Atypical Presentations of Alzheimer Disease

Abstract

OBJECTIVE:

This article provides a comprehensive review of the distinct features of four atypical Alzheimer disease (AD) variants: dysexecutive AD, behavioral variant AD, posterior cortical atrophy, and the logopenic variant of primary progressive aphasia. It also elucidates their clinical presentations, underlying pathophysiologic pathways, diagnostic indicators, and management requirements.

LATEST DEVELOPMENTS:

Recent research has revealed that these atypical AD forms vary not only in clinical manifestations but in their functional neuroanatomy spanning a common pathophysiologic spectrum. Imaging techniques, such as MRI, fludeoxyglucose positron emission tomography (FDG-PET), and tau PET, have identified distinct abnormalities in specific brain regions associated with each variant. This same variability is less tightly coupled to amyloid imaging. Emerging diagnostic and therapeutic strategies should be tailored to each variant’s unique features.

ESSENTIAL POINTS:

Atypical forms of AD often present with symptoms that are predominantly nonmemory related, distinguishing them from the more common memory-centric presentation of the disease. Two distinct clinical and pathologic entities, dysexecutive AD and behavioral variant AD, have replaced the outdated term frontal AD. Posterior cortical atrophy is another variant that mainly affects higher-order visual functions, which can lead to misdiagnoses because of its atypical symptom profile. Logopenic primary progressive aphasia is marked by difficulties in word retrieval, a challenge that may not be readily apparent if the person compensates by using circumlocution. Modern diagnostic techniques, such as MRI, PET, and biomarker analysis, have proven crucial for the accurate diagnosis and differentiation of these atypical AD variants. In treating these forms, it is critical to use tailored therapeutic interventions that combine pharmacotherapy with nonpharmacologic strategies to effectively manage the disease.

INTRODUCTION

Alzheimer disease (AD) is typically characterized by a multidimensional cognitive decline, with memory impairment often emerging as the most salient feature. However, other cognitive domains, including language, visuospatial and perceptual abilities, perceptual-motor skills, and executive functions, are not spared and can display variable deterioration. Beyond these characteristic presentations, certain atypical forms of AD are marked by predominant symptoms in nonmemory cognitive domains.¹ These atypical presentations not only differ in their clinical manifestations but also exhibit unique pathophysiologic pathways, diagnostic indicators (see FIGURE 2-1 for distinct functional anatomy), and management requirements. These atypical presentations are thought to represent extremes of a common underlying functional spectrum.² This article reviews the distinctive features of four such atypical AD variants: dysexecutive AD, behavioral variant AD, posterior cortical atrophy (PCA), and the logopenic variant of primary progressive aphasia (PPA).

FIGURE 2–1.

Brain fluorodeoxyglucose positron emission tomography (FDG-PET) for individual case examples of the clinical phenotypes discussed in this article. These global brain patterns represent the extremes of a common global functional state space or phenotypic spectrum of neurodegeneration.² The individual patient images are displayed on a template brain surface with negative values indicating hypometabolism in blue and positive values indicating hypermetabolism in yellow and red relative to a normative database. A, Typical memory predominant Alzheimer disease (AD) showing hypometabolism in temporal and parietal areas, B, Posterior cortical atrophy (PCA) showing hypometabolism in occipital and parietal areas and relative hypermetabolism in the frontal lobe. C, Logopenic variant of primary progressive aphasia (lvPPA) showing hypometabolism in left temporal and parietal areas. D, Dysexecutive AD (dAD) showing hypometabolism in the heteromodal association cortex and hypermetabolism in primary sensory and motor cortices. E, Behavioral variant of AD (bvAD) showing hypometabolism in frontal, medial temporal, and cingulate cortices. F, Behavioral variant frontotemporal dementia (bvFTD) showing hypometabolism in the frontal lobe and hypermetabolism in occipital and parietal areas.

DYSEXECUTIVE AND BEHAVIORAL VARIANT ALZHEIMER DISEASE

Dysexecutive AD is a subtype of AD in which the primary symptom is executive dysfunction. Another, rarer subtype can show symptoms similar to behavioral variant frontotemporal dementia (FTD).³ However, although it mimics behavioral variant FTD clinically, its underlying cause is AD, as confirmed by biomarkers or autopsy. This subtype is termed behavioral variant AD in this context and should be distinguished from dysexecutive AD. Recently proposed criteria for dysexecutive AD⁴ stipulate that the behavioral variant FTD syndrome is exclusionary for diagnosing dysexecutive AD (TABLE 2-1). It is important to note that the brain areas responsible for social, emotional, and motivational functions are different from those involved in executive functions.^5,6 As a result, the clinical manifestations of diseases affecting these areas are distinct. This section delves into the differences in clinical presentations, underlying causes, diagnosis, and management of these two distinct AD subtypes that are often confused.

TABLE 2–1.

Proposed Diagnostic Criteria for Progressive Dysexecutive Syndrome and Dysexecutive Alzheimer Disease

	Clinical criteria	Evidence	Exclusions
Progressive dysexecutive syndrome	Persistent, predominant, and progressive decline for 6 months in core executive functions: working memory, cognitive flexibility, or inhibition in the absence of predominant behavioral features (ie, would not meet criteria for the clinical syndrome of behavioral variant frontotemporal dementia)	Evidence of impaired executive functions is obtained by patient or informant reports in conjunction with formal evaluation of cognitive performance on mentally effortful tasks that require conscious active manipulation of abstract or simultaneous information streams	History of sudden onset or other medical conditions severe enough to account for related symptoms (eg, primary psychiatric, cerebrovascular, infectious, toxic, inflammatory, or metabolic disorders)
Progressive dysexecutive syndrome with Alzheimer disease (AD) neuropathologic change (possible dysexecutive AD)	Meets criteria for progressive dysexecutive clinical syndrome	Must have evidence of Alzheimer pathophysiology with one of the following:  Decreased CSF amyloid-β1 (Aβ1)-Aβ42 or Aβ42/Aβ40 ratio; or  Abnormal tracer retention on amyloid-PET	None
Progressive dysexecutive syndrome due to Alzheimer disease (definite dysexecutive AD)	Meets criteria for possible dysexecutive AD	Must have evidence of one of the following:  Increased CSF phosphorated tau  Abnormal tracer retention on tau PET  Alzheimer disease autosomal dominant genetic variation present  Postmortem diagnosis of AD, high likelihood

Clinical Presentations

The progressive dysexecutive syndrome that defines dysexecutive AD represents a unique challenge in the spectrum of AD subtypes.⁴ This clinical syndrome typically occurs at a younger age than other phenotypes, and most patients are still actively engaged in the workforce and have parental responsibilities. Patients diagnosed with dysexecutive AD predominantly face challenges in areas requiring planning, organization, and decision making. Their cognitive abilities, especially those related to the executive functions of working memory and cognitive flexibility, are noticeably compromised. This often manifests in their daily activities, and tasks that require some executive faculties are markedly impaired whereas more automatic activities are entirely preserved. This manifests in a striking disconnect; multitasking and simple tasks involving sequences and spatial manipulations (eg, making a sandwich, learning a new software program or process at work, or mentally manipulating numbers) are markedly impaired, but more complex processes that are well learned and automatic (eg, driving) continue to be performed with seemingly less difficulty.

There is variability in how much an individual can compensate for executive dysfunction by relying on well-learned behaviors. Therefore, some individuals may be able to continue in their jobs for years without noticeable impacts on their performance. However, once a new process, procedure, or software program is introduced, they are unable to perform their work duties, and their cognitive difficulties become apparent to themselves and their employers and family. A similar phenomenon occurs when learning a new process or procedure introduced into the home (eg, new technology such as a smartphone or television), playing a new board game, or assembling toys or furniture. Other cognitive domains that depend on executive abilities related to working memory and cognitive flexibility can also be impaired (eg, language, visuospatial, and memory) leading to logopenic-type language dysfunction (eg, anomia and impaired repetition) or apparent topographagnosia and episodic memory loss. For some individuals, memory loss can be the focus of the reported clinical symptoms as patients and care partners may be unfamiliar with the broader impact of executive function for tasks that are being performed daily. Many of these difficulties are referred to as “forgetting” how to do the task or procedure, which can lead to a mistaken focus on memory encoding and retrieval as the primary cognitive deficit.

