Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay (ARSACS)

Autosomal recessive neurodegenerative disease caused by biallelic variants in the SACS gene, combining cerebellar ataxia with onset usually in early childhood, pyramidal spasticity and mixed sensorimotor polyneuropathy; described in north-eastern Quebec and now recognised worldwide as one of the most frequent recessive ataxias.

Literature and regulatory information updated: 6 November 2026.

0. Initial Reference Table

1. Introduction

1.1. Original description in Quebec

Autosomal recessive spastic ataxia of Charlevoix-Saguenay was delineated in 1978 by Jean-Pierre Bouchard, André Barbeau and colleagues, from French-Canadian patients from the Charlevoix and Saguenay–Lac-Saint-Jean regions of north-eastern Quebec [23]. The founding paper described a syndrome that the authors considered homogeneous: spasticity, dysarthria, distal amyotrophy, foot deformities, truncal ataxia, absent sensory evoked potentials in the lower limbs, a retinal striation reminiscent of the early phase of Leber optic atrophy and mitral valve prolapse in 57% of patients; biochemically, many patients had impaired pyruvate oxidation, some had hyperbilirubinaemia and others had low levels of beta-lipoproteins and of high-density lipoprotein apoproteins, findings that the authors likened to those of Friedreich ataxia [23]. According to OMIM, the initial series included 42 patients from 24 sibships, with a further 24 known affected individuals; no patient had ever walked normally, the disease had a long course with little progression after age 20 years and the oldest patient was 52 years old [2]. GeneReviews refers to the same description as a cohort of about 325 French Canadians from 200 families, a figure corresponding to the set of patients identified by the Quebec group over the following decades [1, 24]. The authors proposed that the allele originated in a couple who settled in Quebec City around 1650 [2].

The following year, the same group compared the electrophysiology of ARSACS with that of Friedreich ataxia: ARSACS had more electromyographic signs of denervation and slower nerve conduction, with the same sensory conduction abnormality in both diseases, and more electroencephalographic abnormalities and a lower intelligence quotient than Friedreich ataxia; the acronym ARSACS was proposed at that time [2]. Computed tomography showed, in the 9 patients studied, cerebellar atrophy confined mainly to the superior vermis and the anterior lobes [2]. By 1985, Bouchard knew of almost 200 patients and highlighted the "increased visibility of the retinal nerve fibres" as a characteristic feature [2].

1.2. Milestones of the molecular discovery

In 1999, Richter and colleagues mapped the gene to chromosome region 13q11 by shared homozygosity and haplotype analysis in Quebec families, identifying two groups of ancestral haplotypes suggesting two mutations of independent origin [2, 25]. In 2000, Engert and colleagues cloned the SACS gene, whose 11,487-base-pair open reading frame was contained in a single giant exon of 12,794 base pairs, the largest then known in a vertebrate; the two truncating variants identified corresponded to the two ancestral haplotypes [26]. The protein was named sacsin. In 2006, Ouyang and colleagues identified 8 exons upstream of the giant exon, which raised the predicted protein to 4579 amino acids and made it necessary to sequence all exons, and not only the giant exon, in patients with a compatible phenotype [5, 27]. In 2009, Parfitt and colleagues showed that sacsin is a functional co-chaperone of the Hsp70 machinery, with a ubiquitin-like domain that binds the proteasome [28].

From 2003 onwards, the disease was confirmed in Tunisia, Italy, Japan, Turkey, the Netherlands, Belgium, Germany, Brazil and, later, in practically every region studied [10, 11, 14, 29–31]. European studies from 2008 to 2013 showed that ARSACS accounts for a relevant fraction of early-onset recessive ataxias outside Quebec and broadened the phenotype to adult-onset forms, forms without spasticity or without neuropathy and patients initially diagnosed with Charcot-Marie-Tooth disease [10, 11, 14, 32].

1.3. Nomenclature and classification

OMIM names the disease "Spastic ataxia, Charlevoix-Saguenay type" and assigns it the symbol SPAX6 in the series of autosomal spastic ataxias, whose reference entry is SPAX1 (#108600) [2]. The Task Force on the Classification and Nomenclature of Genetic Movement Disorders of the International Parkinson and Movement Disorder Society (MDS) proposed in 2018 a system combining the predominant phenotype and the gene: 62 recessive ataxias with the prefix ATX, 30 disorders with a double prefix when ataxia consistently coexists with another movement disorder and 89 entities in which ataxia is occasional; ARSACS received the designation ATX/HSP-SACS, reflecting the combination of ataxia with hereditary spastic paraplegia (HSP) [1, 3]. In 2019, a Task Force of the Society for Research on the Cerebellum and Ataxias (SRCA) published a classification consensus that identified 59 primary autosomal recessive cerebellar ataxias, organised by clinical presentation and pathophysiology, and 48 multisystem disorders to be included in the differential diagnosis [33]. The overlap between ataxias and spastic paraplegias, exemplified by ARSACS, has led to the group of spastic ataxias (SPAX) now being studied together, with ARSACS and spastic paraplegia type 7 (SPG7) as the two most frequent genotypes [34, 35].

The case of the twins described in 1987 as "Fitzsimmons-Guilbert syndrome" (spastic paraplegia, dysarthria, brachydactyly and cone-shaped epiphyses) illustrates molecular reclassification: the exome showed compound heterozygosity in SACS and, simultaneously, a heterozygous truncating variant in TRPS1, so that the phenotype resulted from two distinct genetic diseases, ARSACS and trichorhinophalangeal syndrome type I [2, 4].

1.4. Definition of the entity

ARSACS is defined by the presence of biallelic pathogenic variants in SACS in a patient with suggestive clinical findings [1]. The classic triad comprises progressive cerebellar ataxia, pyramidal signs with lower-limb spasticity and sensorimotor polyneuropathy; the most useful supporting signs are the linear hypointense stripes in the paramedian pons on T2, superior-predominant vermian atrophy and thickening of the peripapillary retinal nerve fibre layer [1, 6]. Orphanet defines it as a rare neurodegenerative disease characterised by early-onset cerebellar ataxia, pyramidal syndrome and peripheral neuropathy [6]. Neither spasticity, nor the Babinski sign, nor ataxia is mandatory: in a German series they were absent in 22%, 33% and 11% of patients, respectively [14]. GeneReviews does not recognise distinct allelic phenotypes [1]; isolated hereditary motor and sensory neuropathy with biallelic SACS variants is regarded as an extreme of the same spectrum, and not as a separate allelic disease [13, 32].

This article covers ARSACS as a single entity. Early-onset recessive ataxias in general, Friedreich ataxia, hereditary spastic paraplegias and ataxias with oculomotor apraxia appear in the differential diagnosis.

2. Molecular Genetics and Pathophysiology

2.1. The SACS gene

The SACS gene is located at 13q12.12, at GRCh38 coordinates 13:23,328,830–23,433,702 [5]. The 13q11 location given in the cloning paper and in earlier reference works corresponds to the 1999 mapping and has been superseded by the current cytoband [5, 26, 36]. The MANE Select reference transcript is NM_014363.6 (ENST00000382292.9), with the coding sequence between positions 590 and 14,329 of the transcript, corresponding to 13,740 nucleotides and a 4579-amino-acid protein (NP_055178.3) [7]. Engert and colleagues described an 11,487-base-pair open reading frame in a single 12,794-base-pair exon and a 3829-amino-acid protein; the discovery of 8 upstream exons by Ouyang and colleagues brought the structure to 9 exons, the giant exon being exon 9 [5, 26, 27]. Some recent publications number this exon as exon 10, according to the transcript annotation they use, and it harbours most of the variants described [37]. Northern blotting detected a 12.8 kb transcript in fibroblasts, brain, skeletal muscle and, in smaller amounts, pancreas, and in situ hybridisation showed intense labelling in all areas of the central nervous system [5]. Sacsin is most highly expressed in large neurons, including cerebellar Purkinje cells [28].

2.2. Structure and function of sacsin

Sacsin, of about 520 kDa, is a multimodular protein [38]. It has an N-terminal ubiquitin-like (UbL) domain, which binds the proteasome; three large repeated regions, called "sacsin repeating regions" and later "Sacsin Internal RePeaTs" (SIRPT1 to SIRPT3), each of more than 1100 residues and organised into sub-repeats (sr1, sr2, sr3 and srX), related to Hsp90; a DnaJ domain, which interacts with Hsp70; and a C-terminal HEPN (higher eukaryotes and prokaryotes nucleotide-binding) domain [1, 5, 39]. Romano and colleagues found a relationship between the position of pathogenic variants in the sub-repeats and clinical severity, measured with a severity scale [39]. In 2026, RNA-binding activity of the isolated HEPN domain was demonstrated [40]. The simultaneous presence of UbL and J domains suggests that sacsin integrates the ubiquitin–proteasome system and Hsp70 function; RNA interference silencing of SACS increased the toxicity of ataxin-1 with 82 glutamines, but not that of normal ataxin-1, indicating a protective role against proteins with polyglutamine expansion [5, 28].

Sacsin localises near mitochondria, in non-neuronal cells and in primary neurons, and interacts with dynamin-related protein 1 (DRP1), which is involved in mitochondrial fission [1, 41]. It acts as a co-chaperone in the organisation of the intermediate filament cytoskeleton, needed for organelle transport along axons and dendrites [42, 43]. In the absence of sacsin, the mitochondrial network becomes hyperfused, with abnormal morphology and reduced oxygen consumption in fibroblasts, and vimentin intermediate filaments and neurofilaments aggregate into bundles ("bundling") [1, 41, 43]. These changes have been proposed as cellular trait markers: Pilliod and colleagues observed a drastic and recurrent alteration of the mitochondrial network in 10 of 11 patients whose fibroblasts were studied, and vimentin "bundling" is also increased in asymptomatic heterozygous carriers [38, 44].

In the cerebellum of the Sacs−/− mouse, lack of sacsin impairs the transport of mitochondria and endoplasmic reticulum to the distal dendrites of Purkinje cells and reduces calcium-buffering proteins, causing pathological rises in cytosolic calcium in response to stimuli, Purkinje cell degeneration and secondary neuroinflammation [45, 46]. The Sacs knockout mouse reproduces the ataxic behaviour, the mitochondrial changes, neurofilament aggregation and abnormal Purkinje cell firing, with cell loss most marked in the anterior folia of the vermis [47, 48]. The homozygous Sacs R272C knock-in mouse develops ataxia, supporting a partial loss-of-function mechanism for missense variants [49]. Human neuropathology shows atrophy of the superior and anterior vermis with Purkinje cell death, degeneration of the pyramidal tracts, cortical atrophy, motor neuron loss and demyelinating neuropathy [50].

