Alzheimer’s Disease in Detail

Alzheimer's disease in detail 

Topics covered:

  • Introduction
  • From discovery towards understanding
  • Clinical features
    • Cognitive impairment
      • Amnesia
      • Aphasia
      • Agnosia
      • Apraxia
      • Other cognitive impairments
    • Functional impairment
    • Neuropsychiatric symptoms
      • Mood
      • Psychosis
      • Personality
      • Other behavioural manifestations
  • Classification
    • Diagnosis
    • History
    • Examination
    • Assessment of function
    • Global assessment
    • Investigations
  • Aetiology and molecular neurobiology
    • Neuropathology
    • The cholinergic hypothesis
    • The amyloid cascade hypothesis
    • The presenilin genes
    • Tangle formation and tau phosphorylation
    • Molecular genetics
  • Treatment
    • Managing the patient
    • Managing the family
    • Managing the environment
  • Preventing AD and future treatments
  • Conclusions 
  • References

Introduction

Alzheimer's disease (AD) and other dementias incur huge costs to society, to the families of those affected, and to the individuals themselves. Costs to society include both direct costs to health and social services and indirect economic costs in terms of lost productivity, as carers are taken out of the workplace, and the economic costs to those families caring for or funding the care of their relative. Increasingly, as treatments become available, these costs are targets for change and are part of the cost–benefit analysis of new compounds, especially the largest single direct cost, that of the provision of nursing and other forms of continuing care. Apart from the financial cost to families there is the emotional impact resulting in distress and psychiatric morbidity.

As the population ages, these costs pose substantial social and economic problems. Although lifespan itself has remained static, the numbers of elderly in both developed and developing societies is increasing rapidly. In the developed world the sharpest projected growth is in the very elderly cohort—precisely the one that is at most risk of AD. Within the developing world the total number of elderly people is projected to rise substantially, reflecting to a large part better child health and nutrition. For countries in South America and Asia, with large and growing populations, the costs involved in caring for people with dementia in the future will become an increasing burden on health and social services budgets. In the absence of such services families will inevitably shoulder the main part of providing care, although the very process of development is associated with increasing urbanization and, to some degree, a diminution of the security provided by extended family structures.

From discovery towards understanding

In the early part of the twentieth century, Alois Alzheimer described his eponymous disorder in a middle-aged woman who suffered not only cognitive deterioration and functional decline but psychotic experiences, including delusions and auditory hallucinations. Neuropathology included gross atrophy and plaques and tangles on microscopy. Although all the important features of AD were described at this stage, two important developments came much later. First, in the 1960s with the studies of Roth and colleagues in Newcastle (1) and others elsewhere, it was appreciated that much dementia in the elderly has an identical neuropathological appearance to that of AD in younger people. The other development was the rediscovery that AD has a rich phenomenology. The non-cognitive symptomatology of AD is integral to the clinical manifestation of this disease, and is a major cause of carer burden and medical intervention. This second phase of research—the recognition that both the neuropathology and clinical phenomenology described by Alzheimer occur in what had previously been though of as senile dementia or, worse, just ageing, was accompanied by a growing understanding of the neurotransmitter deficits in AD. The cholinergic hypothesis provided the first glimpse of possible interventions, and remains the most important finding from this period of AD investigations. The third phase of AD research encompasses the use of molecular approaches to understanding pathogenesis. The techniques of molecular biology have been applied to understanding the formation of plaques and tangles, to a growing understanding of the genetic aetiology of much of AD, and, through the use of transgenic approaches, to developing animal and cellular models of pathogenesis.

Just as research can broadly be seen to have three phases—discovery, neuropathology, and molecular aspects—so too does the clinical response to AD. For many years cognitive impairment in the elderly was perceived as senility. As a process thought to be an inevitable consequence of ageing it was difficulty to establish medical-care models. Hence the needs of the elderly with AD were not seen as requiring specialist intervention, carers needs were not realized, and public appreciation of the impact of dementia on the elderly themselves or on the family was negligible. The change in perception of AD from ‘just ageing' to a disease was accompanied, and to some degree led, by the development of ‘old age psychiatry' as a specialism on the one hand and by the rapid growth of the Alzheimer disease societies on the other. During this second phase of AD treatment, the goals have been to ensure that the care needs of patients are met, that families' concerns are addressed, and that behavioural disturbance is minimized. The third phase of AD treatment began with the arrival of specifically designed interventions. Compounds have been introduced that were designed to ameliorate some of the deficits incurred by the disease process, and other approaches are being developed to treat those disease processes themselves.

Clinical features

Cognitive impairment

Dementia is an acquired and progressive cognitive decline in multiple areas; AD is one cause of dementia and the core clinical symptom of AD is cognitive impairment. However, as noted above, AD is clinically heterogeneous and includes diverse non-cognitive symptoms and inevitable functional impairment. Cognitive decline is manifested as amnesia, aphasia, agnosia, and apraxia (the 4As).

Amnesia

Memory loss in AD is early and inevitable. Characteristically, recent memories are lost before remote memories. However, there is considerable individual variation, with some patients able to recall specific and detailed events of childhood and others apparently having few distant memories accessible. With disease progression, even remote and emotionally charged memories are lost. The discrepancy between recent and remote memory loss suggests that the primary problem is of acquisition or retrieval of memory rather than a destruction of memory, and this is confirmed in early AD, (2) although as the disease progresses it is likely that all memory processes are impaired. Retrieval of remote memory is assumed to be preserved for longer because of rehearsal over life.

Aphasia

Language problems are found in many patients at presentation, although the language deficits in AD are not as severe as those of the frontotemporal degenerations (3) and may only be apparent on detailed examination. Word-finding difficulties (nominal dysphasia) are the earliest phenomena observed and are accompanied by circumlocutions and other responses, for example repetitions and alternative wordings. As the disorder progresses, syntax is affected and speech becomes increasingly paraphasic. Although harder to assess, receptive aphasia, or comprehension of speech, is almost certainly affected. In the final stages of the disorder, speech is grossly deteriorated with decreased fluency, preservation, echolalia, and abnormal non-speech utterances.

Agnosia

Patients with AD may have difficulty in recognizing as well as naming objects. This can have implications for care needs and safety if the unrecognized objects are important for daily functioning. One particular agnosia encountered in AD is the loss of recognition of one's own face (autoprosopagnosia). This distressing symptom is the underlying cause of perhaps the only clinical sign in AD—the mirror sign. Patients exhibiting this will interpret the face in the mirror as some other individual and respond by talking to it or by apparent fearfulness. Autoprosopagnosia can present as an apparent hallucinatory experience, until it is realized that the ‘hallucination' is fixed in both content and space, occurring only when self-reflection can be seen.

Apraxia

Difficulties with complex tasks that are not due to motor impairment become apparent in the moderate stages of AD. Typically, difficulties with dressing or tasks in the kitchen are noticed first, but these are inevitably preceded by loss of ability for more difficult tasks. Strategies to avoid such tasks are often acquired as the disease progresses, and it is only when these fail that the dyspraxia becomes apparent.

Other cognitive impairments

There appear to be no cognitive functions that are truly preserved in AD. Visuospatial difficulties commonly occur in the middle stages of the disorder and may result in topographical disorientation, wandering, and becoming lost. Difficulties with calculation, attention, and cognitive planning all occur.

