Corticobasal degeneration: a pathologically distinct 4R tauopathy

Abstract

Corticobasal degeneration (CBD) is a rare, progressive neurodegenerative disorder with onset in the 5^th to 7^th decade of life. It is associated with heterogeneous motor, sensory, behavioral and cognitive symptoms, which make its diagnosis difficult in a living patient. The etiology of CBD is unknown; however, neuropathological and genetic evidence supports a pathogenetic role for microtubule-associated protein tau. CBD pathology is characterized by circumscribed cortical atrophy with spongiosis and ballooned neurons; the distribution of these changes dictates the patient’s clinical presentation. Neuronal and glial tau pathology is extensive in gray and white matter of the cortex, basal ganglia, diencephalon and rostral brainstem. Abnormal tau accumulation within astrocytes forms pathognomonic astrocytic plaques. The classic clinical presentation, termed corticobasal syndrome (CBS), comprises asymmetric progressive rigidity and apraxia with limb dystonia and myoclonus. CBS also occurs in conjunction with other diseases, including Alzheimer disease and progressive supranuclear palsy. Moreover, the pathology of CBD is associated with clinical presentations other than CBS, including Richardson syndrome, behavioral variant frontotemporal dementia, primary progressive aphasia and posterior cortical syndrome. Progress in biomarker development to differentiate CBD from other disorders has been slow, but is essential in improving diagnosis and in development of disease-modifying therapies.

Introduction

Corticobasal degeneration (CBD) is a rare, progressive neurodegenerative 4-repeat (4R) tauopathy (Box 1) associated with a wide variety of motor, sensory, behavioral and cognitive symptoms that can also occur in other conditions. This clinical heterogeneity and lack of specific features make CBD difficult to diagnose before the patient’s death.¹ Owing to the wide spectrum of clinical presentations of CBD, neuropathological examination is required for a definitive diagnosis. Little is known about the epidemiology of CBD, and the available studies offer only crude estimates of its incidence. CBD has an estimated prevalence of 4.9–7.3 cases per 100,000 individuals² and an annual incidence rate of 0.02 cases per 100,000 individuals on the basis of an Eastern European and Asian population study.³ Disease onset is typically in the 5^th to 7^th decades of life, with an average disease duration of 7 years. No sex bias has been observed. The pathogenesis of CBD is unknown, and no effective therapy is currently available to treat this disease or even to slow its progression.

Box 1 |. Tauopathies.

When tau pathology is considered to be the main contributing factor to neurodegeneration, the disorder is classified as a tauopathy. Tauopathies are characterized by their distinct biochemical protein isoform profiles, with regard to the presence of either three or four repeats (3R or 4R, respectively) in the tau microtubule-binding domain.

• In Alzheimer disease, tau tangles are composed of equimolar 3R and 4R isoforms

• Corticobasal degeneration, progressive supranuclear palsy and many variants of frontotemporal dementia and parkinsonism linked to chromosome 17 (and associated with MAPT mutations) have predominantly 4R tau pathology¹⁰⁵

• Pick disease has predominantly 3R tau pathology,¹⁰⁶ and alterations from normal tau isoform levels contribute to its pathogenesis¹⁰⁷

CBD was originally described as a movement disorder, but over the past few decades, as more clinical and pathological studies have been performed, the disease is now considered a disorder of both movement and cognition. CBD was first reported in 1968 by Rebeiz et al., who described clinical and neuropathological features of three patients in whom the disorder was termed corticodentatonigral degeneration with neuronal achromasia.⁴ Gibb and Marsden coined the term CBD in 1989,⁵ and this pathology has also been referred to as corticobasal ganglionic degeneration by some authors.^6–8 The initial descriptions of CBD emphasized progressive movement abnormalities, which began as a unilateral slowing of voluntary movements with concurrent involuntary movements that eventually became generalized but remained asymmetric. In their paper, Rebeiz et al. noted, “…[the patients’] intellect was relatively preserved until the end.”⁴ However, their well-documented report describes cognitive as well as behavioral dysfunction in all three patients. On neuropathological examination, the researchers found distinct features common among the three patients, which included frontoparietal atrophy with neuronal loss, gliosis, ‘achromatic’ or ballooned neurons, and pigment loss in the substantia nigra.