2.3. Disease mechanism and explanation for the lack of genotype–phenotype correlation

Most variants described are loss-of-function [1, 51]. Longo and colleagues studied a large series of fibroblasts from patients with variants of different types and in different domains and found that sacsin was almost absent in all, even in compound heterozygotes with missense variants; they excluded messenger RNA decay, defective translation and accelerated post-translational degradation, and showed that nascent mutant sacsin is ubiquitinated and degraded during translation itself [52]. This co-translational degradation mechanism explains why the nature and position of the variant predict the phenotype poorly [52]. GeneReviews identifies no clear genotype–phenotype correlations, and Baets and colleagues reached the same conclusion in a Belgian series [1, 11]. Two observations partially diverge: the systematic review by Xiromerisiou and colleagues found an association between the predicted pathogenicity of variants, severity and age at onset, and Romano and colleagues related position in the sub-repeats to severity [37, 39]. Thiffault and colleagues observed that compound heterozygotes with one missense variant retain minimal sacsin expression, and proposed that phenotypic variability outside Quebec reflects partial losses of function [51].

One exception shows the limits of this rule: in a patient with uniparental isodisomy and a homozygous nonsense variant near the 3′ end of the gene, a stable sacsin lacking the last 202 amino acids was formed, through escape from nonsense-mediated decay [53].

2.4. Inheritance

ARSACS is autosomal recessive [1, 2]. Heterozygotes are asymptomatic and are not at risk of developing the disease [1]. Paternal uniparental isodisomy of chromosome 13, with homozygosity for p.Arg4378Ter, showed that not every parent of an affected individual is a carrier [1, 8]. The SACS deletions described are mostly whole-gene deletions that include neighbouring genes; reported examples are a de novo 1.54 Mb microdeletion at 13q12.12 involving 6 genes, a homozygous contiguous-gene deletion that simultaneously caused ARSACS and SGCG-related limb-girdle muscular dystrophy (LGMD2C) and an intragenic deletion of exons 3 to 5 [1, 11, 54, 55]. In two half-siblings, a shared paternal microdeletion including SACS was combined, in each, with a different point variant, probably of maternal origin [56]. None of the sources consulted describes dominant or digenic inheritance for ARSACS; a 2023 paper proposed that a heterozygous missense variant, p.Val1335Ile, might cause a late-onset form through haploinsufficiency, an isolated hypothesis that contradicts the absence of disease in heterozygotes documented by GeneReviews and by carrier studies [1, 53, 57].

2.5. Variant spectrum and founder variants

More than 200 pathogenic variants have been described, most of them missense, and ClinVar lists 1319 entries classified as pathogenic or likely pathogenic for SACS [53, 58]. The size of the gene explains the diversity: even in Quebec, 12 different variants were found [51]. The historical nomenclature of the Quebec variants (6594delT, 5254C>T) uses the numbering of the original 2000 open reading frame, which predates the identification of the upstream exons; the current designations derive from the complete transcript [1, 5].

The equivalence between historical and current designations is given by the GeneReviews authors; the current designations and protein consequences were validated in VariantValidator on the MANE transcript [1, 7]. A 2021 review writes the c.7504C>T variant as "p.Arg2502Cys", which is a typographical error: the substitution creates a stop codon [7, 24]. In the Portuguese series of Santos and colleagues, one family was homozygous for c.814C>T, which affects the same codon 272 as the R272C mouse model [49, 63].

2.6. Penetrance, expressivity and modifiers

In patients with biallelic variants the disease manifests, but expressivity varies widely, within and between families, even on the homogeneous genetic background of Quebec: some patients had no signs of neuropathy or spasticity at age 18 years [1, 64]. Retinal striation is marked in Quebec patients and often faint or absent in Tunisian, Italian and Japanese patients, suggesting variation between population groups [2, 29]. Sex and age at onset modulate progression: in a multicentre registry of 173 patients, men had 25% lower baseline scores on the SARA (Scale for the Assessment and Rating of Ataxia) and a slightly higher maximum, and the time to half of the progression was 35% shorter when the disease began after age 15 years [65]. The coexistence of other genetic diseases, such as trichorhinophalangeal syndrome type I in the "Fitzsimmons-Guilbert" twins or the TTPA and FXN ataxias in the same consanguineous Tunisian family, may modify the picture and complicate counselling [4, 66].

2.7. Position in metabolic classifications

Saudubray includes ARSACS in the table of recognised mitochondrial syndromes, with spasticity, ataxia, amyotrophy, nystagmus and dysarthria, and places SACS in the group of defects of mitochondrial transporters, protein import, quality control, fusion and mitochondrial DNA depletion, in the ataxia column [67, 68]. The grouping is justified by the mitochondrial dysfunction of fibroblasts and models, but ARSACS has no routine metabolic biomarkers and the diagnosis is molecular [1, 68].

3. Clinical Spectrum and Manifestations by System

3.1. Overview and course by age

In the classic form, the disease presents in early childhood with gait unsteadiness, often with delayed acquisition of independent walking, between 12 and 18 months [1]. The mean age at onset in the series compiled by GeneReviews is about 6 years, with a range of 0 to 40 years [1]. In the Saguenay cohort, the mean age at presentation was 3.41 ± 1.55 years (n = 40); at onset, brisk reflexes were more frequent than spasticity, and the neuropathy became clinically apparent only in the second decade [64]. The usual sequence is ataxia, then lower-limb spasticity during childhood and neuropathy in adolescence; dysarthria appears in late childhood and progresses slowly; eye movements, dysarthria and upper-limb ataxia progress more slowly than gait ataxia, spasticity and neuropathy [1]. The neuropathy may, however, be the first sign [14]. Distal amyotrophy, with absent ankle reflexes and sometimes bilateral foot drop, is present in most patients after age 21 years [1]. In the original Quebec series, no patient had walked normally and progression was small after age 20 years; most patients eventually need a wheelchair [2].

Outside Quebec, the spectrum has broadened to adolescent- and adult-onset forms. In the Belgian series, 5 of 11 index patients had onset after age 20 years, one of them at 40 years [5, 11]. In a Japanese series of hereditary neuropathies, onset ranged from 1 to 49 years [13]. Adult-onset forms more often present with predominant neuropathy or without one of the features of the triad [11, 14].

In the French SPATAX and ATAXIC series (47 patients from 39 families; 62% female; consanguinity in 26% of families), the mean age at onset was 5.1 ± 7.4 years, between birth and 30 years; 72% had onset before age 5 years and 83% before age 10, and 2 patients had onset after age 25 years [44]. Unsteadiness was the first sign in 96% (43/45), isolated in 82% (37/45) or with leg stiffness in 11% (5/45); only 2 patients began with isolated leg stiffness [44]. In 2 patients, lower-limb spasticity remained isolated for 16 and 8 years before the onset of ataxia [44]. In this series, spasticity was present in the lower limbs in 63% and in the upper limbs in 12%, school difficulties compatible with mild intellectual disability in 51%, almost always without neuropsychological assessment, and deafness, abnormal auditory evoked potentials and epilepsy were rare [44]. Six patients (5 index patients and one relative) had a congenital ataxia phenotype: onset at a mean of 1.1 ± 0.1 years, assessment at 10.7 ± 2.3 years, mild disability (SDFS 2) in the 3 with data and a stable course over up to 14 years [44]. The oldest, a 14-year-old boy, had mild, non-progressive static and kinetic ataxia (SARA 6/40), partial epilepsy since age 10 years, demyelinating sensorimotor neuropathy, linear pontine hypointensities and superior vermian atrophy [44].

3.2. Frequency of manifestations

GeneReviews compiles the frequencies from the main published series, based on the table by Pilliod and colleagues, with the caveat that not all features were assessed in all groups and that some denominators are small [1, 44].

The primary sources for this table are the series of Baets (Belgium), Duquette (Quebec), Pilliod (France), Prodi (Italy), Synofzik (Germany) and Vermeer (Netherlands) [1]. The frequency of visible retinal fibres on fundoscopy is higher in series with targeted examination: 70.6% (12/17) in the United Kingdom and 82% (peripapillary striations) in a Brazilian series of 29 patients [15, 69].

3.3. Cerebellar system

Ataxia is the first sign in almost all patients, with broad-based gait, truncal ataxia, dysmetria, dysdiadochokinesia, intention tremor, gaze-evoked nystagmus, saccadic pursuit and scanning speech [1, 6, 60]. Nystagmus and abnormal pursuit distinguish ARSACS from Troyer syndrome [2]. The dysarthria of ARSACS has coordination and timing deficits that also affect swallowing [70]. In the Quebec longitudinal studies, upper-limb coordination and fine dexterity declined little over 2 years, unlike balance and gait [71].

3.4. Pyramidal system

Spasticity predominates in the lower limbs, with knee hyperreflexia, clonus and Babinski sign; in Tunisia, the pattern was brisk knee jerks with absent ankle jerks and Babinski sign [29]. In a series of 25 adults who were still walking (mean age 32.2 ± 10.4 years; 45.7% with a walking aid), spasticity and limitation of passive range of motion were mild, but co-contraction of lower-limb muscles was excessive, suggesting that walking disability does not depend on tone alone [72]. Severe neuropathy may mask spasticity: two Japanese siblings homozygous for p.Phe304Ser had neither spasticity nor hyperreflexia, but had bilateral Babinski signs [5, 73]. Urinary dysfunction is mainly due to pyramidal involvement [1].

3.5. Peripheral nerve

The neuropathy is sensorimotor, mixed, with axonal and demyelinating components, begins in childhood and causes distal atrophy and weakness, foot deformities, impaired touch and vibration sensation and, over time, reduced or absent reflexes in the legs [1]. Sensory evoked potentials show severe involvement of the posterior columns [29]. In the Pilliod series, the neuropathy was demyelinating in 53%, axonal in 25% and mixed in 14% [1]. In a German series of 11 patients (39.0 ± 14.1 years; disease duration 30.6 ± 12.5 years), the neuropathy was primarily demyelinating in 9 and unclassifiable in 2, because of unrecordable potentials, and high-resolution ultrasound showed no nerve enlargement in any patient, contrary to what is usual in hereditary demyelinating neuropathies [74]. Patients with isolated hereditary motor and sensory neuropathy and biallelic SACS variants had no spasticity or pyramidal signs; 3 of 9 had ataxia, of sensory type; MRI was normal or showed only superior vermian atrophy [32]. Closer review showed cerebellar oculomotor abnormalities and/or vermian atrophy in 8 of those 9 patients [1]. In Japan, in a panel of 3353 cases with suspected hereditary neuropathy or Charcot-Marie-Tooth disease, 9 had confirmed pathogenic or likely pathogenic variants (0.268%); all had motor and sensory neuropathy, 6 had pyramidal signs, 8 ataxia or cerebellar atrophy, 4 cognitive impairment and one postural hypotension, and the polyneuropathy was mainly sensory and length-dependent [13].

3.6. Eye and retina

The characteristic ophthalmological sign is the presence of yellow striations of hypermyelinated nerve fibres radiating from the optic disc and partially obscuring the vessels [1, 2]. Optical coherence tomography (OCT) shows thickening of the peripapillary nerve fibre layer, even when the fundus appears normal [1, 15]. In a British study of 292 individuals, all 17 patients with ARSACS had thickening, whereas all other patients with ataxia and the controls, except one, had normal or reduced thickness; the 13 heterozygous carriers studied had no thickening [15]. A Brazilian series of 29 patients described on OCT foveal hypoplasia in 100%, a sawtooth appearance of the inner retina in 89%, a papillomacular fold in 86% and macular microcysts in 18%, with all macular layers thicker than in controls; nerve fibre layer thickness in the inferior macular sector correlated with the SARA score [69]. Most patients have no visual complaints, although older patients may have visual impairment [1, 15]. Thickening on OCT without hypermyelinated fibres visible on fundoscopy occurred in 5 of the 17 British patients, making OCT more sensitive than fundoscopy [15]. In the French series, fundoscopy, performed in 16 patients, showed fibre hypertrophy in only 3 (19%), more marked near the superior and inferior poles of the optic disc than in the nasal and temporal sectors, and confirmed by OCT in 2 [44].