Functional impairment

Although the cognitive decline in AD is the core symptom, it is the functional deterioration that has the most impact on the person themselves and it is the functional loss that necessitates most of the care needs of patients with AD, including nursing-home residency. (4) Increasingly, abilities to function in ordinary life (activities of daily living (ADLs)) are lost, starting with the most subtle and easily avoided and progressing to the most basic and essential. In general, functional abilities decline alongside cognitive abilities. However, the precise correlation between these functions is not perfect, suggesting that factors other than disease severity account for part of the variance between patients. (5) Functional abilities are related to gender; for example, cooking abilities are rehearsed more frequently in women, and home-improvement skills in men. However, the overall pattern shows some similarities between groups of patients with similar disease severity. This is exploited in the Functional Assessment Staging (FAST) scale; (6) in the original form, this is a seven-point scale of functional impairment, with stage 1 as no impairment and stage 7 as severe AD. A sequential decline is mapped by descriptions of the abilities that are lost: stage 2, difficulties with language and finding objects; stage 4, difficulties with finances; stage 6, incontinence and inability to dress or wash oneself.

ADLs are divided into those that relate to self-care and those that concern instrumental activities. Instrumental ADLs, those related to the use of objects or the outside world, are lost first and can be subtle.(7) A change in the ability to use the telephone properly or to handle finances accurately may not be apparent. Self-care ADLs include dressing and personal hygiene and are also lost gradually; for example, untidiness in clothing progresses to difficulties in dressing. Personal hygiene becomes poor as dentures are not cleaned and baths taken less often, before finally assistance is required with all self-care tasks.

Neuropsychiatric symptoms

Mood

The relationship between AD and depression is complex. Depression is a risk factor for AD, depression can be confused with dementia (pseudodementia), depression occurs as part of dementia, and cognitive impairments are found in depression. Depression occurring as a symptom of dementia will be considered here. Assessing the mood of a person with dementia is difficult for obvious reasons. However, psychomotor retardation, apathy, crying, poor appetite, disturbed sleep, and expressions of unhappiness all occur frequently. The rates of depression found in cohorts of patients with AD vary widely, reflecting changes in prevalence at different levels of severity and difficulties in the classification of symptoms suggestive of depression in those with cognitive loss. A major depressive episode is found in approximately 10 per cent of patients, minor depressive episode in 25 per cent, some features of depression in 50 per cent, and an assessment of depression by a carer in up to 85 per cent. (8,9 and 10) It is commonly believed that depression is more common in the early than in the later stages of AD, although this may reflect the difficulties of assessing depression in the more severely affected and least communicative patients. Indeed, severely affected patients in nursing homes may be particularly prone to depression. (11) Elation, disinhibition, and hypomania all occur in AD but are relatively infrequent, elevated mood being found in only 3.5 per cent of patients by Burns et al. (10)

The underlying cause of mood change in AD is not known. However, loss of serotonergic and noradrenergic markers accompanies cholinergic loss; some studies have found a greater loss of these markers at postmortem in AD patients with depression than in non-depressed patients. (12,13)

Psychosis

Psychotic symptoms occur in many patients, although, as with depression, the difficulty in determining the presence of delusions or hallucinatory experiences in the moderately to severely demented gives rise to a very wide range of frequency rates. Few studies have been able to determine the rates of psychosis in community-dwelling, fully representative samples of patients with AD. However, of those known to, largely, psychiatric services, between 10 and 50 per cent suffer from delusions and between 10 and 25 per cent experience hallucinations. (14,15 and 16) Delusions are frequently paranoid and the most common delusion is one of theft. In the context of the confusion and amnesia of dementia, it is easy to appreciate how the experience of mislaying an object becomes translated into conviction of a theft. Other patients become convinced that someone, often a family member, is trying to harm them.

Hallucinations are only somewhat less frequent than delusions—the median of one series of studies being 28 per cent. (17) Visual hallucinations are reported more commonly than auditory ones, and other modalities are rare. (16) Most studies of the non-cognitive symptomatology of AD precede the wide recognition and accepted criteria of dementia with Lewy bodies, one of the cardinal symptoms of which is visual hallucinations. (18) It is probable that a large number of those AD patients experiencing visual hallucinations reported in the studies would now be classified as having dementia with Lewy bodies.

Phenomena falling short of delusions or hallucinations, such as persecutory ideas or intrusive illusionary experiences, are common in AD as are misidentification syndromes. Capgras' syndrome may occur, but frequently the symptom is less fully evolved with the patient mistaking one person for another. Failure to recognize one's own face may be due to visuospatial difficulties or to a true misidentification syndrome—distinguishing between the two is difficult.

Various factors have been associated with psychosis in AD, but few have been substantiated in multiple studies. Burns et al.(16) found that more men than women suffered delusions of theft, although others find that psychosis occurs more often or earlier in women. An association with polymorphic variation in serotonin receptors has been reported. (19) Patients with psychotic symptoms show regional metabolic differences on functional neuroimaging. (20) The relationship between psychosis and dementia severity is not as clear cut as that between functional ability and dementia severity. Psychosis can occur at any stage of the disease process, although most studies find the maximal rate of psychosis in those with at least moderate dementia. (16,17)

Although the biological basis of psychosis within AD is not fully understood, it is probable that psychosis symptoms impact upon carers causing increased distress, (21) and that underlying psychosis accounts for much of the behavioural disturbance and aggression encountered in AD. (22)

Personality

Changes in personality are an almost inevitable concomitant of AD. Indeed, it is difficult to envisage how profound cognitive impairment resulting in the loss of recognition of loved ones, and an understanding of and ability to react with the outside world, could not result in a change in personality. Family members have described the loss of personality as a ‘living bereavement'—the person remains, but the person once known has gone. Personality change is most frequently one of loss of awareness and normal responsiveness to the environment. Individuals may become more anxious or fearful, there is a flattening of affect, and a withdrawal from challenging situations. Catastrophic reactions are short-lived emotional reactions that occur when the patient is confronted, and cannot avoid, such a challenging situation. Less commonly, personality changes may be of disinhibition with inappropriate sexual behaviours or inappropriate affect. Aggressiveness is, as noted above, often accompanied by psychosis, but it may be part of a more general personality change.

Other behavioural manifestations

Behavioural complications in AD have become a target of therapy. However, the term encompasses a wide rage of behaviours, some of which include neuropsychiatric syndromes, some caused by neuropsychiatric syndromes, and some of which have little apparent relationship to mood or to thought content. Behavioural complication is itself a largely subjective term that relies to a great extent on informer evaluation: but a behaviour may be a complication in one context, although not in another.

Behaviours exhibited in AD include wandering, changes in eating habit, altered sleep or circadian rhythms, and incontinence. These behaviours are closely linked to disease severity and occur to some extent in the majority of patients with AD. Wandering may be a manifestation of topographical confusion, a need for the toilet, or it may reflect hunger, boredom, or anxiety. Sleep is frequently disturbed, with many patients exhibiting altered sleep–wake cycles and others experiencing increased confusion towards evening (‘sundowning'). (23) A central defect in the regulation of circadian rhythms underlying these phenomena is postulated. (24) Excessive or inappropriate vocalizations (grunting and screaming) occur in the late stages.