The classic clinical movement disorder described by Rebeiz et al. in 1968 is now referred to as corticobasal syndrome (CBS), and its cardinal features are progressive asymmetric rigidity and apraxia with additional signs of cortical and basal ganglia involvement, such as aphasia and dystonia.^5,8 Although CBS has some predictive value for finding CBD pathology, autopsy studies demonstrate considerable heterogeneity in the pathology of patients who present with CBS.⁹ Over the years, clinical and pathological studies have highlighted several focal cortical syndromes associated with CBD pathology and, therefore, defining widely accepted clinical diagnostic criteria for research purposes is nearly impossible.^7,10–12 Tau pathology is also found in other neurodegenerative disorders, such as Alzheimer disease (AD)¹³ and progressive supranuclear palsy (PSP), and as tau-directed therapies undergo clinical trials, determining the correct underlying pathology in patients with CBS has become of more than merely academic interest.

This Review focuses on the range of clinical presentations associated with CBD pathology and the latest advances in our understanding of its pathogenesis. We describe progress in the development of therapeutic agents for CBD and briefly outline efforts to develop biomarkers that can distinguish tauopathies from tau-unrelated degenerative disorders. The latter will become increasingly important for designing future pharmacological trials in this rare disorder.

Clinical syndromes associated with CBD

The clinical syndrome in patients with neurodegenerative diseases is a representation of the topography of the lesions but not necessarily the nature of the underlying pathology. Since CBD is a disease with prominent focal cortical atrophy and variable subcortical pathology, patients with CBD can exhibit a range of distinct clinical syndromes, including CBS, Richardson syndrome (the most common clinical presentation of PSP), progressive aphasia, frontotemporal dementia (FTD), posterior cortical atrophy (PCA), and dementia with features similar to AD (Table 1).^6,14–23 Although diagnostically far from perfect, clinical features can help to distinguish CBD from its so-called mimickers in many patients (Box 2).

Table 1 |.

Clinical heterogeneity of pathologically confirmed CBD

Reference	Female:male ratio	Age at onset/disease duration (years)	Common features at onset (% of patients)					Clinical diagnosis* (% of patients)
			Limb apraxia	Memory impairment	Behavioral changes	Gait disturbance	Speech disorder	CBS	AD or PD	FTD	RS
Schneider et al. (1997)18	10:1	67/7.9	64	18	9	9	9	64	27	0	0
Litvan et al. (2000)26	14:13	64/7.0	37	33	33	26	22	22	30	7	4
Murray et al. (2007)16	8:7	61/5.4	33	33	40	27	40	40	13	33	13
Ling et al. (2010)17	9:10	66/6.0	26	37	26	63	16	26	10	5	42
Totals	41:31	65/6.6	38	32	29	33	22	33	21	11	15

^*The total does not equal 100% because some series had final diagnoses other than the diseases listed here. Abbreviations: AD, Alzheimer disease; CBD, corticobasal degeneration; CBS, corticobasal syndrome; FTD, frontotemporal dementia; PD, Parkinson disease; RS, Richardson syndrome.

Box 2 |. Diagnosis of corticobasal degeneration.

Several clinical features can help to distinguish corticobasal degeneration from its mimickers.

Progressive supranuclear palsy

• Postural instability with falls usually occurring within the 1^st year of disease onset

• Vertical (especially downward) supranuclear gaze palsy

• Usually symmetric parkinsonism, although cases of asymmetric parkinsonism associated with progressive supranuclear palsy have been described

• Absent cortical features, such as myoclonus and apraxia

Parkinson disease

• Marked and sustained benefit from dopaminergic therapy

• Resting tremor

Dementia with Lewy bodies

• Visual hallucinations (not drug-related)

