Advancing Alzheimer's prevention: Are clinical trials ready for earlier interventions?

Last week, the FDA announced the launch of the Neuroscience Advancement and Innovation Nexus (NAIN), a new cross-centre programme designed to advance the science, tools and regulatory approaches supporting medical product development for conditions affecting the brain and nervous system. Its first initiative, the Alzheimer’s Prevention Accelerator, will focus on treatments intended to prevent or delay Alzheimer’s-related cognitive decline. 

This initiative addresses some of the most persistent challenges in Alzheimer’s drug development, with particular emphasis on identifying appropriate trial populations, developing stage-appropriate endpoints and establishing the evidence needed to demonstrate that treatments prevent or delay cognitive decline. 

Advances in blood biomarkers are making it possible to identify Alzheimer’s pathology years before noticeable changes in memory or thinking. Earlier intervention could delay the onset of symptoms, but it poses a difficult question for clinical research: how can we demonstrate that treatment has prevented cognitive decline in someone who has yet to develop measurable impairment? 

Addressing this challenge requires a better understanding of who is likely to progress, how cognition changes in the earliest disease stages and how those changes relate to meaningful treatment benefits. 

Digital cognitive measurement plays an important role in this, particularly the potential for sensitive, longitudinal digital assessments to complement biomarkers in generating more detailed and nuanced evidence of disease progression and treatment efficacy. 

Rethinking endpoints for prevention trials 

Traditional Alzheimer’s trials have largely evaluated whether treatment slows deterioration in patients already experiencing cognitive impairment. In prevention trials, participants may perform within normal ranges for years, making it difficult to detect meaningful differences between treatment groups using conventional pen and paper cognitive scales. 

Developing suitable endpoints therefore requires a more targeted approach to measuring cognitive change, tailored to the disease stage and intended treatment effect. 

In very early symptomatic disease, sensitive measures of memory and learning may detect subtle changes that broader clinical scales miss. In cognitively unimpaired populations, researchers may need to evaluate the emergence of impairment or consider alternative endpoints. 

Digital cognitive assessments could help by capturing more precise performance measures repeatedly over time, providing greater insight into individual cognitive domains and trajectories. 

For example, CANTAB Paired Associates Learning (PAL) — a visual learning and recall task —  tracked amyloid status in older adults without dementia where standard memory measures did not [1] and related to biomarkers measured a decade earlier. [2]  

Integrating biomarkers and cognition 

Biomarkers provide evidence of Alzheimer’s pathology and whether treatments are affecting their intended biological targets. Cognitive assessments provide insight into how these biological changes relate to memory, thinking and other cognitive functions, and can separate them by domain. For example, in patients with mild cognitive impairment, different CANTAB® subtests were associated with different pathological markers, with some tracking hippocampal volume and others tracking cerebrospinal fluid amyloid and tau.[3] 

Combining these measures could help address two important challenges in prevention trials. 

First, identifying appropriate participants. Individuals with similar biomarker profiles may experience very different rates of cognitive decline. Integrating cognitive performance with biological data could improve predictions of disease progression and help researchers select populations more likely to demonstrate measurable treatment effects. 

Second, evaluating treatment efficacy. Longitudinal digital cognitive assessment alongside biomarkers could provide complementary evidence of whether changes in underlying pathology translate into preserved cognitive function. 

The priority is to establish where combining these measurements improves clinical development decisions and strengthens the evidence of treatment benefit.

Building the scientific evidence

Cambridge Cognition is contributing to this research through studies examining the relationship between biomarkers and cognitive performance.  

The AI-Mind study has investigated associations between CANTAB® cognitive measures and plasma p-tau217 in individuals with mild cognitive impairment, helping identify measures with promising sensitivity to Alzheimer’s pathology and longitudinal reliability.[4] 

Meanwhile, the INTUITION Brain Health Study, conducted in collaboration with Biogen and Apple, demonstrated CANTAB®’s sensitivity to clinical status as well as the feasibility of remote cognitive monitoring at scale, enrolling more than 23,000 US participants. 

These studies contribute to the scientific foundation for more sensitive, scalable cognitive measurement, while further research is needed to establish the suitability of specific measures as efficacy endpoints in prevention trials. 

Advancing Alzheimer’s prevention 

The FDA’s new initiative presents an important opportunity to advance the methods used to evaluate preventative treatments. 

As Alzheimer’s research moves towards earlier intervention, developing sensitive, clinically meaningful endpoints and understanding how cognitive and biological measurements can be used together will be increasingly important. 

The ability to detect Alzheimer’s pathology earlier has opened new possibilities for treatment. Now we need the scientific tools and evidence to demonstrate that earlier intervention can meaningfully change the course of cognitive decline.

References:

1. Pettigrew C, Soldan A, Brichko R, et al; BIOCARD Research Team. Computerized paired associate learning performance and imaging biomarkers in older adults without dementia. Brain Imaging Behav. 2022;16(2):921–929.
2. Soldan A, Pettigrew C, Moghekar A, Albert M; BIOCARD Research Team. Computerized cognitive tests are associated with biomarkers of Alzheimer’s disease in cognitively normal individuals 10 years prior. J Int Neuropsychol Soc. 2016;22(10):968–977.
3. Nathan PJ, Lim YY, Abbott R, et al; PharmaCog Consortium. Association between CSF biomarkers, hippocampal volume and cognitive function in patients with amnestic mild cognitive impairment (MCI). Neurobiol Aging. 2017;53:1–10.
4. Study – AI-Mind
5. Butler PM, Yang J, Brown R, Hobbs M, Becker A, Penalver-Andres J, Syz P, Muller S, Cosne G, Juraver A, Song HH, Saha-Chaudhuri P, Roggen D, Scotland A, Silveira N, Demircioglu G, Gabelle A, Hughes R, Erkkinen MG, Langbaum JB, Lingler JH, Price P, Quiroz YT, Sha SJ, Sliwinski M, Porsteinsson AP, Au R, Bianchi MT, Lenyoun H, Pham H, Patel M, Belachew S. Smartwatch- and smartphone-based remote assessment of brain health and detection of mild cognitive impairment. Nat Med. 2025 Mar;31(3):829-839. doi: 10.1038/s41591-024-03475-9. Epub 2025 Mar 4. PMID: 40038507; PMCID: PMC11922773.