Autosomal dominant spinocerebellar ataxias

INTRODUCTION — The spinocerebellar ataxias (SCAs) are a genetically and clinically heterogeneous group of neurodegenerative disorders characterized by a slowly progressive cerebellar syndrome in association with oculomotor, retinal, pyramidal, extrapyramidal, sensory, and cognitive/behavioral symptoms that vary with the affected gene. Most SCAs are adult-onset, although childhood presentations occur in some forms.

A genetic diagnosis is possible for many but not all of the recognized SCAs, and most patients have one or more affected family members. Several forms share a common pathophysiology related to expansion of cytosine-adenine-guanine (CAG) repeats.

The autosomal dominant SCAs are reviewed here. Other causes of hereditary and nonhereditary ataxia are reviewed separately. (See "Overview of cerebellar ataxia in adults" and "Overview of the hereditary ataxias".)

OVERVIEW

Nomenclature and classification — The nomenclature of the autosomal dominant ataxias is complicated and has evolved over time.

SCAs were originally organized around phenotype (Harding classification) and later in order of the identification of their genetic loci (numeric SCA classification). Yet another proposal from the International Parkinson and Movement Disorder Society (MDS) spans a variety of movement disorders and attempts to classify genetic forms based on the most prominent movement disorder and the gene name, using the ATX prefix for ataxias [1].

Until the SCA and MDS classifications are harmonized, we will continue to use SCA nomenclature in this topic where relevant. Autosomal dominant hereditary ataxias recognized in the MDS classification that do not have an SCA designation are also included here. (See 'Others' below.)

There are yet other genetic disorders that are inherited in an autosomal dominant manner, which often present with prominent and progressive ataxia. These are discussed in this review to complete the differential diagnosis of these chronic, progressive ataxic syndromes. (See 'Other autosomal dominant ataxias' below.)

A number of other hereditary ataxias with other modes of inheritance or pathobiology are discussed separately. (See "Overview of the hereditary ataxias".)

Clinical spectrum — Over 40 types of SCAs have been identified. Cerebellar ataxia is a feature of each type; other distinguishing features may suggest a particular type (table 1).

However, only 60 to 75 percent of patients with SCA have variants in the known loci [2]. Furthermore, among patients with apparently idiopathic sporadic cerebellar ataxia (ie, no family history), an SCA variant (most often SCA6) or Friedreich ataxia can be identified in approximately 20 percent [3]. Online resources such as the Online Mendelian Inheritance in Man (OMIM) database can be used to keep up with advances in this area.

CAG repeat disorders — Several types of SCAs (SCA1, SCA2, SCA3, SCA6, SCA7, and SCA17) are associated with expansion of cytosine-adenine-guanine (CAG) repeats in the region that encodes for polyglutamine tracts in the protein products, similar to that seen in Huntington disease. (See "Huntington disease: Genetics and pathogenesis".)

Wildtype chromosomes with a stable CAG repeat have 6 to 34 repeat units; more than 36 repeats results in an unstable, expanded, disease-causing allele [4]. Expansion of CAG repeats is thought to produce a toxic "gain of function" (ie, disease develops because the mutant form of the protein gains a new function, not because the protein loses its normal function).

The disorders associated with expansion of CAG repeats share several clinical features:

●They typically present in middle age, with progressive ataxia evolving over the ensuing 10 to 20 years.

●The greater the number of CAG repeats on expanded alleles, the earlier the age of onset and more severe the disease. Thus, juvenile-onset disease typically is associated with very large expansions.

●As examples of the variability in onset, a prospective study reported the mean age of onset for several common types of SCAs as follows [5]:

•SCA1, 37 years (range 5 to 65)

•SCA2, 35 years (range 7 to 66)

•SCA3, 37 years (range 5 to 66)

•SCA6, 55 years (range 31 to 77)

●The repeats show both somatic and germline instability. As a result, successive generations of affected families experience anticipation, a phenomenon characterized by earlier onset and a progressively worse phenotype in subsequent generations.

●Only a certain subset of neurons is vulnerable to dysfunction, even though the relevant protein is expressed widely throughout the brain and other tissues.

●Cerebellar atrophy is the most common reported finding. Brainstem atrophy is variable, being more characteristic of SCA1, SCA2, and SCA7. Neurodegeneration of the reticulotegmental nucleus of the pons has been reported for patients with SCA1, SCA2, and SCA3; this nucleus plays a role in the performance of horizontal smooth pursuit eye movements and in the accuracy of horizontal saccades [6].

Diagnostic evaluation — Issues specific to the evaluation of a patient with suspected SCA are discussed here. The etiologic evaluation of a patient with ataxia more generally is reviewed separately. (See "Overview of cerebellar ataxia in adults", section on 'Etiologic evaluation'.)

Genetic testing — In the setting of a patient who has a positive family history of cerebellar ataxia, genetic testing is the most efficient and definitive way to determine the cause of the symptoms and to identify the ataxia subtype. The few clinical differentiating features (table 1) may suggest the likely candidate(s) for a stepwise screen. More commonly, however, an ataxia repeat expansion gene panel is an appropriate initial test. Most such panels including testing for SCA1, SCA2, SCA3, SCA6, SCA7, SCA8, SCA10, SCA17, and dentatorubral pallidoluysian atrophy (DRPLA).

Some patients are referred having previously undergone genome sequencing with uninformative results. Importantly, repeat expansions will not be reliably picked up by most clinical whole-exome or whole-genome sequencing tests unless enhanced bioinformatics or advanced sequencing technologies are used [7]. Online resources (eg, the Genetic Testing Registry) can be used to determine which SCA genes can be tested and the appropriate type of test.

Additional information on genetic testing for patients with progressive ataxia of unknown etiology is presented separately. (See "Overview of cerebellar ataxia in adults", section on 'Genetic testing'.)

Neuroimaging features — Cerebellar atrophy is the most common structural neuroimaging feature in patients with SCA. Among the SCAs, atrophy is most prominent in SCA2 and least prominent in the milder diseases, SCA5 and SCA6.

Brainstem atrophy, which would be expected to be almost universal, is characteristic of SCA1, SCA2, and SCA7. In one study, the degree of atrophy correlated with the neurologic deficit in SCA1 but not SCA2 [8]. Brainstem atrophy is minimal in SCA3 and DRPLA and is rare in SCA6.

