Alzheimer Disease and Other Dementias

History

The history is the most important component of the dementia diagnosis. Ideally, the history should be taken from not only the patient but also an individual who knows the patient well as lack of awareness of impairment commonly accompanies dementia. This individual can provide invaluable information regarding impairment in activities of daily living (ADLs). Other key elements of the history include identifying the presenting symptom, mode of onset, duration of symptoms, and rate of progression. Neurodegenerative causes of dementia typically present with an insidious onset and slow rate of progression, while the sudden onset of symptoms should raise suspicion for stroke, medication effect, infection, autoimmune process, or psychosocial stressors. Patients may have a subacute onset over weeks or months. The dementias presenting with this course will be discussed in Chapter 94 . Thorough evaluation of patients referred for cognitive symptoms to a memory clinic can identify a potentially reversible or partially reversible disorder in up to 9% of cases ( ) and a treatable coexisting disorder in up to 23% ( ). Several key components of the dementia history are reviewed in Table 95.3 . Background information such as age, level of education, occupation, social stressors, and cultural background can influence the presentation of dementia and should be taken into consideration. As a general framework, when an older patient presents with a progressive amnestic disorder with subsequent decline in other cognitive domains, AD dementia is the most common diagnosis. Alternatively, if the initial presentation is one of a change in language, personality, or behavior with relatively spared memory, FTD is the most likely consideration. The presence of parkinsonism, hallucinations, fluctuations, and REM sleep behavior disorder with dementia are most suggestive of DLB. Vascular cognitive impairment (VCI) ( ) can be temporally related to a stroke or develop gradually with prominent cognitive slowing and executive dysfunction with cerebral small-vessel white matter disease. Often vascular disease occurs together with other causes of dementia as a comorbid component of the clinical picture. In the setting of subacute dementia, consider Creutzfeldt-Jakob disease, autoimmune dementia, and their differential diagnosis.

Cognitive Assessment

Many useful cognitive screening instruments have been developed. Common well validated instruments include the Mini-Mental State Exam (MMSE) ( ), the Blessed Orientation Memory Concentration Test ( ), the Kokmen Short Test of Mental Status (STMS) ( ), and the Montreal Cognitive Assessment (MOCA) ( ). Detailed cognitive testing by a neuropsychologist can be very helpful. The neuropsychologist provides an in-depth cognitive evaluation by administering a standardized battery of tests. These tests evaluate important cognitive domains such as attention and concentration, memory, language, visuospatial abilities, and executive function. They also gauge the psychiatric contributions to the clinical picture. Patients with different dementias have different strengths and weaknesses on these tests. The pattern of performance helps determine if the person is impaired, the severity of impairment, and the likely brain areas that are damaged. Neuropsychology can also be helpful in following a patient’s progression over time.

General Neurological Examination

A general neurological exam is a key part of the evaluation of dementia. While the general neurological exam is typically normal in early AD, abnormalities on exam may indicate other neurodegenerative processes. The presence of parkinsonism may suggest DLB or another parkinsonian dementia. Focal findings on exam such as asymmetric reflexes or other lateralizing signs may suggest a vascular component to the dementia. A coexisting peripheral neuropathy may suggest a metabolic disturbance. Fasciculations can be seen in patients with suspected FTD to suggesting coexisting motor neuron disease. A language screening exam and testing for apraxia should also be performed. The presence of a gait abnormality may indicate normal pressure hydrocephalus (NPH), parkinsonism, or vascular disease.

Laboratory Evaluation

The AAN practice parameter recommends routine assessment for vitamin B ₁₂ deficiency and thyroid hormone abnormalities because these conditions are common, and can affect cognitive function. Treatment of vitamin B ₁₂ deficiency and hypothyroidism may not completely reverse cognitive symptoms, but recognition of these conditions is important. Other routine lab tests include a complete blood cell count, electrolyte panel, glucose, liver function tests, and creatinine. Screening for syphilis should be done based on clinical suspicion ( ). Special circumstances, including age less than 65 years, seizures, rapidly progressive dementia, history of cancer or autoimmune disease, suspicion of central nervous system (CNS) infection, constitutional symptoms, history of drug abuse or immunosuppression, systemic infection, suspicion of vasculitis, or other atypical features, can guide further laboratory evaluation, including CSF examination. In neurodegenerative disease, the cell count, protein, and glucose concentrations in the CSF are within normal limits and specific markers of AD pathology; for example, Aβ42 total and phospho-tau, may be useful.

