Keywords

Introduction

The two most common clinicopathologic subtypes of frontotemporal lobar degeneration (FTLD) are characterized by TDP-43 or tau pathology [1]. Tau is a microtubule-associated protein important for stability and functional properties of microtubules. The gene that encodes tau protein (MAPT) is located on chromosome 17, and it undergoes alternative splicing of exons 2, 3, and 10 to generate six isoforms of tau [2]. Alternative splicing of exon 10 generates two major classes of tau protein that contain either three (3R) or four (4R) ≈30-amino acid repeats in the microtubule-binding domain of tau. Neurodegenerative tauopathies can be subclassified based upon the predominant type of tau that accumulates in cellular lesions [3]. Pick’s disease, a rare frontotemporal dementia with lobar cortical atrophy and neuronal Pick bodies, is characterized by tau composed predominantly of 3R tau, while neurofibrillary tangles that characterize the pathology in Alzheimer’s disease and chronic traumatic encephalopathy are composed of a mixture of 3R and 4R tau with distinct ultrastructural properties [4, 5]. Disorders associated with 4R tau are clinically and pathologically heterogeneous and include aging-related disorders, such as aging-related tau astrogliopathy (ARTAG) [6] and argyrophilic grain disease (AGD) [3, 7]. The most common of the neurodegenerative 4R tauopathies are progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD), which is the focus of this chapter.

Progressive Supranuclear Palsy

PSP was described by Steele, Richardson, and Olszewski in a small autopsy series of patients with postural instability, vertical supranuclear gaze palsy, facial and cervical dystonia, as well as dementia. Despite some clinical variability, they shared distinctive pathologic features, including argyrophilic neurofibrillary tangles in select subcortical and brainstem nuclei. [8]. With the advent of tau biochemistry and molecular biology, the pathologic features of PSP have been expanded to include not only neuronal lesions but also glial lesions [3, 9]. The clinical syndromes associated with the characteristic tau pathology of PSP have also expanded from the original descriptions and is described later in the chapter.

Epidemiology of Progressive Supranuclear Palsy

The prevalence of PSP is thought to be approximately 6/100,000 patients [10,11,12,13]; however, there is a growing understanding that PSP pathology is associated with multiple clinical phenotypes, suggesting that the above figure may require revision. Increased awareness of this fact led to increased age-adjusted prevalence estimates in Europe (8.8–10.8/100,000 patients) [11, 14]. Of note, age-adjusted prevalence estimates from the same city in Japan (Yonogo) adjusted to the census of the earlier study increased from 5.8/100,000 patients in 1999 to 17/100,000 patients in 2010 [15, 16]. This is, in part, due to identification of more phenotypes, since the previous studies used the National Institute of Neurologic Disease and Stroke and Society for PSP (NINDS-SPSP) criteria that only identified the classical PSP phenotype (also named PSP-Richardson syndrome [PSP-RS]) .

Clinical Features of Progressive Supranuclear Palsy

In addition to the typical presentation described by Richardson and colleagues (PSP-RS), other phenotypes associated with PSP pathology have been described, including an extrapyramidal disorder mimicking Parkinson’s disease (PSP-P), corticobasal syndrome (PSP-CBS), dementia with predominantly frontal characteristics (PSP-F), dementia with speech and language disturbances (PSP-SL), and others. Consequently, the newest clinical criteria for PSP, supported by the International Parkinson and Movement Disorder Society (MDS-PSP criteria), include a wider clinical spectrum [17]. Typical age of onset of PSP is in the seventh decade of life [17,18,19], and average survival is 5–6 years; however, certain phenotypes are associated with much longer disease durations [19, 20].

Several criteria for PSP were proposed based upon clinical case series [21,22,23,24], but the first widely used criteria that were based on autopsy-confirmed cases was reported by Litvan et al. [18] and supported by the NINDS-SPSP. The NINDS-SPSP criteria outlined several core features of PSP-RS. Mandatory features included a gradually progressive disorder with age of onset 40 years of age or later, presence of vertical supranuclear gaze palsy, and/or postural instability with falls within the first year of disease. Both features had to be present for a diagnosis of “probable PSP,” and only vertical supranuclear gaze palsy or slowing of saccades and postural instability with falls within the first year of disease was consistent with a diagnosis of “possible PSP.”

Regarding vertical supranuclear gaze palsy, restricted downward gaze has been considered most specific for PSP because restricted upward gaze can be seen to a lesser degree in aging [25], Parkinson’s disease [26], and other conditions [27,28,29,30,31] like severely restricted upward gaze and slowing of vertical saccades. At more advanced stages, horizontal supranuclear gaze palsy may develop, as well [32]. Vertical supranuclear gaze palsy may be preceded by subtle ocular motor abnormalities, including loss of vertical optokinetic nystagmus [33], “stair casing,” and the “round the house sign” [34], where horizontal saccadic excursions interrupt vertical eye movements. Other ocular motor movement abnormalities include hypometric saccades, breakdown of smooth pursuit, and square wave jerks [35]. Loss of vergence is observed early and may contribute to frequent complaints of diplopia [36]. Other eye findings include blepharospasm and eyelid-opening apraxia [37], although these are not usually early features.

Early loss of postural reflexes and falls are common and often an early complaint in PSP-RS, usually occurring within the first year of illness. Falls tend to be backwards, but it can occur in any direction and may be compounded by freezing of gait. Falls can result in significant morbidity due to lacerations, fractures, or intracerebral bleeding [32, 38].