People with dysexecutive AD may become more irritable and anxious because they can no longer perform simple tasks and become overwhelmed with multitasking. This may also lead to a withdrawal from activities that were previously enjoyable but are now too difficult to engage in and raises clinical concern for apathy or other emotional disturbances. These circumstances can also understandably produce a depressed mood in many individuals with this type of executive dysfunction. The various combinations of these psychiatric symptoms (eg, irritability, anxiety, apathy, poor attention, and depressed mood) often lead to several psychiatric diagnoses before the diagnosis of dysexecutive AD (eg, major depressive disorders, anxiety disorders, and attention deficit hyperactivity disorders). However, the root cause of these social, emotional, and motivational disturbances is secondary to the primary executive dysfunction. Over time, their ability to solve problems and perform simple tasks becomes increasingly compromised, leading to job loss and increased dependence on care partners or family members for nonautomatic daily functions.

In contrast, behavioral variant AD paints a different clinical picture. People with behavioral variant AD typically undergo noticeable shifts in their personality and behavior with root causes stemming from social, emotional, and motivational disturbances. It is not uncommon for them to behave inappropriately in social contexts, making interactions challenging for both the patient and those around them. Their emotional landscape also changes, with many displaying a diminished emotional response or a blunted affect as the leading clinical symptom. A pronounced lack of motivation becomes evident and is the driving source of clinical impairment in some patients with behavioral variant AD, just as it is for some individuals with behavioral variant FTD. Activities that were once pursued with enthusiasm are now met with indifference. This decline in interest can extend to essential daily tasks including basic hygiene. Obsessive, compulsive, and repetitive behaviors can also occur and be indistinguishable from those of behavioral variant FTD. Memory and executive dysfunction may or may not occur but are not the prominent or driving force of the clinical disturbances. Therefore, clinically there is little difference in the syndromes of behavioral variant FTD and behavioral variant AD, but this is an area of ongoing study as behavioral variant AD is not routinely distinguished from dysexecutive AD, with some suggesting they are on one continuum⁷ and others suggesting they are on a multidimensional spectrum (TABLE 2-2).⁶

TABLE 2–2.

Clinicopathologic Features of Dysexecutive Alzheimer Disease, Behavioral Variant Alzheimer Disease, and Behavioral Variant Frontotemporal Dementia

Clinicopathologic feature	Dysexecutive Alzheimer disease	Behavioral variant Alzheimer disease	Behavioral variant frontotemporal dementia
Clinical	Progressive dysexecutive syndrome with younger age of onset (typically between 45 and 70 years old)	Progressive behavioral syndrome with variable age of onset	Progressive behavioral syndrome with younger age of onset (typically between 40 and 65 years old)
Neuropsychological testing	Typically, very abnormal with executive dysfunction affecting performance on most tests (performance validity testing may be falsely indicative of poor validity)	Can demonstrate normal to variable memory and executive dysfunction	Can demonstrate normal to variable memory and executive dysfunction
Social, emotional, and motivational	Minor and secondary feature that may not be present	Major and primary feature that must be present	Major and primary feature that must be present
Imaging and neuroanatomy	Heteromodal association cortex, but parietal dominant focus with frontal involvement is typical; sparing of the hippocampus is common but not always present	Frontotemporal that typically extends to inferior temporal and parietal	Frontotemporal
Alzheimer disease biomarkers	Positive	Positive	Negative

Pathophysiology

Dysexecutive AD stands out not only for its clinical symptoms but also for the specific brain changes causing these symptoms. Although the term mechanism is often used to describe such causes, causal explanation is used in this article to better capture the broader interplay of anatomy and dynamic physiology behind the deterioration of mental functions in the setting of dysexecutive AD and behavioral variant AD.⁸ All AD subtypes share common causal explanations such as misfolded proteins and the accumulation of amyloid-β (Aβ) and tau. However, the causal explanations for the distinct clinical symptoms of each subtype relate to the degeneration of specific functional brain regions and systems, leading to observable behavioral differences.

Executive functions are key mental skills that are affected in patients with dysexecutive AD.⁵ These skills include working memory (holding and working with information in your mind), cognitive flexibility (switching between tasks or thoughts), and inhibition (controlling impulses). Common models of working memory refer to a few different parts: the phonologic loop (dealing with sounds and words), the visuospatial sketch pad (handling visual and spatial information), and the episodic buffer (linking information across domains to form integrated units of visual, spatial, and verbal information with time sequencing).⁹

Recent research points to a system in the brain where there is a general space for short-term memory. This space is spread across different brain areas, but the main operations, particularly those related to working memory, happen mainly in the parietal and frontal regions. Put simply, the left side of the brain handles information step by step, which helps with logical thinking. However, the right side of the brain can juggle multiple pieces of information quickly, aiding in pattern recognition and abstraction.¹⁰

The symptoms seen in patients with dysexecutive AD, such as trouble with new tasks, planning, or organizing, come from problems in these working memory areas or how these areas communicate with each other. Researchers using brain imaging to study these areas use various names such as executive control networks, working memory networks, parietofrontal networks, task-positive networks, and multidemand networks, among others.^11–16

In contrast, behavioral variant AD presents a different pathophysiologic profile. Although it shares the foundational causal explanations of misfolded proteins and the accumulation of Aβ and tau, the clinical manifestations of behavioral variant AD are more closely tied to the degeneration of brain regions responsible for behavior, emotion, motivation, and personality. Specifically, the anterior temporal lobes and certain frontal regions play pivotal roles in motivation, social behavior, and emotional processing, and are more affected in behavioral variant AD.

The behavioral changes observed in patients with behavioral variant AD, such as impulsivity, social inappropriateness, apathy, and personality shifts, can be attributed to the degeneration of the anterior temporal lobes and affected frontal lobe regions. Moreover, the overlap in symptoms of behavioral variant AD and behavioral variant FTD can be explained by the similar brain systems affected in both conditions. However, the underlying molecular pathology distinguishes the two, with behavioral variant AD rooted in AD pathology.

Functional neuroimaging studies have highlighted various networks disrupted in behavioral syndromes, including the salience network, default mode network, and emotional processing circuits. These networks, when compromised, can lead to the characteristic motivational, behavioral, and emotional symptoms of behavioral variant AD. Understanding the distinct pathophysiologic underpinnings of dysexecutive AD and behavioral variant AD is crucial for accurate diagnosis and targeted counseling and therapeutic interventions.

Diagnosis

Diagnosing dysexecutive AD demands a comprehensive approach, emphasizing the differential diagnosis of a progressive dysexecutive syndrome. It is essential to recognize features indicative of degenerative etiologies, such as a gradual and insidious progression. The process begins with a thorough history and physical examination, considering nondegenerative causes of executive dysfunction. Although many of these features are common to all degenerative syndromes and are discussed elsewhere, it is crucial to manage nondegenerative factors influencing executive functioning. These often include sleep disorders, medications, vascular diseases, CSF dynamics disorders, and psychiatric conditions.

Neuropsychological testing is vital to determine the exact nature and severity of executive dysfunction. The Trail Making Test Part B, for instance, is particularly sensitive to various forms of executive dysfunction. However, interpreting these tests requires caution. Most tasks demand some level of executive function, leading to a profile of impairment across multiple domains for individuals with a progressive dysexecutive syndrome. Furthermore, poor performance on validity measures can mistakenly suggest malingering or psychiatric causes.¹⁷ This misinterpretation is especially problematic for working-age patients with dysexecutive AD undergoing testing because of job loss, which can appear as a motivation for secondary gain. Additionally, psychiatric symptoms resulting from lost executive function can lead to misdiagnoses of primary psychiatric disorders.