3.7. Central nervous system: imaging

MRI shows atrophy of the superior vermis and cerebellar hemispheres, linear hypointense stripes in the paramedian pons on T2 and FLAIR, hyperintensity of the lateral pons at the junction with the middle cerebellar peduncles, which are thickened, and a T2-hyperintense rim around the thalami [1, 14, 75]. In a German series, all patients had lateral pontine hyperintensity, 63% bilateral parietal atrophy and 63% circumscribed thinning of the posterior part of the body of the corpus callosum [14]. In the Brazilian series of 13 patients, cerebellar atrophy and linear pontine hypointensities were present in all and biparietal atrophy in 12 [31]. Cervical cord atrophy has been described in Quebec and in a Thai case [60]. In the French series, MRI of 41 patients showed superior-predominant vermian atrophy in 39, cerebellar hemisphere atrophy in 23, linear pontine hypointensities on T2/FLAIR in 19, lateral hyperintensity of the middle cerebellar peduncles in 7 and increased peduncle calibre in 9, and perithalamic hyperintensities in 5; it was normal in 2, one of whom was imaged at age 7 years [44]. In some patients MRI was reported as normal, but a review of the published images showed pontine stripes in cases reported as negative [1]. In a Moroccan patient with epilepsy, the initial MRI was normal [61]. Orphanet also lists arachnoid cyst and hypoplasia of the corpus callosum as frequent findings [6].

3.8. Cognition, behaviour and epilepsy

ARSACS is not associated with severe intellectual disability; mild intellectual disability or school difficulties are present in about 29% of patients [1]. In the Belgian series, one of 17 patients had mild intellectual disability and 2 had a reduced intelligence quotient [2]. A Sicilian female patient homozygous for p.(Arg272His) had a full-scale intelligence quotient of 60, early axial hypotonia and progressive vermian atrophy on serial MRI over 13 years [59]. The cerebellar cognitive affective syndrome (CCAS) scale showed, in a Saguenay cohort (n = 31; median age 57 years) and in a younger Tübingen cohort (n = 17; median 35 years), that all patients failed several subtests and exceeded the threshold for definite CCAS; even the youngest patients failed more subtests than controls (median 5 versus 1), indicating early and progressive deficits [76]. Two siblings studied with detailed neuropsychological assessment had motivational, cognitive and behavioural deficits compatible with this syndrome [1]. Epilepsy occurs in about 7% of patients; progressive myoclonic epilepsy in two patients compound heterozygous for missense variants, focal epilepsy in a patient without spasticity and epilepsy with onset at age 5 years with a good response to an antiseizure medication have been described [1, 61, 77, 78].

3.9. Hearing, heart, sphincters and other manifestations

Deafness occurs in about 13% of patients; it is not known whether it is part of the disease or results, in some cases, from deletion of neighbouring genes, as in a 26-year-old patient with bilateral sensorineural deafness and a 13q12.12 deletion [1, 54]. Mitral valve prolapse was frequent in Quebec patients (57% in the original series) and has been described in a single patient of other origin; severe dilated cardiomyopathy in one patient has no established causal relationship; GeneReviews does not recommend routine cardiac assessment [1, 23]. Urinary urgency and incontinence are present in about one third of patients and, in the German series, were associated with erectile dysfunction [1, 14]. Chronic pain was reported by 30% of 50 Quebec patients, and cramps, spasms and spasticity by more than half [79]. Pes cavus, hammer toes and scoliosis are frequent [1, 2].

3.10. Quantitative natural history

Prospective data come mainly from Quebec and from European multicentre cohorts. In 19 adults assessed 2 years apart, there was a significant decline, above measurement error, in lower-limb coordination, balance, walking speed and distance and the SARA score, without significant decline in the upper limbs [71]. In 40 patients followed for 4 years, balance reached a floor of about 6 points on the Berg scale after age 40 years, and the remaining patients lost about 1.5 points per year; walking speed decreased by a mean of 0.044 m/s per year and the 6-minute walk distance by 20.8 m per year, a progression greater than that of normal ageing [80]. In 60 adults followed for 2 years, the disease-specific DSI-ARSACS severity index worsened by 1.5 points, SARA increased by 1.6 points, the Barthel index fell by 7.4 points and grip strength decreased by 2.3 kg; the Timed Up and Go increased by 7.8 s (24%), walking speed fell by 17% to 20% and the Berg scale by 4.3 points (24%), with greater progression in patients already using a walking aid or a wheelchair [81, 82].

In the international ARCA registry, modelling of 173 patients showed a logistic SARA trajectory: progression starts at about 0.1 points per year and reaches a maximum of 0.8 points per year about 36.8 years after onset, with wide individual prediction intervals at 5 years (median 7.4 points) [65]. In the French series, assessed at 29.7 ± 12.9 years of age and 24.3 ± 12.5 years of disease, the SDFS functional scale (0, asymptomatic, to 7, bedridden) classified 17% of patients as mildly affected (SDFS 1–2; mean age 21.3 years), 28% as moderately affected (SDFS 3; 27.7 years) and 55% as severely affected (SDFS 4–7; 35.2 years), including 8 in a wheelchair and 1 bedridden (mean age 36.5 years; 16 to 53 years); after age 30 years, 25% (6/24) used a wheelchair [44]. These data are cross-sectional and do not describe the individual trajectory. A 2025 review estimates mean SARA progression at about 0.6 points per year in ARSACS and SPG7 [46]. In the Portuguese spastic ataxia cohort, after a median of 31 years of disease, 42.9% of patients retained independent walking, 19.6% used an aid and 30.4% did not walk; these figures refer to the spastic ataxias as a whole, and not only to ARSACS [12]. All these data describe the course under usual care, including physiotherapy, without disease-modifying therapy.

4. Clinical Recognition and Red Flags

4.1. When to suspect

GeneReviews recommends suspecting ARSACS in the presence of slowly progressive cerebellar ataxia, with walking difficulty and unsteadiness from 12–18 months or later; lower-limb spasticity; peripheral neuropathy with distal atrophy and weakness; MRI with superior-predominant atrophy of the vermis and/or cerebellar hemispheres and bilateral hypointense stripes in the paramedian pons; and hypermyelinated retinal fibres, visible as yellow striations from the edges of the optic disc or as a mean peripapillary nerve fibre layer thickness greater than 119 µm on OCT [1]. The combination of ataxia with spasticity and neuropathy, especially with pontine stripes and retinal fibre layer thickening, is the profile that points most strongly to the diagnosis [1, 15].

The Oxford Desk Reference highlights, in the history of the ataxic patient, ethnic origin, because some ataxias cluster in geographical regions, such as ARSACS in Quebec, consanguinity, which suggests recessive inheritance, and a three-generation pedigree [83]. It recommends considering ARSACS when genetic testing for Friedreich ataxia is negative and including nerve conduction studies and ophthalmological assessment in the investigation of the ataxic child [36]. In a Portuguese cohort of hereditary ataxias, spastic ataxias begin earlier, have recessive inheritance and, in more than half of cases, present with non-cerebellar signs, such as motor delay, spastic gait or cognitive regression [12].

4.2. Clinical contexts by age

4.3. Red flags and supporting signs

The most specific supporting sign is OCT: a mean peripapillary nerve fibre layer thickness greater than 119 µm had a sensitivity of 100% and a specificity of 99.4% among patients with ataxia [15]. The linear hypointense pontine stripes on T2, described by some authors as pathognomonic, lateral pontine hyperintensity with thickening of the middle cerebellar peduncles and thalamic rims are the imaging signs to look for [1, 14, 60, 75]. The ratio of pontine area to superior vermis area, with a cut-off of 1.67, distinguished ARSACS from SPG7 with an area under the curve of 0.98 [35]. On neurological examination, the combination of brisk knee jerks, Babinski sign and absent ankle jerks, with nystagmus, is typical [29, 60].

ARSACS has no acute presentations of its own. Ataxia of acute or subacute onset, progressing over weeks or months, suggests another cause, such as paraneoplastic disease, prion disease or multiple system atrophy, and requires urgent investigation; the Ataxia UK guideline recommends urgent paediatric assessment of a child with ataxia [16]. Rapid worsening of gait in a patient with known ARSACS should prompt a search for treatable superimposed causes, such as urinary tract infection, which can worsen neurological disability, adverse effects of antispasticity drugs that unmask weakness, or fractures and fall-related injuries [16].

4.4. Newborn screening and carrier screening

There is no newborn screening for ARSACS. In Quebec there has been, since 2010 as a pilot project and since 2018 for the whole province, a free carrier testing programme for the founder variants for people with at least one grandparent born in Saguenay–Lac-Saint-Jean, Charlevoix or Haute-Côte-Nord who are planning to have children (see section 8) [24, 85].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic criterion

The diagnosis is established in a proband with suggestive clinical findings and biallelic pathogenic variants in SACS identified by molecular testing [1]. Pilliod and colleagues proposed, from 321 index patients with spastic ataxia from the French SPATAX network and 156 patients with congenital ataxia from the ATAXIC project, new diagnostic definitions combining clinical, genetic and cellular criteria, using the alteration of the fibroblast mitochondrial network as a trait biomarker when the molecular result is difficult to interpret, as with missense variants or a single heterozygous truncating variant; they identified biallelic variants in 33 SPATAX patients and in 5 patients with non-progressive ataxia [44]. Inclusion criteria were, in the SPATAX cohort, cerebellar ataxia with lower-limb pyramidal signs, onset before age 45 years, absence of ataxia in the parental generation and Friedreich ataxia excluded; in the ATAXIC cohort, ataxia with onset before age 2 years with a stable course or progressive improvement [44]. Variants were classified with the 5-class system of Plon and colleagues, which predates the ACMG/AMP criteria: class 5, pathogenic; class 4, likely pathogenic; class 3, uncertain significance; class 2, likely not pathogenic; class 1, not pathogenic [44]. The proposed definitions are three:

Source: Pilliod and colleagues, table 3 [44].

The alteration of the mitochondrial network consists of a significant increase in the proportion of cells with hyperfused, balloon-shaped or bulbous mitochondria, seen with MitoTracker Green in fibroblasts from an arm skin biopsy, cultured between the 3rd and 5th passage and compared with controls; in 8 controls aged 18 to 54 years, morphology did not vary with sex or age [44]. Of the 11 patients studied, 10 had the alteration, with a reduction in mitochondrial mass of 17% to 55% and in mitochondrial DNA of 14% to 67%; basal oxygen consumption was preserved in 4 of 5 patients tested and reduced by 51% in a patient homozygous for p.(Arg272His), who also had a 30% reduction in membrane potential [44]. The only patient without the alteration, a girl with stable congenital ataxia, normal MRI and a deletion of about 1.5 Mb at 13q12.12 encompassing the whole gene, with no second variant on DNA and complementary DNA sequencing, was considered negative for ARSACS [44].