While these features define the core clinical features of PSP-RS, a number of other clinical features are often observed. Parkinsonism manifested by symmetric akinesia and rigidity with an axial predominance is common. Neck stiffness with retrocollis has been described in early descriptions of PSP, but it is rare [8]. Facial dystonia produces the so-called PSP stare, with decreased blink rate, furrowed and raised eyebrows, and a look of surprise. Inappropriate laughter and crying episodes are often observed (pseudobulbar affect). Early hypokinetic and spastic dysarthria is a secondary feature, which can progress to anarthria in severe cases [39]. Dysphagia occurs relatively early, and it is frequently implicated as a cause of death due to aspiration pneumonia [40, 41]. Cognitive manifestations associated with PSP overlap with corticobasal syndrome and frontotemporal dementia (FTD). The clinical course of PSP is relentless and nearly always is associated with a frontal-subcortical-type dementia.

PSP-RS phenotype is the clinical syndrome most likely to have PSP pathology at autopsy. Because of this, the NINDS-SPSP criteria proved to be specific for PSP pathology [42, 43], but to have relatively low sensitivity [43,44,45]. This is because PSP pathology can present with other clinical syndromes, and eye movement abnormalities seen in PSP often occur later in the course of the disease and sometimes not at all [19, 20, 46,47,48,49,50,51,52,53,54,55,56,57]. In one autopsy series, 76% of pathologically confirmed PSP had a clinical syndrome other than PSP-RS [58].

The most common clinical PSP variant mimics idiopathic Parkinson’s disease (PSP-P) and makes up about one-third of pathologically confirmed cases [46, 59,60,61,62]. These patients have asymmetric resting tremor and asymmetric appendicular bradykinesia and rigidity, making the distinction between PSP-P and Parkinson’s disease challenging [46, 59, 60, 62, 63]. As many as one-third of these patients will respond to levodopa and show greater than 30% reduction in the Unified Parkinson’s Disease Rating Scale [46, 64,65,66,67]. Some also develop levodopa-induced dyskinesias [46]. Most PSP patients have minimal or no response to levodopa therapy, and if a response occurs, it is typically mild and not sustained [20, 24, 68]. Robust and prolonged response to levodopa therapy is an exclusionary criterion for PSP and makes Parkinson’s disease a more likely diagnosis [17]. It can be 3–4 years into the disease course before supranuclear gaze palsy is present to aid in refining the diagnosis in PSP-P [19, 62]. PSP-P patients also have a longer disease duration than PSP-RS, with an average survival of 10–15 years [19, 46, 62].

Other syndromes have been described in autopsy-confirmed PSP. Some present with impulsivity and behavioral changes, including apathy, impulsivity, and social inappropriateness akin to behavioral-variant frontotemporal dementia (PSP-F) [53, 69, 70]. Others present with progressive non-fluent aphasia or apraxia of speech (PSP-SL) [48, 52, 70, 71]. About 10% have a corticobasal syndrome with asymmetrical dystonia, myoclonus, apraxia, and cortical sensory loss (PSP-CBS) [55, 56, 70, 72]. Another rare presentation, but one that is highly predictive of PSP pathology, is pure akinesia with gait freezing (PSP-PAGF) [47, 73, 74]. Early presentations currently considered to be “suggestive” of PSP in MDS-PSP criteria are isolated postural instability (PSP-PI) [19, 75] and isolated oculomotor dysfunction (PSP-OM) [19, 20]. The most uncommon presentations are progressive cerebellar ataxia (PSP-C) [51, 76, 77] and primary lateral sclerosis (PSP-PLS) [50, 57]. It is important to note that while some patients present with discrete syndromes, it is common for considerable overlap, and patients also acquire new signs and symptoms as the disease progresses. Regardless of the initial syndrome, most patients develop vertical supranuclear gaze palsy and postural instability, which are core features of PSP-RS, that make diagnosis obvious, but these may occur only later in the disease course in some of the PSP clinical variants [19].

Recognition of the spectrum of clinical heterogeneity in PSP, led the MDS-PSP criteria to incorporate a broader set of symptoms and signs, as well as levels of certainty that would be associated with PSP pathology [17]. These criteria are more sensitive, but they are less specific than the NINDS-SPSP criteria [78, 79]. The implementation of “multiple allocation extinction” rules (MAX rules) have been necessary to help disentangle patients who may be classified into more than one clinical MDS-PSP category [79]. Even so, these MAX rules may fail to separate up to 40% of patients with PSP-P and PSP-RS overlap syndromes [80]. These issues highlight the ongoing need for specific biomarkers to improve diagnostic accuracy of PSP during life.

Neuropathology of Progressive Supranuclear Palsy

The external appearance of PSP at postmortem evaluation depends upon the clinical syndrome. PSP-RS may have no significant cortical atrophy or mild atrophy affecting the dorsolateral frontal lobe. PSP-F and PSP-CBS usually have more marked frontal atrophy, especially affecting the superior frontal gyrus, while PSP-SL may have more significant frontal atrophy, especially affecting the peri-Sylvian inferior frontal gyrus. Asymmetry, which is not often assessed with research protocols that evaluate only one side of the brain for histology, can be notable in PSP-SL and PSP-CBS. PSP-PLS has focal atrophy affecting the precentral gyrus; it can be asymmetrical as well. The most striking macroscopic finding in PSP-RS (and PSP-P) is midbrain atrophy (Fig. 1a) with loss of neuromelanin pigment on transverse sections of the brainstem (Fig. 1d). The subthalamic nucleus invariably has atrophy (Fig. 1b), and there is also atrophy of the superior cerebellar peduncle (Fig. 1e) and atrophy of the hilus of the cerebellar dentate nucleus (Fig. 1c). Atrophy of subthalamic nucleus and midbrain is usually less severe in PSP-F and PSP-CBS, and often very severe in PSP-PAGF. In the latter, atrophy is frequently accompanied by similar changes in the globus pallidus and with reddish-brown discoloration due to deposition of iron pigment (pallido-nigro-luysial “pigment-spheroid degeneration” [81]).