Structural neuroimaging plays a dual role: excluding specific nondegenerative causes and providing evidence of degenerative changes in working memory systems. Atrophy in the heteromodal association cortex, particularly the parietal lobe, is common. The frontal lobe may be involved but always to a lesser degree than the parietal lobe. The term frontal AD is anatomically incorrect and syndromically ambiguous, so this term should not be used for dysexecutive AD or behavioral variant AD. A normal-appearing hippocampus might mislead clinicians to rule out AD, but the hippocampus is commonly spared in dysexecutive AD. Functional imaging, such as fludeoxyglucose positron emission tomography (FDG-PET), can offer more definitive insights, helping differentiate dysexecutive AD from other degenerative conditions and even subtyping dysexecutive AD.^2,5,10 Even when changes are relatively mild on structural imaging, they are often clearly abnormal on FDG-PET. These imaging patterns guide further testing, medical counseling, staging, prognosis, and management.

Despite comprehensive clinical assessments, neuropsychological testing, and neuroimaging, some conditions can mimic dysexecutive AD, leading to diagnostic uncertainty. Therefore, additional AD biomarker data, preferably from CSF, are indispensable.¹⁸ However, interpreting CSF phosphorylated tau (pTau) levels requires caution, because some patients with dysexecutive AD might present with normal levels.⁴ Tau PET imaging in dysexecutive AD usually demonstrates the highest magnitude of signal increases of any AD subtype.⁴ The parietal and frontal regions show the most characteristic increase in tau PET signal alongside inferior temporal signal changes seen in most AD clinical phenotypes. This regional information informs phenotypic characterization, but this information is also present in FDG-PET and to a lesser degree in structural imaging. From a diagnostic perspective, tau PET may be helpful only when diagnostic uncertainty remains or other AD biomarkers are not feasible.^4,18 Apart from diagnosis, tau PET may have an emerging role in staging.

Behavioral variant AD is relatively less common and presents its own diagnostic challenges. Clinically, behavioral variant AD can resemble behavioral variant FTD, making differentiation based solely on symptoms challenging. The distinction between the disorders is in the underlying pathology. Although the clinical presentation might echo behavioral variant FTD, biomarkers will indicate AD pathology. Autosomal dominant genetic etiologies are relatively common in the setting of behavioral variant FTD, but this is not a common consideration in behavioral variant AD. However, genetic counseling and consideration of genetic testing should be done for all patients with a younger onset and individuals with a strong family history of multiple affected relatives, especially first-degree relatives with a younger age of onset. Neuropsychological testing, although not diagnostic, remains essential for documenting severity, strengths, and weaknesses. It may reveal varying degrees of executive and memory impairments in both behavioral variant FTD and behavioral variant AD. Functional neuroimaging, such as FDG-PET scans, can be invaluable for diagnosis and staging, potentially showing patterns in the temporal and frontal regions indicative of behavioral variant AD. Differentiating between behavioral (ie, behavioral variant FTD and behavioral variant AD) and executive (dysexecutive AD) syndromes on neuroimaging requires recognizing distinct functional neuroanatomic patterns, with dysexecutive AD typically sparing the medial frontal region relative to parietal and dorsal lateral prefrontal regions, unlike the behavioral syndromes.

In diagnosing these subtypes, a combination of clinical assessments, neuropsychological testing, structural and functional imaging, and AD biomarkers is paramount. This multifaceted diagnostic approach ensures accurate identification and differentiation among dysexecutive AD, behavioral variant AD, behavioral variant FTD, and other etiologies, guiding patients toward the most appropriate care, interventions, and clinical trials.

Management

The management of dysexecutive AD and behavioral variant AD combines pharmacologic with nonpharmacologic approaches. Cholinesterase inhibitors and N-methyl-d-aspartate (NMDA) receptor antagonists, traditionally used in the treatment of other forms of AD, likely offer similar benefits to patients with dysexecutive AD and behavioral variant AD. However, their efficacy in affecting executive and behavioral features in this setting has not been specifically evaluated. Similarly, amyloid-lowering monoclonal antibodies have not been specifically evaluated in these clinical phenotypes. Appropriate use recommendations suggest that this absence of treatment-related information for atypical AD is not a contraindication for amyloid-lowering monoclonal antibody use, but they do suggest this limitation should be acknowledged in discussions with therapy candidates and their care partners.¹⁹ Data in 2023 suggested that patients with high tau levels may not benefit from anti-amyloid therapy.²⁰ Therefore, further research is necessary to evaluate the possibility that there may be less of a benefit in certain patients with dysexecutive AD who commonly have very high tau levels measured with PET imaging.^4,5,11 Future clinical trials should emphasize the importance of detailed phenotyping that includes functional neuroimaging, such as FDG-PET (FIGURE 2-1), to provide a better understanding of phenotypic variability in responses to therapeutics. This approach will facilitate the shared decision-making process when clinicians are considering the risks and benefits of different therapeutic strategies.

The social, emotional, and motivational symptoms, whether primary in behavioral variant AD or secondary in dysexecutive AD, may respond to typical pharmacologic strategies for these symptoms (eg, antidepressants, antipsychotics, and anxiolytics), but this has not been systematically investigated, and vigilance for over-medicating should be maintained throughout the disease course.

Alongside medications, nonpharmacologic strategies should also be emphasized. Tailored counseling to enhance executive function performance (eg, simplifying environments, avoiding multitasking, and emotion control) or relying on well-learned behaviors to compensate can equip patients with strategies to cope with daily tasks, improving their quality of life and reducing care partner burden. Behavioral redirection and distraction can also help with problematic behaviors just as they can in behavioral variant FTD. Structured routines, environmental modifications, and care partner training can all contribute to managing executive and behavioral challenges.

Regardless of the subtype, care partner education remains a cornerstone of management. Given the unique challenges posed by both dysexecutive AD and behavioral variant AD, it is essential for care partners to understand the nuances of these conditions. Providing them with the knowledge and tools to manage symptoms, handle behavioral challenges, and offer emotional support can significantly enhance the care provided to patients, ensuring their well-being and dignity are maintained (CASE 2–1).

POSTERIOR CORTICAL ATROPHY SYNDROME

PCA syndrome due to AD can be difficult to recognize and diagnose without a keen awareness of the higher-order visual deficits that dominate this atypical presentation. PCA is characterized by prominent cortical visual dysfunction with relative sparing of memory, language, executive functions, behavior, and personality at presentation or early in the course. It is defined by its clinical features, and the diagnosis is supported by abnormal posterior brain imaging findings as reviewed later in this section.

The term posterior cortical atrophy was coined by Benson and colleagues²¹ when they described five patients who developed progressive dementia after presenting with higher-order visual dysfunction and relative sparing of memory, insight, and judgment in association with predominant parieto-occipital cortical atrophy. Autopsies were not performed. Controversy then ensued about whether PCA was a unique neurodegenerative disease or simply an atypical syndrome of AD. In 1993, Levine and colleagues²² published a clinicopathologic case study of a patient, similar to those described by Benson and colleagues, with postmortem examination that revealed AD pathology, leading to naming the condition the visual variant of AD. The term posterior cortical atrophy is now favored, particularly after the publication of the PCA consensus criteria by Crutch and colleagues²³ in 2017, which are reviewed later in this article, and with the discovery that non-AD pathologies can also present with PCA syndrome including dementia with Lewy bodies, corticobasal degeneration, Creutzfeldt-Jakob disease, and copathology of Lewy bodies and AD.