Applying the definitions to 40 families retained after Sanger sequencing, next-generation sequencing or array comparative genomic hybridisation changed the classification: with variant classification alone, 35 families were positive and 5 inconclusive; with the new definitions, 25 families (31 patients) had definite ARSACS, 8 families (8 patients) probable, 6 families (8 patients) possible and 1 was considered negative [44]. The altered mitochondrial network allowed two families left undiagnosed by molecular analysis to be classified as definite: a patient homozygous for p.(Ile513Val), class 2, with onset at age 22 years and dystonia, and a patient with a single class 5 frameshift variant inherited from the mother, also present in the affected sister, with no second variant on targeted microarray or on RNA study [44]. The first case does not strictly fit table 3, which requires at least one class 5 variant or two of class ≥ 4, showing that the authors gave the cellular biomarker greater weight than variant classification [44]. The authors recommend testing SACS in patients with ataxia, spastic ataxia or spastic paraplegia, regardless of age at onset, course and the results of ancillary investigations, because MRI and electromyography, although highly suggestive, may be normal and do not serve as a trait biomarker [44].

These definitions have limitations. The mitochondrial biomarker was studied in 11 patients and compared with 2 controls in the final assays, with no estimate of sensitivity or specificity in a population of patients with other ataxias; other disorders of mitochondrial dynamics also alter the fibroblast network: in the spastic paraplegias SPG7, SPG31 and SPG56 an elongated network has been described, in the same direction as ARSACS, and in spinocerebellar ataxia SCA28 mitochondrial fragmentation [44]. The technique requires a skin biopsy and fibroblast culture in a research laboratory. The definitions have not been adopted by GeneReviews, which requires biallelic pathogenic variants, and predate the ACMG/AMP criteria, in which functional evidence obtained in the patient's cells enters as a criterion for the pathogenicity of a variant and not as a stand-alone diagnostic criterion [1, 44]. Longo and colleagues proposed sacsin quantification as an unequivocal criterion, based on the almost complete absence of the protein in patients, whatever the variant [52].

5.2. Genetic testing algorithm

GeneReviews describes two pathways [1]. When the phenotype is suggestive, it recommends single-gene testing in patients from high-risk populations or with an affected relative, and a multigene panel, with deletion and duplication analysis, in the others. In single-gene testing, sequencing is done first; if only one or no variant is found, targeted deletion and duplication analysis follows. When the phenotype is indistinguishable from other ataxias or spastic paraplegias, comprehensive genomic testing is the best option, generally exome and, alternatively, genome; if the exome is not diagnostic, an "exome array", when available, can detect deletions or duplications of one or more exons [1]. In patients of north-eastern Quebec ancestry, targeted testing for the two founder variants, c.8844del and c.7504C>T, may be done first [1, 86]. Because even in Quebec 12 different variants have been found, a negative result for the founder variants does not exclude the disease [51].

The percentages of 95% and 5% refer to the proportion of known pathogenic variants that each method detects, and not to the clinical sensitivity of the test in a patient with suspected ARSACS [1]. When sequencing finds only one variant, this may reflect true homozygosity, compound heterozygosity with a large deletion on the other allele or uniparental isodisomy, and parental testing and copy-number analysis resolve most of these situations [1]. In the Japanese neuropathy series, read-depth copy-number analysis identified complete gene deletions in 2 of the 10 cases [13].

5.3. Variant interpretation and functional tests

The report should state the gene, the transcript with version (NM_014363.6), the variants in HGVS nomenclature (c. and p.), zygosity, phase (cis or trans, when known) and the classification according to the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria [1, 7]. The interpretation of missense variants is the main difficulty, because most SACS variants are missense and bioinformatic prediction of pathogenicity is unreliable [53]. Three functional approaches have been described: quantification of sacsin by western blot in fibroblasts or lymphoblasts, which shows almost complete absence of the protein in patients [51, 52]; the same quantification in peripheral blood mononuclear cells, which can be cultured and cryopreserved, in which patients have almost absent sacsin and carriers of a missense variant have a reduction of about 50% [53, 87]; and analysis of the mitochondrial network and vimentin "bundling" in cultured fibroblasts [43, 44]. Sacsin may be preserved with nonsense variants very close to the 3′ end, which escape nonsense-mediated decay, and a normal protein level does not exclude the disease in that context [53]. These techniques are available in research laboratories and not routinely in most countries.

5.4. Supporting ancillary investigations

In the initial assessment of a patient with ataxia of unexplained cause, Friedreich ataxia, ataxia with vitamin E deficiency and treatable metabolic causes should be excluded; the Oxford Desk Reference stresses the need for vitamin E measurement when Friedreich ataxia testing is negative [16, 36].

5.5. Evaluation after diagnosis

GeneReviews recommends, after diagnosis: full neurological examination, brain MRI and electromyography; physiotherapy and occupational therapy assessment as a baseline for mobility and activities of daily living; developmental or cognitive assessment to plan educational support; baseline assessment of dysarthria; audiological assessment if there are complaints; and clinical genetics or genetic counselling consultation to discuss the risks to offspring and reproductive options [1]. The Ataxia UK ataxia guideline, endorsed by ERN-RND, adds referral to a specialist ataxia centre once the diagnosis of progressive ataxia is established, coordination between neurology and genetics and genetic counselling before and after testing [16]. Measurement of post-void residual volume and exclusion of urinary tract infection are part of the assessment of urinary symptoms [16].

5.6. Prenatal and preimplantation diagnosis

Once the two pathogenic variants have been identified in an affected relative, prenatal diagnosis in at-risk pregnancies and preimplantation genetic testing are possible [1]. There are no fetal ultrasound markers of ARSACS in the sources consulted.

6. Differential Diagnosis

6.1. Principles

The differential diagnosis of ARSACS covers three groups: early-onset autosomal recessive cerebellar ataxias, especially those accompanied by pyramidal signs or neuropathy; hereditary spastic paraplegias complicated by ataxia or neuropathy; and hereditary motor and sensory neuropathies, when the neuropathy predominates [1, 32, 34]. The Ataxia UK guideline groups among the ataxias with prominent spasticity at presentation late-onset Friedreich ataxia, ARSACS, SPG7, adult Alexander disease, spinocerebellar ataxia type 3 (SCA3, Machado-Joseph disease), cerebrotendinous xanthomatosis and fatty acid hydroxylase-associated neurodegeneration, and recalls metabolic causes with spasticity, such as ataxia with vitamin E deficiency, coenzyme Q10 deficiency, abetalipoproteinaemia, metachromatic leukodystrophy, adrenomyeloneuropathy, vanishing white matter disease, Krabbe disease, Gaucher disease type III and GM2 gangliosidosis [16]. In the Portuguese spastic ataxia cohort, the most frequent genetic causes after ARSACS were SYNE1, ANO10, KIF1C, PGN, ZFYVE26, ATP13A2 and KCNA2 [12]. Treatable causes should be excluded first [16, 36].

6.2. Differential diagnosis table

6.3. Main differentials

Friedreich ataxia is the most important differential, because it is the most prevalent recessive ataxia worldwide and because it shares ataxia, dysarthria, the Babinski sign and loss of position and vibration sense [1]. It is distinguished by later onset, cardiomyopathy, the absence of overt spasticity in the typical form, the absence of hypermyelinated retinal fibres and white matter changes, and a normal or thin retinal nerve fibre layer on OCT [1, 15, 84]. The late-onset form of Friedreich ataxia, with retained reflexes and spasticity, comes closer to ARSACS and justifies testing for the FXN expansion in any recessive spastic ataxia [16, 88].

SPG7 is, together with ARSACS, the most common spastic ataxia genotype [35]. SPG7 usually begins later, is accompanied by optic atrophy and has no significant neuropathy; the MRI of ARSACS is the only one with an increased pontine area, and a pons/superior vermis ratio greater than 1.67 separates the two with a sensitivity of 97% [1, 35]. SPG7 may cause spastic ataxia with onset in mid-adulthood, sometimes with progressive external ophthalmoplegia or ptosis, and SPG31 is only rarely accompanied by mild cerebellar ataxia after a long course; because in the spastic paraplegias SPG7, SPG31 and SPG56 the fibroblast mitochondrial network is also elongated, the mitochondrial alteration described in ARSACS does not distinguish them, and they are separated by genotype and clinical picture, which in these paraplegias is dominated by the progressive pyramidal syndrome [44].

When the neuropathy predominates, the differential is Charcot-Marie-Tooth disease. A demyelinating neuropathy without nerve enlargement on ultrasound, mild pyramidal signs, superior vermian atrophy, cerebellar oculomotor abnormalities or thickening of the retinal fibre layer on OCT should prompt the inclusion of SACS in the panel [14, 32, 74]. In ataxia with vitamin E deficiency and the other metabolic causes, serum measurement precedes genetic testing, because treatment changes the course [16, 36]. In highly consanguineous families, more than one recessive ataxia may coexist: in a Tunisian family, 7 patients had ataxia with vitamin E deficiency, 2 Friedreich ataxia and 2 ARSACS [66]. In the young child with apparently non-progressive ataxia, ARSACS may be mistaken for a congenital ataxia [44].

7. Treatments and Risk Management Strategies

7.1. Principles and sources of recommendations

There is no approved curative or disease-modifying therapy for ARSACS; treatment is symptomatic and tailored to each patient's needs [1]. No ARSACS-specific clinical practice guideline has been published by a scientific society, by a European Reference Network or by the French Haute Autorité de Santé, which has no PNDS (French National Diagnosis and Care Protocol) for this disease. The available recommendations are those of the 2020 GeneReviews, specific to ARSACS but without dosages, and those of the Ataxia UK guideline for the management of the progressive ataxias (3rd edition, 2016), endorsed by ERN-RND in 2018 and updated in 2019, published in a journal by de Silva and colleagues, which includes ARSACS among the spastic ataxias and gives consensus recommendations, mostly of GPP (good practice point) grade, for spasticity, pain, bladder, dysphagia, physiotherapy and speech and language therapy [1, 16, 90]. Ataxia UK stresses that there is little evidence for the efficacy of antispasticity interventions specifically in the spastic ataxias and that decisions are made by extrapolation from more common conditions, such as stroke, multiple sclerosis, cerebral palsy and spinal cord injury [16]. For this reason, the doses that follow are those of the guidelines and labels for these indications, applied to ARSACS by extrapolation.

7.2. Rehabilitation

Rehabilitation is the intervention with the most direct evidence in ARSACS. In an exploratory study in Quebec (NCT05479656), a supervised group programme, 3 times a week for 8 weeks (two gym sessions and one pool session), targeting motor control of the trunk and lower limbs, improved balance, trunk control, walking speed, ataxia severity measured by SARA and specific activities of daily living in ambulatory patients; all participants improved beyond the standard error of measurement on at least two measures, but the study had no control group [92, 93]. The PACE-ARSACS (Promoting Autonomy by exerCisE) home programme, developed by physiotherapists with a patient partner and available online, brings together exercises for trunk control, sitting and standing balance and sit-to-stand transfers, applicable to patients who walk unaided, with an aid or in a wheelchair, in sessions of 15 to 20 minutes [17, 94]. The starting level is defined by a standardised instrument; the physiotherapist teaches the programme at a home visit and follows up by telephone in the first week and then every 2 weeks [17]. In a non-randomised, open-label study, with a 12-week control phase followed by 12 weeks of training 3 times a week, 14 of the 22 participants completed the programme; measures remained stable in the control phase and, after training, SARA improved by a mean of 1.5 points, equivalent to about one year of natural progression, and participation (LIFE-H) improved; 11 of the 14 patients had more tests improved than worsened, especially those who did not walk, whereas self-selected walking speed worsened beyond measurement error [17]. The programme was considered safe and acceptable [17].