Fig. 1
figure 1

Macroscopic findings in PSP . (a) A sagittal section of the brainstem shows marked atrophy of the midbrain (arrows). (b) A coronal section of the diencephalon shows marked atrophy of the subthalamic nucleus (arrowheads). (c) A section of the cerebellum at the level of the middle cerebellar peduncle shows marked atrophy and discoloration of the dentate nucleus of the cerebellum (arrow). (d) A transverse section of the midbrain shows atrophy and marked neuromelanin pigment loss in the substantia nigra (asterisk). (e) A transverse section of the pons shows marked atrophy of the superior cerebellar peduncle (arrowheads)

Full size image

Histopathologic findings in PSP are similar in the various subtypes. The clinicopathologic subtypes differ in the relative distribution of the neuronal loss and gliosis, and in the density of tau pathology [82]. There are no distinctive cellular pathologies in PSP clinicopathologic variants. The major histopathologic lesions in PSP are neurofibrillary tangles, which often have a globose shape in vulnerable subcortical nuclei, such as the subthalamic nucleus (Fig. 2a) and substantia nigra (Fig. 2b). The morphology and distribution of tangles in PSP is different from the most common disorder with neurofibrillary tangles, Alzheimer’s disease (AD), in that subcortical and brainstem nuclei are preferentially affected. The tangles are positive for phospho-tau (Fig. 2d). Using antibodies specific to tau isoforms, the tangles in PSP preferentially accumulate 4R tau (not shown). Tau immunohistochemistry also shows distinctive glial pathology in PSP, including tufted astrocytes (Figs. 2d and 3e) and oligodendroglial coiled bodies (Fig. 2f). Tufted astrocytes are most frequent in neocortex, neostriatum, and midbrain tectum. Coiled bodies are widespread in affected cerebral white matter and vulnerable subcortical fiber tracts in the basal telencephalon, diencephalon, brain stem, and cerebellum. A common neurodegenerative change in the cerebellar dentate nucleus that is not associated with tau pathology is the presence of irregularly swollen cell processes around apical dendrites and cell bodies of cerebellar dentate nucleus neurons (Fig. 2c), a process referred to as grumose degeneration [83]. Glial pathology is increasingly recognized to play a significant role in pathogenesis of neurodegenerative disease, and in PSP microgliosis and astrogliosis parallels the systems affected by neurodegeneration [84], with little evidence to suggest that it precedes tau pathology.

Fig. 2
figure 2

Microscopic findings in PSP . (a) An H&E stained section of the subthalamic nucleus shows severe neuronal loss and astrocytosis, with neurofibrillary tangles (arrow) in residual neurons. (b) An H&E stained section of the substantia nigra shows neuronal loss and gliosis with extraneuronal neuromelanin pigment and globose neurofibrillary tangles (arrowheads). (c) An H&E stained section of the cerebellar dentate nucleus shows granular eosinophilic swollen cell processes (arrowhead), obscuring the outlines of the neuron, findings characteristic of grumose degeneration (arrow). (d) Phospho-tau immunohistochemistry of the caudate nucleus shows a globose neurofibrillary tangle (arrowhead) and a tufted astrocyte (arrow). (e) Phospho-tau immunohistochemistry of the caudate nucleus shows several tufted astrocytes (arrows) with morphologic heterogeneity. (f) Phospho-tau immunohistochemistry of the internal capsule shows oligodendroglial coiled bodies (arrowhead). All images are of same magnification, bar in (f) is 20 μm

Full size image
Fig. 3
figure 3

Macroscopic findings in CBD . (a) The medial surface of left hemibrain shows atrophy of the superior frontal gyrus (asterisk indicates area of greatest pathology) and focal atrophy of the corpus callosum (arrows). (b) A coronal section of the brain at the level of the fornix shows marked enlargement of the frontal horn of the lateral ventricle (large asterisk). There is also atrophy and discoloration of the globus pallidus (small asterisk). (c) A coronal section of the diencephalon and anterior medial temporal lobe shows no hippocampal atrophy and minimal-to-no atrophy of the subthalamic nucleus (arrowheads). (d) A section of the cerebellum at the level of the middle cerebellar peduncle shows no atrophy and normal myelin in the hilus of the dentate nucleus (arrow). (e) A transverse section of the pons shows no atrophy of the superior cerebellar peduncle (arrowheads). (f) A transverse section of the midbrain shows mild atrophy and marked neuromelanin pigment loss in the substantia nigra (asterisk)

Full size image

Corticobasal Degeneration

The term corticobasal degeneration was coined by Gibb, Luthert and Marsden [85] to describe the pathology of a rare disorder associated with cognitive and motor features affecting the neocortex and basal ganglia. The clinically defined corticobasal syndrome (CBS) is characterized by progressive cognitive decline associated with asymmetrical rigidity, dystonia, myoclonus, and alien-limb phenomenon. Early autopsy studies reported focal cortical atrophy and swollen achromatic neurons (“ballooned neurons” [86]), as well as neuronal loss in the substantia nigra and cerebellar dentate nucleus—“corticodentatonigral degeneration with neuronal achromasia” [87]. These descriptions did not recognize the tau pathology in CBD because neuronal lesions in CBD are weakly positive or negative with traditional silver impregnation methods. It was not until the early 1990s that widespread tau pathology in CBD was shown to be distinct from Alzheimer’s disease, using immunohistochemistry and ultrastructural methods [88,89,90]. The pathognomonic astrocytic lesion of CBD (“astrocytic plaques”) was described in 1995 [91].