Cerebral atrophy with compensatory enlargement of the lateral ventricles can be seen in SCA2, the infantile variant of SCA7, and DRPLA.

Magnetic resonance spectroscopy and positron emission tomography (PET) are sensitive for deterioration in patients with SCA when examined longitudinally and quantitatively; in this way they can demonstrate abnormalities not seen on magnetic resonance imaging (MRI). At present, these modalities are primarily being explored as potential biomarkers for clinical trials.

Electrophysiologic testing — Most SCAs cause defects in nerve conduction, especially SCA4. The predominant axonal neuropathy primarily affects sensory neurons, and sural nerve action potentials often are absent. However, these studies are rarely useful in determining the type of ataxia.

On the other hand, visual-evoked potentials may suggest SCA7 because of the characteristic macular degeneration, whereas interictal electroencephalographic (EEG) abnormalities may suggest DRPLA. Seizures also are common in SCA10, but the interictal EEG is typically normal.

Prognosis — The age of onset and rate of disability progression for the more common SCAs vary according to individual differences in the length of the causative CAG trinucleotide repeat. Most of the available data concern the four most common types of SCAs, which are SCA1, SCA2, SCA3, and SCA6.

Generally, patients require a wheelchair by 10 to 15 years after symptom onset. Several studies have found that SCA1 is associated with more rapid disease progression and shorter survival compared with SCA2, SCA3, and SCA6 [5,9-11]. In one of these reports, the median age at death for patients with SCA1 was 56 years [9]. Higher baseline scores on the Scale for the Assessment and Rating of Ataxia (SARA) were associated with shorter survival for all four subtypes [10].

Disability and mortality data for the other types of SCAs are limited. However, in most there is disability progression over one to two decades on average, and lifespan is shortened, though some types are compatible with a normal lifespan [12].

Treatment — No effective treatment is available for the SCAs. Investigational approaches include interventions to suppress polyglutamine neurotoxicity and genetic therapies to reduce polyglutamine gene products [13]. Treatments with some promise include antisense oligonucleotides, ribonucleic acid (RNA) interference, and stem cell therapies.

SPECIFIC DISORDERS

SCA1 — In different populations, SCA1 accounts for 3 to 16 percent of autosomal dominant cerebellar ataxias [14-18].

●Clinical features – SCA1 typically presents in the third to fourth decade of life, but tremendous variability in onset and clinical severity exists. It is characterized clinically by progressive cerebellar ataxia, dysarthria, and bulbar dysfunction. Other findings include hyperreflexia, increased tone, extensor plantar responses, and, in some patients, masking of these upper motor neuron findings by peripheral nervous system disease that results in wasting of the extremities and generalized fasciculations.

●Pathology – Pathologically, SCA1 is characterized by degeneration of cerebellar Purkinje cells, brainstem cranial nerve and inferior olivary nuclei, and the spinocerebellar tracts [19].

●Pathogenesis – SCA1 is caused by a cytosine-adenine-guanine (CAG) repeat expansion on chromosome 6p22.3 in the coding region of the ataxin 1 (ATXN1) gene [20]. Expanded ATXN1 genes contain from 36 to 81 uninterrupted CAG repeats (normal alleles have 6 to 44 repeats; alleles with 36 to 44 CAG repeats are considered normal if they are interrupted by a few cytosine-adenine-thymine [CAT] units). Mutant ATXN1 aggregates into single nuclear inclusions [21]. This aggregation reflects, at least in part, resistance of the mutant protein to degradation. The pathogenetic role of the nuclear inclusions themselves is uncertain [22].

The pathogenesis of SCA1 involves primarily a gain-of-function mechanism and, to a lesser degree, a partial loss-of-function mechanism for ATXN1 [23]. In addition to the expanded polyglutamine tract, which is located towards the N-terminus of ATXN1, other important mediators include the ATXN1 and HMG-box protein 1 (AXH) domain (amino acids 570 to 689), the nuclear localization sequence (NLS; amino acids 771 to 776), and the C-terminal amino acid of the NLS, serine 776 (Ser776). The AXH domain forms an oligonucleotide-binding fold [24] and also interacts with a number of transcription factors and proteins [24,25], including the capicua transcriptional repressor (CIC) protein [26]. The NLS directs mutant ATXN1 to the nucleus, which is necessary for disease to occur [22]. Phosphorylation of Ser776 is also necessary for toxicity of mutant ATXN1 [27] and is important for the interaction of ATXN1 with a protein called RNA-binding motif protein 17 (RBM17) [28,29].

One model of ATXN1 toxicity postulates that wildtype ATXN1 exists in an equilibrium of two main protein complexes, one involving ATXN1 (not phosphorylated at Ser776) associated with CIC, and the other involving ATXN1 (phosphorylated at Ser776) associated with 14-3-3, RBM17, and possibly other proteins [30,31]. The expanded polyglutamine tract of mutant ATXN1 enhances the function of these complexes (gain of function) to drive toxicity.

●Investigational therapies – Interventions that lower ATXN1 levels could be useful for identifying therapeutic targets in SCA1. One approach is illustrated by a study that screened for kinase genes involved in the regulation of ATXN1 protein levels [32]. The investigators found that multiple components of the RAS-MAPK-MSK1 signaling cascade play a role in modulating ATXN1 protein levels in Drosophila and human cell lines. In addition, pharmacologic inhibitors of the pathway reduced ATXN1 levels. Further studies are needed to identify additional pathways that may be involved in ATXN1 regulation and to determine whether manipulating these pathways is beneficial in SCA1.

The same type of approach may be useful to look for therapeutic targets in other neurodegenerative diseases characterized by the toxic accumulation of mutant proteins. Other approaches that may prove to be of clinical relevance include the proof-of-concept preclinical data using either antisense oligonucleotides to lower ATXN1 or a microRNA that targets ATXN1 RNA [33,34].

SCA2 — In different populations, SCA2 accounts for 6 to 18 percent of SCA kindreds [14-17,35].

●Clinical features – SCA2 is distinguished from SCA1 by the presence of slow saccadic eye movements [36]. Clinical features are otherwise similar to those of SCA1. (See 'SCA1' above.)