The AAN practice parameter recommends against electroencephalogram (EEG) in the standard evaluation of dementia ( ). However, EEG may be very useful in a patient with a history of seizures, assessment of rapidly progressive dementia, a history of spells, or suspicion of transient epileptic amnesia ( ).

Neuroimaging

Structural neuroimaging modalities including computed tomography (CT) and MRI can identify potentially treatable causes of dementia including subdural hematoma, hydrocephalus, and intracranial neoplasm. The AAN practice parameter paper recommends screening with either CT or MRI to identify these conditions ( ). In a study evaluating the usefulness of the AAN dementia guidelines, 3% of dementia patients had a surgically treatable finding on neuroimaging (NPH, subdural hematoma, neoplasm). Neuroimaging changed management in 15% of cases and clinical diagnoses in 19%–28% ( ). In the recent “Imaging Dementia-Evidence for Amyloid” Scanning (IDEAS) study, the use of amyloid PET among Medicare beneficiaries with cognitive impairment of unclear etiology led to a change in diagnosis and clinical management in a significant proportion of the participants ( ).

DSM-5

Recently, the DSM-5 was released, updating the prior criteria for dementia. DSM-5 introduces the terms “mild neurocognitive disorder,” which is similar to MCI, and “major neurocognitive disorder,” which is analogous to dementia. Major neurocognitive disorder represents a significant cognitive decline in at least one cognitive domain that interferes with daily function that is recognized by the individual, informant, or clinician and documented by neuropsychological testing. Mild neurocognitive disorder represents a cognitive decline which does not impair daily activities ( ). DSM-5 recommends a two-tiered approach: (1) syndrome characterization as outlined above and (2) an etiological determination.

Prevalence

According to the , 5.7 million Americans have AD dementia (2018) which represents 70% of dementia in the United States (not autopsy confirmed) ( ). The prevalence of Alzheimer dementia increases with age from 3% of people between the ages of 65 and 74 to 32% of people age 85 and older (Alzheimer’s disease facts and figures 2018). While MCI is more common in men, the prevalence of AD dementia is higher in women. This in part can be explained by a longer life span in women than men. The lifetime risk of developing Alzheimer dementia from the age of 45 is approximately 10% for men and 20% for women (Alzheimer’s disease facts and figures 2018). By 2060, the prevalence will increase to an estimated 15 million individuals in the United States ( ).

Incidence

Age is the most important risk factor for AD dementia. The incidence of AD dementia increases with age: 2 new cases per 1000 for individuals age 65 to 74, 11 new cases per 1000 people age 75 to 84, and 37 new cases per 1000 people age 85 and older (Alzheimer’s disease facts and figures 2018).

Societal Cost and Future Projections

Anticipated increases in cost to society for AD dementia are notsustainable. The total payments in 2018 for all individuals with dementia are estimated to be $277 billion (Alzheimer’s disease facts and figures 2018). The Alzheimer Association estimates that by 2050 annual costs for AD dementia will reach approximately $1.2 trillion ( ). Therefore, the burden on society will continue to be enormous unless a prevention or treatment is developed. The Health and Retirement Study has reported that AD is the costliest chronic disease in the United States exceeding those of cancer and heart disease ( ).

Risk Factors

In addition to age and female gender, many other risk factors for AD dementia have been reported.

Protective Factors

Alzheimer Clinical Features

AD dementia survival is shorter than predicted based on US population estimates, with a median survival from diagnosis of 4.2 years for men and 5.7 years for women ( ).