Epidemiology of Corticobasal Degeneration

Like PSP , pathologically confirmed CBD has a range of clinical presentations, and CBS may not be the most common. Moreover, the pathologic substrate of CBS is mixed, with PSP being as common as CBD [56, 92], but other disorders, particularly atypical presentations of Alzheimer’s disease, can also present with CBS [56, 85, 93,94,95,96,97,98]. Estimates of prevalence of CBD are inherently flawed. For these reasons, the term corticobasal syndrome (CBS) is now preferred to refer to the clinical presentation described earlier, whereas corticobasal degeneration (CBD) is reserved for the neuropathological diagnosis. The incidence of CBD is estimated to be 0.62–0.92/100,000 [93, 99,100,101].

Clinical Features of Corticobasal Degeneration Presenting as Corticobasal Syndrome

The onset of CBS is typically in the sixth or seventh decade of life, with a mean survival of about 7 years from diagnosis [93, 99,100,101]. The motor manifestations of CBS include an asymmetric parkinsonism manifested predominantly by rigidity and bradykinesia [93]. While asymmetry in parkinsonian features is common in Parkinson’s disease, the asymmetry in CBS can be striking. There is frequently additional dystonic posturing of the limb. Superimposed may be ideomotor and limb-kinetic apraxia [55, 99, 102]. Alien-limb phenomenon affecting the arm or leg has been described and often results in an unawareness of a levitating hand or leg due to feeling the limb alien, and more rarely, intermanual conflict [103]. Myoclonus is often present, and it may affect limbs or, rarely, the face [99, 104]. Myoclonus is worsened by action, posture, or stimuli [55, 99, 104]. At times, myoclonus can be difficult to differentiate from tremor, although the quality of myoclonic tremor is jerky rather than the smooth oscillatory tremor observed in Parkinson’s disease and other parkinsonian disorders [105]. Postural instability and falls are common, but usually later in the disease course than in PSP, unless the symptoms start in lower extremities [93]. Parkinsonism associated with CBS may benefit from levodopa therapy, but improvement in symptoms is rare and levodopa-induced dyskinesias are also rare [55]. Sustained and robust levodopa responsiveness is an exclusionary criterion to the diagnosis of CBS [93, 106].

Several cognitive features and other signs referable to higher-order cortical function are common in CBS. As previously mentioned, apraxia is a core feature. Ideomotor apraxia is usually one of the first disease features. Some patients develop orobuccal apraxia or apraxia of eyelid opening [99, 104, 107]. Cortical sensory loss with astereognosis and agraphesthesia are frequently observed [108, 109]. Visual neglect may be seen, and it is related to parietal lobe dysfunction [95, 107, 110]. A progressive non-fluent aphasia is also described in CBS, with occasional overlay of apraxia of speech from frontal lobe dysfunction [95, 104, 107, 111]. Other features of frontal lobe dysfunction, such as apathy and disinhibition, are common and early [55, 93].

The clinical presentation of autopsy-confirmed CBD is varied, with some presenting with a cognitive syndrome, and some primarily with a motor phenotype. Other neurodegenerative disorders, PSP and Alzheimer’s disease in particular, can present with CBS. Unlike PSP, these initial presentations may not necessarily coalesce into a common phenotype over time, making diagnostics even more challenging. Concomitantly, the clinical diagnosis of CBS has relatively poor predictive value for CBD pathology at autopsy compared to other neurodegenerative disorders. The sensitivity of clinical findings predicting CBD at autopsy is between 26% and56%. The majority of these studies were performed using older criteria; recently, more specific criteria have not been fully vetted [55, 59, 70, 95]. Current clinical criteria for CBD define a gradual progressive disorder with insidious onset and several possible phenotypes, including CBS, a frontal behavioral-spatial syndrome, a variant of primary non-fluent aphasia, and a PSP syndrome. The clinical syndrome of probable CBS is defined as having two of the following signs: limbs with asymmetric rigidity and akinesia, limb dystonia or limb myoclonus, and two of the following signs and symptoms: orobuccal or limb apraxia, cortical sensory deficits, or alien-limb phenomena. Possible clinical CBS involves having one limb with rigidity or akinesia, limb dystonia, or limb myoclonus with one of the above supportive features. A frontal behavioral spatial syndrome is described with the attendant cognitive features. Non-fluent primary progressive aphasia and a PSP phenotype are recognized but considered as possible CBD. Patients with a PSP clinical syndrome must have at least one additional symptom or sign (limb rigidity/akinesia, limb dystonia or myoclonus, apraxia, and cortical sensory loss) [93].

