The importance of early Alzheimer's detection

Alzheimer's disease (AD) is a progressive neurodegenerative disorder affecting millions worldwide and is the most common cause of dementia in older adults [1]. A key factor is that the underlying neuropathological changes can begin 15 to 20 years before noticeable memory loss, cognitive decline, or difficulties with daily functioning appear [2][3][4]. Early detection is valuable because of the recent development of disease-modifying drugs for AD, such as Leqembi [5] and Kinsula [6]. These medications offer the potential to slow disease progression, but their effectiveness is greatest when administered during the early stages. However, identifying patients in these early stages is difficult. Individuals may not realize they are affected and may not seek medical attention [7]. Furthermore, traditional diagnostic methods can be costly and invasive, limiting their widespread use.
 

RADAR-AD: Exploring remote monitoring solutions

To address these challenges, the RADAR-AD project investigated the potential of remote monitoring technologies (RMTs) as a cost-effective, user-friendly approach to complement existing diagnostic procedures. Remote monitoring technologies include a range of devices and applications designed to passively or actively collect data related to an individual's health and behavior. Examples from our study include:

• Gait analysis: Using wearable sensors to measure gait parameters during specific walking tests.
• Simulated banking task: Using a smartphone application to simulate an ATM cash withdrawal, assessing cognitive and motor skills.
• Longitudinal activity tracking: Using activity trackers to continuously monitor, e.g., activity levels and sleep quality over extended periods.
• Cognitive assessments: Using apps like Altoida [8] and Mezurio [9] to track memory and cognitive domains.

229 participants were recruited for the RADAR-AD study and categorized into four groups according to their Alzheimer’s disease status:

• Healthy controls
• Preclinical AD (individuals with biomarkers of AD but no clinical symptoms)
• Prodromal AD
• Mild-to-Moderate AD

The data collected from the technologies were then analyzed using statistical methods and machine learning techniques to determine their ability to differentiate between the different disease stages (see Figure 1).

Key findings: Distinguishing between AD Stages

Our analyses revealed several key findings:

• Preclinical AD detection is challenging: Distinguishing healthy controls from individuals with preclinical AD proved difficult using all the assessed technologies.
• Discrimination in later stages: Statistically significant differences were found between healthy controls and individuals with prodromal or mild-to-moderate AD, indicating more potential in these populations. Reduced physical activity, decreased REM sleep duration, altered gait patterns, and cognitive decline (as assessed by apps like Altoida and Mezurio) were identified as potentially important indicators.
• Machine learning performance: Machine learning models showed varying degrees of success in distinguishing between AD stages. For instance, Mezurio-based models effectively differentiated healthy controls from mild-to-moderate AD patients, while Altoida and Fitbit-based models provided moderate performance in detecting the earlier prodromal stage. Notably, the Amsterdam iADL questionnaire demonstrated strong performance across stages. It was completed by a caregiver in this study, but it holds potential for remote application and could be a cost-effective and time-efficient supplement to traditional assessments.

For more detailed information, you can:

• Read our recently published paper
• Visit the RADAR-AD project website
• Learn about the PREDICTOM project, another of our group's projects that focuses on earlier dementia detection.

Citations

[1] Scheltens P, Strooper BD, Kivipelto M, Holstege H, Chételat G, Teunissen CE, et al. Alzheimer’s Disease. Lancet (London, England). 2021;397(10284):1577–90. https://doi.org/10.1016/S0140-6736(20)32205-4

[2] Jack CR Jr, Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA Research Framework: Toward a Biological Definition of Alzheimer’s Disease. Alzheimers Dement. 2018;14(4):535–62. https://doi.org/10.1016/j.jalz.2018.02.018

[3] Fortea J, Vilaplana E, Carmona-Iragui M, Benejam B, Videla L, Barroeta I, et al. Clinical and Biomarker Changes of Alzheimer’s Disease in Adults with Down Syndrome: A Cross-Sectional Study. Lancet. 2020;395(10242):1988–97. https://doi.org/10.1016/S0140-6736(20)30689-9

[4] Bateman RJ, Xiong C, Benzinger TLS, Fagan AM, Goate A, Fox NC, et al. Clinical and Biomarker Changes in Dominantly Inherited Alzheimer’s Disease. N Engl J Med. 2012;367(9):795–804. https://doi.org/10.1056/NEJMoa1202753

[5] van Dyck CH, Swanson CJ, Aisen P, Bateman RJ, Chen C, Gee M, et al. Lecanemab in Early Alzheimer’s Disease. N Engl J Med. 2023;388(1):9–21. https://doi.org/10.1056/NEJMoa2212948

[6] Kang C. Donanemab: First Approval. Drugs. 2024;84(10):1313–8. https://doi.org/10.1007/s40265-024-02087-4

[7] Langbaum JB, Zissimopoulos J, Au R, Bose N, Edgar CJ, Ehrenberg E, et al. Recommendations to Address Key Recruitment Challenges of Alzheimer’s Disease Clinical Trials. Alzheimers Dement J Alzheimers Assoc. 2023;19(2):696–707. https://doi.org/10.1002/alz.12737

[8] https://altoida.com/

[9] https://mezur.io/