Early Presentation

Typical AD dementia initially presents with an episodic memory impairment reflecting the selective vulnerability of the medial temporal lobe to AD pathology. Episodic memory relates to our ability to remember information specific to a time and place when that memory was formed (e.g., “What did you eat for dinner? What did you do on a trip?”). Recent episodic memory is particularly impaired in early AD.

Pattern of Progression

While episodic memory is the hallmark of early AD, subsequent cognitive decline is heterogeneous as the pathology spreads to the association cortices. While individual presentations vary significantly, Feldman and Woodward have described the typical symptom progression in AD dementia: Mild AD (recent memory impairment, repetitive questions, loss of interest in hobbies, anomia, impaired instrumental ADLs), moderate AD (aphasia, executive dysfunction, impaired basic ADLs), severe AD (agitation, complete loss of independence, sleep disturbance) ( ).

Common Clinical Features

Semantic memory dysfunction can be an early feature of AD but occurs after episodic memory involvement. Semantic memory is factual knowledge not linked to time or space context. Examples include naming or recalling animals, objects and tools, or landmarks. The preservation of semantic memory in a subset of early AD cases indicates that transentorhinal dysfunction is inadequate to disrupt semantic memory, which likely requires extension of pathology into the temporal neocortex ( ). A category fluency test (asking the subject to generate as many items as possible from a given category such as fruits, animals, or vegetables) is commonly used as a brief test of semantic knowledge. This not only tests the ability to remember the names of objects but the ability to search one’s mind for a category of objects. This test is often impaired early in the course of the disease and also tests executive function.

Executive function (planning, organization, problem solving, set switching) decline occurs in mild AD ( ). The executive dysfunction in AD dementia is mild until later in the disease course compared to bvFTD.

Language disturbance often occurs in the mild-moderate stage of AD dementia. Initial complaints often include word-finding difficulty. In time, other features of aphasia may develop, and in the late states, language output can be limited.

Decline in visuospatial skills is common, often manifesting early with the complaint of becoming lost or being disorientated in unfamiliar places.

Apraxia often occurs later in the course of typical AD, although it may be an early feature in atypical AD.

Strikingly, several abilities are preserved until very late in the development of AD dementia. For example, AD dementia patients have preserved motor learning (procedural) ( ), motor, and sensory skills.

A useful way to think of the clinical picture of AD dementia is to look at the pattern of both deficits and strengths the patient exhibits. Patients with AD dementia have episodic memory loss and develop anomia, executive dysfunction, and visuospatial difficulty in the setting of preserved ability to walk, see, hear, and feel. AD dementia is also characterized by the juxtaposition of “knowing how” (procedural memory) and “not knowing what” (declarative memory). The clinical picture of deficits and preserved abilities is explained by the anatomical distribution of the NFT pathology. demonstrated that tangle burden is greatest in the medial temporal lobe but spares the primary sensory and motor cortices until very late in the course, which is why AD dementia patients can see, hear, and move but not remember. Early AD is characterized by an isolated memory impairment resulting from significant pathology in the entorhinal cortex and hippocampus, which serves to disconnect the medial temporal lobe from other cortices. The spread of pathology to other association cortices results in accumulation of other symptoms: semantic memory involvement from spread to the anterior temporal lobes, executive dysfunction from spread to the frontal lobes, and visual-spatial dysfunction from spread to the occipitotemporal lobes. In contrast, the primary motor and sensory cortices are only affected late in the course. The preserved basal ganglia and cerebellum are involved in the procedural/motor learning (or knowing how).