Individualised training with "exergames" improved postural control in a trial with a blinded assessor and intra-individual control in advanced degenerative spinocerebellar ataxias, evidence that GeneReviews applies to ARSACS by extension [1, 95]. The Ataxia UK guideline recommends early referral to physiotherapy, to maintain function and prevent falls, considering gait and balance programmes case by case, rehearsing steps visually before crossing spaces with obstacles (grade D), using coordinative video-game programmes in children who walk unaided, under supervision (grade C), assessing walking aids individually, reducing weight-bearing through the upper limbs and encouraging exercise as health promotion [16]. The relatively mild spasticity and excessive co-contraction of lower-limb muscles observed in ambulatory patients suggest that rehabilitation should target motor control and not tone alone [72]. In the wheelchair phase, posture and seating should be assessed, and manual wheelchair skills training is being studied in a randomised waiting-list trial [16, 96].

Intensive 4-week home-based speech and language therapy, adapted to cerebellar dysarthria, improved intelligibility, the primary outcome assessed blind, and speech naturalness in 7 patients, in an intra-individual control design [91]. The Ataxia UK guideline recommends open access to speech and language therapy, individualised programmes for dysarthria, attention to cognitive and hearing deficits that interfere with communication, augmentative or alternative communication when intelligibility is below 50% (grade D) and instrumental swallowing examination when clinical assessment is insufficient to guide the management of dysphagia [16].

7.3. Spasticity: strategy

The Ataxia UK guideline recommends that a neurologist, advised by a physiotherapist, decides on the type of treatment; that physiotherapy be the first step, followed by drugs if the benefit is incomplete and by surgery when both fail; that generalised spasticity be treated with oral drugs in the usual order baclofen, tizanidine, gabapentin, clonazepam, dantrolene or diazepam, chosen according to adverse-effect profile and tolerability, starting with low doses and titrating slowly; and that focal spasticity, especially of small muscles, be treated with intramuscular botulinum toxin in a specialist clinic, followed by physiotherapy, because stretching, functional taping, casting, dynamic orthoses and electrical stimulation prolong the benefit [16]. Chronic diazepam is not recommended, except in very severe cases, and intrathecal baclofen is reserved for selected patients, in a specialist centre [16]. GeneReviews adds that early treatment of spasticity may prevent tendon shortening and contractures and delay major functional disability until weakness or ataxia becomes severe [1]. Ataxia UK warns of paradoxical worsening of mobility through unmasking of underlying weakness: the patient should be warned, titration should be slow and the drug reduced or stopped if mobility worsens [16]. This risk is particularly relevant in ARSACS, in which distal weakness from the neuropathy coexists with spasticity and in which spasticity may support walking [1, 72].

7.4. Antispasticity drugs

Baclofen (Lioresal®, in Portugal; Baclon® and generics in Brazil; generics in the USA). A centrally acting GABA-B receptor agonist, available as 10 mg and 25 mg tablets in Portugal [97, 98]. The Ataxia UK guideline recommends it as the first oral choice in generalised spasticity, without dosage, so the reference dose is that of the Portuguese summary of product characteristics (SmPC), for spasticity of spinal or cerebral origin, applied to ARSACS by extrapolation [16, 97]. In adults, it is started at 15 mg per day in 2 to 4 doses and increased by 15 mg per day every 3 days up to the desired dose, generally 30 to 80 mg per day; in sensitive patients it is started at 5 to 10 mg per day; doses of 100 to 120 mg per day only in hospital with monitoring [97]. In children and adolescents, it is started at about 0.3 mg/kg/day in 2 to 4 doses, preferably 4, and increased at intervals of one week; the usual maintenance is 0.75 to 2 mg/kg/day, with a maximum of 40 mg/day below age 8 years and 60 mg/day above age 8 years; the tablets are not suitable below 33 kg [97]. The FDA label proposes 5 mg three times a day for 3 days, followed by 10, 15 and 20 mg three times a day in 3-day periods, with a usual dose of 40 to 80 mg/day and a maximum of 80 mg/day (20 mg four times a day), and does not establish safety and efficacy below age 12 years [99]. Withdrawal must always be gradual, over 1 to 2 weeks, because of the risk of a withdrawal syndrome with seizures and hallucinations; doses should be reduced in renal impairment [97, 99]. In pregnancy, it is used only if the benefit outweighs the risk, because the drug crosses the placenta and withdrawal seizures have been described in exposed newborns; it passes into breast milk in minimal amounts, with no effects expected in the breastfed infant [97].

Botulinum toxin type A (onabotulinumtoxinA, Botox®; also Dysport®, Xeomin® and others in Brazil). It inhibits acetylcholine release at the neuromuscular junction. The units of each product are not interchangeable between preparations [98, 100]. The FDA label approves Botox® for spasticity in patients aged 2 years or older, without restriction of cause: in adults, lower limb, 300 to 400 units divided among the ankle and toe muscles; in children, lower limb, 4 to 8 units/kg (maximum 300 units) and upper limb, 3 to 6 units/kg (maximum 200 units); maximum cumulative dose over 3 months of 400 units in adults and the lesser of 10 units/kg and 340 units in children [100]. The European SmPC, consulted at INFARMED, restricts the spasticity indications to paediatric focal spasticity associated with cerebral palsy, from age 2 years, and to focal spasticity of the upper and lower limbs after stroke in adults, with equivalent doses (adults, lower limb, 300 to 400 units in up to 6 muscles, maximum 400 units per treatment; children, lower limb, 4 to 8 units/kg up to 300 units), reinjection no sooner than 12 weeks and electromyographic or nerve stimulation guidance when useful [101]. Injection in ARSACS is therefore within the FDA-approved indication and outside the indications of the European SmPC [100, 101]. The FDA boxed warning describes distant spread of toxin effect, with asthenia, generalised weakness, diplopia, ptosis, dysphagia, dysphonia, dysarthria, urinary incontinence and breathing difficulties, hours to weeks after injection, with a risk probably greater in children treated for spasticity and in patients with conditions predisposing to these symptoms [100]. In ARSACS, pre-existing dysphagia and dysarthria justify monitoring for these effects.

Other antispasticity drugs. Tizanidine, gabapentin, clonazepam and dantrolene are part of the Ataxia UK second-line sequence [16]. The sources consulted do not describe their use in series of patients with ARSACS, and the choice follows the national labels of these substances for spasticity of other causes. Cannabinoids reduced perceived pain and spasticity in multiple sclerosis studies, but the largest trials showed no objective reduction in spasticity [16]. In Quebec, 18% of 50 patients with ARSACS used cannabis and 40% had used it at least once, more often those with chronic pain; the study did not assess efficacy [79].

7.5. Orthoses and surgery

Orthoses are part of the spasticity treatment recommended by GeneReviews and, combined with physiotherapy, prolong the effect of botulinum toxin [1, 16]. Surgery, orthopaedic (lengthening, tenotomy or tendon transfer) or neurosurgical (peripheral neurotomies, dorsal rhizotomies, DREZotomy), is reserved for fixed, irreducible deformities or failure of physical and pharmacological measures [16]. Scoliosis, present in 15% of patients, should be monitored and referred for spinal surgery if severe, according to the Ataxia UK recommendations formulated for Friedreich ataxia and applicable by analogy [1, 16].

7.6. Urinary dysfunction

Urinary urgency and incontinence, present in about one third of patients, result mainly from pyramidal dysfunction [1]. The Ataxia UK guideline recommends measuring post-void residual volume, by ultrasound or catheterisation, and excluding urinary tract infection with a dipstick before starting antimuscarinics; when the residual is 100 mL or more, clean intermittent self-catheterisation is taught, and when it is lower, antimuscarinics are used [16]. General measures include reducing caffeine, fizzy drinks and alcohol, timed voiding and bladder retraining, an individualised fluid intake of 1 to 2 litres per day and pelvic floor exercises for mild symptoms [16]. Urodynamics is not routine, and referral to urology is indicated for haematuria, suspected concomitant urological disease, recurrent infections, refractoriness or for intravesical botulinum toxin or a suprapubic catheter [16]. Antimuscarinics require caution in heart disease and cognitive impairment; in patients with cognitive impairment, Ataxia UK prefers the more selective agents, such as trospium or darifenacin [16]. GeneReviews cites tolterodine, oxybutynin and low-dose amitriptyline and referral to urology [1].

Intravesical botulinum toxin is approved by the FDA for urinary incontinence due to neurogenic detrusor overactivity in adults and, from age 5 years, for neurogenic detrusor overactivity, when antimuscarinics are ineffective or poorly tolerated; Ataxia UK places it within the remit of urology [16, 100].

7.7. Pain, cramps and sexual dysfunction

Chronic pain affects about 30% of patients and is associated with spasms, cramps and a neuropathic component [79]. Ataxia UK recommends physiotherapy and/or drugs and, for neuropathic pain, amitriptyline, nortriptyline, carbamazepine, pregabalin, gabapentin or duloxetine, with referral to a pain clinic when pain is severe or limits activities [16]. For amitriptyline, the reference dosage for neuropathic pain in adults is that of the Portuguese SmPC: start with 10 to 25 mg at night, increases of 10 to 25 mg every 3 to 7 days, usual dose of 25 to 75 mg per day at night, caution above 100 mg and no single dose above 75 mg; the analgesic effect appears within 2 to 4 weeks; in the elderly and in patients with cardiovascular disease it is started at 10 to 25 mg; an electrocardiogram should be performed before starting [104]. The FDA label approves amitriptyline only for depression and carries a boxed warning on suicidality in children, adolescents and young adults [105]. Quinine for nocturnal cramps is not recommended in patients with heart disease, because of serious adverse effects [16]. Erectile dysfunction, described in ARSACS, should be addressed in men and may be treated with phosphodiesterase 5 inhibitors, with caution in hypotension and heart disease [14, 16].

7.8. Other manifestations

Epilepsy is treated with antiseizure medications according to seizure type; a Moroccan patient with epilepsy beginning at age 5 years had a good response to an antiseizure medication and physiotherapy, and progressive myoclonic epilepsies, which are more difficult to control, have been described [61, 77]. The sources consulted do not describe a preferred antiseizure medication in ARSACS. Deafness is treated with hearing aids [1]. For dysphagia, Ataxia UK recommends speech and language therapy, a dietitian and supplements if there is weight loss, and discussing percutaneous gastrostomy if intake cannot be maintained; the approach is multidisciplinary, with the speech and language therapist, dietitian, physiotherapist and occupational therapist adjusting positioning and assistive devices [16]. School difficulties require neuropsychological assessment and curricular adaptation [1], and cerebellar-type cognitive-affective deficits, consistently present, justify neuropsychological assessment in adults too [76].