Clinical Presentation

Frequently, individuals experiencing PCA-related symptoms attribute their visual issues to ocular causes and first seek the expertise of ophthalmologists or optometrists. A common scenario is for a patient to receive multiple prescriptions for eyeglasses or undergo cataract extraction without alleviation of their symptoms. In these instances, patients remain undiagnosed for an extended duration, often spanning many months or even years, until advanced higher-order visual impairments occur or with the apparent onset or worsening of memory deficits or other cognitive impairments.

TABLE 2-3 lists examples of initial patient concerns and their evolution over time. Symptoms may initially manifest as ordinary visual blurriness, gradually evolving into difficulty reading and problems perceiving objects that are conspicuously within view. Eventually, patients report poor performance with tasks that are visually demanding or require oculomotor coordination, and even at that stage, some cognitive screening batteries, such as the Mini-Mental State Examination (MMSE), do not capture visual brain dysfunction. After 1 to 2 years, symptoms can extend beyond the visual realm to include difficulty with calculations, writing, and praxis. An exception to this pattern occurs in patients with dominant biparietal dysfunction, who initially have limited visual symptoms related to spatial awareness but have prominent symptoms consistent with impaired praxis (eg, difficulty with object manipulation such as using a computer mouse), dysgraphia, and dyscalculia. Descriptions of variations in the PCA presentation include the biparietal (or dorsal) variant as noted, an occipital variant, and an occipitotemporal (or ventral) variant, although the majority of patients express mixed characteristics.²⁴

TABLE 2–3.

Presenting Visual Concerns From Patients With Posterior Cortical Atrophy Syndrome^a

	Patient comments
Early vision concerns, year 1	I can’t see clearly.
	My vision seems blurry.
	It is getting harder to see while I am reading.
	I have difficulty seeing things on the computer, such as a spreadsheet.
Subsequent vision concerns, year1–2	I am still having difficulty reading, even after several pairs of new reading glasses.
	I can’t find things that turn out to be right in front of me.
	I can’t read traffic signs.
	I have problems seeing when I am using email (or phone or other specific circumstance).
Later-stage vision concerns, year 2 and beyond	My depth perception seems off.
	I can’t read anymore.
	I can’t see to sign my name on a straight line.
	I might have to stop driving if I can’t get to the bottom of what is wrong with my vision.

^aAny of these symptoms can be the first symptom discussed with a neurologist when an ophthalmologist or optometrist finds no ocular cause for the visual concerns.

Onset of PCA generally occurs at an early age, usually between 50 and 64 years, although late-onset (at 65 years or older) presentations can occur in up to 15% of people with PCA.^25,26 The atypical nature of the initial symptoms (ie, visual and not memory), the young onset, satisfactory performance on screening measures of global cognition, and preserved insight can lead to a mistaken diagnosis of anxiety or depression as the underlying cause of symptoms. Although it is common for patients to experience symptoms of anxiety or depression, attributing all symptomatology to a mood disorder may result in diagnostic delays, potentially continuing until patients reach a point at which they can no longer engage in activities such as driving, reading, or working.

Diagnostic Criteria and Classification

In 2017, an international PCA work group published consensus criteria for PCA syndrome as well as a PCA classification framework that incorporates underlying pathology and other syndromes that can occur alongside PCA (eg, corticobasal syndrome) (TABLE 2-4).²³ In brief, a patient must have 3 of the 16 core cognitive features as an early or presenting sign. Features include alexia, space and object perception impairment, all elements of Balint syndrome (oculomotor apraxia, optic ataxia, simultanagnosia) and Gerstmann syndrome (left-right disorientation, dysgraphia, finger agnosia, acalculia), environmental agnosia, impaired praxis (limb, dressing, and constructional), apperceptive prosopagnosia, and a homonymous visual field defect. Neuroimaging findings that support the diagnosis include atrophy of posterior cortical structures on MRI, hypometabolism on FDG-PET, and hypoperfusion on single-photon emission computed tomography (SPECT). The treating or consulting neurologist should personally review the brain images because subtle, disproportionate posterior findings are not always reported. Patients with PCA are classified as having PCA-pure when they have no other associated syndromes or PCA-plus when PCA occurs in association with another clinical syndrome such as corticobasal syndrome.²⁷ Further classification can occur when biomarkers or tissue confirmation is available. For instance, if CSF AD biomarkers are present or an amyloid PET study is positive, then the patient can be classified as having PCA-AD. More information about this can be found in the 2017 article by Crutch and colleagues.²³

TABLE 2–4.

Summary of the 2017 Consensus Posterior Cortical Atrophy Criteria and Classification Framework^a

Core posterior cortical atrophy (PCA) syndrome features (all three must be present)

⧫ Insidious onset, gradual progression, and prominent early disturbance of visual functions or other posterior cortical functions

Core PCA cognitive features (at least threemust be present as an early or presenting feature)

⧫ Space perception deficit

⧫ Simultanagnosiab

⧫ Object perception deficit

⧫ Constructional dyspraxia

⧫ Environmental agnosia

⧫ Alexia

⧫ Left-right disorientationc

⧫ Acalculiac

⧫ Apperceptive prosopagnosia

⧫ Agraphiac

⧫ Homonymous visual field defect

⧫ Finger agnosiac

⧫ Oculomotor apraxiab

⧫ Optic ataxiab

⧫ Limb apraxia (not limb-kinetic)

⧫ Dressing apraxia

Core PCA neuroimaging features (supportive of diagnosis)

⧫ Prominent parieto-occipital or occipitotemporal atrophy or hypometabolism or hypoperfusion on MRI, fludeoxyglucose positron emission tomography (FDG-PET), or single-photon emission computed tomography (SPECT)

Relatively spared features (all must be evident)

⧫ Anterograde memory function, speech and nonvisual language functions, executive functions, behavior, and personality

Exclusions

⧫ No other explanation for symptoms based on the following: afferent visual dysfunction, afferent visual lesions, vascular lesions, brain tumor or other mass lesions, or any other causes

PCA designations using a three-level classification framework

⧫ Level 1

◊ Designation of PCA

⧫ Level 2

◊ If the patient meets criteria for PCA and does not meet criteria for other clinical syndromes, the designation is PCA-pure; if the patient meets criteria for PCA and meets criteria for other clinical syndromes (eg, dementia with Lewy bodies, corticobasal syndrome, prion disease), the designation is PCA-plus

⧫ Level 3

◊ Designation of disease pathology causing PCA, for example, a patient is designated as having PCA-AD when Alzheimer disease (AD) biomarkers are present and as having definite PCA-AD with autopsy evidence

^aData from Crutch SJ, et al, Alzheimers Dement.²³

^bBalint syndrome features.

^cGerstmann syndrome features.

Evaluation

A targeted history and a tailored neurologic examination are necessary for the detection of several of the core features including environmental agnosia, alexia, dysgraphia, dyscalculia, oculomotor apraxia, optic ataxia, and left-right confusion. Other features demand specific assessment tools to enhance their detection, and this necessity extends to formal neuropsychological testing, which is important for an accurate diagnosis. Valuable tools and visual stimuli that can be used during an office visit or formal neuropsychological testing are reviewed in TABLE 2-5.^28–35 The PCA assessment working group provided practical approaches to the assessment of core features.³⁶ For the detection of a homonymous visual field defect when concern for PCA arises, computerized threshold visual field testing is recommended, which is done routinely in eye clinics and is a more sensitive method than confrontation visual fields (ie, finger-counting in each quadrant). Homonymous defects can be detected in up to 62% of patients with PCA.³⁷

TABLE 2–5.

