Cognition in Trinucleotide Repeat Spinocerebellar Ataxias: A Review

Abstract

Spinocerebellar ataxias (SCAs) comprise a group of complex and heterogeneous hereditary neurodegenerative disorders characterized by cerebellar ataxia, with ophthalmoplegia, pyramidal and extrapyramidal features, peripheral neuropathy, motor neuron disease, pigmentary retinopathy, epilepsy, and dementia in varying proportions. Cognitive impairment is not frequent in SCAs but is rarely noticed since it gets camouflaged behind the exorbitant ataxic manifestations of the disease. The exact incidence and extent of cognitive impairment in these rare disorders are not known due to the heterogeneity between different SCA types and different modalities of testing employed in different studies. Through our review, we have summarized the cognitive aspects of SCA and can safely conclude that cognitive dysfunction is common in some SCA types when compared to others. Not only is it important to appreciate its presence as a symptom complex in SCA but also is the need to actively search and treat it to improve the patients' quality of life.

INTRODUCTION

Spinocerebellar ataxias (SCAs) comprise a group of complex and heterogeneous hereditary neurodegenerative disorders characterized by cerebellar ataxia with ophthalmoplegia, pyramidal and extrapyramidal features, peripheral neuropathy, motor neuron disease, pigmentary retinopathy, epilepsy, and dementia in varying proportions.[1,2,3,4,5] These are rare (incidence between 1 and 5 per 100,000) disorders having the onset between 30 and 50 years of age.[1,2] Although ataxia is a criterion standard, different SCA types have variations in their pathological topography leading to differences in their clinical, radiological, and cognitive profile at a subtype level. The exact incidence and extent of cognitive impairment in these rare disorders are not known due to the heterogeneity between different SCA types and different modalities of testing employed in different studies.[1]

COGNITIVE IMPAIRMENT IN SPINOCEREBELLAR ATAXIA

SCAs are characterized by the involvement of prominent cerebellar but extra-cerebellar structures as well. As a result, various plausible theories have been postulated regarding cognitive dysfunction in the same.

Cerebellar cognitive affective syndrome[5]

The use of anatomical investigations and trans-synaptic tracing techniques in primates provided the first evidence of the presence of cortico-ponto-cerebello-thalamo-cortical loops, which form the basis for the proposition that the cerebellum plays a role in cognition. These loops connect the prefrontal cortex (executive function) with the cerebellar dentate nuclei and associated posterolateral cerebellar cortex. The same loops have since been demonstrated in human beings through diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) studies.

Schmahmann and Sherman in their landmark article provided the framework for the constellation of non-motor symptoms associated with localized cerebellar insult and termed it as “cerebellar cognitive affective syndrome (CCAS).” These include impairments in executive functions, language, visuospatial functions, and personality. These were associated with posterior cerebellar lesions and the manifestations were ascribed to disruption of circuitry linking the same to cerebral cortical association areas and paralimbic regions.

Prefrontal Cortico-Striato-Thalamo-Cortical loops[6,7]

Since many SCAs like SCAs 2, 3, and 17 have concomitant parkinsonism, an analogous involvement of these loops was suggested. The orbitofrontal and dorsolateral prefrontal cortex relay in the striatum and the involvement of this circuitry may be the cause of cognitive dysfunction in these SCA types. This usually presents with a typical “frontal- subcortical” profile of cognitive deficits.

Basal forebrain nuclei cholinergic projections[8]

The major cholinergic input to the hippocampus, amygdala, and neocortex is provided by the projections from the basal forebrain nuclei. Their involvement was hypothesized when the evidence of dysfunction in verbal and visual memory was documented in the SCA patients, which remains unexplained by (a) and (b). It was thought to be contributory in a way similar to that seen in Alzheimer's disease. Cortical cholinergic denervation has since been demonstrated predominantly in SCAs1 and 2.

Direct cortical involvement[9]

Radiological and post-mortem assessments from a few studies have revealed that certain SCA types (like SCA12) directly affect the neurons of the association cortex leading to neuronal loss and cortical atrophy.

Crossed cerebellar diaschisis[9]

Crossed cerebellar diaschisis (CCD) cannot be ruled out as a definite contributor for cognitive impairment in the SCA patients in the absence of functional imaging as unilateral cerebellar damage has been shown to have accompanied decreased blood flow in the contralateral basal ganglia and frontoparietal cortex.

TESTING COGNITIVE IMPAIRMENT IN SCA

A few studies have looked at cognitive dysfunction in SCA but have suffered from many confounding factors.

First is the inconsistency of test selection from the battery of standard neuropsychological tests available. Not only does this result in difficulty in comparing study results and conducting a meta-analysis but also sometimes the choice of the test is not sensitive enough to identify the cognitive deficit in cerebellar dysfunction alone.[9]

Sometimes, the tests chosen might depend on the rapid motor or verbal response, which may be hampered in ataxic patients leading to them being wrongly classified in the impaired cognition group.[9] Examples of such tests include the Tower of London and Tower of Hanoi tests, which are timed and require a substantial degree of motor accuracy.