• Rapid eye movement sleep behavior disorder

• Fluctuating cognition

Multiple system atrophy

• Pronounced dysautonomia

• Cerebellar ataxia

• Disruption of rapid eye movement sleep behavior

Alzheimer disease

• Initial prominent short-term memory impairment

Creutzfeldt–Jakob disease

• Rapid disease progression with a disease course <1 year

Corticobasal syndrome

CBS, the classic motor presentation of CBD as first described by Rebeiz and co-workers,⁴ accounts for approximately 50% of autopsy-confirmed cases (D. W. Dickson, unpublished observations from the CurePSP Brain Bank). Core features of CBS include levodopa-unresponsive parkinsonism, asymmetric akinesia, and rigidity, accompanied by other cortical and basal ganglia dysfunction, such as limb and oculomotor apraxia, cortical sensory deficits, dystonic posturing of a limb, myoclonus, and alien limb phenomenon.^10,24 These cortical signs and symptoms are attributed to a predominant frontal and parietal involvement. Patients can present with either motor or cognitive complaints or with a combination of the two. In a combined series of 27 individuals with CBD confirmed by autopsy, the most frequent features at onset of the disease were limb clumsiness in 37% of the patients, followed by memory loss and behavioral changes in another 33%.^19,25,26 A separate longitudinal study of 15 individuals with CBD reported an exclusively cognitive disturbance presentation in about half of the patients; another 20% had an exclusively motor abnormality presentation, and the remaining 30% had both motor and cognitive complaints.¹⁶ Owing to the remarkably low sensitivity and positive predictive value of these features for diagnosing CBD, studies of the neuropsychiatric features of patients with a clinical diagnosis of CBD should be interpreted with caution.

The accuracy of clinical diagnosis of CBD does, however, improve with disease progression. In a review of 27 autopsy-confirmed CBD by Litvan and co-workers, 22% of the patients had CBS as the initial diagnosis, increasing to 41% before death.²⁶ Similarly, in a study by Kertesz and co-workers of 13 autopsy-confirmed cases of CBD, 35% had CBS as an intial diagnosis, increasing to 54% before death.²⁷ Being able to predict CBD pathology as early in the disease course as possible is of great importance owing to the growing number of tau-directed therapies currently being developed.

The CBD mimickers

The CBD mimickers, which resemble CBS but have different underlying pathologies, include AD, PSP and frontotemporal lobar degeneration with TDP-43 pathology (FTLD–TDP, sporadic and familial), among others.^9,28–32 Whitwell et al. reported that although a region of atrophy was common to all pathologies that result in CBS, CBS caused by FTLD-TDP was associated with notable frontotemporal gray matter atrophy; CBS caused by AD was associated with prominent temporoparietal atrophy; and in CBS caused by CBD or PSP gray matter atrophy was focused in the premotor cortex with additional involvement of superior posterior frontal cortex.³³

Richardson syndrome

Richardson syndrome is the typical presentation of PSP,^14,34 and is characterized by severe postural instability leading to falls soon after disease onset, vertical supranuclear gaze palsy, dysarthria, dysphagia, and frontal cognitive disturbances.³⁵ The occurrence of PSP presenting as CBS has been attributed to either greater cortical tau pathology than is observed in the typical PSP presentation³⁶ or an asymmetrical tau burden.³⁷ A study published in 2010, which assessed differences in cortical atrophy in patients with CBS by use of MRI, showed that different patterns of atrophy are associated with specific histopathological abnormalities.³³ As discussed below, CBD and PSP were originally believed to be distinct clinical and pathological entities, but as additional studies have been performed, their overlapping features have become increasingly apparent. When patients with CBD present with bilateral motor disturbances and early frontal dementia, they are frequently misdiagnosed as having PSP since both disorders involve levodopa-unresponsive parkinsonism and can share several other clinical features.^38,39 Asymmetry of clinical signs and symptoms might have been overemphasized in the CBD literature, as several studies of autopsy-confirmed cases have reported the occurrence of symmetric motor disturbances.^17,38,40,41

Patients with CBD can develop abnormal eye movements similar to those seen in patients with PSP. Decreased vertical saccadic velocities preceding vertical supranuclear gaze palsy are typically seen in patients with PSP, whereas increased saccadic latencies with preserved velocity are seen in patients with CBD, and the horizontal and vertical planes are similarly affected.^42–44 In 2010, Ling et al. described clinical features of patients with CBD presenting with Richardson syndrome and confirmed that differences in these eye movement abnormalities are diagnostically helpful in differentiating CBD from PSP.¹⁷