Examples of Screening Stimuli and Tools to Capture Core Features of Posterior Cortical Atrophy Related to Higher-order Visual Functions

Screening tools	Format	Targeted posterior cortical atrophy core features	Source
Poppelreuter-Ghent overlapping figures	Images of shapes or objects that overlap each other; the complexity of the figure can Increase with more images and images at noncanonical angles	Not specific to one feature but relies on figure-ground discrimination and captures space and object perception impairments	Della Sala et al, 199528
Judgment of line orientation	Two angled lines are shown with a set of lines oriented at different angles within a semicircle; the task is to match the angle of each of the two lines to one of the angled lines arranged in the semicircle	Space perception impairment	Benton et al, 198329
Navon figures	A large letter or shape made up of smaller letters or shapes; the task is to recognize both the global and the local shapes	Simultanagnosia	Morris et al, 202130
Cookie Theft picture	Drawing of a kitchen scene from the Boston Diagnostic Aphasia Examination and used as part of the National Institutes of Health (NIH) stroke scale; of note, first published in 1972, there is a call for retiring this image for a more appropriate image that is inclusive and nonbiased	Space and object perception impairments; simultanagnosia	NIH31
Design copy tests	Simple: intersecting pentagons Complex: Rey-Osterrieth Complex Figure; the task is to copy the design	Constructional apraxia	Simple: Mini-Mental State Examination32 Complex: Zhang et al, 202133
Visual scanning or visual search tests	Letters or figures are embedded with other letters or figures; the task is to identify the target letter or figure	Space perception impairments	Kaplan et al, 199134
Fragmented letters	Letters that are visually degraded with missing parts; the task is to identify the letter with the partial information	Object perception impairment	Addenbrooke’s Cognitive Examination-III35

CASE 2–1.

A 52-year-old woman, a financial analyst, presented with a year-long history of increasing difficulties in tasks requiring planning and organization. Both her husband and colleagues observed a marked decline in her performance, particularly when adapting to new software programs or mentally manipulating numbers. She began struggling with the sequential steps of vacation planning. In general, her challenges often manifested as “forgetting” procedural steps in tasks, and her family blamed these errors on memory loss. This led to frequent frustration with, and subsequent avoidance of, activities she once enjoyed, such as hosting dinner parties. The emotional strain also manifested as increased irritability and anxiety, particularly when confronted with decision making or planning tasks.

An evaluation by her primary care provider led to a brain MRI, which was interpreted as normal age-related changes, and the quantitative volumetrics of the hippocampi were in the normal range. A neuropsychological assessment underscored significant impairments in executive functions, notably in working memory and cognitive flexibility. However, her subpar performance across various tests, including performance validity measures, raised concerns about secondary gain in the setting of work-related issues. In addition, the observed degree of impairment seemed inconsistent with her ability to drive and manage many daily activities without notable difficulties. This discrepancy resulted in a provisional diagnosis of depression and anxiety, with recommendations to destress and reduce her work hours.

Subsequently, she was referred to a subspecialist, who identified mild atrophy in the parietal lobe on the brain MRI and arranged for further testing. Given her young age of onset, executive predominant cognitive dysfunction, normal-appearing hippocampi, and reports of changes in her personality by her family, the differential diagnosis included frontotemporal dementia (FTD). However, it was noted that executive dysfunction could be at the root of her clinical symptoms, and her changes in personality and memory could be understood as a secondary manifestation of the changes in executive abilities. Abnormalities observed on brain fludeoxyglucose positron emission tomography (FDG-PET) supported this interpretation (FIGURE 2-1), demonstrating pronounced hypometabolism, especially in the parietal and frontal regions linked with working memory, but medial frontal regions linked to emotion, personality, and motivations were normal, as was medial temporal lobe metabolism. Further, spinal fluid analysis revealed low amyloid-β (Aβ) levels, but phosphorylated tau (pTau) levels remained within the normal range, with an elevated pTau to Aβ ratio.

Considering her clinical manifestations, neuropsychological findings, spinal fluid analysis, and imaging outcomes, a diagnosis of dysexecutive Alzheimer disease (AD) was established. Therapeutically, she was started on a cholinesterase inhibitor, with discussions about N-methyl-d-aspartate (NMDA) receptor antagonist and amyloid-lowering monoclonal antibodies. Although she believed her driving was unaffected, she was counseled to stop driving given her executive cognitive impairment. The patient and her family were educated regarding available social support resources, genetic counseling, clinical trials, and current research. A social worker also supported her in addressing disability and associated concerns. Additionally, she and her family participated in counseling sessions to arm them with coping strategies tailored to her executive function challenges, encompassing environmental modifications, checklist usage, multitasking avoidance, understanding anxiety’s role in executive function impairment, and capitalizing on familiar routines.

COMMENT

This case exemplifies the nuanced nature of atypical AD, specifically dysexecutive AD, by highlighting the importance of nonmemory cognitive symptoms as primary manifestations. The patient’s impaired planning, organization, and decision-making abilities, alongside preserved routine functions, illustrate the characteristic executive dysfunction of dysexecutive AD. Her perceived memory loss underscores a common misattribution in which difficulties in daily tasks due to executive dysfunction are mistakenly blamed on memory impairment. This case further demonstrates how behavioral changes, such as increased irritability and anxiety, are secondary to impaired executive function rather than primary behavioral issues as seen in behavioral variant AD and behavioral variant FTD. This case also illustrates the pitfalls of neuropsychological assessment and how performance validity testing can be misinterpreted in the context of dysexecutive AD. Younger age of onset and job loss are also common and contribute to this misinterpretation. The integrated treatment approach aligns with the principles that atypical AD forms such as dysexecutive AD require a comprehensive and individualized management plan.

As with any cognitive disorder, evaluation for treatable causes and contributing factors should be completed and tailored to the individual patient and to the potential pathologic causes of PCA. CT or MRI of the brain is necessary to rule out structural lesions as the cause of symptoms, and studies such as FDG-PET can reveal focal metabolic changes even in the absence of definitive atrophy. Occipital hypometabolism with relative sparing of posterior cingulate metabolism (ie, the cingulate island sign) is a feature of PCA,³⁸ even in the absence of dementia with Lewy bodies. CASE 2–2 demonstrates a typical presentation of PCA with the progression of symptoms and signs over time and findings on neuroimaging that support the diagnosis.

CSF AD biomarkers for amyloid and tau in PCA are indistinguishable from those for typical AD, as is true for amyloid PET findings.^25,39 However, tau PET imaging highlights distinctive posterior regional deposition patterns that can differentiate among various AD clinical phenotypes, which is akin to the regional metabolic variances observed with FDG-PET, as depicted in FIGURE 2-1.

Progression and Prognosis

All patients with PCA who meet the 2017 criteria will progress to dementia, and the pattern of progression can be heterogeneous. Limited data exist regarding specific longitudinal profiles of progression. However, as a group, impaired posterior functions appear to progress at a faster rate than domains that are initially relatively spared in keeping with the regional progression of cerebral atrophy.⁴⁰ This might explain findings from the limited existing data that reveal a slower progression on measures of global cognitive function (ie, the MMSE) for patients with PCA compared with other atypical AD phenotypes and typical AD. People with PCA are not spared the behavioral changes that occur in the later stages of disease when global impairment and moderate to severe dementia are present. Few studies have evaluated PCA survival, but early evidence shows that survival can range widely with an approximate median survival of 8 to 10 years.³⁶

Management

The mainstay of care is symptom management and patient, family, and care partner education, which should be tailored to meet individual needs and follow the principles of care recommended for patients with AD.⁴¹ Treatment with acetylcholinesterase inhibitors and NMDA receptor antagonists for PCA is a common practice and often recommended,³⁹ but data to guide their use in PCA are not available. Anxiety and depression should be managed appropriately, which can improve daily function. For patients who develop visual hallucinations, rapid eye movement (REM) sleep behavior disorder, parkinsonism, or other signs of Lewy body disease, management and evaluation should be directed appropriately as symptoms emerge. If a patient with PCA qualifies for anti-amyloid therapies, then appropriate treatment with US Food and Drug Administration (FDA)–approved drugs might be indicated. Physicians should follow appropriate use guidelines for decision making regarding anti-amyloid treatment.^19,42 However, the lack of representation of PCA in clinical trials, compounded by the lack of longitudinal outcome measures for PCA, hinders the ability to thoroughly understand the risks and benefits of anti-amyloid therapy for PCA.