Posterior Cortical Atrophy

In 1988, Benson used the term posterior cortical atrophy to describe five cases with progressive dementia involving visual-spatial function, alexia, and partial Balint and Gerstmann syndromes with relatively preserved memory ( ). Later, autopsy series of PCA revealed Alzheimer pathology as the most common underlying etiology. While AD is the most common pathology, other pathologies in clinically diagnosed PCA cases include corticobasal degeneration and prion disease. PCA patients are often initially referred to optometry or ophthalmology for difficulty seeing, which may manifest as driving impairment or trouble reading. Not uncommonly, PCA patients undergo procedures such as cataract removal without significant benefit, or alternatively, they may be told there is nothing wrong with their eyes prior to definitive diagnosis. The mean age of onset of PCA is approximately 60 ( ). Compared to typical AD, insight is preserved ( ). In a large series of 40 PCA cases, complete or partial Balint syndrome (optic ataxia, oculomotor apraxia, simultanagnosia) was present at diagnosis in 88% of patients, complete or partial Gerstmann (left-right confusion, finger agnosia, agraphia, acalculia) was present in 62%, and visual field loss was present in 48%. Simultanagnosia is characterized by only being able to see one object at a time while viewing a scene. Ishihara plates used to assess color perception and complex scenes are good screening tools for the presence of simultanagnosia. Other important signs and symptoms include ideomotor apraxia, alexia, prosopagnosia, hemineglect (sensory or visual), achromatopsia, and dressing apraxia ( ). PCA is associated with the APOE ε4 allele ( ). Structural MRI reveals parieto-occipital atrophy and FDG-PET demonstrates associated parieto-occipital hypometabolism ( Figs. 95.3 and 95.4 , respectively). The distribution of amyloid is similar between PCA and typical AD ( ). In contrast, PCA patients have significantly more tau deposition in the occipital lobe compared to typical AD (see Fig. 95.4 ), corresponding to their clinical symptoms ( ). Adaptive equipment for the blind or those with low sight can be helpful. No randomized controlled trial supports the use of any drug therapies in PCA, although acetylcholinesterase inhibitors are commonly used since the most common underlying pathology is AD. Recent consensus criteria ( Table 95.6 ) for the diagnosis of PCA have been published emphasizing PCA as a clinicoradiological syndrome ( ). The new classification recognizes that PCA can present in a “pure” form or overlap with other degenerative diseases such as corticobasal syndrome.

Logopenic Primary Progressive Aphasia

LPA is one of the primary progressive aphasias. The distinguishing features of LPA include impaired naming, repetition, and word retrieval with phonological errors. Motor speech and grammar are spared. have suggested impairment in the phonological loop, which stores and rehearses verbal memory as the basis of the clinical presentation. In a study by , 64% of logopenic patients had AD pathology at autopsy, and these patients had greater tangle burden in the left hemisphere language areas and fewer tangles in the entorhinal cortex when compared to typical AD pathology. In a large multicenter study, amyloid pathology was present in 86% of those with LPA compared to 20% of those with nonfluent/agrammatic PPA and 16% of those with semantic variant PPA ( ). Phonological errors may be the strongest predictor of underlying AD pathology ( ). Structural MRI reveals left temporal-parietal atrophy. FDG-PET scans reveal temporal-parietal hypometabolism and tau PET demonstrates left greater than right temporal and parietal tau deposition ( Fig. 95.5, A and B ).

Behavioral/Frontal or Dysexecutive Variant Alzheimer Disease

Rarely, patients with pathologically confirmed AD present with early impairment on tests of frontal lobe function including verbal fluency and Trails A. These patients may have behavioral and personality changes similar to patients with FTD. However, those with behavioral presentation of AD develop apathy as the most common behavioral feature, in contrast to hyperorality and perseverative/compulsive behaviors seen more commonly in bvFTD ( ). Interestingly, the NFT burden in these patients is increased in the frontal lobes ( ).

Others

Other focal presentations with underlying AD pathology include corticobasal syndrome and, rarely, progressive agrammatic aphasia or semantic variant primary progressive aphasia, which are typically due to FTLD pathology ( ).

Cerebrospinal Fluid Biomarkers

The combination of reduction in the CSF Aβ42 and elevation in the CSF tau protein has a sensitivity of 85% and specificity of 86% for the diagnosis of AD dementia ( ). More recent studies using Aβ42 and tau or phosphorylated tau suggest these biomarkers can improve diagnosis in difficult cases and predict conversion from MCI to AD dementia ( ; ; ). The current research framework includes elevated phosphorylated tau as evidence of pathological tau accumulation while the less specific total tau is used as a neurodegenerative biomarker ( ). While these biomarkers can provide important information, normal CSF Aβ42 and tau have been reported in autopsy-proven AD dementia patients ( ). Recent data suggest that novel CSF markers, for example, neurofilament light protein, may be good index of neurodegeneration ( ).