●Pathogenesis – SCA2 results from expanded polyglutamine tracts in the ataxin 2 (ATXN2) gene on chromosome 12q24, which encodes a protein called ataxin 2 [37,38]. Ataxin 2 contains structural elements that appear to be important in RNA splicing. In contrast to the nuclear inclusions in SCA1, mutant ataxin 2 is associated with cytoplasmic microaggregates [39].

●Genotype-phenotype correlations – Disease-causing alleles usually contain 35 to 64 CAG repeats (normal 14 to 31 repeats). An inverse relationship exists between disease onset and severity and the number of CAG repeats [40,41].

•When ATXN2 alleles contain more than 200 repeats, the disease can present as early as infancy with hypotonia, developmental delay, infantile spasms, autonomic dysfunction, dysphagia, and retinitis pigmentosa [42,43].

•On the other hand, disease onset in late adulthood, manifested as ataxia, slow saccades, and hyporeflexia, has been described in two patients with 33 CAG repeats [44].

In one ethnic Chinese family, SCA2 has been associated with a predominant phenotype resembling dopa-responsive Parkinson disease or an ataxia-parkinsonism phenotype similar to progressive supranuclear palsy [45]. (See "Diagnosis and differential diagnosis of Parkinson disease".)

In addition, occasional patients with polyglutamine expansions in ATXN2 have a phenotype of motor neuron disease similar to or indistinguishable from amyotrophic lateral sclerosis. (See "Epidemiology and pathogenesis of amyotrophic lateral sclerosis", section on 'Genetic susceptibility in sporadic ALS'.)

Genetic background also influences the phenotype in geographically distinct families. For example, cognitive decline has been described in an Italian family [46,47], and chorea and dystonia have been described in families from Tunis and Martinique [48,49].

SCA3 (Machado-Joseph disease) — SCA3, also known as Machado-Joseph disease (MJD), is the most common of the autosomal dominant SCAs. It is present in 21 to 23 percent of kindreds in the United States [15,16,18], 12 percent in Australia [17], and as many as 48 percent in China [14].

●Clinical features – MJD can be associated with a variety of symptoms other than ataxia. They include:

•Slow saccades and saccadic pursuit

•Lid retraction that gives the impression of a persistent stare

•Signs of brainstem dysfunction, such as dysarthria, difficulty in swallowing, poor cough, and tongue fasciculations

•Signs of upper and lower motor neuron dysfunction; thus, tone can range from hypotonia to rigidity, reflexes can range from absent to exaggerated, and the plantar response is usually extensor

•Extrapyramidal features including rigidity and dystonia

•Muscle cramps and fasciculations [50]

•Cognitive impairment, including verbal and visual memory deficits, impairment of verbal fluency, and visuospatial and constructional dysfunction [51]

•Autonomic dysfunction with symptoms such as cold intolerance, nocturia, and orthostatic dizziness [52]

•Sleep disorders, including rapid eye movement sleep behavior disorder, restless legs syndrome, insomnia, and excessive daytime sleepiness [53,54]

●Pathogenesis – MJD is caused by CAG expansions in the ataxin 3 (ATXN3) gene (previously known as the SCA3 or MJD gene) on the long arm of chromosome 14 (14q32) [55,56]; it encodes a protein called ataxin 3 [55]. The mutant gene has an increased number of CAG repeats (40 to more than 200 versus 12 to 41 [interrupted] in the wildtype allele).

Wildtype ataxin 3 is predominantly cytoplasmic, whereas mutant ataxin 3 is localized within the nucleus of neuronal cells [57]. The mutant protein forms nuclear inclusions and is associated with cell degeneration [58,59].

●Genotype-phenotype correlations – Phenotypic variation is most likely a reflection of the number of CAG repeats and other genetic factors [60].

SCA4 — SCA4 has few features that distinguish it from the other hereditary ataxias, except perhaps for an exaggerated sensory axonal neuropathy and extensor plantar reflexes [61]. The gene locus maps to chromosome 16q22 [61,62].

SCA5 — SCA5 is characterized by an almost pure cerebellar syndrome. It is a relatively mild disorder that typically begins between the ages of 20 and 30 and progresses slowly [63], though infantile onset with a more severe phenotype has also been reported [64]. MRI reveals global cerebellar atrophy [63].

The gene locus is in the centromeric region of chromosome 11 [65], and the disease has been associated with variants in the spectrin beta, nonerythrocytic 2 (SPTBN2) gene [66]. The original kindred was descended from the paternal grandparents of Abraham Lincoln.

SCA6 — SCA6 accounted for 15 to 17 percent of dominant cerebellar ataxias in two series [16,17]. A population-based molecular genetics study found that the point prevalence of SCA6 in the northeast region of England was 1.59 in 100,000, and the number of individuals who had or were at risk of developing SCA was at least 1 in 19,000 [67].

●Clinical features – SCA6 shares many clinical features with SCA5 [68-70]. It is associated with slowly progressive cerebellar ataxia that begins between the ages of 20 and 60; global cerebellar atrophy is seen on MRI [69,70]. Other clinical features that may be seen include horizontal and vertical nystagmus and an abnormal vestibulo-ocular reflex [68].

●Pathogenesis – SCA6 is a polyglutamine disorder caused by an expanded CAG repeat in the calcium voltage-gated channel subunit alpha1 A (CACNA1A) gene, but the CAG expansion is relatively small (21 to 33 repeats versus fewer than 18 in unaffected subjects).

In one series of Japanese patients with SCA6, 9 of 35 had no apparent family history of ataxia; these patients had smaller CAG repeats (21 or 22) and later age of onset (65 years) [70]. A similar finding was noted in another report of patients with apparently idiopathic sporadic cerebellar ataxia. A variant could be identified in 19 percent; SCA6 was the disorder most commonly identified in patients with late-onset disease (47 to 68 years of age) [3]. One patient who was homozygous for the SCA6 variant had an earlier age of onset and more severe disease than did her heterozygous sister [70].

The product of the CACNA1A gene on 19p13 is the alpha-1A subunit of the P/Q type calcium channel [71]. Possible mechanisms of neuronal cell degeneration include increased calcium entry and cytoplasmic aggregates of channel protein [72,73].