There are multiple exclusion criteria that, if present, make CBD a less likely cause of the clinical presentation. The most important are the presence of genetic mutations in GRN, FUS, TARDBP, PSEN1/2, and APP genes. Another exclusionary criterion is a cerebrospinal fluid (CSF) Aβ42/tau ratio consistent with Alzheimer’s disease [112]. Classic 4–6 Hz parkinsonian resting tremor, hallucinations, dysautonomia, cerebellar signs, the presence of both upper and motor neuron signs, or the semantic or logopenic variants of primary progressive aphasia are also considered exclusionary; they are more likely to indicate Parkinson’s disease, dementia with Lewy bodies, multiple systems atrophy, ALS, or FTLD . Lastly, because there are occasional reports of fulminant presentations of CBD [113, 114], imaging consistent with Creutzfeldt-Jakob disease is also exclusionary.

Neuropathology of Corticobasal Degeneration

The external appearance of the CBD brain at postmortem evaluation depends upon the clinical syndrome. For patients presenting with CBS or frontotemporal dementia syndromes, there is usually focal atrophy, especially affecting the medial superior frontal gyrus (Fig. 3a). Language-predominant syndromes often have inferior frontal gyrus (peri-Sylvian) atrophy. There is often atrophy of the corpus callosum (Fig. 3a), which tends to parallel the distribution and severity of the focal cortical pathology. Atrophy can be asymmetrical, but this is often difficult to assess at autopsy, given that half the brain is usually frozen for research purposes. Some cases, particularly patients with long tract signs, may have atrophy that extends to the motor cortex. Coronal sections frequently show enlargement of the frontal horn of the lateral ventricle (Fig. 3b). The most common finding in the basal ganglia is atrophy and reddish-brown discoloration of the globus pallidus (Fig. 2b). Unlike PSP, there is usually no significant atrophy of the subthalamic nucleus (Fig. 3c). Similarly, the hilus of the cerebellar dentate nucleus (Fig. 3d) and the superior cerebellar peduncle (Fig. 3e) do not have atrophy. Similar to PSP, there is usually loss of neuromelanin pigment in the substantia nigra (Fig. 3f).

Microscopic examination of atrophic cortical sections shows neuronal loss with superficial spongiosis, gliosis, and usually achromatic or ballooned neurons, which are readily detected with routine histology stains, such as hematoxylin-and-eosin (Fig. 4a). Ballooned neurons are found in middle and lower cortical layers of affected neocortices and have diffuse phospho-tau immunoreactivity (Fig. 4d), as well as intense immunoreactivity with antibodies to alpha-B-crystallin, a small heat-shock protein (not shown), and for neurofilament.

Fig. 4
figure 4

Microscopic findings in CBD . (a) An H&E stained section of superior frontal gyrus shows ballooned neurons (arrow). (b) An H&E stained section of the subthalamic nucleus shows mild neuronal loss, but more marked gliosis. (c) An H&E stained section of the substantia nigra shows focal neuronal loss (extraneuronal neuromelanin—asterisk) and several neurons with so-called corticobasal bodies (arrowheads). (d) Phospho-tau immunohistochemistry of the superior frontal gyrus shows many neuropil threads and a ballooned neuron with diffuse cytoplasmic tau immunoreactivity (arrow). (e) Phospho-tau immunohistochemistry of the caudate nucleus shows an astrocytic plaque (asterisk). (f) Phospho-tau immunohistochemistry of the subthalamic nucleus shows morphologic heterogeneity of neuronal inclusions (arrowheads). Panels a and c–f are of same magnification, bar in (f) is 20 μm. Panel (b) is a lower magnification, bar is 50 μm

Full size image

In addition to ballooned neurons, the neocortex and neostriatum in CBD have widespread deposition of tau in both neurons and glia [3, 9]. Glial inclusions are found in both oligodendroglia and astrocytes. The astrocytic lesions have a characteristic plaque-like morphology (“astrocytic plaques” [91]) (Fig. 4e) that is morphologically distinct from tufted astrocytes of PSP. The pathologic feature that best discriminates PSP from CBD is pervasive thread-like cell processes in affected gray and white matter in CBD, to the extent that the difference can be seen by examining the slide with the naked eye (Fig. 5).

Fig. 5
figure 5

Comparison of tau burden in PSP and CBD. Sections of the neostriatum in PSP and CBD, immunostained under the same conditions with a sensitive phospho-tau antibody (CP13 from Peter Davies, Feinstein Institute, Long Island, NY), show a clear distinction between PSP and CBD, due to dense tau pathology, mostly thread-like processes (not visible at this magnification), in CBD

Full size image

The subthalamic nucleus often has at least mild neuronal loss and gliosis (Fig. 4b), but it is rarely as severe as in PSP. Similarly, the substantia nigra has neuronal loss in CBD, but it can be mild (Fig. 4c). Neurons in the substantia nigra may have so-called corticobasal bodies [85] (Fig. 4c). Cortical neurons in atrophic areas have pleomorphic tau-immunoreactive lesions. In some neurons, tau is densely packed into small irregular inclusion bodies. In other neurons, the inclusions are more diffuse (“pre-tangles”). Neurofibrillary lesions in subcortical nuclei, such as the subthalamic nucleus, also typically have marked morphologic heterogeneity (Fig. 4f), while those in the locus ceruleus and substantia nigra can resemble globose neurofibrillary tangles (Fig. 4c).