Neuroimaging Biomarkers

Various neuroimaging biomarkers are sensitive to different stages and markers of AD.

Structural imaging

The AAN practice parameter recommends that a neuroimaging examination, either CT or MRI, be performed at the time of the initial dementia assessment ( ). While this recommendation was primarily suggested to exclude reversible and treatable causes of dementias, MRI provides much higher resolution than CT and has proven very useful in the differential diagnosis of dementia and as a biomarker of neurodegeneration in AD dementia. Medial temporal lobe atrophy of the hippocampus and entorhinal cortex with concomitant dilatation of the temporal horns is an early characteristic of AD dementia ( Fig. 95.7 ) and can predict conversion from normal cognition to MCI and MCI to AD dementia ( ). Reduction in hippocampal volumes correlates with NFT pathology at autopsy and cognitive decline ( ). In aMCI, atrophy is limited to the medial temporal lobe structures while with AD onset, atrophy spreads to the lateral temporal and parietal cortices. This change corresponds with Braak staging, which is why structural MRI serves as a biomarker of neurodegeneration ( ).

The utility of volumetric measurements of the entorhinal cortex, while promising, is controversial. Although medial temporal lobe structures become atrophic in early AD and correlate with episodic memory performance, later in the disease atrophy rates are greater in the temporal, parietal, and frontal cortices and are associated with deterioration in other cognitive domains including language, praxis, and visuospatial ( ). Medial temporal lobe atrophy is not specific for AD and can be seen in other degenerative and vascular processes. In particular neurodegenerative hippocampal sclerosis of aging can have a similar imaging appearance to Alzheimer dementia ( ). During the dementia stage, significant global atrophy occurs typically most significantly in a temporal-parietal distribution with coinciding ventriculomegaly.

The presence of white matter hyperintensities observed by FLAIR or T2 MRI also appears to contribute to cognitive impairment in AD ( ).

Functional imaging

Functional brain imaging using single-photon emission computed tomography (SPECT) and FDG-PET can suggest disease specific patterns. While functional imaging studies have not been endorsed by criteria for MCI or AD due to dementia ( ; ), their value is recognized in selected cases. Especially when the structural imaging scan is not informative, functional imaging modalities may provide additional useful information ( ; ).

Decreased blood flow in a temporal-parietal distribution seen on SPECT correlates with hypometabolism seen on FDG-PET and is suggestive of AD.

In aMCI, hypometabolism is primarily in the hippocampus and posterior cingulate. In AD dementia, the hypometabolism includes these regions as well as the temporal-parietal regions ( ). Recent studies have indicated that FDG-PET can be used as an aid in the diagnosis of AD dementia, in particular in differentiating AD from FTD ( ; ), which has led to the Centers for Medicare and Medicaid Services approving reimbursement for evaluating this differential diagnosis. Several studies have validated the use of FDG-PET as a biomarker in AD, resulting in its inclusion as a biomarker in the most recent AD criteria ( ). In MCI ADNI participants, FDG-PET predicted conversion to AD dementia ( ). Of particular interest, FDG-PET in cognitive normal homozygous carriers for APOE ε4 demonstrates hypometabolism in the posterior in the cingulate, temporal, and parietal cortices ( ). Fig. 95.8 demonstrates an FDG-PET in a patient with typical AD dementia.

Amyloid imaging

The development of Pittsburgh Compound B (PiB) has allowed measurement of amyloid burden in living subjects ( ). While not currently recommended for routine clinical use, this imaging modality is already being used in clinical trials for identifying preclinical AD, MCI due to AD, and monitoring effectiveness of amyloid-targeted therapies. Appropriate-use criteria have been published which make recommendations regarding the clinical setting in which amyloid PET could be considered ( ). The deposition of amyloid as measured by PiB-PET occurs primarily in the frontal and temporal-parietal regions (see Fig. 95.9 for examples of PiB-PET imaging ) . Since its discovery, numerous studies have demonstrated possible clinical utilities. PiB-PET outperformed FDG-PET in discriminating FTD from AD ( ). PiB binding cross-sectionally correlates with cognitive function ( ). Longitudinally, however, once, symptomatic, the deposition of β-amyloid plateaus demonstrating that PiB utility may be greatest in the preclinical stages ( ).

