Global Clinical Trials for Alzheimer's Disease: Design, Implementation, and Standardization

I

Alzheimer’s Disease: The Twenty-first Century Epidemic

Outline

Chapter 1 The Epidemiology and Prevention of Alzheimer’s Disease and Projected Burden of Disease

Chapter 2 Developing a National Plan to Address Alzheimer’s Disease

Chapter 3 Preventing Alzheimer’s Disease

Chapter 1

The Epidemiology and Prevention of Alzheimer’s Disease and Projected Burden of Disease

Tiia Ngandu¹, ², Francesca Mangialasche³ and Miia Kivipelto¹, ², ³, ⁴,    ¹Department of Chronic Disease Prevention, National Institute for Health and Welfare, Helsinki, Finland,    ²Alzheimer’s Disease Research Center, Karolinska Institutet-Stockholm University, Stockholm, Sweden,    ³Aging Research Center, Karolinska Institutet-Stockholm University, Stockholm, Sweden,    ⁴Department of Neurology, University of Eastern Finland, Kuopio, Finland

Today there are more than 36 million demented persons worldwide. With the aging of populations, the prevalence of dementia and Alzheimer’s disease (AD) is expected to more than triple in the coming 40 years. Besides high age and genetic factors, evidence from observational studies shows that many modifiable vascular and lifestyle-related risk factors at midlife affect the risk of the disease. Control of vascular and metabolic risk factors, healthy diet, and maintenance of AN active lifestyle (physical, mental, and social components) may thus be the key issues on the road to prevent or delay the onset of AD. As the pathologic processes leading to AD may take decades to develop, it is crucial to identify optimal time windows for preventive efforts. Targeting several risk factors simultaneously and long duration of the intervention may be needed for the optimal preventive effect. As many persons are affected, relatively small effects of an integrative intervention on common risk factors may have a huge public health impact. Usage of biomarkers may help in identifying different risk phenotypes and in monitoring effects of the interventions. International collaboration is necessary to initiate future large-scale dementia prevention studies that are needed to formulate evidence-based preventive measures in cognitive decline and dementia.

Keywords

epidemiology; prevalence; incidence; risk factor; prevention

1.1 Occurrence of Dementia and Alzheimer’s Disease

The occurrence of dementia, and its major subtype, Alzheimer’s disease (AD), is strongly influenced by age. The prevalence of dementia nearly doubles every five years after the age of 60. The World Alzheimer Report 2009 by Alzheimer’s Disease International (ADI) estimated that currently the age-standardized prevalence for those aged 60 or older is 5–7%. Among persons aged 60–64 years the prevalence is 0.7–1.8%, but among those aged over 90 years, the prevalence is 29–64% [1,2]. Several other meta-analyses have reported fairly similar prevalence estimates even if methods of individual studies have not been identical [3,4]. With the aging of populations, the prevalence of dementia and AD is increasing rapidly. It is estimated that in 2010 there were 35.6 million demented people worldwide. This number is expected to increase to 115.4 million by 2050 if efforts to prevent the disease fail [2]. The greatest increase is expected in less developed countries, in which the annual growth rate of older adults (aged 60+) is higher than in more developed countries. Indeed, by 2050, nearly 80% of the world’s older population is expected to live in less developed countries, making dementia a global challenge [5].

The incidence of dementia and AD increases almost exponentially with age [6] (Figure 1.1). The pooled incidence from European studies among people aged 65 or above was 19.4 per 1,000 person-years [6]. The total number of new cases of dementia each year worldwide is nearly 7.7 million [7]. There are indications that the incidence rate might plateau or even decline among very old people [8]. A recent report also proposed that there may be a decline in the dementia incidence over time, but this remains to be confirmed [9].

Figure 1.1 Incidence of dementia by age per 1000 person years [90].

AD is the most important cause of dementia. It accounts for between 50 to 70% of all dementia cases [6,10]. The majority of AD cases are sporadic, while the familial forms caused by autosomal dominant mutations are very rare. There is increasing evidence from population-based neuropathological and neuroimaging studies showing that mixed brain pathologies (AD, vascular) account for most dementia cases, especially among very old people [11]. This has relevant implications for the planning of preventive and therapeutic measures, since in the general population 70% of all dementia cases occur in people aged 75 or older [6].