Classification

AD is classified, as with all other disorders, by DSM-IV and by ICD-10. In addition, it also has a specialized classification system resulting from the National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer's Disease and Related Disorders Association ( NINCDS–ADRDA).(25) This clinical diagnostic system is internationally accepted and widely observed. There are other classification systems for neuropathological diagnosis, most notably the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) criteria. (26)

DSM-IV stipulates that a dementia syndrome is characterized by a decline in multiple cognitive deficits, including amnesia, resulting in impairment. A gradual onset and decline in the absence of other conditions sufficient to cause dementia indicates AD. ICD-10 shares with DSM-IV the definition of a dementia syndrome as a deterioration in more than one area of cognition, but including memory that is sufficient to impair function. Again an emphasis on insidious onset and slow decline in the absence of other disorders sufficient to cause dementia indicates AD. The NINCDS-ADRDA criteria defines possible, probable, and definite categories; the latter being restricted to neuropathological confirmation of a clinical diagnosis. (25) It is important to note that both clinical and neuropathological data are required—no single neuropathological lesion is pathognomonic of AD, and it is still uncertain how often or to what extent the neuropathological lesions of AD also occur in normal ageing. Probable AD, according to NINCDS–ADRDA, requires a dementia with progressive decline in memory and other cognitive areas, cognitive impairment established by formal testing, no disturbance of consciousness, and absence of other disorders sufficient to cause dementia. Supporting features include decline in function, change in behaviour, positive family history, and decline in specific cognitive areas including aphasia, apraxia, and agnosia. Non-specific change on electroencephalography (EEG) and progressive changes on CT are supporting, but not necessary, features. Possible AD should be diagnosed if there are variations in the clinical presentation, another disorder sufficient to cause a dementia (even if it is not thought to do so in this case), or a restricted cognitive decline.

A number of studies have attempted to determine the accuracy of diagnostic criteria against postmortem diagnosis. One of the difficulties in these studies is that because AD is the most common dementia (by some way), such studies are very likely to find a high-positive predicative value. Kukull et al. (27) found the specificity of DSM-III to be higher than NINCDS–ADRDA (0.8 versus 0.65), but NINCDS–ADRDA had a higher sensitivity (0.92 versus 0.76); others find an even lower specificity. (28)

Diagnosis

AD is the most common of the dementias, occurring in some 60 to 70 per cent of cases. However, this oft-stated figure must be treated with some caution for two reasons. First, cases that come to postmortem represent a biased sample, and the proportion of pathologically confirmed AD in community-dwelling representative samples is unknown. Second, even at postmortem the distinction between different dementias is not clear cut—many AD brains show the presence of Lewy bodies and others have considerable evidence of vascular damage. The proportion of mixed pathologies is actually rather high, between 15 and 30 per cent of all dementias.

History

Making a clinical diagnosis of AD is a positive process and not one of exclusion. The most valuable diagnostic assessment is a careful informant history, paying attention to the pattern and timing of onset and progression. In the research context, a family history interview conducted by telephone provides a degree of accuracy compatible with a full clinical assessment. (29,30) Detailed semistructured family informant diagnostic schedules are available, such as CAMDEX. (31) A history should be taken for the presence of risk factors for AD (e.g. a positive family history) and vascular and other risk factors (e.g. hypertension and head injury). Taking a family history for late-onset disorders such as AD requires special attention. Because of attrition due to other illness, many elderly people have had too few relatives reach the age of onset of dementia to make a pedigree analysis informative. The ages at death of all relatives should be established, together with cause of death and the presence or absence of dementia or memory problems in late life. The term ‘sporadic' dementia should be avoided, and is misleading when applied to an individual with a dementia where one parent died young and where no sibling reached the age of 65 to 70 years. The history should also screen for the presence of other illnesses sufficient to cause a dementia, and for systemic health in general. The presence of any significant physical illness, from chronic pain to delirium, may significantly alter cognitive abilities in the elderly, and especially so in those with AD.

A careful history should also establish the presence of any behavioural disturbance that has occurred. The relationship of aggression, wandering, agitation, or other behaviours to care tasks and other recent changes in the provision of the care package should be established. As the mainstay of the management of behavioural disturbance in all dementias is behavioural, establishing the antecedents to behaviour is an absolute prerequisite to effective management.

Examination

In addition to an examination of the mental state to establish the presence of disorders of mood and thought content, the examination will establish the specific pattern of cognitive impairment and the degree of impairment. Screening tests used to establish the presence of cognitive impairments include the Mini Mental State Examination; (32) this is a 30-point scale routinely used in all clinical trials of drugs for the treatment of AD, which is also a useful proxy measure for severity. It should be accompanied by other cognitive testing, including supplementary examination for aphasia and apraxias. Other cognitive and physical examinations will be necessary where the differential diagnosis is between a lobar dementia (e.g. frontotemporal dementia) or a subcortical dementia (e.g. that accompanying Huntington's disease).

In addition to the cognitive examination, a physical examination should be conducted in all patients with AD, although this might not be most effectively and conveniently performed at the initial assessment. Physical illness, including chronic pain, infection, cardiac insufficiency, or anaemia are all common in the elderly and can both complicate the diagnosis of AD and increase confusion in those known to have AD.

Assessment of function

Clinical assessment of function can be performed by informant history and by direct observation. The occupational therapist fulfils an invaluable role in establishing the detailed functional ability of those with AD, in addition to implementing changes in the home designed to maximize function. The FAST scale (6) is based on the premise that the pattern of decline in function is relatively uniform in AD, and hence establishes a staging of severity on function rather than cognition. As in most instances functional severity is of more relevance for the provision of services, there is much to recommend such an approach.

Global assessment

Driven largely by the United States Food and Drugs Administration, global assessment has become part of the assessment of all patients with AD in clinical trials and is finding its way into clinical practice. The underlying premise is that an assessment by a clinician, often supplemented by an informant history, provides information on severity that neither a cognitive assessment nor a functional assessment alone can provide. One scale, the Clinicians Interview of Change, has become widely used in this context and is an interesting attempt to semiformalize the routine clinical impression without operationalized criteria.

Investigations

At the initial assessment, patients with dementia should be investigated for other disorders that could complicate, exacerbate, or be confused with AD. A dementia screen might include routine biochemistry, thyroid function tests, vitamin B 12 and folate estimations, and a full blood count; many would also include syphilis serology, although the frequency of abnormal findings is low. A CT brain scan is not necessary in many cases and is not a required investigation in the NINCDS–ADRDA classification, although worsening atrophy on CT is supportive evidence. Functional scanning (single-photon emission CT (SPECT) in particular) can be useful where regional dementias are suspected, and magnetic resonance imaging can provide supportive evidence where vascular dementia is a possibility. An EEG is nearly always non-specifically abnormal even in the early stages of AD, in contrast with frontotemporal degenerations where an EEG remains unaffected at a broadly equivalent severity. This can help to distinguish the conditions, particularly where there is neuroimaging evidence of regional insufficiency.

Aetiology and molecular neurobiology

AD is the most common dementia, affecting more than 20 per cent of the population over the age of 85 years. Epidemiological evidence has suggested risk factors and putative protective factors, but the greatest advances in understanding its pathogenesis have come from the combination of molecular and epidemiological approaches.

