect subtle changes in gait and tremor, for instance. Such a combination of technologies could enable researchers to predict disease progression months before traditional scales.

With the growth of wearables and AI/ML in areas like Parkinson’s, sponsors gain access to sensitive, continuous digital biomarkers. Meanwhile, data becomes more representative of the patient’s overall condition as opposed to a snapshot assessment skewed by daily fluctuations.

Increasing use of wearables also supports a growing population of digitally enabled patients with a desire to better understand their disease and symptoms. In one recent US-based survey of Parkinson’s patients, 91% said they were very or somewhat interested in using wearable technologies, with 67% already having had experience with consumer wearable devices.[i] 97% of the respondents were over the age of 50, challenging the common misconception that older generations are not willing to embrace digital technologies. Empowered patients are more likely to remain engaged with the research from start to finish, offering further benefits to sponsors considering this approach.

Wearables are just the beginning for AI in Parkinson’s trials

With the advancement of deep learning, models are now capable of processing raw and unstructured data from MRI scans, EEG signals, and voice and video recordings, opening up further avenues for detecting subtle biomarkers with enhanced precision.[ii] As these capabilities progress and the range of available digital biomarkers expands, their integration into Parkinson’s trials will enable researchers to gain a more holistic view of a patient’s health.

With enhanced clinical assessments and AI analytics used to quantify brain atrophy and structural changes from MRI scans[iii], future trials can make use of an increasing number of objective biomarkers that correlate with clinical progression, heralding a better, more accurate future for data integrity in Parkinson’s trials.

The increasing advancement of AI tools represents a major opportunity to enhance and streamline neuroscience clinical trials. Download our new whitepaper, below, to understand what this opportunity could mean across various neurological indications and trial operations.

[i] Hirczy, S, Zabetian, C, Lin, YH. The current state of wearable device use in Parkinson’s disease: a survey of individuals with Parkinson’s. Front. Digit. Health, 23 December 2024. https://doi.org/10.3389/fdgth.2024.1472691
[ii] https://pmc.ncbi.nlm.nih.gov/articles/PMC12336134/
[iii] Reddy S, Giri D, Patel R. Artificial Intelligence Diagnosis of Parkinson’s Disease From MRI Scans. Cureus. 2024 Apr 23;16(4):e58841. doi: 10.7759/cureus.58841. PMID: 38784299; PMCID: PMC11114626.