1.2 Risk Factors for Dementia and AD

AD is a multi-factorial syndrome resulting from genetic–environmental interactions. Advanced age, familial aggregation, and the susceptibility gene apolipoprotein E (ApoE) ε4 allele have long been recognized as significant risk factors for AD. Mutations in genes APP, PSEN1, or PSEN2 have been identified behind the familial, early-onset form of AD, but the sporadic form differs from the familial type and presents considerable heterogeneity in terms of risk factor profiles, pathogenesis, and neuropathological findings. In recent years, several possibly modifiable risk and protective factors have been examined, and many vascular and lifestyle-related factors have been linked to the disease [3]. Most of the current evidence is based on observational studies. The lessons learned from the few randomized controlled trials are discussed later on in section 1.3.

The pathologic processes leading to AD probably start decades before the first symptoms are manifested [12]. The life-course approach is important: The role of different risk factors may vary in different periods of life. This stresses the importance of investigating the risk factors in long-term cohort studies with preferably both the exposure and outcome information collected several times during the lifespan. This would also enable us to find optimal time windows for preventive interventions.

Table 1.1 presents a summarized overview of these main putative risk or protective factors, which are discussed in this section. The accumulated evidence on risk or protective factors for dementia and AD has also been summarized in a recent report by the National Institutes of Health (NIH) [13]. The report indicated that there is still a need to refine knowledge about modifiable risk factors for dementia and AD, and in particular multidimensional intervention studies are needed.

Table 1.1

The Main Risk and Protective Factors for Dementia and AD

ApoE: apolipoprotein E. CR1: complement component receptor 1. PICALM: phosphatidylinositol binding clathrin assembly protein. CLU: clusterin. TOMM40: Translocase of Outer Mitochondrial Membrane 40 homolog (www.alzgene.org). ELF-EMFs: extremely-low-frequency electromagnetic fields. HRT: hormone replacement therapy. NSAIDs: non-steroidal anti-inflammatory drugs. SES: socio-economic status.

1.2.1 Genetic, Socio-demographic, and Socio-economic Risk Factors

Age and family history were the first established risk factors for sporadic dementia and AD. Those individuals who have at least one first-degree relative with dementia are at an increased risk of developing dementia compared with persons without a family history [14]. Thus far ApoE ε4 allele has been the only established susceptibility gene for sporadic or late-onset AD. It has been estimated that it accounts for about 60% of the genetic component of late-onset AD [15]. With the era of genome-wide association studies, new candidate risk genes have been identified, but the associations are less consistent [16]. These are mainly genes involved in the metabolism and processing of the amyloid precursor protein (APP) and β-amyloid, as well as tau protein, including the GSK3β, DYRK1A, Tau, and CLU genes [17,18]. Until now, mutations in APP have not been implicated in the late-onset form of AD, with the exception of the rare variant, N660Y, which was recently identified in one case from a late-onset AD family [19]. Another recent study identified a mutation in the APP gene that can be protective against AD and age-related cognitive decline. This mutation is associated with a reduced production of amyloidogenic peptides [20].

Other genes that have been associated with increased risk of AD are TOMM40, CR1 and PICALM. The TOMM40 gene is located in a region of chromosome 19, which is in linkage disequilibrium with ApoE, and its polymorphism affects the age of onset of AD in subjects with an ApoE genotype [21]. CR1 is involved in the complement cascade, while PICALM encodes a protein that is involved in clathrin-mediated endocytosis, an essential step in the intracellular trafficking of proteins and lipids such as nutrients, growth factors, and neurotransmitters [22].

Sex may also play a role in the risk of AD; especially among the older age groups, women have higher rates of AD compared with men [6]. Somewhat higher rates of dementia have been reported among African American and Latino populations than in Caucasians [23], but the role ethnic background plays is still controversial. Higher educational level and socio-economic status are associated with a lower risk of dementia and AD [24–26]. One of the main hypotheses behind this is that persons with higher education may have greater cognitive reserve capacity that postpones the clinical manifestation of the disease.

1.2.2 Vascular Risk Factors

Since the late 1990s, when the first prospective cohort studies became available, the research on vascular and lifestyle-related risk factors of dementia and AD has been advancing rapidly. The hope of identifying modifiable risk factors that could be used as targets for preventive efforts is the key driving force behind this line of research.

