Periodic Reporting for period 2 - DIVERT-AD (Exploit population imaging to unravel resistance to Alzheimer's disease)
Période du rapport: 2023-09-01 au 2024-08-31
Prevention of late-life cognitive impairment and dementia is one of the greatest global challenges in the 21st century. Currently, there are 47 million people worldwide living with dementia. This number is projected to double by 2030 due to rising life expectancy. In Europe alone, the annual cost of dementia is estimated to exceed €250 billion. These alarming figures underscore the critical need for effective dementia prevention strategies.
The Lancet Commission on Dementia Prevention has proposed fourteen potentially modifiable risk factors whose successful elimination could prevent up to 45% of dementia cases. These factors are less education, hypertension, high cholesterol, hearing impairment, smoking, obesity, depression, physical inactivity, diabetes, low social contact, excessive alcohol consumption, head trauma, visual impairment and air pollution (Livingston et al, Lancet, 2024). In order to personalise dementia prevention and make it more effective, we need to understand how these risk factors influence the development of dementia, and who would benefit most from risk management.
In DIVERT-AD, I investigated two mechanisms. Managing risk factors may firstly reduce the burden of brain pathology, like Alzheimer’s disease. It may secondarily increase a person’s ability to cope with brain pathology, so that cognitive performance is less affected relative to disease burden. This knowledge is important for the effective implementation of dementia prevention strategies, for example in terms of timing and which factors are more appropriate for primary and secondary prevention. In DIVERT-AD, I also investigated which groups are more vulnerable to dementia risk factors. Older age, female sex and carrying an apolipoprotein E4 allele (a common genetic variant) are key non-modifiable risk factors of dementia and critical for tailoring prevention strategies.
The Lancet Commission on Dementia Prevention has proposed fourteen potentially modifiable risk factors whose successful elimination could prevent up to 45% of dementia cases. These factors are less education, hypertension, high cholesterol, hearing impairment, smoking, obesity, depression, physical inactivity, diabetes, low social contact, excessive alcohol consumption, head trauma, visual impairment and air pollution (Livingston et al, Lancet, 2024). In order to personalise dementia prevention and make it more effective, we need to understand how these risk factors influence the development of dementia, and who would benefit most from risk management.
In DIVERT-AD, I investigated two mechanisms. Managing risk factors may firstly reduce the burden of brain pathology, like Alzheimer’s disease. It may secondarily increase a person’s ability to cope with brain pathology, so that cognitive performance is less affected relative to disease burden. This knowledge is important for the effective implementation of dementia prevention strategies, for example in terms of timing and which factors are more appropriate for primary and secondary prevention. In DIVERT-AD, I also investigated which groups are more vulnerable to dementia risk factors. Older age, female sex and carrying an apolipoprotein E4 allele (a common genetic variant) are key non-modifiable risk factors of dementia and critical for tailoring prevention strategies.
I conducted an in-depth investigation of the relationship between established modifiable risk factors of dementia and Alzheimer’s disease pathology. I used amyloid PET imaging, which is the gold-standard to quantify amyloid-beta plaques in the brain. For this analysis, I drew on data from the Rotterdam Study, a large, population-based cohort with repeated measurements of modifiable risk factors and one of the largest collections of amyloid PET scans in Europe, comprising over 600 participants. My findings show that diabetes was associated with greater amyloid plaque burden after an average follow-up of seven years. The association was similar in both genetic risk (APOE4) carriers and non-carriers. In contrast, hypertension was associated with higher amyloid plaque burden primarily among APOE4 carriers, indicating a significant gene-environment interaction. Many other established modifiable risk factors for dementia were not significantly associated with amyloid plaque burden in this cohort, including low education, physical inactivity (self-reported and accelerometer-based), smoking, excessive alcohol use, obesity, high cholesterol, hearing loss, depressive symptoms, head trauma, and social isolation.
In two other projects I identified factors that were associated with vascular brain disease and neurodegeneration, two other key causes of dementia. The first project was conducted within the Cross-Cohort Collaboration including six population-based cohorts with more than 16,000 participants. We found that a higher education level was not associated with magnetic resonance imaging markers of vascular disease (white matter hyperintensities) or neurodegeneration (e.g. hippocampal volume). Married participants had a greater hippocampal volume compared to unmarried individuals. Being physically active versus inactive (self-reported) was associated with fewer white matter hyperintensities. In the second project, I zoomed in on different types of physical activity. The novel aspect of this work was to leverage the UK Biobank’s longitudinal data collection for studying the bidirectional relationship between self-reported physical activity, vascular brain disease, and neurodegeneration. To our surprise, we found overall more consistent evidence that a healthier brain predicted higher physical activity levels (especially household activities) than for the inverse, more established relationship (higher strenuous physical activity predicted a healthier brain).