In a population-based study of cognitively normal individuals over age 70, approximately one-third have significant β-amyloid load ( ). The prevalence of a positive amyloid PET scan among cognitively unimpaired individuals in the general population increased from 2.7% in individuals between ages 50 and 59 years to 41% in those individuals between ages 80 and 89 years ( ). The high rate of amyloid-positive scans in the cognitively normal population along with the absence of any disease-modifying therapy would suggest amyloid imaging in cognitively normal elderly individuals should be reserved for research purposes until disease-modifying therapy is available.

Several F-18 analog amyloid tracers have been developed and approved for clinical use by the Food and Drug Administration (FDA) florbetapir, flutemetamol, and florbetaben. The F-18 agents provide comparable results to PiB imaging, but a longer half-life allows for transportation to clinical centers. Florbetapir performs well compared to autopsy confirmation ( ). In the ADNI study, Aβ deposition, as measured with florbetapir, correlated with cognitive decline in cognitive normal and MCI participants. In the cognitive normal group this decline sometimes occurred in the absence of FDG abnormality ( ). Multiple studies have demonstrated the prognostic importance of amyloid PET. The incident risk of developing mild cognitive impairment increased greater than twofold among cognitively unimpaired individuals who were amyloid PET positive compared to those who were amyloid PET negative ( ).

Longitudinal Tracking of Biomarkers

The advent of biomarkers for tracking the progression of AD has vastly increased our knowledge about its temporal progression and will play a key role in clinical trial design and execution. Several key biomarker studies in conjunction with data of unselected autopsies published over a short period of time have provided evidence that the pathophysiological processes underlying AD start decades before cognitive decline. This knowledge has shifted the focus of disease therapies to the presymptomatic or early symptomatic phases of the disease. One influential hypothetical model for the sequence of these biomarkers was first proposed by Cliff Jack in 2010 ( ) and revised in 2013 ( ). This model represents the pathophysiological processes underlying AD, and is summarized in Fig. 95.10 . The Dominantly Inherited Alzheimer Network (DIAN) study has provided important information regarding the timing of biomarkers in autosomal dominant AD largely consistent with the model described above ( ). This sequence can be summarized by the following: CSF Aβ42 declines over two decades before clinical symptoms; Aβ-PET abnormalities begin about 15 years before symptoms; brain volume loss and increased CSF tau also occur 15 years before symptoms; FDG-PET abnormalities occur 10 years before symptoms ( ).

While this sequence is predictable for dominantly inherited AD, the Mayo Clinic Study of Aging has demonstrated that a substantial proportion of cognitively normal subjects have neurodegeneration biomarkers but not amyloid biomarkers, as was noted earlier in this chapter ( ; ). In preclinical AD, 42% of incident Aβ-PET positive cases also have positive neurodegeneration biomarkers first. This suggests that at least two biomarker profile pathways to preclinical AD exist, and some of the cases labeled sNAP are on the AD pathway but with a different biomarker progression ( ). In the same framework, progression in MCI to dementia has also been characterized and a similar group of MCI sNAP subjects have been identified, implying that neurodegenerative pathologies other than AD may be operating in some MCI subjects, and while most progress to AD dementia, some do not ( ).

Recently, longitudinal tau-PET studies have demonstrated the rate of accumulation of tau over time. Cognitively unimpaired individuals accumulate tau at a rate of 0.5% per year compared to cognitive impaired individuals who accumulate at a rate of 3% per year ( ).