The Mini-Mental State Examination (MMSE) is one of the most commonly used screening tools for cognitive impairment in research and clinical practice, but it is insensitive to executive dysfunction, and therefore, is likely to underestimate the cognitive dysfunction of the SCA patients, if used alone.[10]

The cognitive tests that are not dependent on the performance speed pressure and do not place undue importance on accuracy and motor/visual speed can be used to study cognition in ataxic subjects including those with SCA. Examples of such tests include the Wisconsin Card Sorting Test and Raven's Progressive Matrices. Even Trail making or Stroop tests may be used (although they are timed tests per se) because their timed internal component can be used to account for the articulatory and upper limb motor deficiencies along with visual scanning.

Individual SCAs Cognitive domains found affected in various studies in different SCA types have been mentioned in Table 1.

Table 1.

SCA studies on cognition with the domains affected and their correlations

Publication	SCA type	Number of patients	Cognitive domains studied	Affected	Correlation
Kish et al., 1988[8]	SCA1	14		Mildly reduced verbal and non-verbal intellectual ability, memory, and executive function - 11/14 cases
Bürk K, et al., 2001[11]	SCA1	14	IQ, attention, verbal and visuospatial memory, executive function	Verbal memory and executive dysfunction	None
Storey et al., 1999[10]	SCA2	8	MMSE, executive function, FAS	Frontal lobe function- verbal fluency, Stroop and WCST
Gambardella et al., 1999[12]	SCA2	17	Conceptual reasoning, memory, language, depression	Conceptual reasoning- WCST	Decrease in the saccadic velocity
Bürk K, et al., 1999[13]	SCA2	17	IQ, attention, verbal, visuospatial memory, and executive functions	verbal memory and executive dysfunction	Disease duration for both
Maruff P, et al., 1996[6]	SCA3	6	Visual attention, learning and visual memory	Visual attentional function	None
Zawacki TM, et al., 2002[14]	SCA3	6	MMSE and other neuropsychological batteries	Relative impairments on timed verbal attention tasks and verbal fluency, executive impairments, depression	None
Kawai Y, et al., 2004[15]	SCA3	16	Executive functions, visuospatial perception, and verbal memory, attention, immediate and delayed recall, logical thinking function, and orientation function	Executive functions, visuospatial perception, and verbal memory, attention, immediate and delayed recall, logical thinking function, and orientation function	Verbal fluency correlated with ataxia severity
Braga-Neto P, et al., 2012[16]	SCA3	38	MMSE, all cognitive domains	Executive and visuospatial function
Roeske S, et al., 2013[17]	SCA3	11	All cognitive domains	Verbal and learning memory
Bürk K, et al., 2003[18]	SCAs1, 2 and 3	SCA1-11	IQ, attention, executive function, verbal and visuospatial memory	SCA1: Executive dysfunction	None
		SCA2-14		Mild deficits of verbal memory were present in SCA1, SCA2, and SCA3
		SCA1-11
Ma J, et al., 2014[19]	SCAs1, 2 and 3	SCA1-8	Executive functions, visuospatial perception, and verbal memory, attention, immediate and delayed recall, logical thinking function, and orientation function	SCA1: executive function, visuospatial perception, and attention	Severity of ataxia
		SCA2-2		SCAs 2 and 3: immediate and delayed recall, executive function, and verbal memory
		SCA3-8
Globas C, et al., 2003[20]	SCA6	12	IQ, attention, verbal and visuospatial memory as well as executive function	Mild impairment in most categories but none reached a statistical significance	None
Suenaga M, et al., 2007[21]	SCA6	18	Attention, verbal memory, visuospatial memory, and executive function	Verbal fluency and immediate visual memory	None
Cooper FE et al., 2010[22]	SCA6	27	All cognitive domains	Executive function; phonemic and category fluency
van Gaalen J, et al., 2014[23]	SCA6	29	Language functions	Mild-to-moderate linguistic impairment, most distinct on the writing and comprehension sub-tests	Severity of ataxia
Rentiya ZS et al., 2017[24]	SCA6	21	General intelligence (MMSE), executive functions, visuospatial perception, and verbal memory, attention, immediate and delayed recall	Visuospatial executive function, phonemic verbal fluency, and semantic-verb word generation	Verbal working and immediate visuospatial memory correlated with disease duration for SCA6
Klinke I, et al., 2010[25]	SCA1, 2, 3 and 6	32	All cognitive domains	SCA1s 2 and 3: Poor frontal attention and executive deficits SCA6: Mild impairment	Age of onset; disease duration
Agrawal A, et al., 2021[26]	SCA12	30	All cognitive domains	Executive function and new learning ability	None
Bruni AC, et al., 2004[27]	SCA17	16

SCA1

It is caused by CAG repeat expansion on the ataxin-1 gene[28] resulting in an expanded polyglutamine tract of 39-91 residues, and histopathologically, it is associated with marked atrophy of the brainstem, cerebellar cortex, and deep nuclei, red nuclei, VPL Ventroposterolateral nucleus of the thalamus, and Betz cells of the motor cortex.[29] Of particular relevance to our review is the fact that there is also a loss of cholinergic forebrain nuclei and volume loss of the frontal and prefrontal cortex.[30] However, radiologically, brainstem atrophy is present but cortical atrophy is absent.