Postural instability and falls occurring within the first year of symptom onset are indicative of PSP.⁴⁵ By contrast, if patients with CBD do experience falls, they usually occur late in the disease course; when the patient has asymmetric motor disturbances, falls tend to incline towards the affected side of the body. Few studies have compared the neuropsychiatric profiles of patients with CBD and PSP. In a clinical series, Cummings and Litvan used the Neuropsychiatric Inventory to assess behavioral symptoms in patients with dementia and found that depression and irritability were notably more frequent in patients with CBD than in those with PSP. The latter, by contrast, exhibited an increased frequency of apathy.⁴⁶

Patients with pathologically confirmed CBD and PSP show differing patterns of brain atrophy on MRI.⁴⁷ When grouped according to their presenting feature (extrapyramidal versus cognitive impairment), patients with CBD who had an extrapyramidal presentation had greater cortical atrophy and less brainstem atrophy compared to patients with PSP who had an extrapyramidal presentation. Among patients who presented with cognitive impairment, those with CBD had substantially greater atrophy of cortical and subcortical gray matter, and less atrophy of subcortical white matter, than those who had PSP.⁴⁷

Behavioral variant frontotemporal dementia

Patients who present with behavioral variant FTD (bvFTD) exhibit marked behavioral and personality changes, including apathy, disinhibition, perseveration, executive dysfunction, obsessive–compulsive behaviors and dramatically impaired insight.⁴⁸ In approximately 50% of such patients, the underlying pathology is tau-positive (PSP, CBD or Pick disease). The other 50% have tau-negative FTLD with ubiquitin immunoreactive inclusions (FTLD-U), which in the majority of cases are TDP-43-positive.^49,50 Although imaging signatures have been identified for specific pathologies, in one large study of patients with bvFTD no specific signature pattern of atrophy could distinguish tau-positive individuals from tau-negative patients.⁵¹ Josephs et al. found that bvFTD was the syndrome least predictive of underlying pathology.⁵² In spite of this finding, some studies have identified clinical and neuropsychiatric features that could help to distinguish tau-positive from tau-negative bvFTD. Hodges et al. found that patients with tau-positive bvFTD were substantially older at the time of diagnosis (mean 61.7 years) than the tau-negative cases (mean 53.7 years).⁴⁹ Forman et al. found that the presence of both a behavioral disorder and aphasia at disease onset was more predictive of FTLD-U than were AD or other tauopathies.⁵⁰ Grossman et al. found that tau-positive FTLD was associated with visual perceptual and spatial difficulties, whereas tau-negative FTLD was linked to language and executive dysfunction.⁵³ In a cluster analysis, Hu et al. found that poor planning and judgment was associated with tau pathology, whereas impaired personal conduct and an absence of dysexecutive symptoms were predictive of FTLD-U pathology.³¹ These observations suggest that longitudinal evaluations of neuropsychological features can be useful in the antemortem differentiation between tau-positive and tau-negative pathology.

The nearly equal distribution of tau-positive and tau-negative underlying pathologies in bvFTD syndrome highlights the considerable overlap in clinical features of CBD, PSP and FTLD. Indeed, some neurologists argue in favor of grouping these three conditions under the single designation of FTD–Pick complex.²⁰ Kertesz et al. have shown that bvFTD followed by the late development of progressive aphasia is usually caused by tau-negative FTLD.²⁷ In another distinct syndrome progression, which is highly predictive of tau-positive pathology, the patient presents with a movement disorder (such as CBS) or the movement disorder develops after primary progressive aphasia. The differential diagnosis, however, is complicated because some patients with CBD develop only minimal CBS features late in the disease course, but present with notable features of FTD.³⁸ The latter CBD phenotype is rarely mentioned in the literature.

Primary progressive aphasia

Primary progressive aphasia is a focal dementia disorder where the dominant deficit is progressive language impairment.⁵⁴ A distinct form of primary progressive aphasia—termed progressive nonfluent aphasia—is most often associated with tau-positive pathology, including CBD, PSP and Pick disease.^22,52,55 The other forms of primary progressive aphasia—semantic dementia and logopenic progressive aphasia—are more often associated with FTLD–TDP and AD pathology, respectively.⁵⁶ The characteristics of progressive nonfluent aphasia include errors in language production with agrammatism and grammatical simplification.⁴⁸ Speech production is typically strained or halting, with speech-sound errors occurring as a result of coexisting motor speech problems, referred to as apraxia of speech.