Neuropathology

At postmortem, the brain in AD is lighter than aged-unaffected controls with more prominent sulci and a larger ventricular volume. Microscopic examination reveals the most prominent lesions described by Alzheimer—the extracellular plaque and intracellular neurofibrillary tangle. (26) No consensus has developed regarding which of these lesions is responsible for the cognitive impairment of AD. Plaques, or more precisely amyloid load, might correlate with the degree of cognitive impairment, (33) although a significant amyloid deposition is also found in normal, unimpaired, aged individuals. (34) However, there is a high degree of correlation between dementia severity and neurofibrillary tangle formation, (35) although it is possible that some of the features of AD are more stable than others; for example, extracellular neurofibrillary tangles persist after the neurone has died, whereas extracellular Lewy bodies are not found.

The plaque consists of an amyloid core surrounded by dystrophic neurites, which are themselves filled with highly phosphorylated tau protein. Studies of Down syndrome brains have suggested a temporal course to plaque formation. First, peptides derived from the amyloid precursor protein ( APP) are deposited in a diffuse plaque. (36) Over time this becomes organized as the amyloid peptides become fibrillar and form the amyloid deposit, neuritic change then occurs, and the plaque becomes fully mature.

Neurofibrillary tangles are composed of paired helical filaments, structures which are also found in the dystrophic neurites around mature plaques, and together with straight filaments, in neuropil threads. These filaments are themselves composed of the microtubule-associated protein, tau, which is present in a stably and highly phosphorylated state. (37) Tau is a neuronal-specific protein, found predominantly in the axon, that functions to stabilize microtubules, a property that is regulated by phosphorylation. Phosphorylated tau is less effective in promoting tubulin polymerization into microtubules, and in cells highly phosphorylated tau does not stabilize microtubules. (38) In normal adult brain a proportion of tau is highly phosphorylated, but this proportion is considerably greater in AD. Tau deposits are a feature of other disorders, such as progressive supranuclear palsy and some frontotemporal degenerations. Mutations have been found in frontotemporal degenerations with parkinsonism (FTDP-17), (39) and progressive supranuclear palsy has also been associated with changes in the tau gene (40) thereby emphasizing the importance of this molecule to neurodegeneration.

Braak and Braak (41) studied large numbers of brains from individuals who died at various ages and at different stages of dementia severity, which has resulted in the wide acceptance of the neuropathological staging of AD. The very earliest stages, before the clinical manifestation of dementia, are characterized by the appearance of highly phosphorylated tau in the hippocampus. In later stages, neurofibrillary tangles appear in the same brain regions and then become more widely distributed.

The cholinergic hypothesis

The pathological changes in AD are localized both structurally and functionally. Plaques and tangles first occur in the hippocampus before spreading to involve other regions. Some areas of the brain are relatively preserved—the occipital lobe is affected relatively late and the cerebellum appears to be spared from neuritic change (neurofibrillary tangles and the fully matured plaques, although diffuse amyloid deposits do occur). Functional localization was demonstrated by evidence of the relatively greater and earlier loss of cholinergic neurotransmission. At postmortem there is evidence of significantly greater neuronal loss in the cholinergic nucleus basalis of Meynert and loss of cholinergic markers. (42,43 and 44) These observations led to the cholinergic hypothesis, which stated that the cognitive impairment of AD was due to a disorder predominantly affecting cholinergic neurones. It was this hypothesis that led to the development of pharmacological strategies to rectify cholinergic loss and the introduction of the first compounds specifically designed for and efficacious in AD. However, the cholinergic hypothesis was something of a simplification as other neurotransmitter systems (e.g. serotonergic and noradrenergic) are also affected in AD.

The amyloid cascade hypothesis

In 1984, the protein deposited in blood vessels (congophilic angiopathy) in AD was shown to be a 4-kDa peptide known as b-amyloid. (45) This peptide, which is identical to the amyloid in plaques, is derived from a larger peptide, APP, the gene for which is coded on chromosome 21. After a series of misleading linkage studies, mutations in the APP gene were found in a family with autosomal dominant early-onset AD. (46) These two discoveries—the identification of b-amyloid and the discovery of mutations in the parent APP gene—led the way to the amyloid cascade hypothesis, which has remained the dominant molecular model of the disorder. (47)

Many subsequent molecular observations have been consistent with this model, which posits the formation of b-amyloid as the initiating, or at least early event, leading to all the other changes observed including tau aggregation and phosphorylation, neuronal loss, cholinergic deficits, and clinical symptoms. Perhaps the most convincing evidence that there is such a unidirectional cascade comes from the observation that mutations in the APP gene give rise to plaque formation and also to neurofibrillary tangle pathology, whereas mutations in the tau gene give rise to tangle formation but not to plaque formation in FTDP-17.

Much subsequent research has concentrated upon understanding the metabolism of APP and the formation of b-amyloid peptide. (48,49) APP is a ubiquitous single-pass cell-membrane protein expressed in many cell lines with a high degree of evolutionary conservation. At least three putative secretases cleave APP and the metabolic products can be detected in individuals unaffected by AD; the processing is not pathological in AD, but the balance between different metabolic routes may be shifted in the disease state. a-Secretase cleaves APP at the outer cell-membrane surface at a site within the b-amyloid moiety itself. Clearly, a-secretase cannot therefore yield intact b-amyloid, and this metabolic route, resulting in a secreted product, APPs, and other fragments, is termed non-amyloidogenic. a-Secretase activity is increased following stimulation of protein kinase C. (50) This might have some clinical relevance as certain neuronal receptors are coupled to protein kinase C and, indeed, muscarinic agonists do increase non-amyloidogenic metabolism. These studies predict that therapies designed to correct the cholinergic deficit in AD might have a disease modification effect. (51)

Amyloidogenic metabolism is the result of b-secretase cleaving APP beyond the amino terminus of b-amyloid and of g-secretase cleaving the resulting peptides at the carboxy terminus in the cell. The b-amyloid products vary in length, with predominant species having a length of 40 or 43 amino acids. The longer peptides are somewhat more prone to forming fibrils in vitro. It is probable that the proportion of b-amyloid (42 and 43) peptides relative to b-amyloid (40) peptides is critical in pathogenesis,(49) and that mutations in the APP gene increase these longer amyloid peptides. (52,53) Transgenic mice overexpressing the mutated APP gene also produce more b-amyloid peptide and have amyloid deposits in brain. (54) Interestingly, these animals do not develop other aspects of AD pathology.

The presenilin genes

Mutations in presenilin-1 (PS-1) and presenilin-2 (PS-2), two very similar genes on chromosome 14 and chromosome 1 respectively, also cause early-onset autosomal dominant AD.(55) The function of these genes is not fully understood, but homology with genes in flies and worms suggests that the presenilins participate in NOTCH signalling—a complex signal-transduction cascade critical, amongst other things, in determining neuronal cell fate. Mutations in the presenilin genes, and hence their role in AD pathogenesis, may result in an interference with the normal functioning of the protein or may induce a gain in a novel pathogenic function. Whatever the mechanism, it is clear that mutations in the presenilins result in an increase in the production of b-amyloid. (49) Therefore the finding of mutations in these genes adds to, rather than detracting from, the amyloid cascade hypothesis although, as with any hypothesis, the complexity of an originally simple idea increases.

Tangle formation and tau phosphorylation

Tangles are composed of paired helical filaments, themselves composed of hyperphosphorylated tau. It is not fully determined whether tau phosphorylation precedes tau aggregation, but tau is highly phosphorylated in fetal brain and, albeit to a lesser extent, in normal adult brain. (37) However, neuropathological evidence suggests that highly phosphorylated tau does begin to accumulate in the brain before the formation of tangles, and before the clinical manifestation of AD. (56) It seems as though, if not the only event in paired helical-filament formation, increased tau phosphorylation is at least an early event.