The CACNA1A gene has been implicated in two other disorders: episodic ataxia type 2 and familial hemiplegic migraine [74,75]. Although the pathogenic variants are different, some overlap of symptoms exists among these disorders. SCA6 can present with intermittent ataxia in the early stages, similar to episodic ataxia, and all three conditions often are associated with cerebellar atrophy. (See "Overview of the hereditary ataxias", section on 'Episodic ataxia type 2' and "Hemiplegic migraine", section on 'Familial hemiplegic migraine'.)

SCA7 — SCA7 accounts for approximately 2 to 5 percent of dominant SCAs [16,17].

●Clinical features – SCA7 has a variable clinical expression, depending in part upon the age at onset [76].

•When the onset is in childhood, seizures, myoclonus, and cardiac involvement accompany the ataxia, and visual loss develops early in the disease course. Extremely long (>150) CAG repeats are associated with onset in infancy and evidence of systemic disease [77]. A severe infantile phenotype associated with over 200 CAG repeats has also been described [76].

•In adult-onset cases, smaller repeat expansions tend to manifest first as ataxia, whereas larger repeats can cause pigmentary macular degeneration, leading to visual loss before ataxia develops [78].

This visual loss had been thought to be unique to SCA7 [79], but retinal degeneration has been documented in childhood-onset cases of SCA2 and SCA1 caused by very large repeats. Nevertheless, in adults, defects in color vision and electroretinogram abnormalities might suggest the diagnosis early in the disease course [78].

●Pathogenesis – SCA7 is a glutamine repeat disorder that maps to chromosome 3p12 [78,79]. The number of CAG repeats varies between 37 and 306 (normal 4 to 35). Marked intergenerational instability is present, with expansion particularly likely upon paternal transmission [79,80].

The product of the SCA7 gene is called ataxin 7 and is preferentially expressed in neurons [81]. Mice overexpressing mutant ataxin 7, but not wildtype ataxin 7, have features similar to those of the human disease, with neurodegeneration involving the cerebellum and retina [82].

SCA8 — SCA8 accounts for approximately 3 to 5 percent of SCAs worldwide [83].

●Clinical features – SCA8 is difficult to distinguish clinically from the other SCAs. Affected patients typically have a slowly progressive, predominantly cerebellar ataxia affecting gait, swallowing, speech, and limb and eye movements, with variable age at onset [84,85]. Marked cerebellar atrophy is found on MRI or computed tomography (CT) scan [85].

●Pathology – Neurodegeneration of Purkinje, inferior olivary, and nigral cells, accompanied by periaqueductal gliosis, was described in one Japanese family [86].

●Pathogenesis – SCA8 is unusual because the pathogenesis involves a cytosine-thymine-guanine (CTG) expansion in the noncoding region of the ATXN8 opposite strand IncRNA (ATXN8OS) gene on chromosome 13q21 (70 to 250 repeats compared with 15 to 50 in unaffected subjects [87]) and a complementary CAG repeat in the ataxin 8 (ATXN8) gene [88,89]. Oddly, it is this particular range of repeats that is pathogenic, as opposed to any expansion; some unaffected individuals bear hundreds of repeats.

The CTG expansion is transcribed in both the forward and reverse directions to yield RNA with a noncoding cytosine-uracil-guanine (CUG) expansion from the ATXN8OS gene and a CAG transcript from the ATXN8 gene that encodes a string of polyglutamines [90]. It is likely that both these products act in dominant toxic pathways. Genetic analysis of 37 families with SCA8 ataxia found that SCA8 expansions arose independently on three different haplotypes, supporting the direct role of CTG expansion in disease pathogenesis [91].

The CTG expansion found in SCA8 shows dramatic genetic instability and age-dependent reduced disease penetrance. As an example, only one or two individuals in a given family may be affected by the disease, despite dominant inheritance. A maternal penetrance bias may exist, as CTG repeats tend to expand in mothers and contract in fathers (ie, fewer repeats in sperm than blood) [84,89]. The reduced penetrance in fathers may make the disease appear recessive or sporadic [84].

SCA types 9 to 20 — The remaining types of SCA are rare and less well characterized. Types 21 and higher are described in the sections that follow. (See 'SCA types 21 to 48' below.)

●SCA9 has not been assigned to a clinical disorder.

●SCA10 exhibits phenotypic variability and is characterized by seizures and pure cerebellar ataxia in Mexican families [92], cerebellar ataxia without epilepsy in Brazilian families [93], and extrapyramidal signs, seizures, and ataxia in an Argentinean family [94].

SCA10 exhibits anticipation, and the locus maps to chromosome 22q13 [95,96]. The disease is caused by expansion of adenine-thymine-thymine-cytosine-thymine (ATTCT) pentanucleotide repeats in the ataxin 10 (ATXN10) gene initially identified in five Mexican families; normal alleles bear 10 to 22 repeats, while disease-causing alleles contain 800 to 4500 repeats [97].

●SCA11 is a relatively mild, pure cerebellar ataxia caused by variants in the tau tubulin kinase 2 (TTBK2) gene with a locus at 15q14-q21.3 [98,99]. The TTBK2 gene encodes tau tubulin kinase 2, and neuropathologic examination of one affected individual revealed tau deposition and cerebellar degeneration [99].

●SCA12 is a trinucleotide repeat disorder caused by a CAG trinucleotide expansion in the protein phosphatase 2 regulatory subunit Bbeta (PPP2R2B) gene encoding a brain-specific regulatory subunit of protein phosphatase 2A [100]. The normal range of repeats is 7 to 28; disease alleles bear 43 to 78 repeats [101]. This ataxia often is complicated by tremor and dementia in the later stages. The genetic locus is 5q31-q33.

●SCA13 is characterized by cerebellar ataxia and intellectual disability [102]. The gene locus is on chromosome 19q13, and heterozygous variants of the potassium voltage-gated channel subfamily C member 3 (KCNC3) gene are associated with the disorder [103].

●SCA14 is not a trinucleotide repeat disorder but rather is caused by variants in the protein kinase C gamma (PRKCG) gene, located on chromosome 19q13 [104-108]. SCA14 is characterized by slow progression and long duration. The disorder can have either a late onset (≥39 years) as pure cerebellar ataxia or an earlier onset with intermittent myoclonus followed by ataxia.