Clinical and pathological studies in patients with progressive nonfluent aphasia have highlighted the difficulty of obtaining an accurate antemortem diagnosis. As with the other clinical syndromes discussed above, several different pathologies have been reported to occur at variable frequencies in patients with progressive nonfluent aphasia. Some researchers have suggested that quantitative studies of neuropsychological features can be useful to predict the underlying pathology in patients with progressive language deterioration. Apraxia of speech, with or without progressive nonfluent aphasia, is often observed in patients with a tauopathy, but not in those with FTLD–TDP.²² Josephs et al. suggested that one could differentiate between patients with CBD and PSP since apraxia of speech without nonfluent aphasia was more frequent in the patients with PSP, whereas those with CBD tended to have both apraxia of speech and nonfluent aphasia. In three combined series of 93 pathologically confirmed individuals, 80% of patients with progressive nonfluent aphasia had tau-positive pathology^.49,52,57

Posterior cortical atrophy syndrome

Patients with posterior cortical atrophy primarily exhibit disturbances in visual perception with features of Balint syndrome, Gerstmann syndrome, visual agnosia and alexia.⁵⁸ The occipitoparietal cortices are predominantly affected. Posterior cortical atrophy is most often associated with AD or Creutzfeldt–Jakob disease,⁵⁹ although posterior cortical atrophy syndrome (also known as Benson syndrome), has been described, albeit rarely, in patients with pathologically confirmed CBD.^59–61 Two patients with CBD who had posterior cortical atrophy syndrome carried the brunt of the tau pathology in the posterior parietal cortex and Brodmann areas 17 and 18.⁶⁰

Classifying the 4R tauopathies

When they were originally described, CBD and PSP were considered distinct clinical and pathological disorders,^4,35 and they have remained distinct nosological entities over the years. However, a consensus exists that CBD and PSP could represent a disease spectrum, since they have overlapping clinical, pathological, genetic and biochemical features. Both disorders involve an accumulation of the abnormal 4R tau isoform in neurons and glial cells (oligodendroglia and astrocytes). They also share a similar genetic basis, with a higher frequency of the microtubule-associated protein tau (MAPT) H1 haplotype (and the H1c subhaplotype) than is found in healthy controls (Box 3). These characteristics of CBD and PSP support grouping them as the 4R tauopathies. This approach could prove useful in the design of tau-directed therapeutic clinical trials. Whether CBD and PSP share the same disease mechanism remains unknown; nevertheless, diagnostic accuracy is substantially improved when CBD and PSP are grouped as a single entity.

Box 3 |. Tau protein and corticobasal degeneration.

Microtubule-associated protein tau, encoded by the MAPT gene, is required for microtubule assembly and stability, and is highly expressed in neuronal axons.^108–110 The phosphorylation status of tau regulates its function. In tauopathies, tau protein is aberrantly phosphorylated and has an abnormal conformation that favors aggregation and inclusion formation.¹³

The human MAPT gene is located on chromosome 17q21.3.¹¹¹ Alternative splicing of exons 2, 3 and 10 produces six tau protein isoforms in the adult CNS.^111,112 Moreover, genetic analyses of MAPT have revealed two major tau haplotypes, H1 and H2.^113,114 Either the H1 haplotype or a specific H1 sub-haplotype (H1c) is associated with an increased risk of corticobasal degeneration or progressive supranuclear palsy.^{113,115–118}

Patients with MAPT mutations can exhibit clinical and pathological features associated with corticobasal degeneration.^119–121 Over 40 pathogenetic MAPT mutations have been identified in more than 100 families affected by frontotemporal dementia and parkinsonism linked to chromosome 17.¹²²

Clinical presentations

In the archetypal presentations of CBD and PSP (CBS and Richardson syndrome, respectively), the two disorders have distinct clinical features; however, the differential diagnosis is often complicated and misdiagnosis is not infrequent, even at autopsy. Some studies suggest PSP has a slightly later age of onset than CBD, but few meaningful differences in any clinical or demographic feature have been demonstrated.