Protein phosphorylation is a product of kinase and phosphatase activity. It is likely that many such enzymes may participate in the regulation of tau phosphorylation in the brain, but two have been shown to be predominant. In cells, and in vitro, glycogen synthase kinase-3 is the main tau-kinase and protein phosphatase 2A is the predominant tau phosphatase. (57,58)

Molecular genetics

Mutations in three genes have been found to cause early-onset familial AD, which is inherited in an autosomal dominant fashion. (59) Mutations in the APP gene (on chromosome 21) are the least common, only affecting perhaps 20 families worldwide. Mutations in PS-1 (on chromosome 14) are somewhat more frequent, although are still a rare cause of AD. Mutations in PS-2 (on chromosome 1) appear to be largely restricted to an ethnic German people residing in the United States, suggesting an individual founder effect. Mutations in these genes have not been identified in true late-onset AD. Individuals with Down's syndrome are at extremely high risk of AD, with neuropathological evidence being present in virtually all individuals living to middle age, probably because of trisomy APP.

The genetic component of late-onset AD has been demonstrated by epidemiological studies, showing that a family history of dementia is the largest single risk factor for AD. (60) However many, perhaps most, patients with AD do not have a positive family history, thus giving rise to the idea of ‘sporadic' AD with a separate aetiology to ‘familial' AD. For late-onset AD this concept is outmoded and redundant. Many patients with AD do not have a family history because of attrition of family members due to death by other causes. For the cohort currently suffering from AD their parents were born in the latter part of the nineteenth century or early years of the twentieth, lived through two major world wars, and reached adulthood before the discovery of antibiotics. It is not surprising that few patients with late-onset AD have two parents and more than one sibling living to the age of onset of AD, and if one parent died young and there are no elderly siblings then the family history is non-informative. Well-designed studies examining the rate of AD in first-degree relatives by age find a cumulative incidence reaching 50 per cent or higher by the age of 90 years. (61,62)

One gene has been unequivocally associated with late-onset AD, although even this gene accounts for only something like 50 per cent of the genetic variance. (63) The apolipoprotein E gene (APOE, gene; apoE, protein) on chromosome 19 has three common alleles, coding for three protein isoforms that differ by the substitution of an amino acid at just two positions. Of the three alleles e3 is the most frequent and e2 the least; following linkage to chromosome 18 it was demonstrated that the e4 allele confers risk, whilst the e2 may be protective. (64) This finding has been replicated in a huge number of studies and in many different populations, although there are some, as yet, unexplained differences as black African-Americans apparently do not show an increased risk with the e4 allele. (65)

The mechanism of action of the APOE gene in increasing the risk of AD is not known. As APOE variation is a major genetic influence on serum cholesterol (people with the APOE e4/* genotype have higher serum cholesterol levels), it is possible that an altered lipid metabolism—either peripherally or locally—might affect the pathogenesis of AD. (66) Alternative theories arise from in vitro studies, which show a differential binding of APOE protein isoforms both to amyloid protein and to tau protein.(64) Certainly it does seem as though APOE isoforms affect neurones in culture in different ways, with apoE4 isoforms inducing shorter neurites and microtubule collapse. (67)

Other genes have been associated with AD, but none have been replicated in as many studies as APOE. It is likely that a combination of linkage and association studies using large populations will identify the other genes that influence AD, either alone or in interactions with other genes or the environment.

Treatment

For many conditions the goals of treatment or intervention are self-evident—cure, prevention of relapse, and resolution of symptoms. For AD, however, the goals of treatment can be less obvious and differ between patients and for individual patients over time. Ultimately, the quality of life of the patient should be improved, but assessing quality of life is difficult in those with dementia, and given the early loss of insight who is to judge such issues? (68) Quality of life may appear poor—patients may have diminished emotional repertoires, few pleasurable activities, and considerable handicap—but they may share none of the negative cognitions experienced by others with a similarly questionable quality of life induced by different illnesses. Other patients may appear content or happy, despite the loss of the autonomy and self-awareness normally considered an essential component of a good quality life. Equally, the treatment unit in AD includes carers, and there are times when the patient's quality of life is in conflict with the quality of life for other members of the family. Resolving such conflicts of interest and other moral and ethical issues is part of the treatment process in AD. With the arrival of specific treatments for AD and the prospect of disease-modifying therapies, an even harder question arises regarding prolonging life for those with dementia: if quality of life appears poor to observers, is it right to prolong the process, can quality of life in those with dementia truly be assessed, or should carers and families be allowed to assess for themselves the benefits and costs of treatment? (69)

There is no single model of management of patients with AD. In many countries management is the role of the gerontologist or neurologist. In others, as in the United Kingdom, the old-age psychiatry team provide the core specialist services. Many, perhaps even the majority, of those with AD are managed within primary care with the support of social services. Referral from primary care to specialist services will be according to local agreements, but most would concur that behavioural disturbance or the use of specific drugs to treat AD warrant referral to secondary care. Interventions for AD, whether provided in primary or secondary care, can be thought of as directed towards the patient, the patient's family, and the patient's environment. Guidelines on the identification and management of patients with dementia have been produced and may be a constructive approach to ensuring best clinical practice. (70,71 and 72)

Managing the patient

Management of the patient with dementia is discussed in greater detail here: The management of dementia. Management starts with the assessment and diagnosis, and perhaps the difficult dilemma is how much of the diagnosis and prognosis to discuss with the patient. (73) Most practitioners do not discuss the diagnosis with the patient themselves, although especially in the early stages a frank consultation can be beneficial. For most patients, however, cognitive impairment renders an appreciation of the diagnosis and prognosis difficult.

A large part of managing the patient is directed towards managing mood and behavioural disturbance. Accurate assessment of the disturbance is critical, and includes determining the antecedents and responses to the behaviour as well as a full description of the behaviour and any associated abnormalities in the mental state. Treatments of behavioural disturbance in AD are most often behavioural and sometimes restricted to giving information to careers. However, pharmacological interventions are an important part of the management of behavioural disturbance, even though caution regarding the use of psychotropic medications in those with dementia is necessary.

Specific treatments for AD have been developed, concentrating in clinical trials on ameliorating the core symptom of cognitive impairment. The cholinomimetic approaches are the most advanced, but other therapies are receiving extensive evaluation. Although designed for the large part as strategies to enhance cognition, these compounds also favourably appear to affect function and may reduce behavioural disturbance. Further evaluation is being conducted to determine whether there are disease-modifying effects. Drug treatments for AD are described here: Drugs for cognitive disorders.

Managing the family

Although patients may not appreciate or be able to follow a detailed discussion of the diagnosis and prognosis, their relatives, spouses, and other carers will. This is an important part of the treatment process; as the carer provides the main interventions for much of the period of the disease process, care should be taken to ensure that appropriate and sufficiently complete information is given.