●SCA15 is a very slowly progressive, pure cerebellar ataxia, with onset occurring from mid-childhood to middle age [109-112]. Additional clinical features may include a mild postural or action tremor and titubation. Neuroimaging typically reveals atrophy of the cerebellum, especially the vermis, but the cerebral hemispheres and brainstem are unaffected. Although initial analysis seemingly excluded the inositol 1,4,5-trisphosphate receptor type 1 (ITPR1) gene [111], later studies confirmed that SCA15 is caused by deletions or missense variants in the ITPR1 gene [113,114].

●SCA16 is now thought to be the same disorder as SCA15 [115,116]. It was originally identified as an autosomal dominant cerebellar ataxia in four generations of a Japanese family [110,117]. While initial linkage analysis suggested the locus was on chromosome 8q22.1-24.1 [110], a later analysis suggested that SCA16 links with 3p26.2pter, a region that partly overlaps the SCA15 region [117]. A heterozygous deletion in the ITPR1 gene, the causative gene for SCA15, was then demonstrated in a family with SCA16, confirming that SCA16 is the same disorder as SCA15 [115].

●SCA17 is an autosomal dominant cerebellar ataxia first described in four Japanese pedigrees [118]. It is caused by an abnormal CAG or CAA expansion in the TATA-box binding protein (TBP) gene, a general transcription initiation factor. The age of onset ranges from 19 to 48 years and is inversely related to the number of CAG repeats in the TBP gene (≤40 normally, 41 to 45 reduced penetrance, 46 to 66 full penetrance/disease causing) [119]. Most individuals present between the ages of 20 and 30 with gait ataxia and dementia, progressing over several decades to include bradykinesia, dysmetria, dysdiadochokinesis, hyperreflexia, and paucity of movement. One family with SCA17 has been described with a Huntington disease-like syndrome, including generalized chorea [120]. Diffuse cortical and cerebellar atrophy is present on neuroimaging in all patients.

Mouse models of SCA17 have been generated by expressing TBP with expanded polyglutamine tracts under the mouse prion promoter. These mice demonstrate a neurodegenerative syndrome. Moreover, biochemical experiments suggest that the expanded polyglutamine tracts reduce TBP dimerization but enhance the interaction of TBP with the general transcription factor IIB, implicating the polyglutamine domain of TBP in transcriptional regulation [121].

●SCA18 is an autosomal dominant disorder described in an American family of Irish ancestry with ataxia, pyramidal tract signs, muscle weakness, sensory axonal neuropathy, and mild cerebellar atrophy [122]. Haplotype analysis links the disorder to a region at 7q22-q32.

●Both SCA19, originally identified in a Dutch family [123,124], and SCA22, originally identified in a Taiwanese family [125], are autosomal dominant allelic conditions caused by variants in the potassium voltage-gated channel subfamily D member 3 (KCND3) gene that encodes for Kv4.3, a voltage-gated potassium channel [126-128]. SCA19/22 have been identified in multiple unrelated families of varied ethnicities. The phenotype is predominantly a cerebellar syndrome, but more severe cases can be associated with cognitive impairment or myoclonus.

●SCA20 is described in a family of Anglo-Celtic ancestry with a slowly progressive, autosomal dominant SCA [129]. Age of onset in 14 family members was 19 to 64 years (mean 46.5). Dysarthria is the most common initial symptom. Gait or upper limb ataxia follows in most. Palatal tremor and dysphonia are distinguishing clinical features. Dentate nucleus calcification can be visualized by brain imaging with CT or MRI. The disease locus maps to chromosome 11, where a duplication containing at least 12 genes has been identified [130].

SCA types 21 to 48 — Additional rare types of SCA include those numbered from 21 through 37 (table 1).

●SCA21 was first identified in four generations of a French family [131]. Characteristics include early onset, mild to severe cognitive impairment, motor clumsiness, and slow progression [132,133]. The disorder is caused by variants in the transmembrane protein 240 (TMEM240) gene located on chromosome 1p36.33.

●As noted above, SCA22 and SCA19 are allelic conditions identified in multiple unrelated families of varied ethnicities. The phenotype is predominantly a cerebellar syndrome, but more severe cases can be associated with cognitive impairment or myoclonus. SCA19/22 are caused by variants in the KCND3 gene that encodes for Kv4.3, a voltage-gated potassium channel [126-128].

●SCA23 is described in a Dutch family and is characterized by a late-onset (age >40 years), slowly progressive SCA with variable dysarthria, ocular dysmetria, and distal sensory deficits [134]. Variants in the prodynorphin (PDYN) gene located on chromosome region 20p13 are causative [135].

●SCA24 is a recessively inherited ataxia and has been redesignated as autosomal recessive spinocerebellar ataxia type 4 (SCAR4). This condition is described in a Slovenian family with progressive ataxia and disruption of visual fixation by large horizontal saccadic intrusions [136,137]. Corticospinal signs and an axonal sensorimotor neuropathy are additional manifestations. The disorder is due to a variant in the vacuolar protein sorting 13 homolog D (VPS13D) gene on chromosome region 1p36.

●SCA25 is described in a large French family and is characterized by cerebellar ataxia and sensory neuropathy [138]. The responsible locus maps to chromosome 2. The phenotype is highly variable regarding age of onset, severity, and clinical manifestations, ranging from infantile-onset cerebellar ataxia with pure sensory neuropathy suggestive of Friedreich ataxia to a form with mild cerebellar ataxia and prominent sensory neuropathy suggestive of Charcot-Marie-Tooth disease.

●SCA26 is described in a Norwegian family and is characterized by a pure, slowly progressive cerebellar ataxia with autosomal dominant inheritance [139]. The responsible variant is a coding variant in the eukaryotic translation elongation factor 2 (EEF2) gene that maps to chromosome 19p13.3 [140].

●SCA27A is described in a large Dutch family and is characterized by early-onset tremor, dyskinesia, slowly progressive cerebellar ataxia, cognitive impairment, and a variant in the fibroblast growth factor 14 (FGF14) gene on chromosome 13q34. [141,142]. A variant in FGF14 has also been reported in a German patient with familial ataxia [143].