Alzheimer Genetics

In a large twin study from Sweden based on 392 pairs of twins, the genetic component of AD was estimated to be 58%–79% ( ). Family history of AD can provide important risk information. The lifetime risk of AD dementia in first-degree relatives is approximately 39% and this risk increases to 54% by age 80 if both parents have AD dementia ( ).

Early-Onset Genes

Three rare, early-onset, fully penetrant gene mutations have been described to cause Alzheimer dementia. While mutations in these genes are rare, studying them has been instrumental in our understanding of AD. All three increase brain Aβ levels and form an important part of the amyloid hypothesis for AD (discussed later). This provided the basis for several animal models and biomarker development including development of CSF and PET Aβ.

Late-Onset Genes

Apolipoprotein E (APOE) is the most important genetic risk factor for late-onset AD ( ). APOE has three isoforms (E4 associated with high risk, E3 associated with neutral risk, and E2 which is protective). About 20% of all late-onset AD is thought to be related to APOE ε4 ( ). APOE ε4 affects AD risk and age of onset in a dose-dependent way; for example, E4 homozygotes have a mean age of onset of 68 with a lifetime AD risk of 91%; in E4 heterozygotes, the mean age of onset is 76 with a 47% lifetime risk ( ). In contrast, E4 noncarriers have a mean age of onset of 84 with an approximately 20% lifetime frequency ( ; ).

Trem2 variants have recently been identified as rare risk variants for AD dementia ( ). The odds ratio is similar to APOE ε4 but it is rare, which is why it was not seen in previous genome-wide association studies. Its population-attributable risk is lower than APOE ε4 due to its lower frequency.

Many other risk loci have been associated with AD risk, but their overall impact is thought to be small. These include CD33 molecule, ATP-binding cassette, subfamily A, member 7 (ABCA7), sortilin-related receptor L (SORL1), clusterin (CLU), phosphatidylinositol binding clathrin assembly protein (PICALM), as well as many others ( ).

Genetic Testing

Genetic testing for APP, PSEN1, PSEN2, and APOE is commercially available in Clinical Laboratory Improvement Amendments (CLIA) laboratories. Routine genetic testing is not recommended by the practice parameter of the AAN, but if patients have testing, genetic counseling prior to testing is essential. Testing may have significant implications for patients and families in terms of family planning, financial costs, and risk of depression. Genetic testing currently does not change treatment of the patient but may provide opportunities to participate in research studies such as DIAN.

Amyloid Hypothesis

The amyloid hypothesis proposes that AD can result from too much Aβ production or a change in ratio with more Aβ42 than Aβ40 produced, or impaired clearance of Aβ. Understanding APP processing is central to the amyloid hypothesis ( ).

APP is a type 1 protein with the amino terminal in the extracellular space. APP is cleaved by three enzymes (α, β, and γ secretase) ( ) and can undergo processing either through the amyloidogenic pathway (γ and β secretases) which produces Aβ, or the nonamyloidogenic pathway (α secretase) ( Fig. 95.11 ).

In the amyloidogenic pathway, APP is first cleaved by β secretase. This cleavage produces a β C-terminal fragment, which stays on the membrane and soluble APP Aβ. This β C-terminal fragment is then cleaved by γ secretase in the transmembrane region producing APP intracellular domain (AICD) and Aβ peptide, which is then released into extracellular space. PSEN 1 and 2 are part of the γ secretase complex ( ). Other components of the γ secretase complex include nicastrin, APH-1. This aforementioned cleavage occurs by sequential events eventually producing β-amyloid fragments, either Aβ1-42 (about 10%) or Aβ1-40 (about 90%). Aβ can aggregate and these aggregates may form oligomers which may be toxic. Aβ1-42 has a greater tendency to aggregate and is found in greater concentration in plaques, while Aβ1-40 is the predominant form in vascular amyloid deposits.

In the nonamyloidogenic pathway, APP cleavage is mediated by α secretase. This cleavage is in the middle of the β-amyloid peptide above the surface of the membrane, thereby preventing the formation of Aβ. This cleavage generates soluble APP α (sAPPα) and α C-terminal fragments, which can also undergo cleavage by γ secretase, producing a nontoxic peptide called P3.