LOGOPENIC PRIMARY PROGRESSIVE APHASIA

PPA is a language-based dementia syndrome caused by neurodegenerative diseases that selectively target and erode the language network, usually located in the left hemisphere of the brain.⁴³ The syndromic name, described by Mesulam⁴⁴ in the 1980s, intuitively embeds the key diagnostics features, that is, the PPA diagnosis is made when neurodegenerative disease causes relatively isolated cognitive impairment (ie, primary) that becomes more severe over time (ie, progressive) in the domain of language (ie, aphasia).⁴⁵ Other syndromes in which neurodegenerative disease can extend into the language network include amnestic-onset AD, PCA, progressive dysexecutive syndrome, and behavioral variant FTD; however, the diagnosis of PPA would not apply here as these syndromes have early prominent deficits in memory, higher-order visual functions, executive function, and behavior, respectively.

PPA has created an elegant model for extending our understanding of the language network. For PPA, neurodegeneration is partial and progressive and can occur outside of the boundaries of vascular inputs resulting in presentations distinct from stroke-based aphasia. As such, the characteristic subtypes of stroke-based aphasia do not accurately encompass the language impairment presentations for those with PPA. There are at least three recognized subtypes of PPA, logopenic, agrammatic, and semantic, defined by both the language features that are impaired and those relatively spared. Logopenic PPA is characterized by impairments with word retrieval and relative sparing of word comprehension and grammar, whereas agrammatic PPA is characterized by impairments in grammar but relative sparing of word comprehension. Semantic PPA is characterized by impairments in word comprehension and relative sparing of grammar. Each subtype is associated with clinically concordant distributions of peak cortical atrophy (TABLE 2-6⁴⁶). The tripartite subtyping system allows for the classification of some but certainly not all presentations of PPA. A mixed subtype characterized by a combination of comprehension and grammar impairments has been described, as well as unclassifiable presentations.^47,48 As the disease progresses over time, additional impairments in language and eventually other aspects of cognition and behavior may emerge, resulting in a PPA-plus syndrome.⁴⁹ At the individual level, variation in the location and severity of neurodegeneration and pace of progressive decline result in a spectrum of deficits. Thus, the ability to identify subtype distinctions tends to be most prominent during a “goldilocks” period in which impairments are neither too mild to detect nor too severe where deficits may be present in multiple aspects of language.

TABLE 2–6.

Three Subtypes of Primary Progressive Aphasia^a

Variant	Description
Logopenic	Characterized by loss of fluency due to word retrieval failures in spontaneous speech and commonly in naming, with relative sparing of grammar and word comprehension. Current research criteria list repetition as a core required feature; however, reports of mild impairment question whether it is a better fit as an ancillary feature. Peak atrophy is asymmetric in the language-dominant (usually left) hemisphere including the superior temporal gyrus and may extend to the temporoparietal junction
Agrammatic	Characterized by impairments in grammar in language production, which is commonly accompanied by low fluency; however, single-word comprehension is relatively preserved. Naming, repetition, and comprehension of syntactically complex sentences may be impaired. Peak atrophy commonly includes the language-dominant inferior frontal gyrus
Semantic	Characterized by impairment of single-word comprehension, with relatively preserved grammar and repetition with peak atrophy in the language-dominant anterior temporal lobe

^aModified with permission from Mesulam MM, et al, Nat Rev Neurol.⁴⁶ © 2014 Springer Nature Limited.

The primary neuropathologies associated with PPA include AD neuropathologic change or frontotemporal lobar degeneration (FTLD), including tauopathy (FTLD-tau) and FLTD transactive response DNA-binding protein 43 (TDP-43) proteinopathy (FTLD-TDP).^50–52 The neuropathologic entities have shown probabilistic relationships with specific variants of PPA in which the logopenic variant is most commonly associated with AD neuropathologic change (approximately 70%), the agrammatic variant is most commonly caused by FTLD-tau (approximately 60% to 70%), and the semantic variant is most reliably linked to FTLD-TDP (approximately 80%). Historically, the absence of in vivo biomarkers specific to each neuropathic entity gave subtyping prominence as a proxy for determining underlying pathology. The emergence of in vivo biomarkers, including amyloid and tau PET as well as CSF and blood-based biomarkers,⁵³ is transforming this landscape and allowing for greater precision in understanding drivers of disease progression and intervention.

As highlighted earlier, the logopenic variant is commonly but not exclusively associated with AD neuropathologic change. Logopenia refers to a reduced rate of speech output due to word-finding difficulty and was noted as a symptom in the 1982 cases described by Mesulam.⁴⁴ However, it was the last of the three research subtypes to be formally characterized.^54–56 As such, the research literature for logopenic PPA lags relative to the other subtypes, and nuances around the diagnosis, clinical, and anatomic features have been the focus of some systematic reviews since 2019.^57–59 Given that PPA subtypes beyond the logopenic variant can be associated with AD neuropathologic change and because of its later historical description, the remainder of this section provides relevant insights about logopenic PPA, the syndrome of PPA, PPA with elevated amyloid biomarkers, and autopsy-proven PPA due to AD neuropathologic change (sometimes referred to as the aphasic variant of AD).

Clinical and Anatomic Features and Disease Progression

Ascertaining the diagnosis of logopenic PPA can be complex and requires clinical acumen rather than strict reliance on test scores. In the milder stages, fluency may appear normal if the individual has identified strategies to use simpler word choices, phrases, or circumlocution. However, when pressed to provide precise labels or words of lower frequency in conversation or on tests of object naming, hesitations and errors become more prominent. Atrophy on MRI may be subtle early on, and hypometabolism on PET may not be appreciated,⁵⁷ which can contribute to uncertainties and delays in making the diagnosis of PPA.

When language impairment is more prominent, it can be challenging for the clinician to accurately assess the presence or absence of impairment in nonlanguage domains (eg, attention, memory, executive functioning) because many neuropsychological instruments rely heavily on intact language processing for their instructions (eg, Wisconsin Card Sorting Test) or in the stimuli or responses themselves (eg, episodic memory tests of word-list recall or story recall). Using assessments with a lower degree of language dependency and careful ascertainment of patient and family reports from daily life can provide important insights into areas of impairment versus preservation.⁶⁰

Although asymmetric atrophy is a characteristic feature of PPA associated with AD, atrophy tends to be more widespread in PPA with elevated amyloid biomarkers than without, and damage to the contralateral hemisphere tends to emerge earlier than in PPA without elevated amyloid biomarkers (FIGURE 2-6).^61–63 More diffuse atrophy in the logopenic variant has been noted by others.⁶⁴ Functional decline parallels the atrophy findings, where impairment of activities of daily living is more prominent and encompasses more aspects of daily living in PPA with elevated amyloid biomarkers than without.⁶⁵ Decline in naming is more closely linked to atrophy than tau PET burden.⁶³ There is still considerable unexplained variability at the individual level for both atrophy and functional decline.⁶¹ For example, there are reports of patients with logopenic PPA with relatively rapid decline in cognition and daily function whereas others can maintain daily living activities that do not depend on language for more than 10 years. Such variation makes it difficult to provide families with precise prognostic information in the clinic.

FIGURE 2–6.