Neuropathological features

Neuropathological diagnostic criteria for CBD have been refined and validated on the basis of tau immunohistochemical analysis, with an emphasis on tau-positive neuropil threads in gray and white matter of the cortex, basal ganglia, diencephalon and rostral brainstem (Figure 1).⁶² In contrast to the original description by Rebeiz et al., these updated criteria place reduced emphasis on the ballooned neuron because this finding lacks diagnostic specificity.⁶³

Figure 1 |.

The distinct histological lesions of CBD and PSP visualized with tau immunohistochemistry. a | The astrocytic plaque is a hallmark of CBD. b | Tufted astrocytes are found in the superior frontal cortex of patients with PSP. c | Numerous neuropil threads in the internal capsule are found in patients with CBD; d | by contrast, those with PSP have sparse neuropil threads with oligodendroglial coiled bodies. e | Western blot of detergent-insoluble tau proteins (P3 fraction) from patients with CBD and PSP. The CBD and PSP samples both exhibit the characteristic major doublet of hyperphosphorylated full-length tau at 64 kDa and 68 kDa (asterisk). Cleaved tau fragments from a patient with CBD migrate as two bands at approximately 37 kDa (small arrow), whereas cleaved fragments from a patient with PSP migrate as a single band at 33 kDa (large arrow). Abbreviations: CBD, corticobasal degeneration; PSP, progressive supranuclear palsy.

The main difficulty in neuropathological diagnosis arises in patients with atypical CBD who have greater than usual brainstem and cerebellar pathology,⁴⁰ and in patients with PSP who have greater than usual cortical pathology.^36,64 In the majority of individuals, the distribution of tau pathology differs according to whether they have CBD or PSP: tau pathology in patients with CBD occurs predominantly in forebrain structures, whereas in those with PSP tau pathology occurs predominantly in hindbrain structures.^39,65 However, exceptions to these general rules exist in patients with atypical presentations of both disorders. The strongest argument for separating these two diseases comprises their distinct pathological and biochemical features. For example, the astrocytic lesions associated with CBD and PSP are distinct and rarely (if ever) coexist.⁶⁶ The pathognomonic astrocytic lesion in CBD is the astrocytic plaque,^66,67 whereas tufted astrocytes are characteristic of PSP (Figure 1).^39,68

Biochemical features

Although CBD and PSP are both 4R tauopathies, biochemical studies show differences in the tau protein profiles associated with the two disorders. In particular, the tau protein cleavage fragments differ in patients with CBD and PSP. The detergent-insoluble cleaved tau fragments from patients with CBD migrate as two bands (a doublet of around 37 kDa), whereas those from patients with PSP migrate as a single band of 33 kDa (Figure 1).^69,70 These differences in proteolytic processing of tau might be related to the differences in glial lesions that occur in patients with CBD and PSP. Whether differences also exist between these biochemical profiles in various brain regions remains to be determined. Except for differences in the size of tau cleavage fragments, the biochemical properties of tau are similar in the brains of patients with CBD and PSP.

Mechanisms of tau-mediated disease

Under pathological conditions, tau becomes hyperphosphorylated, which reduces its binding affinity for microtubules. This change leads to a loss of proper microtubule functioning and/or a toxic gain of function, as dissociated tau has a greater propensity for multimerization than microtubule-bound tau. The cellular dysfunction that causes tau to become hyperphosphorylated in sporadic tauopathies is currently unknown; however, some evidence supports a role for microglia in sporadic tauopathies,^71,72 and a molecular biology study published in 2010 described a mechanism through which microglial signaling enhances tau hyperphosphorylation in neurons in the setting of neuroinflammation.⁷³ Further work is needed, however, to determine whether this mechanism, which provides a drug target upstream of tau hyperphosphorylation, will lead to the development of new strategies for the treatment of tauopathies.