●SCA27B due to heterozygous pathogenic guanine-adenine-adenine (GAA) repeat expansion deep in the first intron of FGF14 was first identified in 2023 as a cause of familial and sporadic late-onset ataxia in independent cohorts of patients and families from Quebec, Germany, Australia, and South India [144,145]. Additional studies are needed to better define prevalence and phenotypic spectrum, but this appears to be an important cause of late-onset cerebellar ataxia (identified in 13 percent of one French cohort with otherwise unexplained adult-onset cerebellar ataxia) [146]. Rare individuals with biallelic FGF14 GAA repeat expansions have also been identified with the onset of ataxia before the age of 30 years [147].

Initial genotype-phenotype analysis suggests incomplete penetrance of expansions with 250 to 300 repeats and full penetrance of expansions with >300 repeats [144]. The intronic (noncoding) GAA repeat appears to be highly unstable and associated with reduced male transmission due to contraction of repeat expansion during paternal meiosis; this phenomenon can lead to "generation skipping" when a subpathogenic allele is inherited and re-expands in a subsequent parent-offspring transmission.

The phenotypic spectrum is evolving. Among 122 patients with confirmed GAA-FGF14-related ataxia (≥250 GAA repeats), the median age of onset was 55 years, and approximately half of patients had episodic ataxia and/or downbeat nystagmus as early symptoms [144]. In a smaller cohort of sporadic cases, the median age at onset was 67.5 years, and 73 percent were male [146]. The main clinical features that helped to distinguish SCA27B cases from idiopathic late-onset ataxia were older age, vertigo, diplopia, oculomotor disorders (eg, nystagmus), absence of dysautonomia and orthostatic hypotension, and presence of mild sensorimotor axonal neuropathy. Some patients have a phenotype that resembles cerebellar ataxia, neuropathy, and vestibular areflexia syndrome (CANVAS) [148]. (See "Overview of cerebellar ataxia in adults", section on 'Autosomal recessive ataxias'.)

Disease progression appears to be slow and cognitive function relatively preserved in most patients [149]. Symptomatic benefit from 4-aminopyridine has been described and warrants further study [149].

●SCA28 is described in an Italian family and is characterized by a juvenile-onset (mean age 19.5 years), slowly progressive cerebellar ataxia with eye movement abnormalities and, in some cases, pyramidal tract signs [150]. Inheritance is autosomal dominant. The disorder is caused by heterozygous missense variants in the mitochondrial protease gene AFG3-like matrix AAA peptidase subunit 2 (AFG3L2) on chromosome 18p [151]. Intriguingly, the protein encoded by this gene is a homolog of paraplegin, a protein mutated in one form of hereditary spastic paraplegia (SPG7).

●SCA29 maps to chromosome 3p26 and may be an allelic variant of SCA15 [152]. It is characterized by early-onset, nonprogressive ataxia with variable atrophy of the cerebellar vermis.

●SCA30 is an autosomal dominant form described in an Australian family of Anglo-Celtic ancestry [153]. Clinical features include onset in mid- to late life of a slowly progressive, relatively pure ataxia with hypermetric saccades and minor pyramidal signs. Brain MRI shows atrophy of the cerebellar vermis and cerebellar hemisphere. The disorder maps to chromosome 4q34.3-q35.1. While the responsible gene is not established, the teneurin transmembrane protein 3 (TENM3) gene is a candidate.

●SCA31, previously called 16q22-linked autosomal dominant spinocerebellar ataxia, is characterized in Japanese populations by late-onset cerebellar ataxia and reduced muscle tone with or without sensorineural hearing loss [153-155]. Its locus is on chromosome 16q22.1, close to SCA4. It is caused by an insertion containing pentanucleotide repeats, including a long (thymine-guanine-guanine-adenine-adenine [TGGAA])n sequence, within an intronic region shared by the thymidine kinase 2 (TK2) gene and the brain expressed associated with NEDD4 (BEAN) family of genes [156].

●SCA32 is described in a single Chinese family with autosomal dominant ataxia [157]. Cognitive impairment affects those with onset before age 40. All males are infertile due to azoospermia. The disorder links to a locus on chromosome 7q32-q33.

●SCA33 has not been assigned to a clinical disorder.

●SCA34 is reported in a French Canadian kindred with autosomal dominant SCA and skin lesions [158], caused by a heterozygous missense variant in the ELOVL fatty acid elongase 4 (ELOVL4) gene [159]. SCA34 has also been identified in two Japanese families with slowly progressive cerebellar ataxia and pyramidal signs but no skin lesions [160].

●SCA35 is described in several families of Han Chinese descent [161-163]. Onset is at a mean age of 44 years, and the disorder is notable for slowly progressive gait and limb ataxia and mild dysarthria. SCA35 is caused by heterozygous variants in the transglutaminase 6 (TGM6) gene located on chromosome 20p13.

●SCA36 was first reported in a number of families from Japan [164-166] and Spain [167] but has since been found in various ethnic backgrounds [168]. The clinical features include late onset (mean age of 53 years in Japan) with cerebellar dysfunction manifesting as truncal ataxia, dysarthria, and limb ataxia. Motor neuron involvement occurred in most patients from Japan with longer disease duration [164,165], while sensorineural hearing loss occurred as an early feature in most cases from Spain [167]. SCA36 is caused by expansion of an intronic guanine-guanine-cytosine-cytosine-thymine-guanine (GGCCTG) hexanucleotide repeat in the NOP56 ribonucleoprotein (NOP56) gene located on chromosome 20p13.

●SCA37 is a late-onset (mean age 48 years, range 38 to 64 years) cerebellar ataxia identified in a Spanish kindred that links to chromosome 1p32 [169,170]. Prominent clinical features include falls, dysarthria, clumsiness, and abnormal vertical eye movements.

●SCA38, identified in several European families, is characterized by a late-onset (34 to 51 years), pure cerebellar phenotype with truncal ataxia, gait disturbance, and slow progression; associated features include gaze-evoked nystagmus, dysarthria, and limb ataxia. It is caused by variants in the ELOVL fatty acid elongase 5 (ELOVL5) gene [171].