Pathogenesis of Progressive Supranuclear Palsy and Corticobasal Degeneration

There is no single cause of PSP or CBD, but several environmental and genetic factors have been investigated. The Environmental Genetic PSP (ENGENE-PSP) study found that lower educational attainment, exposure to well water and industrial wastes, and firearm use were related to higher risk of developing PSP [115, 116]. These findings are also supported by a cluster of PSPs that emerged in northern France in an area of high industrial waste contamination [117]. Consumption of high levels of annonacin, a mitochondrial complex I inhibitor, found in the paw-paw fruit was associated with developing PSP or other atypical parkinsonian syndromes in studies in the Caribbean island of Guadeloupe [118, 119]. There may be a slight male predominance within PSP patients [22, 46], and one study documented that increased estrogen exposure in women may be protective against developing PSP [120]. Environmental exposures have not been evaluated in CBD to date.

MAPT mutations may lead to either PSP or CBD [121,122,123,124]. Mutations in this gene can also lead to frontotemporal dementia, FTLD with parkinsonism, or primary progressive aphasia [125]. The H1/H1 genotype elevates the risk for developing PSP and CBD [17, 126, 127]. One genome-wide association study in a large cohort of pathologically validated PSP patients additionally identified genetic risk variants at the MOBP, STX6, and EIF2AK3 loci [128]. MOBP , which encodes for myelin oligodendrocyte-binding protein, is also implicated in CBD and highlights potential importance of white matter [121, 129]. STX6 encodes for a SNARE protein implicated in fusing vesicles in the Golgi network [130]. EIF2AK3 encodes for a protein responsible for inhibiting protein synthesis in the face of excess endoplasmic reticulum stress [131, 132]. These genes have been validated in a second genome-wide association study, which additionally identified SLCO1A2 and DUSP10 as other genomic loci of interest [133].

Oxidative stress and inflammation can also be demonstrated in PSP and CBD. Mitochondrial enzymatic activity is decreased in both brain tissue and also in skeletal muscle in PSP patients [134,135,136,137,138,139,140]. Higher IL-1β and other inflammatory cytokines are found in the brains and CSF of PSP patients and lead to microglial activation [141, 142], which has been implicated in tau deposition [84]. Superoxide dismutase and glutathione, essential antioxidants, are often seen to be elevated in PSP brain tissue, possibly as a defense mechanism [139, 143].

Recent data suggest that misfolded tau oligomers are capable of acting as a template and induce further misfolding of normal monomeric tau leading to larger and larger aggregates, causing cellular damage and ultimately death and likely leading to spreading of disease in a ‘prion-like’ manner. In vivo animal studies using preformed fibrils [144, 145], human diseased brain homogenates [146], and other techniques [147, 148] have shown distal spread of tau pathology via trans-synaptic spread [149, 150]. There may be specific “strains” of tau capable of seeding unique tau pathologies [147, 151, 152].

Biomarkers in Progressive Supranuclear Palsy and Corticobasal Degeneration

The clinicopathologic overlap between PSP and CBD and other neurodegenerative diseases makes the discovery of sensitive and specific biomarkers for these diseases of paramount importance.

Magnetic Resonance Imaging

PSP is well described to be associated with several features on structural magnetic resonance imaging (MRI) . Most recognized is the presence of midbrain atrophy, resulting in the “hummingbird sign” best seen on the mid-sagittal section (Fig. 6) [153], as well as “morning glory sign [154]”, or “Mickey Mouse sign [155]”. In one study of an autopsy series of pathologically confirmed cases with PSP, multiple systems atrophy (MSA), or Parkinson’s disease (PD), 16/22 (72.7%) of PSP cases were able to be correctly identified by a radiologist reviewing conventional MRI that had been performed during life, and the presence of a hummingbird sign or morning glory sign was 100% specific but was 68.4% sensitive [156]. One study, however, that included different clinical variants of PSP found midbrain atrophy to be a feature of the Richardson syndrome variant, but midbrain atrophy was not found to be a biomarker of PSP pathology [157]. The superior cerebellar peduncle is also frequently atrophied in PSP and, consequently, several different ratios comparing brain stem, pons, superior cerebellar peduncle, and middle cerebellar peduncle measurements have been studied to differentiate PSP from other parkinsonian diseases and from healthy controls. A frequent problem with these measurements is that they are often insensitive, and the radiologic signs will only manifest at later stages of the disease after neurodegeneration has progressed to the point of causing these recognizable patterns [158,159,160,161,162,163]. A more specific technique to assess the superior cerebellar peduncle is with diffusion tensor imaging (DTI). One DTI study did find the superior cerebellar peduncle to be able to accurately distinguish PSP from normal controls [164]. It is unclear whether atrophy of the superior cerebellar peduncle is a feature of PSP pathology or a feature of Richardson syndrome. Another technique that has also been studied in PSP is resting-state functional magnetic resonance imaging (fMRI). Resting-state fMRI studies have demonstrated disrupted thalamocortical connectivity in PSP [165, 166].

Fig. 6
figure 6

MRI scan in autopsy-confirmed PSP and CBD. MRI scan in PSP shows the classic hummingbird sign on sagittal MRI, while asymmetric atrophy of the posterior frontal cortex is seen in CBD

Full size image

Fewer MRI studies have been performed in CBD, but the most frequently cited sign is asymmetric cortical atrophy, affecting the parietal and frontal lobes (Fig. 6) [167,168,169,170,171]. Corpus callosum atrophy is also cited occasionally. Regrettably, neither of these features are specific for CBD to fully differentiate it from other pathologies that cause CBS clinical phenotypes [70, 167, 172]. In addition, symmetric cortical atrophy has been described in autopsy-confirmed cases of CBD [173]. Research studies have utilized voxel-based morphometry to try to distinguish CBD from Alzheimer’s disease and other neurodegenerative diseases that present with CBS. These studies have found distinguishing features at the group level [174, 175]. No biomarker exists to distinguish CBD from other neurodegenerative diseases at the single subject level.