Although an extensive amount of research has gone into elucidating the mechanism of tau-mediated neurodegeneration, many gaps remain in our current knowledge. Two studies have localized initial tau dysfunction to the synapse, which is the region where cell–cell communication occurs. Ittner et al. showed that tau has a role in targeting fyn kinase to postsynaptic compartments and modulates N-methyl-d-aspartate (NMDA) receptors, resulting in excitotoxic downstream signaling.⁷⁴ Hoover et al. also showed that tau accumulates in dendritic spines, causing a disruption in synaptic function by reducing the levels of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and NMDA receptors.⁷⁵ These studies have advanced our understanding of the pathophysiological processes underlying tauopathies that could provide new strategies for the development of tau-directed therapeutics.

Tau-directed therapeutics

Interest in the development of tau-directed therapeutics is growing. Several different approaches to reducing tau-mediated neurodegeneration focus on compensating for the loss of tau function using microtubule-stabilizing agents, reducing hyperphosphorylation of this protein through the inhibition of tau kinases, inhibiting the toxic aggregation of tau fibrils using agents that block protein–protein interactions, and enhancing intracellular tau degradation by targeting proteins involved in the ubiquitin–proteasome system.

Three compounds have progressed to human clinical testing: methylene blue, lithium chloride and an octapeptide (Asn–Ala–Pro–Val–Ser–Ile–Pro–Gln) known as NAP.⁷⁶ Methylene blue was the first drug to demonstrate inhibition of tau aggregation and to alter the structure of paired-helical filaments that make up neurofibrillary tangles in the brains of patients with AD.⁷⁷ Methylene blue has already been approved by the FDA for several conditions and is currently undergoing phase III clinical testing in patients with AD. Patients with tauopathy might also benefit from methylene blue treatment.

Lithium chloride is an inhibitor of glycogen synthase kinase 3β (GSK3β), one of several kinases that phosphorylate tau.^78,79 GSK3β has been extensively studied as a therapeutic target in AD. Treatment with lithium chloride consistently results in an attenuation of tau hyperphosphorylation and pathology in animal models of tauopathies.^80–84 Transgenic animals that harbor human MAPT mutations recapitulate key features of the neuropathology seen in human tauopathies, and these models are an excellent tool for preclinical drug discovery.⁸⁵

Microtubule-stabilizing compounds, including the anticancer drug paclitaxel, showed improved microtubule density, accelerated axonal transport and increased motor function in a mouse model of tauopathy.⁸⁶ These benefits were accompanied by intolerable adverse effects, however, because paclitaxel does not readily cross the blood–brain barrier and, therefore, high doses are required to observe any therapeutic effect. Efforts to explore other microtubule-stabilizing agents with improved blood–brain barrier permeability led to discovery of the neuroprotective octapeptide NAP, which has shown promise as a therapy for tauopathies by reducing the levels of hyperphosphorylated tau.^87–89 Treatment with another microtubule-stabilizing compound, epothilone D, was reported in 2010 to increase microtubule density and reduce cognitive deficits in an animal model of tauopathy.⁹⁰ Enhancing the cellular machinery responsible for degrading pathological tau species using inhibitors of heat-shock protein 90 reduces levels of hyperphosphorylated tau in vitro and in vivo.^91,92 High-throughput drug library screens have been performed to identify novel molecules capable of inhibiting tau fibrillization. None of these molecules has yet been tested in animal models, but the technique shows promise for future identification of therapeutic compounds.⁹³

Identification of biomarkers

Development of imaging and fluid biomarkers for CBD and related tauopathies is still in the early stages; however, such markers have the potential to improve diagnostic accuracy and specificity. Tau levels in cerebrospinal fluid (CSF) have been studied quite extensively in patients with AD and, since 1997, these levels have been monitored in clinical series of patients with PSP and CBD.^94–97 The authors of some of these studies report an increase in total tau levels in patients with CBD compared with controls,^95,98 whereas others found no notable difference in CSF tau levels in patients with CBD compared with both healthy controls and patients with PSP.^97,99 Although these initial studies have assessed total tau and phosphorylated tau levels, the evidence suggests that evaluating levels of tau fragments and tau phosphorylation patterns could be useful for the differential diagnosis of neurodegenerative diseases and/or in clinical trials.^96,100,101

Levels of other CSF proteins, including amyloid-β₄₂, neurofilament light chain and neurofilament heavy chain, can differentiate patients with PSP from healthy control individuals or those with other parkinsonian disorders.^97,102,103 Further studies are needed to explore the differences in expression of these and other proteins between individuals with CBD and PSP by means of unbiased proteomic methods in brain tissue and CSF.

