Cognitive factor-based selection increases power in Alzheimer's dementia randomized clinical trials
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913747755335680 |
|---|---|
| author | Gallini, Julia Baucom, Zach Tripodis, Yorghos |
| author_facet | Gallini, Julia Baucom, Zach Tripodis, Yorghos |
| contents | Alzheimer's dementia (AD) is of increasing concern as populations achieve longer lifespans. Many of the recent failed AD clinical trials recruiting cognitively intact individuals had a low number of AD events and were thus underpowered. Previous trials have attempted to address this issue by requiring signs of cognitive decline in brain imaging for trial enrollment. However, this method systematically excludes people of color and those without access to healthcare and results in a selected sample that is not representative of the target population. We therefore propose the use of a predictive model based on cognitive test scores to enroll cognitively normal yet high risk participants in a hypothetical clinical trial. Cognitive test scores are a widely accessible tool, so their use in enrollment would be less likely to exclude marginalized populations than biomarkers (such as imaging), which are overwhelmingly available to exclusively high-income patients. We developed a novel longitudinal factor model to predict AD conversion within a 3-year window based on data from the National Alzheimer's Coordinating Center. Through simulation, we demonstrate that our predictive model provides substantial improvements in statistical power and required sample size in hypothetical clinical trials across a range of drug effects compared to other methods of subject selection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_16044 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Cognitive factor-based selection increases power in Alzheimer's dementia randomized clinical trials Gallini, Julia Baucom, Zach Tripodis, Yorghos Methodology Applications 62P10 Alzheimer's dementia (AD) is of increasing concern as populations achieve longer lifespans. Many of the recent failed AD clinical trials recruiting cognitively intact individuals had a low number of AD events and were thus underpowered. Previous trials have attempted to address this issue by requiring signs of cognitive decline in brain imaging for trial enrollment. However, this method systematically excludes people of color and those without access to healthcare and results in a selected sample that is not representative of the target population. We therefore propose the use of a predictive model based on cognitive test scores to enroll cognitively normal yet high risk participants in a hypothetical clinical trial. Cognitive test scores are a widely accessible tool, so their use in enrollment would be less likely to exclude marginalized populations than biomarkers (such as imaging), which are overwhelmingly available to exclusively high-income patients. We developed a novel longitudinal factor model to predict AD conversion within a 3-year window based on data from the National Alzheimer's Coordinating Center. Through simulation, we demonstrate that our predictive model provides substantial improvements in statistical power and required sample size in hypothetical clinical trials across a range of drug effects compared to other methods of subject selection. |
| title | Cognitive factor-based selection increases power in Alzheimer's dementia randomized clinical trials |
| topic | Methodology Applications 62P10 |
| url | https://arxiv.org/abs/2503.16044 |