Asymmetric atrophy and amyloid and tau positron emission tomography (PET) burden in primary progressive aphasia (PPA) associated with biomarker-positive Alzheimer disease neuropathologic change. A (top), An example of asymmetric atrophy patterns when language impairment is relatively mild for PPA with elevated amyloid-β (Aβ) biomarkers (PPA^Aβ+). Note, contralateral right hemisphere involvement may be present. In the moderate to severe stages, atrophy is more diffuse and can include contralateral right hemisphere involvement for patients with PPA^Aß+ (A, middle) relative to nonsemantic PPA without elevated Aβ biomarkers (PPA^Aß−) (A, bottom). Red and yellow indicate significant cortical thinning patterns for PPA relative to controls, with yellow being more severe than red. Aβ (B) and tau (C) PET can show relatively focal asymmetric burden in the language network. For amyloid PET, red indicates the most severe amyloid burden. For tau PET, green coloring indicates elevated burden.

SUVR = standardized uptake value ratio.

Panel A (middle and bottom) modified with permission from Rogalski EJ, et al, Alzheimers Dement.⁶¹ © 2019 Alzheimer’s Association. Panel B modified with permission from Martersteck A, et al, Acta Neuropathol Commun.⁶² © 2022 Springer Nature. Panel C modified with permission from Martersteck A, et al, Alzheimers Dement.⁶³ © 2021 Alzheimer’s Association.

Amyloid and tau PET imaging, CSF analysis, and emerging blood-based biomarkers provide relevant information for discerning the likelihood of AD neuropathologic change as a contributing factor.^53,66–68 The spatial patterns of impairment offered by imaging support the diagnostic process and also inform our understanding of selective vulnerability patterns that are unique and shared across neurodegenerative syndromes caused by AD neuropathologic change. For example, the selective vulnerability of the language network relative to the memory network is highlighted by both atrophy (measured by structural MRI) and functional perturbations (measured by resting-state functional MRI [fMRI]).^69–71 Likewise, both amyloid and tau PET imaging can show relatively focal asymmetric burden in the language network (FIGURE 2-6).^62,72,73 These features further distinguish PPA associated with suspected AD neuropathologic change from amnestic AD.

CASE 2–2.

A 57-year-old man with 16 years of education presented to his optometrist with difficulty reading for 8 months, particularly when reviewing spreadsheets. A pair of reading glasses was prescribed but did not improve his ability to read. His primary care doctor ordered a head CT without contrast, which was normal. His medical history was remarkable for atrial fibrillation and a pacemaker. Escitalopram 10 mg/d was prescribed and helped with anxiety, but his visual symptoms progressed, and 1 year after the initial presentation, he was sent for further evaluation. At that time, he reported that he relied on others to read important material to him, and he had recently driven up an embankment after he misread a temporary construction road sign indicating a lane merge. His other medications included warfarin 2.5 mg/d. He denied visual hallucinations, symptoms of rapid eye movement (REM) sleep behavior disorder, and tremors. He felt slightly forgetful.

Examination revealed 29/30 points on the Mini-Mental State Examination (MMSE) (missing one point for intersecting pentagon copy, FIGURE 2-2). He had normal visual acuities and fundi. He was able to count fingers accurately in all four quadrants but initially struggled. His reading was slow with frequent backtracking. He was unable to identify all four objects on a Poppelreuter-Ghent overlapping figure. The remainder of his neurologic examination was normal. A formal neuropsychological evaluation revealed significant higher-order visual dysfunction with relative sparing of other domains (ranging from normal to mild impairment) with relatively poor performance (moderate to severe) on measures dependent on visual stimuli. A Humphrey threshold visual field test (FIGURE 2-3) revealed a left homonymous hemianopia (inferior more than superior) and fludeoxyglucose positron emission tomography (FDG-PET)/CT (FIGURE 2-4) revealed right more than left parieto-occipital hypometabolism and a cingulate island sign. A lumbar puncture was not performed because of the risks associated with discontinuing warfarin. He met the 2017 criteria for posterior cortical atrophy (PCA) with the demonstration of 5 of 16 PCA core features (deficits in space perception, object perception, and constructional praxis; alexia; and acalculia). Six years later, CT revealed significant posterior atrophy, and he had developed the remainder of the 11 PCA core features and mild to moderate dementia (FIGURE 2-5). Nine years later, he had moderate to severe dementia with no clinical features suggestive of non–Alzheimer disease pathology.

COMMENT

This case exemplifies the early age at presentation, reading difficulty as a common symptom at presentation, and the importance of formal neuropsychological testing that can reveal impairments for posterior functions and relative sparing of memory and other domains. Furthermore, structural brain scans, such as head CT, can appear normal early in the course whereas metabolic scans, such as FDG-PET, can reveal significant abnormalities. As noted in the 2017 PCA criteria, imaging features are supportive of the diagnosis and can include abnormal structural, metabolic, or perfusion findings in the posterior regions.

FIGURE 2–2.

A copy of intersecting pentagons drawn by the patient in CASE 2–2.

FIGURE 2–3.

Visual field testing for the patient in CASE 2–2. The patient’s left eye and right eye Humphrey 24–2 threshold visual field results show left homonymous hemianopia, inferiorly worse than superiorly.

FIGURE 2–4.

Imaging findings for the patient in CASE 2–2. Fludeoxyglucose positron emission tomography (FDG-PET)/CT shows right more than left parieto-occipital hypometabolism (A, red arrows) and a cingulate island sign (B, orange arrow).

FIGURE 2–5.

Imaging findings for the patient in CASE 2–2. Head CT at presentation (A) and 6 years later (B), which shows right more than left posterior cortical atrophy.

Risk Factors

A full understanding of risk factors for PPA, including logopenic PPA, remains largely elusive. The ε4 allele of APOE is an important risk factor for amnestic dementia associated with AD neuropathologic change but does not show the same association with PPA.^74–76 A history of learning disabilities for affected individuals or their first-degree family members has been demonstrated to be higher in those living with a diagnosis of PPA than in those with other dementias and a control population, providing a potential hint for selective vulnerability of the language network.^75,77,78

Care and Interventions

As a younger-onset dementia (average age of onset is younger than 65 years old) with no definitive cure, PPA brings unique and complex communication, family, and psychosocial challenges with an estimated individual economic burden twice that of AD dementia.^79–81 A team approach to care including but not limited to neurology, social work, neuropsychology, neuropsychiatry, speech-language pathologists, and occupational therapy may be appropriate with consideration of nonpharmacologic and pharmacologic interventions. The needs and contributions of the care team may shift over time as the disease progresses and brings new profiles of cognitive, behavioral, and daily life challenges.

Nonpharmacologic interventions, care, counseling, and support programs may consider strategies to address care partner burden, family relationships, and communication breakdowns, as well as language impairment, to maximize the quality of life for people with PPA and their care partners. Speech and language interventions have been the most common nonpharmacologic intervention for those living with logopenic PPA but have historically lacked evidence of efficacy.^82–86 This landscape is changing. One example is the recent completion of the Communication Bridge-2 international, Phase 2, Stage 2, randomized, parallel-group, active-control, clinical trial delivered virtually within a telehealth service delivery model to people with PPA and their communication partners.⁸⁷ The trial enrolled 95 PPA participants, each of the three subtypes were represented, and results are expected in late 2024 or early 2025. The intervention included a custom web application that allowed for asynchronous activities outside of sessions. Additional trials are emerging, including an efficacy trial of a multicomponent and dyadic intervention relative to an impairment-focused intervention.⁸⁸

The National Institutes of Health Phase and Stage model, the Readiness Assessment for Pragmatic Trials tool, and initiatives such as the National Institute on Aging, IMbedded Pragmatic Alzheimer’s disease (AD) and AD-Related Dementias (AD/ADRD) Clinical Trials (IMPACT) Collaboratory (impactcollaboratory.org/) deliver key frameworks and resources to assist in developing nonpharmacologic clinical interventions to successful pragmatic clinical trials and eventual dissemination.⁸⁹ Training fellowships, such as the Institute on Methods and Protocols for Advancement of Clinical Trials in ADRD (IMPACT-AD; impact-ad.org/) are creating training opportunities to increase the multidisciplinary expertise in pharmacologic and nonpharmacologic AD-related dementia clinical trials.