Distinctive patterns of atrophy on brain imaging have been observed in patients with autopsy-confirmed CBD and PSP.⁴⁷ Patients with CBD have a characteristic pattern of posterior frontal atrophy regardless of the clinical syndrome, suggesting that this pattern of atrophy could prove to be a useful biomarker of CBD pathology (Figure 2). Differences in atrophy patterns elsewhere in the brain seem to reflect clinical differences: for example, marked prefrontal atrophy is observed in patients with CBD related to bvFTD, and substantial atrophy in posterior regions of the brain is observed in those with CBD related to posterior cortical atrophy. If CBD and PSP can be differentiated in living patients by use of inexpensive and relatively noninvasive methods, our knowledge of the epidemiology of these disorders will increase, which could ultimately lead to refined treatments.

Figure 2 |.

Patterns of gray matter loss in groups of patients with autopsy-confirmed CBD but different clinical syndromes. The results were generated using voxel-based morphometry and are shown on three-dimensional surface and medial renders of the brain. All clinical groups show gray matter loss as indicated by red coloration in the posterior lateral and medial frontal lobe, suggesting that this is a signature pattern of CBD pathology; however, patterns of gray matter loss that differ between groups extend into other regions of the brain. These differences are thought to cause the various clinical syndromes associated with CBD. Abbreviations: bvFTD, behavioral variant frontotemporal dementia; CBD, corticobasal degeneration; CBS, corticobasal syndrome; PCA, posterior cortical atrophy; PPA, primary progressive aphasia; RS, Richardson syndrome.

Conclusions

The complex and highly variable clinical presentation of CBD is the prevailing difficulty in identifying and accurately diagnosing this disorder in living patients. Efforts are currently being made to improve the accuracy of diagnosis using imaging and CSF biomarkers. Although considerable progress has been made in delineating the clinical picture of CBD, these studies are inherently limited because they can only be performed retrospectively owing to the necessity of obtaining an autopsy-confirmed diagnosis. The first completed genome-wide association study for PSP and CBD is a key endeavor that will help to elucidate the genetic basis of these 4R tauopathies.¹⁰⁴ These genomic data will provide important insight into novel pathways and the fundamental biology involved in the pathogenesis of tauopathy, which might lead to new pharmacological strategies for the treatment of patients with this condition.

Key points.

• Corticobasal degeneration (CBD) is associated with neuronal and glial tau pathology (including astrocytic plaques) in gray and white matter of the cortex, basal ganglia, diencephalon and rostral brainstem

• The most common clinical presentation of CBD is asymmetric progressive rigidity and apraxia with limb dystonia and myoclonus, termed corticobasal syndrome

• CBD pathology can be found in patients with Richardson syndrome (the most common clinical presentation of progressive supranuclear palsy), behavioral variant frontotemporal dementia and primary progressive aphasia

• The underlying pathology in corticobasal syndrome can be Alzheimer disease, progressive supranuclear palsy, frontotemporal lobar degeneration or Pick disease; corticobasal syndrome is not, therefore, a specific phenotype of CBD

• Given the remarkable clinical heterogeneity of CBD and poor accuracy of diagnosis in living patients, the findings of studies based on non-autopsy-confirmed cases should be interpreted with caution

• Current efforts to improve the diagnostic accuracy of CBD include imaging and cerebrospinal fluid biomarkers, as well as genome-wide association studies

Review criteria.

We searched MEDLINE using the terms “corticobasal degeneration” and “cortical basal ganglionic degeneration”, and their variants, synonyms, and acronyms, for articles published between January 1988 and December 2010. We also examined the bibliographies of key articles to identify additional relevant studies. Research papers with pathologically confirmed diagnosis of corticobasal degeneration were specifically sought. Reports based entirely on clinical diagnoses were included only if they were the sole available literature on that particular topic.