Study Published in Nature Digital Medicine Shows Potential of Machine Learning and Augmented Reality-based Digital Biomarkers in Alzheimer’s Detection
In addition to the pandemic, the digitization of healthcare is rapidly shifting clinical trials from analog to hybrid or fully decentralized clinical trials, catalyzing the development and implementation of digital health technologies to support clinical research.
Digital endpoints not only support the decentralization of clinical trials but also provide a means to collect more sensitive, high-frequency data while reducing trial costs and improving overall efficiency.
Below, we discuss the use of digital endpoints in clinical trials for neurology, highlighting the benefits and impediments to adoption.
The assessment of health and disease state requires a robust set of criteria to define health status, disease progression, and therapeutic response. These measurements are known as endpoints. In Alzheimer's disease clinical trials, for example, endpoints are utilized to measure and understand the therapeutic response to a drug or therapy.
Digital endpoints differ from traditional endpoints in that they include the use of sensor-generated data that is often collected outside of a clinical setting during their normal Activities of Daily Living (ADLs). This data can be derived from body-worn sensors, integrated sensors in wearables and portables, and even ingestibles and embeddables. For example, an integrated inertial sensor in a smartphone can be used to measure gait, or a smartwatch can be used to measure blood oxygen levels and heart rate.
Digital endpoints have the potential to offer previously inaccessible accuracy and precision, provide new previously unattainable insights, and ultimately, reduce the cost of drug development.
By using digital endpoints in clinical trials, both patients and pharmaceutical companies will see a breadth of benefits. Patients benefit from easier access to clinical trials and new therapies brought to the market more rapidly, while pharmaceutical companies benefit from more efficient and affordable trials.
Subject screening and selection for neurological disease clinical trials is currently an incredibly costly and time-consuming process. Typically, subject screening relies on expensive imaging and diagnostic technologies. For example, to determine subject eligibility, many Alzheimer’s drug clinical trials require expensive positron emission (PET) scans or invasive cerebrospinal fluid analyses as a part of the key inclusion criteria to establish protein levels, such as beta-amyloid and tau.
With thousands of potential subjects, the screening and selection process can drive clinical trial expenses through the roof. Digital endpoints, along with strong analytical tools, have the potential to be utilized to determine precisely where on the Alzheimer’s disease continuum a given patient lies. This method can serve as an initial tool to narrow down the subject pool, filtering out ineligible subjects before completing costly imaging or diagnostic procedures needed for key inclusion criteria.
Currently, there are few tools available for efficient and accurate longitudinal monitoring of neurocognitive function. These assessments are only performed periodically throughout clinical trials and typically rely on a person to administer and score the assessment, introducing human-to-human variability. Additionally, normal day-to-day human variabilities within the test takers themselves can significantly bias assessment results. By utilizing digital endpoints, we not only gain more accurate, sensitive data but also benefit from a more detailed understanding of therapeutic response.
Understanding at a highly granular level how an individual’s ADLs are changing is arguably the most compelling method for monitoring brain function in neurological disease patients. However, longitudinally assessing the effect of neurological disease-modifying drugs on clinical trial subjects is currently largely dependent on unreliable cognitive assessments that collect limited data on cognition and practically no data on function. These assessments often produce noisy, highly variable results that lack the specificity and granularity to reasonably conclude the efficacy of a drug or therapy, ultimately delaying the go/no-go decision.
While traditional endpoints are still fundamental and necessary for trials, digital endpoints and digital biomarkers provide pharmaceutical companies with a scalable, affordable way to collect the granular, supplemental, and contextual information needed to arrive at a go/no-go decision confidently and efficiently.
It is clear that digital endpoints offer benefits beyond overcoming the shortcomings of traditional endpoints. However, there are barriers to digital endpoint adoption that should be considered.
While there are barriers, the benefits to patients and pharmaceutical companies greatly outweigh the efforts needed to support this adoption.
Altoida’s mission is to accelerate and improve drug development, neurological disease research, and patient care. To learn more about our precision-neurology platform and app-based medical device, contact us!