7.9. Agents and circumstances to avoid

There is no absolute drug contraindication in ARSACS [1]. As in all diseases with a neuropathic component, neurotoxic drugs, such as some chemotherapies, should be used with caution [1]. Antispasticity drugs may unmask weakness and worsen gait, and antimuscarinics may worsen cognition and are less safe in heart disease [16]. Abrupt withdrawal of baclofen should be avoided [97, 99]. Botulinum toxin injections in patients with dysphagia or dysarthria require monitoring for spread of effect [100].

7.10. Pregnancy, lactation, surgery and transition

Women with ARSACS have had normal pregnancies and deliveries; some have a functional decline during pregnancy, which reverses when they regain their previous weight and activity level [1]. If baclofen is continued in pregnancy, the FDA label recommends tapering the dose gradually and stopping it before delivery or, if this is not possible, informing the parents of the risk of neonatal withdrawal syndrome, which includes increased tone, tremor, jitteriness and seizures; the Portuguese SmPC allows use only if the benefit outweighs the risk and considers breastfeeding compatible [97, 99]. None of the sources consulted describes anaesthetic recommendations specific to ARSACS. The transition from paediatric to adult medicine coincides with the onset of neuropathy and decline in walking, and the guidelines recommend continuous follow-up in a specialist ataxia centre, with physiotherapy, occupational therapy and speech and language therapy [16, 64].

7.11. Regulatory status

No medicine is authorised for ARSACS in the USA, the European Union, Portugal or Brazil. The drugs used are symptomatic and approved for other indications.

7.12. Registered clinical trials

In November 2026, the interventional trials registered on ClinicalTrials.gov for ARSACS were all rehabilitation trials; there were no drug, gene therapy or oligonucleotide trials [93].

These studies are small, non-randomised in those that have already published results, and with multiple outcomes; the mean SARA improvement in the home programme was compared with the estimated natural progression in the same population, but not with a concurrent control group [17]. Effects should not be compared between studies with different designs and populations. The observational studies PROSPAX (NCT04297891), with SARA and SPRS as outcomes at 2 years, and the TreatHSP/SPAX platform (NCT07798674), with 4000 participants planned up to 2035, are preparing future trials; the 7-item SPAXCOM composite scale had a 2-year effect size of 0.71, greater than that of SARA (0.43), which reduces the required sample size [107–109]. Pharmacometric modelling of the natural history of ARSACS has been used to compare trial designs, and analysis by non-linear mixed-effects models was more powerful than conventional methods [110]. A 2025 review notes that most patients available for trials already have more than 10 years of disease and that many no longer walk independently, which weighs on the choice of inclusion criteria [46].

7.13. Investigational therapies and developments in the last 12 months

All therapeutic candidates are at the preclinical stage. Replacement gene therapy faces a specific obstacle: the SACS coding sequence, of about 13.7 kb, exceeds the capacity of the usual viral vectors [7, 111]. Benoit Gentil's group at McGill University has developed a gene augmentation therapy strategy with a "minigene" (mini-SACS) containing only the essential domains of sacsin, carried by a viral vector whose type and route of administration have not been disclosed; according to an institutional press release of July 2024, initial studies in animal models delayed the onset of several symptoms, with no published quantitative data [111]. In May 2026, the same institution reported that the effect wanes over time in the mouse and that the project is being converted into an RNA therapy, which would allow repeated and adjusted doses, still without results [18]. Europe PMC contained, in November 2026, no peer-reviewed publications on these approaches. The level of evidence is therefore that of a press release on preclinical work, with no human study, regulatory application or known designation.

Results from the last 12 months (November 2025 to November 2026) include the phenotypic improvement with loss of SARM1 in the mouse, the RNA-binding activity of the HEPN domain, the demonstration of early and progressive cognitive-affective deficits with the CCAS scale, the conversion of the McGill gene therapy project into an RNA therapy and the publication of the SPAXCOM composite scale for trials in spastic ataxias [18, 40, 48, 76, 109]. In the same period there were no marketing authorisation applications, opinions of the Committee for Medicinal Products for Human Use of the European Medicines Agency or FDA decisions relating to ARSACS in the sources consulted.

7.14. Agreement and divergence between sources

GeneReviews and the Ataxia UK guideline agree on the primacy of physiotherapy, on oral baclofen and botulinum toxin for spasticity and on the use of antimuscarinics for the bladder [1, 16]. Ataxia UK is more detailed: it sets an order of oral antispasticity drugs, warns of paradoxical worsening of gait, requires measurement of post-void residual volume before antimuscarinics and prefers selective antimuscarinics in cognitive impairment [16]. GeneReviews cites low-dose amitriptyline for urinary urgency, which Ataxia UK mentions only for neuropathic pain [1, 16]. As for diagnostic definitions, GeneReviews requires biallelic pathogenic variants, Pilliod and colleagues accept the alteration of the fibroblast mitochondrial network as a criterion that makes the diagnosis definite with a single pathogenic variant or with variants of uncertain interpretation, and Longo and colleagues propose the absence of sacsin in the patient's cells as a criterion [1, 44, 52]. Baclofen doses diverge between the FDA label, which gives no dosage below age 12 years and limits adults to 80 mg/day, and the European SmPC, which gives a weight-based paediatric dosage; the botulinum toxin indications diverge between the FDA, without restriction of cause, and the EU, limited to cerebral palsy and stroke [97, 99–101].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks

ARSACS is transmitted in an autosomal recessive manner [1]. The parents of an affected individual are, as a rule, obligate heterozygotes; testing them is recommended to confirm carrier status and estimate the recurrence risk reliably, because uniparental isodisomy of chromosome 13 and large deletions can mimic homozygosity [1, 8]. Heterozygotes have no symptoms and are not at risk of developing the disease [1]. The absence of retinal fibre layer thickening in 13 carriers studied by OCT is consistent with this absence of clinical expression, although carrier fibroblasts show vimentin "bundling" and blood mononuclear cells have about half the normal sacsin [15, 38, 53].

In regions with a high carrier frequency, such as Saguenay–Lac-Saint-Jean, affected individuals may be found in two generations, because an affected person may have a carrier partner; this pseudodominance should not be confused with dominant inheritance [1, 9]. In highly consanguineous families, more than one recessive ataxia may segregate, as in the Tunisian family with ataxia with vitamin E deficiency, Friedreich ataxia and ARSACS, and the same individual may carry variants in different genes, which complicates counselling [66].

8.2. Cascade and carrier testing

Carrier detection among at-risk relatives requires prior identification of the two pathogenic variants in the affected individual [1]. Genetic counselling should be offered to young adults who are affected, carriers or at risk of being carriers, ideally before a pregnancy, which is the optimal time to determine genetic risk, clarify carrier status and discuss options [1]. In partners with no family history, carrier testing by complete SACS sequencing or by an expanded panel allows the couple's risk to be estimated; the interpretation of missense variants of uncertain significance in partners is the main limitation [1, 53]. The Ataxia UK guideline recommends informed consent and genetic counselling before any genetic test, with discussion of the implications for the individual and the family [16].

8.3. Population carrier screening in Quebec

In Quebec, the discovery of the founder variants led to a public carrier testing programme [24]. After years of community work by the Corporation de recherche et d'action sur les maladies héréditaires (CORAMH), founded in 1980, the Quebec Ministry of Health and Social Services launched in 2010, in Saguenay–Lac-Saint-Jean, a free pilot project for four recessive diseases with a high incidence in the region: ARSACS, Andermann syndrome (hereditary motor and sensory neuropathy with agenesis of the corpus callosum), French-Canadian type Leigh syndrome and tyrosinaemia type 1 [24]. The carrier frequency of these diseases ranges from 1/19 to 1/23, and about 20% of the region's inhabitants carry at least one [24]. Eligible were resident adults with at least one grandparent born in the region who were planning to have children or were less than 16 weeks pregnant; before sample collection, a nurse gave a 45-minute face-to-face information session; carriers received their result by telephone and carrier couples were referred for genetic counselling [24]. By 2012, 3915 people had been tested and 846 carriers of the four diseases identified [24]. Since 2018, the test has been offered throughout the province to adults with at least one grandparent born in Saguenay–Lac-Saint-Jean, Charlevoix or Haute-Côte-Nord who are planning to have children, with buccal cell collection at home, sent by post after online registration [24, 85]. GeneReviews attributes to voluntary screening the decline in birth incidence, which was 1/1932 [1].

8.4. Prenatal and preimplantation diagnosis

Once the variants have been identified in the affected individual, prenatal diagnosis in pregnancies at increased risk and preimplantation genetic testing are possible [1]. Views on prenatal diagnosis may differ between professionals and families, and most centres consider it a personal decision [1]. In the Quebec programme, the pre-test information session presents participants with all the reproductive options available to carrier couples [24].

8.5. Testing of minors

Diagnostic testing is indicated in any child with compatible signs. The sources consulted do not specifically address carrier testing or presymptomatic testing in minors; GeneReviews directs counselling on carrier status and reproductive options to young adults who are affected, carriers or at risk, and the Ataxia UK guideline requires informed consent and genetic counselling before any test [1, 16]. Because carriers do not become ill, knowledge of carrier status has only reproductive consequences [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1. What is known and what is measured

The worldwide prevalence of ARSACS is unknown, and Orphanet classifies it as "Unknown" [6]. The available figures are of three types: birth incidence in a Quebec founder population, the proportion of ARSACS among patients with ataxia in hospital series and variant frequency in population databases. These measures are not interchangeable: a proportion among patients referred for ataxia is not a national prevalence, and the absence of reported cases in a country does not mean absence of the disease.

9.2. Quebec: the founder population

In Saguenay–Lac-Saint-Jean, birth incidence was estimated at 1/1932 live births in the period 1941–1985, corresponding to 51.76 per 100,000 births, and the carrier frequency at 1/22 inhabitants [6, 9]. The same study is cited by OMIM and GeneReviews with a carrier frequency of 1/21; the original abstract states 1/22, a value also used in the cloning paper and in the 2021 regional review [1, 2, 9, 24, 26]. The coefficients of inbreeding and kinship were, respectively, 2 and 3 times higher in affected families than in controls, but the patients' birthplaces were not clustered, and genealogy points to a founder effect, with frequent ancestry in the French region of Perche [9]. In a study of 567 probands with recessive diseases frequent in the region, parental consanguinity was present in 13%, but only 5% of couples were second cousins or closer, confirming that the high frequency of these diseases results from the founder effect and not from consanguineous marriages [2, 24]. Quebec has more than 300 identified patients [1]. The c.8844del variant accounts for about 94% of alleles in French-Canadian patients and c.7504C>T for about 4%; in the gnomAD database, the frequencies of these variants are of the order of 6 and 1 per 100,000 alleles [5, 24]. Birth incidence is falling with voluntary carrier screening [1].