APOE functions primarily in the transport of lipids/cholesterol from astrocytes to neurons. The presence of the APOE ε4 allele is associated with decreased CSF Aβ42 and increased brain Aβ burden seen on Aβ ( ). Despite these biomarker changes, the mechanisms of E4 leading to AD are numerous and include both Aβ- related and Aβ-unrelated mechanisms. Aβ-dependent mechanisms include interfering with cerebral Aβ clearance and promoting Aβ aggregation. Aβ-independent mechanisms include promoting abnormal lipid transport, thereby altering synaptic plasticity and the inflammatory response ( ).

Accumulating evidence suggests that a common mechanism across neurodegenerative diseases may include the trans-synaptic spread of tau and other misfolded proteins to anatomically connected regions in a prion-like manner where the protein is released and taken up by the anatomically related neuron ( ).

Recent efforts have been made to investigate how large-scale neural networks may be involved in the pathophysiology of AD and other neurodegenerative disorders. It has long been known that systems of connected neurons are selectively vulnerable to neurodegenerative disease, but the pathophysiology behind this association is debated. In contrast to disease models that emphasize the molecular misfolding of proteins spreading within connected systems, complex systems models emphasize the causal role of dynamic functional activity within brain systems interacting with molecular physiology ( ).

Amyloid Plaques

Aβ deposition in the brain occurs in a sequential manner starting in the cortex, followed sequentially by the hippocampus, basal ganglia, thalamus, and basal forebrain before in the final stages reaching the brainstem and cerebellum ( ). Plaques are complicated heterogeneous lesions composed of extracellular protein deposits and certain cellular components. Aβ derived from APP is the sine qua non of plaques, which are in turn the defining pathological finding in AD. Despite the strong association of Aβ with plaques, in reality plaques contain many other components. Other associated components of the plaque include APOE, alpha-1 antitrypsin, complement factors, and immunoglobulins. Plaques can be divided into diffuse (Aβ mainly) and neuritic plaques (includes damaged tau containing axons and dendrites, i.e., neurites). Cell components seen in plaques include neurites, microglia, and astrocytes. The typical neuritic plaque has a dense core of Aβ and less compact rim consisting of neurites, microglia, and astrogliosis at the periphery. Diffuse plaques have less diagnostic specificity and can be seen in a variety of different disorders. They contain Aβ but do not stain with tau-containing neurites and are not associated with synaptic loss, reactive astrocytes, or microglia.

Neurofibrillary Tangles

The major component of the NFT is tau within neurons and their cell processes. First, pretangles form in the neuron cytoplasm, often concentrated around the membrane. Then, tau is organized into fibrils as NFTs. When the cell dies, the tangle may be situated extracellularly. Normal tau is a microtubule-associated protein and is not visible within the brain. In NFTs, tau has become hyperphosphorylated and abnormally conformed. This tau can be detected with immunostains that bind abnormally conformed tau. have described a fairly predictable spread of NFT pathology through the brain occurring in six stages which can be reduced to three, consisting of transentorhinal (stages I, II), limbic (III, IV), and isocortical (V, VI) stages. Braak staging has recently been modified to include brainstem regions preceding medial temporal involvement ( ).

Several other associated pathological findings may be seen with AD. CAA, often present in leptomeningeal vessels, capillaries, small arterioles, and middle-sized arteries, is seen in over 80% of AD cases. In contrast to cerebral plaques consisting primarily of Aβ42, CAA’s major component is Aβ40. Granulovacuolar bodies, typically found in the hippocampus, are small dense granules within the vacuole and can be labeled with acid phosphatase, tubulin, ubiquitin, and neurofilament. Hirano bodies (eosinophilic rod bodies), also typically found in the hippocampus, are often in neuronal processes and contain actin and actin-binding proteins. The significance of granulovacuolar bodies and Hirano bodies is incompletely understood.

Clinicopathological Correlations

Numerous studies have shown that dementia duration and severity correlate better with NFTs compared to amyloid plaques ( ; ).