In a fair head-to-head comparison, I studied all twelve established modifiable risk factors for dementia in the same participants and their differential associations with brain pathologies and cognitive decline over time. In brief, I found that modifiable dementia risk factors were more consistently associated with vascular brain disease and especially cognitive decline than with Alzheimer’s disease pathology. This suggests that established dementia prevention targets primarily benefit vascular health and other non-Alzheimer’s pathways, underscoring the need to identify modifiable risk factors that specifically influence Alzheimer’s disease pathology. To address this question, I investigated whether sleep health may be such a modifiable risk factor for Alzheimer’s disease. I found that a weaker circadian rhythm, characterized by frequent transitions between sleep and wakefulness, such as daytime napping and nighttime awakenings, was associated with greater Alzheimer’s disease burden at follow-up. The association was significantly stronger in APOE4 carriers than non-carriers. These findings are particularly compelling because circadian rhythms are modifiable through behavioral and pharmacological interventions. Moreover, there is growing experimental evidence from animal models suggesting that disrupted sleep can accelerate amyloid-beta accumulation, providing a plausible biological mechanism for this association.
I disseminated the results of DIVERT-AD in six peer-reviewed articles including top-tier journals (Alzheimer’s & Dementia, BRAIN, JAMA Neurology, Neurology), and two oral talks at the largest international conference for Alzheimer’s Disease (AAIC).
In two other projects I identified factors that were associated with vascular brain disease and neurodegeneration, two other key causes of dementia. The first project was conducted within the Cross-Cohort Collaboration including six population-based cohorts with more than 16,000 participants. We found that a higher education level was not associated with magnetic resonance imaging markers of vascular disease (white matter hyperintensities) or neurodegeneration (e.g. hippocampal volume). Married participants had a greater hippocampal volume compared to unmarried individuals. Being physically active versus inactive (self-reported) was associated with fewer white matter hyperintensities. In the second project, I zoomed in on different types of physical activity. The novel aspect of this work was to leverage the UK Biobank’s longitudinal data collection for studying the bidirectional relationship between self-reported physical activity, vascular brain disease, and neurodegeneration. To our surprise, we found overall more consistent evidence that a healthier brain predicted higher physical activity levels (especially household activities) than for the inverse, more established relationship (higher strenuous physical activity predicted a healthier brain).
In a fair head-to-head comparison, I studied all twelve established modifiable risk factors for dementia in the same participants and their differential associations with brain pathologies and cognitive decline over time. In brief, I found that modifiable dementia risk factors were more consistently associated with vascular brain disease and especially cognitive decline than with Alzheimer’s disease pathology. This suggests that established dementia prevention targets primarily benefit vascular health and other non-Alzheimer’s pathways, underscoring the need to identify modifiable risk factors that specifically influence Alzheimer’s disease pathology. To address this question, I investigated whether sleep health may be such a modifiable risk factor for Alzheimer’s disease. I found that a weaker circadian rhythm, characterized by frequent transitions between sleep and wakefulness, such as daytime napping and nighttime awakenings, was associated with greater Alzheimer’s disease burden at follow-up. The association was significantly stronger in APOE4 carriers than non-carriers. These findings are particularly compelling because circadian rhythms are modifiable through behavioral and pharmacological interventions. Moreover, there is growing experimental evidence from animal models suggesting that disrupted sleep can accelerate amyloid-beta accumulation, providing a plausible biological mechanism for this association.
I disseminated the results of DIVERT-AD in six peer-reviewed articles including top-tier journals (Alzheimer’s & Dementia, BRAIN, JAMA Neurology, Neurology), and two oral talks at the largest international conference for Alzheimer’s Disease (AAIC).
I exploit the unprecedented wealth of brain imaging and cognitive assessments collected over the last 15 years in large population-based cohorts of middle aged and older adults. I specifically investigate which potentially modifiable risk factors (e.g. diabetes) are associated with imaging markers of Alzheimer’s disease, vascular brain disease, and neurodegeneration, which are primary cause of dementia. The resulting factors are likely important for primary prevention of dementia. I additionally determine which modifiable risk factors are associated with cognitive performance taking disease burden into account. The identified factors are likely relevant for secondary prevention. I also explore whether found associations are different for older versus younger participants, women versus men, or high versus low genetic risk carriers. Importantly, I perform this research in asymptomatic individuals without dementia, where prevention would likely be most effective. By pinpointing the most influential risk factors and the populations most vulnerable to them, my research lays the groundwork for precision prevention of dementia.