●SCA40, identified in a family from China, is notable for late onset (fifth decade) of truncal and gait ataxia; additional features include dysarthria, ocular dysmetria, and pyramidal involvement with hyperreflexia and spasticity [172]. A missense variant in the coiled-coil domain-containing 88C (CCDC88C) gene is the cause.

●SCA41 was described in one 40-year-old male who presented with progressive imbalance and gait ataxia. A missense variant in the transient receptor potential cation channel subfamily C member 3 (TRPC3) gene is implicated as the cause [173].

●SCA42, identified in multiple French and Japanese families, has a varied age of onset. The main features are ataxia, dysarthria, saccadic eye movements, and nystagmus. It is caused by a heterozygous missense variant in the calcium voltage-gated channel subunit alpha1 G (CACNA1G) gene [174,175].

●SCA43, identified in five generations of a Belgian family, is characterized by gait ataxia, neuropathy, tremor, and hyporeflexia. It is caused by a missense variant in the membrane metalloendopeptidase (MME) gene [176].

●SCA44 was first identified in three generations of an English family. Characteristics include gait ataxia, dysarthria, dysphagia, dysmetria, and dysdiadochokinesia. A case of early-onset disease has also been described. SCA44 is caused by a missense variant in the glutamate metabotropic receptor 1 (GRM1) gene [177].

●SCA45 has been characterized in two generations of a family as a pure cerebellar syndrome with limb and gait ataxia, downbeat nystagmus, and dysarthria. It is caused by a missense variant in the FAT atypical cadherin 2 (FAT2) gene on chromosome 5q [178].

●SCA46, identified in a Dutch family, is characterized by cerebellar ataxia and sensory neuropathy. Other features include cerebellar dysarthria, nystagmus, jerky pursuit, square-wave jerks, slow saccades, and saccadic dysmetria. It is caused by a phospholipase D family member 3 (PLD3) gene variant [178].

●SCA47 is characterized by gait ataxia, dysmetria, dysarthria, and diplopia. An early-onset form has also been described, manifest by developmental delay, ataxia, chorea, and seizures. It is caused by missense variants in the pumilio RNA-binding family member 1 (PUM1) gene [179].

●SCA48 has been reported in multiple European families with variable degrees of gait ataxia, dysarthria, and dysphagia; deficits in executive function; and psychiatric or affective manifestations, such as depression, anxiety, and apathy. Heterozygous missense, nonsense, or frameshift variants of the STIP1 homology and U-box containing protein 1 (STUB1) gene are seen in affected individuals [180-183].

OTHER AUTOSOMAL DOMINANT ATAXIAS — Other ataxic disorders that may resemble the SCAs include dentatorubral pallidoluysian atrophy (DRPLA), fragile X-associated tremor/ataxia syndrome (FXTAS), and prion disorders.

Dentatorubral pallidoluysian atrophy — DRPLA is an autosomal dominant progressive ataxia syndrome caused by cytosine-adenine-guanine (CAG) repeat expansion in the atrophin 1 (ATN1) gene on chromosome 12p.

●Epidemiology – DRPLA is a disease that is relatively common in Japan [184]. Most reports suggest that DRPLA is rare in European and North American populations [185-187], but data are conflicting, and one study found it to be an important cause of late-onset cerebellar ataxia in Wales [188].

●Clinical features – DRPLA should be suspected when ataxia and rigidity are accompanied by choreoathetosis, myoclonic epilepsy, and dementia [184,189,190]. Other features include hyperreflexia and slowing of saccades. MRI or CT scan often reveals atrophy of the cerebellum and brainstem, calcification of the basal ganglia, and leukodystrophic changes [191].

The age of onset varies widely. The mean age of onset in Japan is 47 years [192], while the typical age of onset as reported in Western literature is between 20 and 30 years [193]. Myoclonic epilepsy is common in juvenile (age <20 years) onset. The prevalence of seizures decreases with increasing age beyond 20 years. In adult-onset cases, cerebellar ataxia, choreoathetosis, and dementia are the predominant manifestations, making DRPLA difficult to differentiate from Huntington disease and other hereditary SCAs. (See "Huntington disease: Clinical features and diagnosis".)

Haw River syndrome, a variant of DRPLA without the characteristic myoclonic epilepsy, has been described in a few families of African American descent in North Carolina. It is caused by the same expanded repeat as DRPLA [194].

●Pathogenesis – DRPLA is caused by CAG expansion of 49 to 88 repeats in the ATN1 gene on chromosome 12p. The expansion can result in anticipation, especially when transmission is via the father [189,195]. The number of CAG repeats in unaffected subjects is 6 to 35. The product of the ATN1 gene is a cytoplasmic protein called atrophin 1 [196]. The action of other enzymes on mutant atrophin 1, such as transglutaminase and caspase, may be involved in the neurotoxicity [197,198].

Fragile X-associated tremor/ataxia syndrome — FXTAS is an adult-onset progressive ataxia syndrome that is seen in some males and females who are carriers of a premutation in the fragile X messenger ribonucleoprotein 1 (FMR1) gene.

Epidemiology — Approximately 20 to 33 percent of adult male FMR1 premutation carriers display the FXTAS phenotype, and the prevalence of the premutation carrier state is as high as 1 in 813 males, suggesting that FXTAS is a frequent genetic cause of ataxia in older males [199]. Both increasing age and larger cytosine-guanine-guanine (CGG) expansions in premutation carriers are associated with greater motor impairment as assessed by severity of tremor, ataxia, and parkinsonism [200].

Although FXTAS was initially thought to affect only male premutation carriers, subsequent reports have described the development of FXTAS in females [201,202], related at least in part to X inactivation [203]. Females are less likely than males to develop FXTAS, and those who do have milder disease than males because female premutation carriers are relatively protected from the disease by the presence of a normal X chromosome [201].

Pathogenesis — Fragile X syndrome is caused by a trinucleotide CGG expansion of greater than 200 repeats in the FMR1 gene, resulting in decreased or absent levels of fragile X messenger ribonucleoprotein (FMRP). It is a common genetic cause of intellectual disability in males [204]. (See "Fragile X syndrome: Clinical features and diagnosis in children and adolescents".)