9.3. Europe and North America outside Quebec

In the Netherlands, complete SACS sequencing in 43 index patients with ataxia beginning before age 25 years identified variants in 16 (37%) [10]. In Belgium, 11 of 85 index patients with a suggestive phenotype (12.9%) had biallelic variants, with a founder variant in several unrelated families [11]. In Germany, ARSACS accounted for at least 11% (9/83) of early-onset ataxias of unknown cause [14]. In France, the SPATAX network found biallelic variants in 33 of 321 index patients with spastic ataxia (10.2%), with a final diagnosis of ARSACS in 34 (10.6%), and the ATAXIC project in 5 of 156 index patients with congenital ataxia (3.2%); 17 other SPATAX index patients had a single heterozygous variant [44]. Of the 39 French probands diagnosed, 22 were from France, 8 from Algeria, 2 from Morocco, 2 from Portugal and 1 each from Australia, Italy and Turkey, and most variants were private [44]. In a prospective cohort from Alsace, 2 of 57 patients with molecularly diagnosed recessive ataxia had ARSACS, compared with 36 with Friedreich ataxia [115]. In Italy, the first cases were described in 2004 [30]. In the USA, an African-American patient with a chromosomal deletion and a novel frameshift variant has been described [116]. GeneReviews reports a relative prevalence of 12% in another cohort of 232 index patients with cerebellar ataxia, not yet published [1]. These figures support the description of ARSACS as one of the most frequent recessive ataxias in the world, although its relative position varies between series: in Alsace, for example, 2 patients were diagnosed with ARSACS, compared with 36 with Friedreich ataxia, 7 with AOA2, 4 with ataxia-telangiectasia and 3 with AOA1 [1, 50, 115].

9.4. Portugal and Brazil

In Portugal, in a prospective cohort of hereditary cerebellar ataxias from a tertiary centre in the north of the country, begun in 2017 and with national referral, 56 of 249 patients (22.5%; 46 families) had spastic ataxia; ARSACS was the most frequent diagnosis in that subgroup, with 17.4% of probands, all with childhood onset except one, and with the complete triad in all but one, who had no neuropathy [12]. In an earlier exome study of 19 Portuguese families with presumed recessive hereditary cerebellar ataxia, 4 families had SACS variants, with presentations through ataxia or motor delay, one of them with seizures in childhood [63]. In the French SPATAX series, 2 of the 39 probands with ARSACS were of Portuguese origin [44]. These studies describe patients referred to specialist centres and do not allow the national prevalence to be calculated.

In Brazil, the first family with a typical clinical and imaging picture was described in 2011, still without molecular confirmation in the abstract [117]. In 2017, two female cousins of German ancestry from Rio Grande do Sul had a novel homozygous frameshift variant [62]. A series of 13 molecularly confirmed Brazilian patients showed ataxia and spasticity in all, neuropathy in 11, retinal fibre layer thickening in all, pontine hypointensities and cerebellar atrophy in all and 14 different variants [31]. The Brazilian authors consider the frequency in the country unknown [62].

9.5. Other populations

In Tunisia, 18 patients from 4 families, 3 of them consanguineous, had a homogeneous phenotype with a mean onset at 4.5 years and infrequent retinal striation [29]. The disease has been described in Japan, including forms without spasticity and without retinal hypermyelination, and in Turkey, where 4 families had different homozygous variants, with no founder variant [5, 27, 73]. Cases have been published in India, including a consanguineous family from a tribal region of Jammu and Kashmir, Australia, Greece, Thailand, Morocco and China [1, 37, 60, 61, 75]. In Japan, SACS variants accounted for 0.268% of 3353 cases with suspected hereditary neuropathy [13].

9.6. Special populations

In populations with a founder effect, testing targeted at the known variants is efficient, but does not replace complete sequencing when the result is negative [1, 51]. In populations with high consanguinity, ARSACS may coexist with other recessive ataxias in the same family [66]. Patients with adult onset or predominant neuropathy tend to be diagnosed later, which contributes to underdiagnosis [11, 64]. In patients who do not walk, who make up a relevant part of the adult population, planning of the wheelchair, posture and transfers is central [16, 46].

9.7. European Reference Network and patient organisations

ERN-RND, the European Reference Network for Rare Neurological Diseases, includes the disease group "Hereditary Ataxias and Spastic Paraplegias"; in 2019 it brought together 32 centres from 13 Member States and endorsed the Ataxia UK ataxia management guideline; it provides a diagnostic flowchart for adult ataxias and has endorsed the SARA scale [16, 19]. The ERN-RND European Rare Neurological Disease Registry is registered on ClinicalTrials.gov (NCT04319796) [93]. The patient organisations providing information and support are the Fondation de l'Ataxie Charlevoix-Saguenay, founded in Montreal in 1972, which maintains an international patient registry, Euro-ATAXIA, the European federation of ataxia associations, and the National Ataxia Foundation, in the USA [1, 20–22, 24]. In Quebec, CORAMH and the Clinique des maladies neuromusculaires de Saguenay–Lac-Saint-Jean provide information, rehabilitation and support to families [24]. The sources consulted do not identify ARSACS-specific associations in Portugal or Brazil.

10. Illustrative Clinical Case

Srikajon and colleagues, from Siriraj Hospital, Mahidol University, in Bangkok, described in 2020 the first confirmed case of ARSACS in Thailand [60]. The patient, a 38-year-old woman, had had progressive walking difficulty and frequent falls since age 2 years; at age 20 years her gait was very unsteady and her speech slurred, and at age 28 years she became wheelchair-dependent [60]. There was consanguinity at the level of the paternal grandparents, with no other relatives with neurological disease [60].

On examination there was saccadic pursuit, hypermetric saccades, horizontal and vertical gaze-evoked nystagmus, scanning speech, finger dysmetria and a broad-based, ataxic and spastic gait, with bilateral pes cavus; she had no scleral telangiectasias, Kayser-Fleischer rings, oculomotor apraxia or vertical supranuclear gaze palsy [60]. Cognition was normal; there was spasticity of both legs, normal tendon reflexes except the ankle jerks, which were absent, Babinski sign and impaired proprioception in the feet and ankles [60]. Fundoscopy showed hypermyelinated nerve fibres radiating from the optic disc (figure 1A) [60].

T2-weighted MRI showed multiple linear hypointensities in the pontine parenchyma (figure 1B), superoanterior atrophy of the vermis and superior cerebellar peduncles and mild cervical cord atrophy (figure 1C); nerve conduction studies revealed a demyelinating sensorimotor polyneuropathy [60]. Exome capture sequencing in a commercial laboratory identified the homozygous variant NM_014363.6:c.382_383del, p.(Glu128SerfsTer2), described by the authors as pathogenic and novel [7, 60]. Figure 1 (fundus, pontine and cerebellar MRI): https://pmc.ncbi.nlm.nih.gov/articles/PMC7394210/figure/F1/

The authors recommend considering ARSACS in patients with early-onset spastic ataxia and negative FXN testing [60]. The case brings together the signs that most help the diagnosis: onset in early childhood, ataxia with spasticity and absent ankle jerks, hypermyelinated retinal fibres and linear pontine stripes on T2. The authors stressed that the patient, despite not being of Quebec origin and having a novel variant, had the typical Quebec phenotype, and that thinning of the cervical cord, occasionally described in Quebec, had not been reported in cases of other origins [60]. The report does not describe treatment or follow-up, so it does not serve as a therapeutic reference; the recommended management is that of section 7. The authors consider linear pontine hypointensities on T2 pathognomonic; in other series these stripes are absent in some patients, and OCT is a more sensitive marker [1, 15, 60].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

ARSACS is slowly progressive [1]. Gait ataxia, spasticity and neuropathy progress faster than dysarthria, oculomotor abnormalities and upper-limb ataxia [1]. In the Pilliod series, 25% of patients assessed after age 30 years used a wheelchair (6/24), and OMIM describes most patients as eventually becoming wheelchair-dependent [1, 2]. In the Thai case in section 10, a wheelchair was needed at age 28 years [60]. Distal amyotrophy with loss of ankle reflexes is present in most patients after age 21 years [1]. Cognition tends to be preserved until late, although there is documented decline in older patients and early and progressive cerebellar cognitive-affective deficits [1, 76]. Vision is usually preserved, except in older patients [1].

The quantitative measures of progression are those described in section 3.10: an increase in SARA of about 0.6 points per year on average, with a slow initial phase (0.1 points per year) and a peak of 0.8 points per year about 37 years after onset; loss of about 1.5 points per year on the Berg scale down to a floor of about 6 points after age 40 years; a decrease in walking speed of 0.044 m/s per year and in the 6-minute walk distance of 20.8 m per year [46, 65, 80]. Onset after age 15 years is associated with faster progression, and men start from lower scores [65]. The only study of life expectancy, in the French-Canadian population, concluded that it was reduced, with a mean of 51 years [1, 118]. The sources consulted do not describe causes of death or survival data in other populations. These data refer to patients under usual care and rehabilitation, without disease-modifying therapy.

11.2. Follow-up

GeneReviews recommends an annual full neurological assessment and assessment of gait, fine motor skills and speech as needed [1]. The Ataxia UK guideline recommends follow-up in a specialist ataxia centre, open access to physiotherapy, occupational therapy and speech and language therapy, bladder and swallowing assessment when symptoms arise and monitoring for scoliosis [16]. There is no recommendation for routine cardiac screening, because cardiac abnormalities are not frequent outside Quebec [1]. OCT and MRI, useful in diagnosis, have no recommended repeat interval; the correlation of macular fibre layer thickness with SARA and longitudinal imaging measures are being studied as biomarkers [35, 69].

11.3. Clinical trial readiness

Systematic follow-up with validated scales has clinical value and allows participation in registries and studies. For ARSACS, the validated instruments are the DSI-ARSACS severity index, with 8 items performed in a routine visit, SARA, the Berg scale, the 10-Meter Walk Test, the Timed Up and Go and grip strength, which were the instruments most sensitive to progression over 2 years [81, 82, 119]. The SPAXCOM composite scale, developed for ARSACS and SPG7, and digital gait measures are being validated for trials [46, 109]. Enrolment of patients in the ERN-RND registries and in natural history platforms, such as PROSPAX and TreatHSP/SPAX, is the route to access future trials [19, 107, 108].

12. Summary and Final Considerations

ARSACS is an autosomal recessive spastic ataxia caused by biallelic variants in SACS, a gene at 13q12.12 encoding sacsin, a 4579-amino-acid molecular chaperone involved in proteostasis, mitochondrial dynamics and the organisation of intermediate filaments [1, 5, 7]. Described in 1978 in north-eastern Quebec, where a founder effect led to a birth incidence of 1/1932 and a carrier frequency of about 1/22, it is now recognised on every continent and accounts for between 11% and 37% of patients in European series of early-onset ataxia of unknown cause or with a suggestive phenotype, besides being the most frequent diagnosis among the spastic ataxias in a Portuguese cohort [9, 10, 12, 14, 23]. Almost all variants, including missense variants, lead to loss of the protein through co-translational degradation, which explains the lack of genotype–phenotype correlation [52].