Caring for a patient with AD can be difficult and stressful and some carers suffer accordingly. (74) The characteristics of both carers and patients influence the impact that this ‘burden' of caring has on the carers themselves. Men in general, and husbands in particular, seem to be less vulnerable to the adverse effects of caring, possibly because of the response seen in many male carers of rapidly and effectively recruiting outside help. (75) Women may be socialized into accepting more caring roles themselves and therefore seek less help. Non-white carers appear to suffer from less adverse consequences of caring, perhaps because of cultural differences in the perception of family bonds. (76) Patient characteristics that increase the burden of caring include behavioural disturbances, (77,78) depression, (79) and unawareness of cognitive impairment, (80) but not the cognitive impairment itself. Although the core outcome variable in clinical trials of AD drugs is the severity of cognitive impairment, it is not the variable that induces most stress in relatives nor is it the variable that predicts entry to residential care. Other variables are almost certainly protective, and caring for a loved one with dementia is not a universally negative experience. Much caring is done willingly, effectively, with love, and without complaint.

Carer support groups offer much to a person with a relative afflicted by AD. Through support groups, and especially through the national AD societies and the umbrella group—Alzheimer's Disease International—carers can obtain up-to-date and useful information regarding all aspects of AD. A support group can help individuals practically and emotionally through difficult times. Many carers talk of the support group as a life-line, although little empirical evidence exists as to the impact on carer well being.

One particular intervention for the family is that of genetic counselling. Many relatives are worried about inheriting AD. This concern might arise from two sources—the frequent discussion of genes ‘for' AD in the media and the observation of familial occurrence of AD in many individual families. For families with clinically apparent familial AD, advice, information, and where appropriate, genetic testing can be arranged through a genetics centre. Where predictive testing is contemplated for genes causing autosomal dominant, early-onset AD this will adhere to guidelines established for Huntington's disease. It is unlikely that true predictive testing will become available for late-onset AD. (81)

Managing the environment

The mainstay of interventions for AD are provided by social services. The goal of the provision of social care in people with AD is to provide an environment that is comfortable, stimulating, and, above all, safe. For most patients, and for all patients in the early stages, this means care at home, perhaps with the support of home-meal delivery and home-helps to provide shopping and cleaning assistance. Further home care may become necessary as the patient requires assistance with basic self-care tasks such as washing and dressing. The carer may require a sitting service, either for periods during the day to allow them time to themselves or in the evening to allow them to attend a carers group or for socializing. Safety issues are especially important for those with dementia living alone. There are inherent risks to the patient themselves if they wander out of the home and risks to others if the gas can be left on or fires started.

Day care is appropriate for many patients, ideally in a specialist unit. In a generic facility for elderly people those with early dementia can receive little input and those with moderate or advanced dementia can necessitate too much input from the day-centre staff. A good dementia specialist day-care facility will provide the staffing ratio appropriate to patients with a range of seventies, in addition to providing a varied programme of group and recreational facilities to maintain interest and stimulation. Day centres, where patients are arrayed around the edge of the room with a television as a focal point, are, or should be, consigned to history. Day care provides essential respite to many carers, and longer periods of occasional or regular respite can prolong the period a patient can remain in their own home.

The multidisciplinary team consisting of care workers, social services, community psychiatric nurse occupational therapist, and psychologist can maintain patients at home more effectively and for longer periods than can clinicians alone. However, long-term care becomes a necessity for many patients at some point. The costs of providing nursing-home care are huge and far outweigh the costs of providing relatively intensive community care or relatively costly drugs. If treatments were shown to reduce the total length of stay in nursing homes then this would affect the cost–benefit ratio of these compounds considerably.

Preventing AD and future treatments

A number of factors such as non-steroidal anti-inflammatory drugs, hormone replacement therapy, and the antioxidant vitamin E, might be of some use in strategies to prevent AD. Prevention could be primary before any signs of the disease or secondary after some manifestation of the process. Primary preventive measures would have to be directed at either the entire population or to groups at risk (identified by family history or genotype, for example), and therefore would have to be entirely benign and almost cost-free to be acceptable. Secondary prevention, possibly in those with memory impairments not amounting to dementia (minimal cognitive impairment), is a more realistic prospect rendering the determination of the very earliest signs of disease or evidence of a prodromal state a high priority. A biological marker for AD would have immense utility in both clinical practice and in clinical trials. Markers suggested have included platelet membrane fluidity and measurement of amyloid, apoE, and tau in cerebrospinal fluid, as well as genetic markers. (82) Of these only cerebrospinal fluid tau appears to have any possible value as a biomarker.

Tertiary prevention or disease modification refers to treatments to arrest or slow down the disease process after it has become clinically evident. Some evidence exists that drugs already available might have a disease-modifying effect, and other compounds designed to reduce amyloid formation or aggregation or tau phosphorylation are in development. Other approaches have been developed to reduce the inflammatory component of pathogenesis, or to enhance function and provide support for the remaining neurones using nerve growth factor. This latter promising approach is made problematic by the fact that oral or parenteral administration of a peptide would result in its rapid degradation.

Given that AD is a chronically deteriorating condition, determining efficacy of disease modification is difficult. Two approaches have been suggested. (83) The randomized start trial assigns patients to drug or placebo at random, and at some predetermined point those on placebo are switched to treatment. If the treatment is symptomatic only it would be expected that on switching to treatment these individuals would ‘catch-up' with those treated from the outset. However, if the treatment has slowed the disease, those treated from the outset would remain relatively improved compared to the switched group. The randomized withdrawal trial is a reverse of this, with patients withdrawn from active treatment failing to fall to the placebo group results if the compound had affected the disease process.

Conclusions

For the foreseeable future, AD will remain a disorder afflicting a large proportion of the world's elderly. The impact on developing countries especially will be considerable. Care for these patients will continue to be provided from many sources, with specialist services being necessary to compliment primary and generic services, particularly for those patients exhibiting the complex psychiatric phenomenology described by Alzheimer and for those patients where specific drugs are indicated. As the molecular pathogenesis of AD is increasingly understood it is to be hoped that this is translated into treatments ever more effective in modifying or preventing the disease process itself.

References

1. Tomlinson, B.E., Blessed, G., and Roth, M. (1970). Observation on the brains of demented old people. Journal of Neurological Science, 11, 205–42.

2. Petersen, R.C., Smith, G.E., Ivnik, R.J., Kokmen, E., and Tangalos, E.G. (1994). Memory function in very early Alzheimer's disease. Neurology, 44, 867–72.

3. Neary, D., Snowden, J.S., Northen, B., and Goulding, P. (1988). Dementia of frontal lobe type. Journal of Neurology, Neurosurgery and Psychiatry, 51, 353–61.

4. Riter, R.N. and Fries, B.E. (1992). Predictors of the placement of cognitively impaired residents on special care units. Gerontologist, 32, 184–90.

5. Reed, B.R., Jagust, W.J., and Seab, J.P. (1989). Mental status as a predictor of daily function in progressive dementia. Gerontologist, 29, 804–7.

6. Reisberg, B. (1988). Functional assessment staging (FAST). Psychopharmacology Bulletin, 24, 653–9.

7. Green, C.R., Mohs, R.C., Schmeidler, J., Aryan, M., and Davis, K.L. (1993). Functional decline in Alzheimer's disease: a longitudinal study. Journal of the American Geriatrics Society, 41, 654–61.

8. Levy, M.L., Cummings, J.L., Fairbanks, L.A., Bravi, D., Calvani, M., and Carta, A. (1996). Longitudinal assessment of symptoms of depression, agitation, and psychosis in 181 patients with Alzheimer's disease. American Journal of Psychiatry, 153, 1438–43.