PPA has been largely overlooked in the pharmacologic landscape, historically driven by the lack of in vivo biomarkers. For example, cholinesterase inhibitors may have limited benefit for patients with PPA with AD neuropathologic change but require further trials in which patients are allocated by suspected underlying pathology. Limited and inconclusive data also exist for memantine, an NMDA receptor antagonist, and for medications to manage symptoms such as depression. Disease-modifying treatments for AD neuropathologic change have been approved. However, individuals with PPA and elevated AD biomarkers were less frequently included in the trials that led to the approval of these treatments.⁹⁰ Historical contributors to the exclusion of AD-biomarker-positive PPA individuals include the absence of clear guidelines regarding appropriate outcome measures, lack of syndrome awareness, and access to biomarker data. The younger age of onset for logopenic PPA relative to late-onset AD offers unique opportunities to examine the effectiveness of pharmacologic treatments targeting the AD pathophysiologic process in a setting where there is a lower likelihood of copathology and other age-related comorbidities. Advancements over the past decade provide a prime opportunity to advocate for the inclusion of atypical forms of AD in clinical trials through judicious use of in vivo biomarkers. The use of robust pharmacologic or combination (pharmacologic and nonpharmacologic) randomized controlled trials for PPA with AD neuropathologic change offers a promising opportunity for advancing treatment options for PPA.

HEALTH DISPARITIES

Because uncommon dementia syndromes can have a younger age of onset, the diagnostic journey for patients with atypical presentations of AD often requires years rather than months with visits to multiple clinicians who may be nonlocal. This process can require extraordinary persistence, perseverance, and financial resources from the family, which widens the gaps in access to care, especially for those with low literacy, few resources, and different ethnocultural norms. The low diversity in clinical settings also contributes to the exceedingly low diversity in research populations. In the United States, AD is up to twice as likely to occur in non-Hispanic Black people and up to 1.5 times as likely to occur in Hispanic older adults compared with White older adults, yet both populations are far less likely to be included in research than White older adults.⁹¹ Highly educated, White research cohorts are the norm rather than the exception. Calls to action and efforts to increase the diversity of dementia cohorts are mounting, including collaborative efforts from multisite research studies (eg, Longitudinal Early-Onset Alzheimer’s Disease Study), the National Institute on Aging–funded Alzheimer’s Disease Research Centers, registries, as well as nongovernmental organizations (eg, the Alzheimer’s Association and the Association for Frontotemporal Degeneration). Addressing these health equity challenges is paramount to obtaining accurate epidemiologic data, understanding of disease, and tailored care plans.

CONCLUSION

Atypical (nonmemory) presentations of AD present distinct clinical challenges that require a heightened level of awareness and a customized approach to both diagnosis and management. Comprehensive diagnostic strategies and targeted clinical interventions for the unique symptoms of atypical AD can ensure prompt diagnosis and the best possible care. Advances in blood-, fluid-, and imaging-based AD biomarkers provide enormous opportunities for early diagnosis and participation in treatment trials for patients with atypical AD.

KEY POINTS.

• Atypical forms of Alzheimer disease (AD) are defined by predominant symptoms in nonmemory cognitive domains.

• Atypical presentations of AD have unique pathophysiologic pathways, diagnostic indicators, and management requirements.

KEY POINT.

• Dysexecutive AD and behavioral variant AD are distinct presentations, and the term frontal AD should no longer be used.

KEY POINTS.

• Patients with dysexecutive AD have impaired planning, organization, and decision making.

• Compromised working memory and cognitive flexibility manifest in impairments of cognitively effortful tasks while automatic activities are more preserved in patients with dysexecutive AD.

• In the setting of dysexecutive AD, memory loss can be the focus of the reported clinical symptoms as many are unfamiliar with the role of executive function for daily tasks.

• For patients with dysexecutive AD, changes in behavior are secondary to the impact of impaired executive function and are not themselves the primary driver of dysfunction.

• In behavioral variant AD, changes in behavior are the root cause of impaired daily functioning, similar to behavioral variant frontotemporal dementia (FTD).

• Amyloid and tau biomarkers are positive in dysexecutive AD and behavioral variant AD, but the spatial distribution of tau positron emission tomography (PET), fludeoxyglucose (FDG)-PET, and MRI changes more closely align with the unique functional anatomy of these two syndromes.

• Function of parietal and frontal regions related to working memory is commonly abnormal in patients with dysexecutive AD.

KEY POINTS.

• Function of medial frontal regions related to behavior is commonly abnormal in patients with behavioral variant AD and behavioral variant FTD.

• Failure on performance validity testing commonly leads to misinterpretations of neuropsychological testing in the setting of dysexecutive AD.

• In patients with dysexecutive AD, the hippocampus is commonly spared and appears normal on MRI.

KEY POINTS.

• The management of dysexecutive AD and behavioral variant AD combines pharmacotherapy with tailored nonpharmacotherapeutic approaches.

• AD is the most common underlying pathology accounting for posterior cortical atrophy (PCA) syndrome.

• PCA is characterized by higher-order visual dysfunction with relative sparing of memory and other cognitive domains, judgment, and insight early in the presentation.

• People with PCA often present to an eye care provider with significant visual concerns that can go unaccounted for in the presence of a normal eye examination.

• Difficulty reading is a common presenting concern for patients with PCA.

• Delays in diagnosis of PCA can occur because of the nature of initial symptoms (ie, visual and not memory), the characteristic young onset (65 years or younger), preservation of insight, and adequate performance on measures of global cognition at presentation.

• The 2017 PCA consensus criteria specify that 3 of 16 core features must be met for the diagnosis of PCA. Features belong to occipitoparietal and occipitotemporal visual pathways including all elements of Balint and Gerstman syndromes.

KEY POINTS.

• Neuroimaging characteristics of PCA are supportive of the diagnosis and include posterior findings of cortical atrophy on MRI, hypometabolism on FDG-PET, or hypoperfusion on single-photon emission computed tomography (SPECT), as are posterior and occipital tau PET abnormalities.

• Patients with PCA are classified as having PCA-pure when they have no other associated syndromes or PCA-plus when PCA occurs in association with another clinical syndrome such as corticobasal syndrome.

• It is important to keep PCA features in mind while conducting a focused history and examination and using specific visual assessment tools in the office and during formal neuropsychological evaluation.

• Recently published recommendations for the clinical assessment of PCA provide several options for assessment tools and visual stimuli for detecting the unique core features of PCA.

• AD CSF biomarkers for PCA are indistinguishable from profiles for typical AD, whereas MRI, FDG-PET, and tau PET can reveal patterns of posterior atrophy, hypometabolism, and tau deposition that reflect the PCA clinical phenotype.

KEY POINTS.

• Symptom management for patients with PCA should follow that established for AD and be guided by patient needs. It is unknown whether the risks or benefits of anti-amyloid therapies are different for the PCA phenotype.

• In vivo biomarkers play a key role in the diagnosis of AD because there is no one-to-one relationship between clinical phenotype and underlying neuropathology.

• Logopenic primary progressive aphasia (PPA) is characterized by word retrieval failures, which can occur in spontaneous speech or confrontation naming. These deficits may be obscured if the individual is adept at using simplified words or circumlocution.

• Many neuropsychological instruments developed to assess nonlanguage domains rely on preserved language for successful performance, which can make it challenging for clinicians to ascertain the presence or absence of impairment when aphasia is prominent.

KEY POINTS.

• The ε4 allele of APOE is an important risk factor for amnestic dementia associated with AD neuropathologic change but does not show the same association with PPA.

• Nonpharmacologic interventions, care, counseling, and support programs may address care partner burden, family relationships, communication breakdowns, as well as language impairment to maximize quality of life for people with PPA and their care partners.

• The diagnostic journey for patients with PPA tends to take years rather than months.