FXTAS occurs in carriers of a premutation in the FMR1 gene, defined as a CGG expansion ranging from 50 to 200 repeats [205,206]. Carriers of the FMR1 gene premutation do not have the more severe neurodevelopmental problems associated with the full variant [207]. However, premutation carriers can present with one or more of the following three clinical syndromes [208]:

●Premature ovarian failure

●FXTAS

●Mild cognitive and behavioral deficits on the spectrum of those seen in fragile X syndrome

The molecular mechanism causing FXTAS differs from that causing the fragile X syndrome. In fragile X syndrome, methylation of DNA (deoxyribonucleic acid) sequences in the promoter where the expanded CGG resides causes silencing of the FMR1 gene and complete loss of FMRP. In FXTAS, fragile X RNA is produced, but it is poorly translated into protein. This translational deficit is thought to cause a compensatory increase in the levels of FMR1 transcripts. Unfortunately, the RNA bearing the expansion now causes a toxic gain-of-function effect, triggering downstream protein accumulation and neurodegeneration [207].

Clinical features — The predominant clinical feature of FXTAS is a late-onset ataxia associated with postural tremor [205,209,210]. Additional clinical features can include short-term memory loss, executive function deficits, cognitive decline, parkinsonism, peripheral neuropathy, lower limb proximal muscle weakness, and autonomic dysfunction [207,211].

Brain atrophy and characteristic high-signal lesions in the middle cerebellar peduncle and splenium of the corpus callosum can be seen with brain MRI using T2 and fluid-attenuated inversion recovery (FLAIR) sequences [205,212,213]. On neuropathologic examination, findings include white matter disease associated with astrocytic pathology, intranuclear inclusions in both neurons and astrocytes in brain and spinal cord, amyloid beta within capillaries, and cerebral microbleeds [214,215].

Misdiagnosis of FXTAS may be common. This point is illustrated by a study of 56 patients with FXTAS and a family history of fragile X syndrome [216]. These patients received a wide variety of prior diagnoses other than FXTAS as an explanation for their symptoms, including various types of parkinsonism, tremor, ataxia, dementia, and cerebrovascular disease.

FMR1 DNA testing — Based on proposed guidelines for genetic confirmation of FXTAS, testing for FMR1 is suggested for patients who have either of the following presentations [216,217]:

●Unexplained cerebellar ataxia in males older than age 50

●Presence of action tremor, parkinsonism, or dementia in males older than age 50 with either a family history of developmental delay, autism, intellectual disability, or premature ovarian failure, or the middle cerebellar peduncle sign on brain MRI

Screening for the FMR1 premutation in patients who have neurologic illnesses, such as Parkinson disease [218], essential tremor [219], or multiple system atrophy [220], has a very low yield. Screening in older males with late-onset cerebellar ataxia may be more fruitful [199].

Gerstmann-Sträussler-Scheinker syndrome — The prion diseases, particularly the ataxic variant of the Gerstmann-Sträussler-Scheinker syndrome (GSS), can mimic SCA.

GSS is a genetic prion disease inherited in an autosomal dominant pattern and caused by pathogenic variants in the prion protein (PRNP) gene. GSS is discussed separately. (See "Diseases of the central nervous system caused by prions", section on 'Gerstmann-Sträussler-Scheinker syndrome'.)

Others — The International Parkinson and Movement Disorder Society (MDS) classification for hereditary movement disorders includes a number of additional autosomal dominant forms of hereditary ataxia with the ATX prefix [1]. Some of these have already been named as SCA types (eg, SCA13, SCA34, SCA37, SCA40, SCA42, and SCA47). Others without an SCA designation continue to be added and are listed in the table (table 1).

SUMMARY

●Definition – The spinocerebellar ataxias (SCAs) are a genetically and clinically heterogeneous group of autosomal dominant neurodegenerative disorders characterized by a slowly progressive cerebellar syndrome in association with various other neurologic features that vary with the affected gene. Most are adult-onset syndromes, although childhood forms exist. (See 'Introduction' above.)

●Nomenclature – The SCAs are classified according to the genetic loci and numbered according to their order of identification: SCA1 through SCA48 (and the number continues to grow) (table 1). (See 'Nomenclature and classification' above.)

●Clinical spectrum – Progressive ataxia and other signs of cerebellar dysfunction are a feature of each type of SCA. Other distinguishing features may suggest a particular type of SCA (table 1). (See 'Clinical spectrum' above and 'Specific disorders' above.)

The four most common SCAs are SCA1, SCA2, SCA3, and SCA6. In these disorders, symptoms typically begin in middle age, and most patients require a wheelchair by 10 to 15 years after symptom onset due to the degree of ataxia.

No disease-modifying treatments are available, but a number of strategies are under investigation in animal models and early-stage human trials. (See 'Prognosis' above and 'Treatment' above.)

●Diagnosis – Genetic testing is the most efficient and definitive way to determine the cause of the symptoms and to identify the ataxia subtype in a patient with suspected SCA based on clinical features and family history. In most cases, an SCA repeat expansion gene panel is the most appropriate first test.

Approximately 60 to 75 percent of patients with SCA have variants in the known loci. Among patients with apparently idiopathic sporadic cerebellar ataxia (ie, no family history), an SCA variant (most often SCA6) or Friedreich ataxia can be identified in approximately 20 percent. (See 'Diagnostic evaluation' above.)

Other genetic ataxic disorders that may resemble SCA include fragile X-associated tremor/ataxia syndrome (FXTAS), dentatorubral pallidoluysian atrophy (DRPLA), and prion disorders, particularly the rare ataxic variant of the Gerstmann-Sträussler-Scheinker syndrome (GSS). (See 'Other autosomal dominant ataxias' above.)

The etiologic evaluation of a patient with ataxia more generally is reviewed separately. (See "Overview of cerebellar ataxia in adults", section on 'Etiologic evaluation'.)

●Pathogenesis – Several types of SCAs (SCA1, SCA2, SCA3, SCA6, SCA7, SCA12, SCA17) and DRPLA are associated with expansion of cytosine-adenine-guanine (CAG) repeats in the region that encodes for polyglutamine tracts in the protein products. The age of onset and rate of disability progression vary according to individual differences in the length of the causative CAG trinucleotide repeat. (See 'CAG repeat disorders' above.)

Pathogenesis of other SCAs is variable and reviewed individually. (See 'Specific disorders' above.)