9. Rovner, B.W., Broadhead, J., Spencer, M., Carson, K., and Folstein, M.F. (1989). Depression and Alzheimer's disease. American Journal of Psychiatry, 146, 350–3.

10. Burns, A., Jacoby, R., and Levy, R. (1990). Psychiatric phenomena in Alzheimer's disease. III: Disorders of mood. British Journal of Psychiatry, 157, 81–6.

11. Rovner, B.W., German, P.S., Broadhead, J., et al. (1990). The prevalence and management of dementia and other psychiatric disorders in nursing homes. International Psychogeriatrics, 2, 13–24.

12. Zubenko, G.S., Moossy, J., and Kopp, U. (1990). Neurochemical correlates of major depression in primary dementia. Archives of Neurology, 47, 209–14.

13. Förstl, H., Burns, A., Luthert, P., Cairns, N., Lantos, P., and Levy, R. (1992). Clinical and neuropathological correlates of depression in Alzheimer's disease. Psychological Medicine, 22, 877–84.

14. Deutsch, L.H., Bylsma, F.W., Rovner, B.W., Steele, C., and Folstein, M.F. (1991). Psychosis and physical aggression in probable Alzheimer's disease. American Journal of Psychiatry, 148, 1159–63.

15. Drevets, W.C. and Rubin, E.H. (1989). Psychotic symptoms and the longitudinal course of senile dementia of the Alzheimer type. Biological Psychiatry, 25, 39–48.

16. Burns, A., Jacoby, R., and Levy, R. (1990). Psychiatric phenomena in Alzheimer's disease. I: Disorders of thought content. British Journal of Psychiatry, 157, 72–6.

17. Wragg, R.E. and Jeste, D.V. (1989). Overview of depression and psychosis in Alzheimer's disease. American Journal of Psychiatry, 146, 577–87.

18. McKeith, I.G., Galasko, D., Kosaka, K., et al. (1996). Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology, 47, 1113–24.

19. Holmes, C., Arranz, M.J., Powell, J.F., Collier, D.A., and Lovestone, S. (1998). 5-HT 2A and 5-HT2C receptor polymorphisms and psychopathology in late onset Alzheimer's disease. Human Molecular Genetics, 7, 1507–9.

20. Kotrla, K.J., Chacko, R.C., Harper, R.G., Jhingran, S., and Doody, R. (1995). SPECT findings on psychosis in Alzheimer's disease. American Journal of Psychiatry, 152, 1470–5.

21. Victoroff, J., Mack, W.J., and Nielson, K.A. (1998). Psychiatric complications of dementia: impact on caregivers. Dementia, 9, 50–5.

22. Gormley, N., Rizwan, M.R., and Lovestone, S. (1998). Clinical predictors of aggressive behaviour in Alzheimer's disease. International Journal of Geriatric Psychiatry, 13, 109–15.

23. Evans, L.K. (1987). Sundown syndrome in institutionalized elderly. Journal of the American Geriatrics Society, 35, 101–8.

24. Satlin, A., Volicer, L., Stopa, E.G., and Harper, D. (1995). Circadian locomotor activity and core-body temperature rhythms in Alzheimer's disease. Neurobiology of Aging, 16, 765–71.

25. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., and Stadlan, E.M. (1984). Clinical diagnosis of Alzheimer's disease: report of the NINCDS–ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology, 34, 939–44.

26. Gearing, M., Mirra, S.S., Hedreen, J.C., Sumi, S.M., Hansen, L.A., and Heyman, A. (1995). The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part X. Neuropathology confirmation of the clinical diagnosis of Alzheimer's disease. Neurology, 45, 461–6.

27. Kukull, W.A., Larson, E.B., Reifler, B.V., Lampe, T.H., Yerby, M.S., and Hughes, J.P. (1990). The validity of 3 clinical diagnostic criteria for Alzheimer's disease. Neurology, 40, 1364–9.

28. Nagy, Z., Esiri, M.M., Hindley, N.J., et al. (1998). Accuracy of clinical operational diagnostic criteria for Alzheimer's disease in relation to different pathological diagnostic protocols. Dementia, 9, 219–26.

29. Shimomura, T. and Mori, E. (1998). Obstinate imitation behaviour in differentiation of frontotemporal dementia from Alzheimer's disease. Lancet, 352, 623–4.

30. Devi, G., Marder, K., Schofield, P.W., Tang, M.X., Stern, Y., and Mayeux, R. (1998). Validity of family history for the diagnosis of dementia among siblings of patients with late-onset Alzheimer's disease. Genetic Epidemiology, 15, 215–23.

31. Roth, M., Tym, E., Mountjoy, C.Q., et al. (1986). CAMDEX. A standardised instrument for the diagnosis of mental disorder in the elderly with special reference to the early detection of dementia. British Journal of Psychiatry, 149, 698–709.

32. Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). Mini-Mental State, a practical method of grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–98.

33. Cummings, B.J. and Cotman, C.W. (1995). Image analysis of b-amyloid load in Alzheimer's disease and relation to dementia severity. Lancet, 346, 1524–8.

34. Haroutunian, V., Perl, D.P., Purohit, D.P., et al. (1998). Regional distribution of neuritic plaques in the nondemented elderly and subjects with very mild Alzheimer disease. Archives of Neurology, 55, 1185–91.

35. Nagy, Z., Esiri, M.M., Jobst, K.A., et al. (1995). Relative roles of plaques and tangles in the dementia of Alzheimer's disease: correlations using three sets of neuropathological criteria. Dementia, 6, 21–31.

36. Mann, D.M., Brown, A., Prinja, D., et al. (1989). An analysis of the morphology of senile plaques in Down's syndrome patients of different ages using immunocytochemical and lectin histochemical techniques. Neuropathology and Applied Neurobiology, 15, 317–29.

37. Lovestone, S. and Reynolds, C.H. (1997). The phosphorylation of tau: a critical stage in neurodevelopmental and neurodegenerative processes. Neuroscience, 78, 309–24.

38. Lovestone, S., Hartley, C.L., Pearce, J., and Anderton, B.H. (1996). Phosphorylation of tau by glycogen synthase kinase-3b in intact mammalian cells: the effects on organisation and stability of microtubules. Neuroscience, 73, 1145–57.

39. Spillantini, M.G., Murrell, J.R., Goedert, M., Farlow, M.R., Klug, A., and Ghetti, B. (1998). Mutation in the tau gene in familial multiple system tauopathy with presenile dementia. Proceedings of the National Academy of Sciences of the United States of America, 95, 7737–41.

40. Higgins, J.J., Litvan, I., Pho, L.T., Li, W., and Nee, L.E. (1998). Progressive supranuclear gaze palsy is in linkage disequilibrium with the t and not the a-synuclein gene. Neurology, 50, 270–3.

41. Braak, H. and Braak, E. (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathologica (Berlin), 82, 239–59.

42. Davies, P. and Maloney, A.J.F. (1976). Selective loss of central cholinergic neurones in Alzheimer's disease. Lancet, 2, 1403–6.

43. Wilcock, G.K., Esiri, M.M., Bowen D.M., and Smith, C.C. (1983). The nucleus basalis in Alzheimer's disease: cell counts and cortical biochemistry. Neuropathology and Applied Neurobiology, 9, 175–9.