Several population-based studies have demonstrated that elevated blood pressure in midlife can increase the risk of dementia and AD in late-life [27–29]. However, a decline in blood pressure is seen in the years preceding the diagnosis of dementia [30–32]. A similar association has been detected between serum cholesterol values and dementia/AD risk, i.e., high cholesterol in midlife has been shown to increase the risk of dementia and AD [27,29,33], but a decline in cholesterol after midlife is observed in those individuals who will develop dementia/AD in late-life [34]. It has been speculated that the reductions in blood pressure and cholesterol levels may reflect the ongoing neurodegenerative disease process in the brain, and it may even represent a risk marker for late-life cognitive impairment. However, the relationships between these factors are complex and may also partly reflect phenomena related to physiological aging and also changes in lifestyle (such as diet, physical activity, or smoking). The role of these vascular risk factors seems to be different in different time points in life. Therefore it is crucial to take into account the life course perspective when investigating these risk factors and interpreting the results.

There is growing evidence that type 2 diabetes mellitus/impaired glucose tolerance is associated with an increased risk of cognitive impairment and AD, even though not all research confirms this relationship [35,36]. There are many mechanisms through which diabetes may influence the risk of dementia, including hyperinsulinemia, cerebrovascular disease, increased oxidative stress, and generation of advanced glycosylated end (AGE) products [37]. Observational studies have also suggested that persons using antihypertensive drugs [38], or statins [39], may have a decreased risk of developing dementia and AD. Until recently, clinical vascular diseases were seldom studied in relation to AD. However, recent studies have proposed an association between various vascular disorders like myocardial infarction, atrial fibrillation, heart failure and AD [40,41].

1.2.3 Lifestyle-related Risk Factors

While earlier reports on smoking and AD showed negative results, the evidence from prospective cohorts clearly shows that smoking is associated with the increased risk of developing dementia and AD, and the risk may increase in a dose-dependent manner [42–44]. Nowadays, obesity is one of the major public health problems in Western countries [45]. Obesity, especially in midlife [46–48] but also in late-life [49], has been linked to an increased risk of dementia and AD in several studies. A diet rich in saturated fats may increase the risk of dementia and AD irrespective of the BMI value, whereas moderate intake of unsaturated fats may be protective [50,51]. A Mediterranean type of diet may, on the other hand, be protective [52]. Moderate alcohol drinking has also been shown to be protective of dementia, while frequent alcohol drinking may predispose one to the disease [53,54]. Regular physical activity [55,56] as well as a rich social network [57], being married [58], and social and mental activities [59] have all been associated with a reduction of the risk of dementia and AD.

An interaction between genetic susceptibility and lifestyles has been observed: Persons with ApoE e4 and an unhealthy lifestyle (smoking, frequent alcohol drinking, lack of physical activity, or a diet with an abundant intake of saturated fatty acids or little unsaturated fatty acids) may be especially vulnerable to developing dementia and AD [60]. Further investigation of gene–environment interactions both in epidemiologic and experimental settings are needed to increase our understanding of the disease process and for planning interventions.

Vascular and lifestyle-related risk factors and diseases often occur simultaneously, and clustering of vascular risk factors has been shown to additively increase the risk of dementia [29,46,61,62]. This has led to the development of risk scores to assess the risk of cognitive impairment or dementia on an individual level. These practical tools are used for identifying individuals who may benefit from intensive lifestyle consultations and pharmacological interventions, and for health education. The first such tool was based on data from the Cardiovascular Risk Factors, Aging and Dementia (CAIDE) study, and this dementia risk score includes easily measurable variables (age, sex, education, midlife hypertension, hypercholesterolemia, obesity, and physical inactivity) that are associated with the risk of dementia later in life [63]. The risk score instruments can be further improved as more comprehensive datasets become available (e.g., by adding new variables and age groups), but initially they are used as educational tools to demonstrate the role of modifiable vascular factors in dementia syndrome.

1.3 Interventions to Prevent the Disease

The multi-factorial and heterogeneous character of late-onset AD allows multiple prevention approaches (Figure 1.2). The intervention studies conducted so far have had somewhat disappointing results. There have been trials for cholinesterase inhibitors (donepezil, galantamine, rivastigmine), vitamin E supplements, vitamin B6, B12 or folic acid supplements, vitamin C, beta carotene or multivitamin supplements, omega-3 fatty acid, gingko biloba, statins, antihypertensive drugs, acetylsalicylic acid, and gonadal steroids that have shown no effect; some trials for estrogen plus progesterone and non-steroidal anti-inflammatory drugs (rofecoxib and naproxen) have even shown an increased risk of dementia/AD in the treatment group [13]. These trials have mostly used a single agent and they have been conducted on older and/or already cognitively impaired populations. Many of these trials were planned for other outcomes and cognitive outcomes were secondary. Therefore, they may have been underpowered to detect true associations.