Given the prominent white matter degeneration that is common to these conditions, diffusion tensor imaging and white matter volumetric measurements may show more degeneration in PSP and CBD than atypical AD or FTLD TDP-43 that may have overlapping presentations [176,177,178,179].

DaTscan

A DaTscan is used to detect dopamine transporters on dopamine neurons. DaTscans are typically utilized to differentiate Parkinson’s disease from essential tremor. However, DaTscans have been performed in PSP and CBS patients and show a reduction in dopamine transporter receptors. Unfortunately, this finding is nonspecific and can also be seen in other parkinsonian disorders, for example, MSA .

Positron Emission Tomography

The most common PET scan is the fluorodeoxyglucose (FDG)-PET scan, which utilizes radioactive glucose to assess for functional integrity of neocortical regions. FDG-PET findings in PSP and CBS tend to mirror findings on MRI. In PSP, hypometabolism is observed in the premotor cortex as well as the midbrain, the latter when present is known as the pimple sign of PSP [180] (Fig. 7). In CBS and CBD, the FDG-PET scan reveals asymmetric frontal and/or parietal hypometabolism (Fig. 7). There are less than a handful of studies on FDG-PET in autopsy-confirmed PSP, CBD, and other 4R tauopathies. One such study found parietal hypometabolism in CBD and premotor hypometabolism in PSP [181]. Several tracers are currently under investigation that bind to the tau proteins, including 18F-5105, 18F-FDDNP, 18F-THK523, 11C-PBB3, and others [182]. 18F-Flortaucipir (formerly AV-1451 and T807) is the most researched tau tracer to date and appears to bind avidly to paired helical filaments in 3R/4R tauopathies, such as AD [183], and exhibits retention patterns in amnestic AD consistent with Braak tau staging [184, 185] and in posterior cortical regions in posterior cortical atrophy patients [186, 187]. However, 18F-Flortaucipir retention appears to be less robust in 4R tauopathies [183, 188, 189]. Increased retention in the basal ganglia and midbrain can be demonstrated in PSP (Fig. 8), but there is off-site binding, which makes individual patient-level distinctions at early stage difficult [184, 190,191,192,193]. Similarly, in CBS, mild increases in retention in cortical regions can be demonstrated (Fig. 8) that correlate with postmortem tau findings [194], although this has been reported to occur predominantly in CBS patients who presented with a motor speech disorder [195]. PET tracers targeting activated microglia (11C-(R) PK11195) may aid in assessing inflammation associated with neurodegeneration in PSP and CBD [196, 197].

Fig. 7
figure 7

FDG-PET in autopsy-confirmed PSP and CBD. FDG-PET in PSP shows the classic “pimple sign” (hypometabolism of the midbrain) on mid-sagittal section. Also seen is mild hypometabolism of medial prefrontal and supplementary motor cortex. In CBD, asymmetric frontoparietal hypometabolism is observed on the lateral view

Full size image
Fig. 8
figure 8

Flortaucipir PET in autopsy-confirmed PSP and CBD. Flortaucipir PET (AV-1451) in PSP shows increased uptake in the midbrain (substantia nigra) and dentate nucleus of the cerebellum. In a case of CBD that presented with progressive speech apraxia, flortaucipir PET demonstrates asymmetric increased uptake in premotor neocortex

Full size image

Biofluid Biomarkers

CSF tau species, including measures of total tau (t-tau) and phosphorylated tau (p-tau) tend not to be elevated in PSP [198,199,200]. One study reported that a ratio of certain tau fragments may aid in distinguishing PSP from healthy controls and other conditions [201], but the findings could not be replicated [202]. CSF neurofilament light chain (NfL) is an intermediate filament, which can be measured from CSF and is a nonspecific measure of neuronal injury [203], but it shows elevation in PSP, CBD, and other parkinsonian syndromes that can aid in differentiating PSP or CBD from Parkinson’s disease [200, 204,205,206,207]. The sensitivity of the next-generation single-molecule-array assays has made blood-based NfL measurements possible now as well [208, 209]. Real-time quaking-induced conversion (RT-QuIC) is an emerging assay that was originally developed to aid in diagnosis of Creutzfeldt-Jakob Disease (CJD), where a biologic sample is placed in wells containing monomeric proteins and a fluorescent marker and through polymerization encouraged by sequential shaking steps, can show the presence or absence of a pathologic “seed” from the patient sample. This technique has been adapted to detect alpha-synuclein [210], 3R/4R tau species [211], 3R tau species [212], and a 4R tauopathy assay is under development as well [213], which may offer molecularly specific aid in diagnosis in the near future.

Treatment of Progressive Supranuclear Palsy and Corticobasal Degeneration

Current treatment strategies for both PSP and CBS are supportive and symptomatic as no disease-modulating therapies are currently available for either condition.

Parkinsonism

Levodopa preparation may still be trialed to treat the parkinsonism associated with PSP and CBS. In one study of pathologically confirmed PSP patients, approximately one-third of PSP patients showed a significant improvement (> 30% improvement in the Unified Parkinson’s Disease Rating Scale) [46], which is a response rate that has been reported in other studies as well [64,65,66,67]. Doses of over 1 gm/day of levodopa for 1 month are proposed to elicit responses. Often, however, responses to levodopa are very mild in PSP and CBS, if present at all, and typically wane over time [20, 24, 55, 68, 99, 214]. Dopamine agonists have been trialed in PSP but are generally less effective than levodopa and are more likely to cause side effects [65, 215, 216]. Smaller studies documented improvement in parkinsonism using amantadine or amitriptyline in PSP, but caution is warranted because of possible anticholinergic side effects, including cognitive and psychiatric disturbances, dry mouth, or difficulty with urination [65, 217,218,219].