44. Francis, P.T., Palmer, A.M., Sims, N.R., et al. (1985). Neurochemical studies of early-onset Alzheimer's disease. Possible influence on treatment. New England Journal of Medicine, 313, 7–11.

45. Glenner, G.G. and Wong, C.W. (1984). Alzheimer's disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochemical and Biophysical Research Communications, 120, 885–90.

46. Chartier Harlin, M.C., Crawford, F., Houlden, H., et al. (1991). Early-onset Alzheimer's disease caused by mutations at codon 717 of the beta-amyloid precursor protein gene. Nature, 353, 844–6.

47. Hardy, J.A. and Higgins, G.A. (1992). Alzheimer's disease: the amyloid cascade hypothesis. Science, 256, 184–5.

48. Selkoe, D.J. (1994). Alzheimer's disease: a central role for amyloid. Journal of Neuropathology and Experimental Neurology, 53, 438–47.

49. Hardy, J. (1997). Amyloid, the presenilins and Alzheimer's disease. Trends in Neuroscience, 20, 154–9.

50. Hung, A.Y., Haass, C., Nitsch, R.M., et al. (1993). Activation of protein kinase C inhibits cellular production of the amyloid b-protein. Journal of Biological Chemistry, 268, 22 959–62.

51. Lovestone, S. (1997). Muscarinic therapies in Alzheimer's disease: from palliative therapies to disease modification. International Journal of Psychiatry in Clinical Practice, 1, 15–20.

52. Kosaka, T., Imagawa, M., Seki, K., et al. (1997). The bAPP717 Alzheimer mutation increases the percentage of plasma amyloid-b protein ending at Ab42 (43). Neurology, 48, 741–5.

53. Citron, M., Vigo-Pelfrey, C., Teplow, D.B., et al. (1994). Excessive production of amyloid b-protein by peripheral cells of symptomatic and presymptomatic patients carrying the Swedish familial Alzheimer disease mutation. Proceedings of the National Academy of Sciences of the United States of America, 91, 11993–7.

54. Price, D.L. and Sisodia, S.S. (1994). Cellular and molecular biology of Alzheimer's disease and animal models. Annual Review of Medicine, 45, 435–46.

55. Da Silva, H.A.R. and Patel, A.J. (1997). Presenilins and early-onset familial Alzheimer's disease. Neuroreport, 8, 1–12.

56. Braak, E., Braak, H., and Mandelkow, E.-M. (1994). A sequence of cytoskeleton changes related to the formation of neurofibrillary tangles and neuropil threads. Acta Neuropathologica (Berlin), 87, 554–67.

57. Lovestone, S., Reynolds, C.H., Latimer, D., et al. (1994). Alzheimer's disease-like phosphorylation of the microtubule-associated protein tau by glycogen synthase kinase-3 in transfected mammalian cells. Current Biology, 4, 1077–86.

58. Goedert, M. (1996). Tau protein and the neurofibrillary pathology of Alzheimer's disease. Annals of the New York Academy of Sciences, 777, 121–31.

59. Blacker, D. and Tanzi, R.E. (1998). The genetics of Alzheimer disease—current status and future prospects. Archives of Neurology, 55, 294–6.

60. van Duijn, C.M., Clayton, D.G., Chandra, V., et al. (1994). Interaction between genetic and environmental risk factors for Alzheimer's disease: a reanalysis of case–control studies. Genetic Epidemiology, 11, 539–51.

61. Korten, A.E., Jorm, A.F., Henderson, A.S., Broe, G.A., Creasey, H., and McCusker, E. (1993). Assessing the risk of Alzheimer's disease in first-degree relatives of Alzheimer's disease cases. Psychological Medicine, 23, 915–23.

62. Huff, F.J., Auerbach, J., Chakravarti, A., and Boller, F. (1988). Risk of dementia in relatives of patients with Alzheimer's disease. Neurology, 38, 786–90.

63. Owen, M., Liddell, M., and McGuffin, P. (1994). Alzheimer's disease. British Medical Journal, 308, 672–3.

64. Higgins, G.A., Large, C.H., Rupniak, H.T., and Barnes, J.C. (1997). Apolipoprotein E and Alzheimer's disease: a review of recent studies. Pharmacology, Biochemistry and Behavior, 56, 675–85.

65. Maestre, G., Ottman, R., Stern, Y., et al. (1995). Apolipoprotein E and Alzheimer's disease: ethnic variation in genotypic risks. Annals of Neurology, 37, 254–9.

66. Sparks, D.L. (1997). Coronary artery disease, hypertension, ApoE, and cholesterol: a link to Alzheimer's disease. Annals of the New York Academy of Sciences, 826, 128–46.

67. Nathan, B.P., Chang, K.C., Bellosta, S., et al. (1995). The inhibitory effect of apolipoprotein E4 on neurite outgrowth is associated with microtubule depolymerization. Journal of Biological Chemistry, 270, 19 791–9.

68. Howard, K. and Rockwood, K. (1995). Quality of life in Alzheimer's disease. Dementia, 6, 113–16.

69. Hollister, L. and Gruber, N. (1996). Drug treatment of Alzheimer's disease. Effects on caregiver burden and patient quality of life. Drugs and Aging, 8, 47–55.

70. Lovestone, S., Graham, N., and Howard, R. (1997). Guidelines on drug treatments for Alzheimer's disease. Lancet, 350, 232–3.

71. Whitehouse, P.J. and Reidenbach, F. (1997). Guidelines for early identification of Alzheimer disease. Alzheimer Disease and Associated Disorders, 11, 61–2.

72. Fisk, J.D., Sadovnick, A.D., Cohen, C.A., et al. (1998). Ethical guidelines of the Alzheimer Society of Canada. Canadian Journal of Neurological Science, 25, 242–8.

73. Meyers, B.S. (1997). Telling patients they have Alzheimer's disease—important for planning their future, and no evidence of ill effects. British Medical Journal, 314, 321–2.

74. Dunkin, J.J. and Anderson Hanley, C. (1998). Dementia caregiver burden: a review of the literature and guidelines for assessment and intervention. Neurology, 51, S53–60.

75. Bedard, M., Molloy, D.W., Pedlar, D., Lever, J.A., and Stones, M.J. (1997). Associations between dysfunctional behaviors, gender, and burden in spousal caregivers of cognitively impaired older adults. International Psychogeriatrics, 9, 277–90.

76. Connell, C.M. and Gibson, G.D. (1997). Racial, ethnic, and cultural differences in dementia caregiving: review and analysis. Gerontologist, 37, 355–64.

77. Coen, R.F., Swanwick, G.R., O'Boyle, C.A., and Coakley, D. (1997). Behaviour disturbance and other predictors of carer burden in Alzheimer's disease. International Journal of Geriatric Psychiatry, 12, 331–6.

78. Donaldson, C., Tarrier, N., and Burns, A. (1998). Determinants of carer stress in Alzheimer's disease. International Journal of Geriatric Psychiatry, 13, 248–56.

79. Teri, L. (1997). Behavior and caregiver burden: behavioral problems in patients with Alzheimer disease and its association with caregiver distress. Alzheimer Disease and Associated Disorders, 11 (Supplement 4), S35–8.

80. Seltzer, B., Vasterling, J.J., Yoder, J.A., and Thompson, K.A. (1997). Awareness of deficit in Alzheimer's disease: relation to caregiver burden. Gerontologist, 37, 20–4.