Cognitive changes have been described in 5–25% of the patients of SCA1, and are usually seen in the advanced stages of the disease.[31] Overall, they were found to have mild cognitive impairment with executive dysfunction being the most common. This dysfunction rarely progressed to dementia. Social cognition was not affected.[9]

SCA2

It is caused by CAG repeat expansion on the ataxin-2 gene resulting in an expanded polyglutamine tract of 33->200 residues and its histopathological involvement resembles SCA1.[32,33,34] However, the radiological brainstem atrophy is more severe than SCA1 and a “hot cross bun” sign can be seen in 25% of the patients.[35] It is the commonest SCA in India and the frequency of cognitive impairment is estimated to be between 5 and 19% among different studies.[13,31] The cognitive dysfunction profile is “frontal-subcortical” and supportive of the hypothesis of neuronal involvement extending beyond the cerebellum, involving the pallidoluysian system and the substantia nigra.[13]

SCA3 (Machado-Joseph disease)

It is caused by CAG repeat expansion on ataxin-3 gene resulting in an expanded polyglutamine tract of 51–89 residues,[36] and histopathologically, it prominently affects the deep cerebellar nuclei and the red nucleus, while the remaining structures are less severely involved and the basal forebrain cholinergic circuit being spared.[29] Like SCA1, radiologically, the cerebral cortex does not show atrophy despite severe neuronal loss, and the “hot cross bun” pattern may be seen in 1% of the cases.[35] Cognitive impairment has traditionally been considered to be uncommon in this type and individual studies have also not reported a significant occurrence of dementia. Some studies have found the presence of impairments in executive function, visual attention, verbal, and visuospatial memory.[14,15,37] Functional MRI studies revealed hand-movement-related cerebellar activation impairment but the absence of significant reduction of signal in the cerebellar cortex or cerebellar nuclei.[38]

SCA6

It is an uncommon type of SCA and occurs due to a short (20–33) CAG repeat expansion on the CACNA1A subunit of the PQ-type calcium channel.[39] It usually has a later age of onset as compared to the other SCAs. Histopathologically, there is a Purkinje cell loss in the cerebellar cortex and Betz cells in the motor cortex with mild variable neuronal loss in the other regions.[29] However, there is no macroscopic cerebral atrophy and radiology typically reveals pure cerebellar atrophy.[40] A detailed neuropsychological assessment revealed marked impairment of the visuospatial memory, semantic and phonemic fluency tasks with some impairment on the response inhibition test and executive function. The intellectual functioning was not impaired.[31,41] The functional MRI studies revealed hand-movement-related cerebellar activation impairment and significant reduction of signal in the cerebellar cortex or cerebellar nuclei.[38]

SCA12

It is caused by CAG repeat expansion in the promoter region of the PPP2R2B gene on chromosome 5q32,[42] presenting characteristically with action and head tremors with dysarthria.[43] Neuropathologically, marked atrophy of the cerebral cortex and Purkinje cells with less prominent pontine and cerebellar atrophy,[44] and radiologically, varying degrees of mild-to-moderate cerebral and cerebellar atrophy (cerebral > cerebellar)[44] are found. Cognitive impairment was found to be a part of the disease spectrum and characterized by executive dysfunction and disability in new learning ability even early in the disease course.[44] This did not correlate with the patient's age, age at onset, disease duration, or CAG repeat length. DTI in SCA12 pedigree patients revealed microstructural changes in the white matter of the brain even in presymptomatic patients. This was found to first occur in the cerebral cortex and cerebellar vermis.[45]

SCA17

It is caused by CAG repeat expansion on the gene encoding TATA Box-Binding Protein (TBP) resulting in an expanded polyglutamine tract of 41–63 residues,[46] and histopathologically, prominently involves the cerebellar and cerebral cortex, basal ganglia, and medial thalamus.[29] Radiologically, it displays both marked cerebral and cerebellar with mild brainstem atrophy.[40] Psychiatric features like hallucinations and personality changes along with early and severe dementia are common. It usually starts with behavioral complaints and decreased verbal fluency, finally culminating in a frontotemporal dementia pattern of cognitive impairment.[47]

CONCLUSION

Cognitive impairment is not infrequent in SCAs but is rarely noticed since it gets camouflaged behind the exorbitant ataxic manifestations of the disease. This subsequent paucity of testing leads to a false sense of cognitive well-being when there is none present. This problem is further compounded by the absence of consistent neuropsychological assessment tools and studies with an adequate number of patients who have been longitudinally studied over long periods to understand the pattern of cognitive disease progressions. Also, cognitive assessment of SCAs has predominantly been done in relatively commoner types. However, we can safely conclude that cognitive dysfunction is commoner in some SCA types when compared to others and it is important to appreciate its presence as a symptom complex in SCA patients with the need to actively search and treat it to improve the patients' quality of life.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.