Ocular Symptoms

Zolpidem showed mild improvements in saccadic speed in one small study of patients with PSP, but those findings have not been replicated [220,221,222]. Botulinum toxin may be used to treat blepharospasm and eyelid-opening apraxia, but high doses are often required to achieve benefits [223, 224]. Artificial tears and ophthalmic ointments may be used to treat dry eyes, and sunglasses may be of use to aid in photosensitivity symptoms. Alternating an eye patch is useful for double vision, and, occasionally, prism lenses may be fashioned, if the deficits are fixed.

Spasticity, Dystonia, and Myoclonus

Muscle relaxants such as baclofen, tizanadine , and cyclobenzaprine may be considered, but they must be carefully weighed against their possible side effects of somnolence [225]. Botulinum toxin may be used for the disabling focal dystonia of the limbs or neck that occurs in both conditions [223, 225, 226]. Clonazepam or levetiracetam can treat the myoclonus associated with CBS as can valproate [214, 227, 228].

Sialorrhea

Again, botulinum toxin may be used to treat sialorrhea [229], as can medications including glycopyrrolate or 1% atropine drops placed sublingually, although the latter, if not carefully applied, can be absorbed systemically and cause anticholinergic side effects [230].

Memory Impairments

Acetylcholinesterase inhibitors such as donepezil, rivastigmine, or galantamine may offer some mild improvement in memory function, but studies showed that it may worsen gait and dysphagia in PSP and worsen behavioral symptoms in FTD, so it should be used with caution [227, 231, 232]. No studies of memantine in autopsy-confirmed CBD have been performed, but multiple studies of memantine for memory dysfunction in FTD have failed to show benefits [233, 234].

Mood Changes

Selective serotonin reuptake inhibitors or serotonin-norepinephrine reuptake inhibitors may be used to treat depression and anxiety, but they are not helpful for the apathy that can accompany PSP or CBS [227]. Dextromethorphan-quinidine is an effective treatment for pseudobulbar affect as are antidepressants [235].

Nonpharmacological Therapies

PSP and CBS patients benefit from multidisciplinary care from providers knowledgeable about these conditions. Physical therapy decreases the likelihood of falls and improves global functioning [227, 236,237,238]. Weighted walkers are often recommended to aid in safer ambulation. Speech therapy may be employed to strengthen vocal muscles but to also provide strategies for more effective communication [239, 240]. Swallowing evaluations are essential if the patient complains of dysphagia or frequent coughing during meals as food consistency or eating habits may be modified. Safety inspections of the home may be helpful and can often be done my occupational therapists who can suggest changes and modifications to promote safety. Social workers are often needed to aid in utilization of resources that may be available to these patients. Lastly, palliative care consultants can help to manage transitions to less aggressive modalities of care and to promote symptom management and navigate end-of-life decision-making in a way that aids in both the patients and the families’ quality of life [241].

Experimental Therapies for Progressive Supranuclear Palsy and Corticobasal Degeneration

Although there are no current disease-modulating treatment for PSP or CBD, several medications are under investigation, many of which target the tau protein by different mechanisms: by decreasing production, stabilizing microtubules, promoting immune system clearance, or modifying post-translational changes.

Tau in PSP and CBD commonly undergoes post-translational phosphorylation and acetylation [242]; unfortunately, trials of the GSK-3β kinase inhibitors lithium, valproate, and Tideglusib failed to show efficacy or were stopped due to poor tolerability [243]. Salsalate inhibits tau acetylation in animal models and is currently under early investigation (NCT02422485) [244]. O-Glc-NAC modification and caspase-mediated cleavage are other potential therapeutic targets [245, 246]

The microtubule-stabilizing agent davunetide failed to show efficacy in a phase IIb/III trial [247], and the taxane derivative TPI-287 inducted anaphylactic reactions, which necessitated trial stoppage [248]. Other compounds still under investigation that are thought to work through this mechanism include epothilone-D and methylene blue [249, 250].

Anti-inflammatory medications have been trialed in PSP, including rasagiline, CoQ10, and riluzole, but studies have failed to show efficacy [251,252,253], although there was significant benefit in a shorter trial using CoQ10 [254].

Tau immunotherapy is actively under investigation. Specifically, in PSP, the BIIB092 antibody product, directed against the N terminus of extracellular tau [255], showed promise in early trials [256, 257], but a phase II study failed to show efficacy (PASSPORT NCT03068468) [258]. Similarly, ABBV-8E12 had favorable early safety results and good target engagement [259, 260] but failed to show efficacy in larger trials. While these results are discouraging, a number of questions remain regarding this strategy, namely if proper epitopes of tau were selected [261, 262], if oligomeric species or intracellular tau should be prioritized although it is technically more challenging [184, 262,263,264,265,266,267], or if alternative delivery systems may increase blood–brain barrier penetration of antibody products and improve efficacy [184].

Gene therapy through small interfering RNA (siRNA) or antisense oligonucleotides are currently being investigated in animal models of tauopathies [268,269,270] and may be of future use in PSP and CBD.