Figure 1.2 Risk and protective factors for Alzheimer’s disease

There are, however, some positive signs that antihypertensive drug treatment, vitamin B supplementation, physical activity, and cognitive training may be beneficial, at least in certain groups of persons. The positive effect of antihypertensives has been suggested in a few clinical trials: In the Systolic Hypertension in Europe (Syst-Eur) trial, active treatment of isolated systolic hypertension with nitrendipine, a calcium-channel blocker, was found to halve the incidence of AD [64]. The Study on Cognition and Prognosis in the Elderly (SCOPE) showed, in sub-group analyses among the subjects with lower baseline cognition, that Mini Mental State Examination (MMSE) score declined less in the candesartan than in the control group [65]. The Perindopril Protection Against Recurrent Stroke Study (PROGRESS) among subjects with a history of stroke or transient ischemic attack showed a decreased risk of decline in MMSE among persons with active treatment with perindopril [66]. Pooled analyses based on the Hypertension in the Very Elderly Trial (HYVET-COG), Systolic Hypertension in the Elderly Program (SHEP), Syst-Eur, and PROGRESS suggested that antihypertensive treatment could reduce the risk of dementia by up to 13% [67].

Single trials on other agents/interventions have also signaled possible beneficial effects, although none of them has had clinical dementia as an outcome. The VITACOG study reported that two-year supplementation with vitamin B12 plus B6 plus folic acid may slow cognitive and clinical decline in people with mild cognitive impairment (MCI) and elevated homocysteine level [68]. The Fitness for the Aging Brain Study (FABS) of adults with subjective memory impairment, proposed that a six-month home-based physical activity program provided a modest improvement in cognition measured with the cognitive section of the Alzheimer Disease Assessment Scale (ADAS-Cog) [69]. The Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) study reported that 10 sessions of cognitive training over six weeks resulted in improved cognitive abilities specific to the abilities trained that continued for five years after the initiation of the intervention [70].

Some key issues that need to be considered in all AD prevention trials are:

 Intervention: Given the diversity of risk and protective factors identified in observational studies, it may be more effective to have multi-domain interventions targeting several of these risk factors simultaneously, instead of just a single component intervention.

 Timing and target group: As the pathologic disease process takes years, even decades, the trials starting only a short time before the manifestation of clinical dementia syndrome may be too late to have results. Also, different risk factors and interventions may have their own optimal time windows, and initiating the treatment in the wrong age group may even lead to an increase in risk as was seen in trials with non-steroidal anti-inflammatory drugs or hormone replacement therapy. On the other hand, long-term interventions are challenging as they need considerable financial resources. Further, drop-out rates may increase as the duration of the intervention is prolonged. The participants that drop out may have a higher rate of cognitive decline, and this would bias the results away from the null effect, especially if the intervention requires much more active participation than the control condition.

 Outcome measures: Due to a very long preclinical phase, it is difficult to determine the exact time of onset of AD. Lately, there has been a shift toward earlier diagnosis of AD [71]. When the symptoms reach the clinical threshold for dementia, the neurodegenerative processes are at a very advanced stage. Cognitive decline may be a better endpoint than conversion to dementia. The appropriate cognitive test battery should be sufficiently sensitive to detect small changes, but at the same time specific to changes that are closely related to the development of AD. The optimal test battery may also depend on the characteristics of the population.

 Ethical issues: Persons with high blood pressure or cholesterol levels need to be treated because of the risk of developing cardiovascular and cerebrovascular disorders, and thus true placebo-controlled trials in regards to this issue may no longer be possible to conduct. The same is true for unhealthy lifestyle factors that need to be addressed. This treatment contamination would decrease differences between groups.

Intervention studies combining several different approaches have not been conducted for AD so far, but at least three large European studies are ongoing. The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) is a multicenter, randomized, controlled trial (RCT) with 1,260 individuals aged 60–77 who are at risk of cognitive decline [72]. The two-year multi-domain intervention consists of nutritional guidance, exercise, cognitive training, social activity, and management of metabolic and vascular risk factors. The primary outcome is cognitive performance as measured by the modified Neuropsychological Test Battery, Stroop test, and Trail Making Test. Main secondary outcomes are dementia, disability, depressive symptoms, vascular risk factors and outcomes, quality of life, and neuroimaging measures. The intervention will be completed during