The typical picture begins between 12 and 18 months with gait unsteadiness, followed by lower-limb spasticity in childhood and mixed sensorimotor neuropathy in adolescence; dysarthria, nystagmus, pes cavus and urinary dysfunction complete the picture [1, 64]. Adult-onset forms and presentations with isolated neuropathy, without spasticity or without ataxia, make the spectrum broader than the name suggests [11, 14, 32]. Diagnosis rests on genetic testing, by panel with copy-number analysis or by exome; OCT with a mean peripapillary fibre layer thickness greater than 119 µm, linear pontine stripes and superior vermian atrophy on MRI and demyelinating neuropathy without nerve enlargement are the supporting signs [1, 15, 74]. Quantification of sacsin in blood cells resolves many variants of uncertain significance [53].

There is no disease-modifying treatment. Management rests on rehabilitation, which has the most direct evidence in ARSACS, treatment of spasticity with baclofen and botulinum toxin, treatment of bladder dysfunction, pain, dysphagia and dysarthria according to the ataxia management guideline endorsed by ERN-RND, and annual neurological monitoring [1, 16, 17, 91]. The disease progresses slowly, with a need for a wheelchair in about a quarter of patients assessed after age 30 years and a mean life expectancy of 51 years in the only available study [1, 65]. Therapeutic research is at the preclinical stage, with strategies of minigene and RNA-based gene replacement, Purkinje cell protection and correction of mitochondrial dysfunction, and with multicentre cohorts and scales preparing the trials [18, 45, 48, 109]. Genetic counselling, parental testing, cascade testing and, in Quebec, population carrier screening are the available preventive measures [1, 24].

13. References

1. Vermeer S, van de Warrenburg BP, Kamsteeg EJ, Brais B, Synofzik M. ARSACS. 2003 Dec 9 [Updated 2020 Jan 2]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1255/
2. OMIM. #270550 Spastic ataxia, Charlevoix-Saguenay type; SACS. Baltimore (MD): Johns Hopkins University; last edited 2024-01-12. Available at: https://www.omim.org/entry/270550
3. Rossi M, Anheim M, Durr A, Klein C, Koenig M, Synofzik M, et al. The genetic nomenclature of recessive cerebellar ataxias. Mov Disord. 2018;33(7):1056-1076. doi:10.1002/mds.27415. PMID: 29756227.
4. Armour CM, Smith A, Hartley T, Chardon JW, Sawyer S, Schwartzentruber J, et al. Syndrome disintegration: Exome sequencing reveals that Fitzsimmons syndrome is a co-occurrence of multiple events. Am J Med Genet A. 2016;170(7):1820-1825. doi:10.1002/ajmg.a.37684. PMID: 27133561.
5. OMIM. *604490 Sacsin; SACS. Baltimore (MD): Johns Hopkins University; last edited 2015-08-18. Available at: https://www.omim.org/entry/604490
6. Orphanet. Autosomal recessive spastic ataxia of Charlevoix-Saguenay (ORPHA:98). Available at: https://www.orpha.net/en/disease/detail/98 (Orphadata/api.orphacode.org data accessed on 2026-11-06).
7. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166. Tool: https://variantvalidator.org (accessed on 2026-11-06, GRCh38, MANE Select NM_014363.6).
8. Anesi L, de Gemmis P, Pandolfo M, Hladnik U. Two novel homozygous SACS mutations in unrelated patients including the first reported case of paternal UPD as an etiologic cause of ARSACS. J Mol Neurosci. 2011;43(3):346-349. doi:10.1007/s12031-010-9448-4. PMID: 20852969.
9. De Braekeleer M, Giasson F, Mathieu J, Roy M, Bouchard JP, Morgan K. Genetic epidemiology of autosomal recessive spastic ataxia of Charlevoix-Saguenay in northeastern Quebec. Genet Epidemiol. 1993;10(1):17-25. doi:10.1002/gepi.1370100103. PMID: 8472930.
10. Vermeer S, Meijer RP, Pijl BJ, Timmermans J, Cruysberg JR, Bos MM, et al. ARSACS in the Dutch population: a frequent cause of early-onset cerebellar ataxia. Neurogenetics. 2008;9(3):207-214. doi:10.1007/s10048-008-0131-7. PMID: 18465152. PMC2441586.
11. Baets J, Deconinck T, Smets K, Goossens D, Van den Bergh P, Dahan K, et al. Mutations in SACS cause atypical and late-onset forms of ARSACS. Neurology. 2010;75(13):1181-1188. doi:10.1212/wnl.0b013e3181f4d86c. PMID: 20876471.
12. Damásio J, Santos M, Costa S, Moura J, Sardoeira A, Lemos C, et al. Deciphering Spastic Ataxia: Clinical and Genetic Profiles. Neurol Genet. 2025;11(6):e200331. doi:10.1212/nxg.0000000000200331. PMID: 41357347. PMC12681266.
13. Yuan JH, Higuchi Y, Ando M, Hiramatsu Y, Yoshimura A, Hobara T, et al. Frameshift and Copy Number Variants in SACS-Related Neuropathy. Neurol Genet. 2025;11(6):e200318. doi:10.1212/nxg.0000000000200318. PMID: 41164123. PMC12563766.
14. Synofzik M, Soehn AS, Gburek-Augustat J, Schicks J, Karle KN, Schüle R, et al. Autosomal recessive spastic ataxia of Charlevoix Saguenay (ARSACS): expanding the genetic, clinical and imaging spectrum. Orphanet J Rare Dis. 2013;8:41. doi:10.1186/1750-1172-8-41. PMID: 23497566. PMC3610264.
15. Parkinson MH, Bartmann AP, Clayton LMS, Nethisinghe S, Pfundt R, Chapple JP, et al. Optical coherence tomography in autosomal recessive spastic ataxia of Charlevoix-Saguenay. Brain. 2018;141(4):989-999. doi:10.1093/brain/awy028. PMID: 29538656.
16. Ataxia UK. Management of the ataxias: towards best clinical practice. 3rd ed. London: Ataxia UK; 2016 (version 2.0, endorsed by ERN-RND in October 2018, updated in April 2019). Available at: https://www.ern-rnd.eu/wp-content/uploads/2023/07/Final-guideline-AtaxiaUK_english.pdf
17. Lessard I, Gagnon C, Tremblay M, Girard-Côté L, Côté I, Aubertin-Leheudre M, et al. A Tailored Home-Based Training Program Improved Ataxia Severity and Participation in Adults With ARSACS. Cerebellum. 2025;24(3):63. doi:10.1007/s12311-025-01816-z. PMID: 40095137. PMC11914335.
18. McGill University, DNA-to-RNA (D2R). Trainee spotlight: Caitlin Atkinson. 2026 May 22. Available at: https://www.mcgill.ca/dna-to-rna/node/1629
19. European Reference Network for Rare Neurological Diseases (ERN-RND). Cerebellar ataxia & hereditary spastic paraplegias. Available at: https://www.ern-rnd.eu/?p=25660 (accessed on 2026-11-06).
20. Fondation de l'Ataxie Charlevoix-Saguenay. Available at: https://arsacs.com (international patient registry: https://arsacs.com/persons-with-arsacs/#patient-registry).
21. Euro-ATAXIA — European Federation of Hereditary Ataxias. Available at: https://euroataxia.org
22. National Ataxia Foundation. Available at: https://ataxia.org
23. Bouchard JP, Barbeau A, Bouchard R, Bouchard RW. Autosomal recessive spastic ataxia of Charlevoix-Saguenay. Can J Neurol Sci. 1978;5(1):61-69. doi:10.1017/s0317167100024793. PMID: 647499.
24. Bchetnia M, Bouchard L, Mathieu J, Campeau PM, Morin C, Brisson D, et al. Genetic burden linked to founder effects in Saguenay-Lac-Saint-Jean illustrates the importance of genetic screening test availability. J Med Genet. 2021;58(10):653-665. doi:10.1136/jmedgenet-2021-107809. PMID: 33910931. PMC8479736.
25. Richter A, Rioux JD, Bouchard JP, Mercier J, Mathieu J, Ge B, et al. Location score and haplotype analyses of the locus for autosomal recessive spastic ataxia of Charlevoix-Saguenay, in chromosome region 13q11. Am J Hum Genet. 1999;64(3):768-775. doi:10.1086/302274. PMID: 10053011. PMC1377794.
26. Engert JC, Bérubé P, Mercier J, Doré C, Lepage P, Ge B, et al. ARSACS, a spastic ataxia common in northeastern Québec, is caused by mutations in a new gene encoding an 11.5-kb ORF. Nat Genet. 2000;24(2):120-125. doi:10.1038/72769. PMID: 10655055.
27. Ouyang Y, Takiyama Y, Sakoe K, Shimazaki H, Ogawa T, Nagano S, et al. Sacsin-related ataxia (ARSACS): expanding the genotype upstream from the gigantic exon. Neurology. 2006;66(7):1103-1104. doi:10.1212/01.wnl.0000204300.94261.ea. PMID: 16606928.
28. Parfitt DA, Michael GJ, Vermeulen EG, Prodromou NV, Webb TR, Gallo JM, et al. The ataxia protein sacsin is a functional co-chaperone that protects against polyglutamine-expanded ataxin-1. Hum Mol Genet. 2009;18(9):1556-1565. doi:10.1093/hmg/ddp067. PMID: 19208651. PMC2667285.
29. El Euch-Fayache G, Lalani I, Amouri R, Turki I, Ouahchi K, Hung WY, et al. Phenotypic features and genetic findings in sacsin-related autosomal recessive ataxia in Tunisia. Arch Neurol. 2003;60(7):982-988. doi:10.1001/archneur.60.7.982. PMID: 12873855.
30. Criscuolo C, Banfi S, Orio M, Gasparini P, Monticelli A, Scarano V, et al. A novel mutation in SACS gene in a family from southern Italy. Neurology. 2004;62(1):100-102. doi:10.1212/wnl.62.1.100. PMID: 14718706.
31. Rezende Filho FM, Parkinson MH, Pedroso JL, Poh R, Faber I, Lourenço CM, et al. Clinical, ophthalmological, imaging and genetic features in Brazilian patients with ARSACS. Parkinsonism Relat Disord. 2019;62:148-155. doi:10.1016/j.parkreldis.2018.12.024. PMID: 30638817.
32. Vill K, Müller-Felber W, Gläser D, Kuhn M, Teusch V, Schreiber H, et al. SACS variants are a relevant cause of autosomal recessive hereditary motor and sensory neuropathy. Hum Genet. 2018;137(11-12):911-919. doi:10.1007/s00439-018-1952-6. PMID: 30460542.
33. Beaudin M, Matilla-Dueñas A, Soong BW, Pedroso JL, Barsottini OG, Mitoma H, et al. The Classification of Autosomal Recessive Cerebellar Ataxias: a Consensus Statement from the Society for Research on the Cerebellum and Ataxias Task Force. Cerebellum. 2019;18(6):1098-1125. doi:10.1007/s12311-019-01052-2. PMID: 31267374. PMC6867988.
34. Synofzik M, Schüle R. Overcoming the divide between ataxias and spastic paraplegias: Shared phenotypes, genes, and pathways. Mov Disord. 2017;32(3):332-345. doi:10.1002/mds.26944. PMID: 28195350. PMC6287914.
35. Scaravilli A, Negroni D, Senatore C, Ugga L, Cosottini M, Ricca I, et al. MRI-ARSACS: An Imaging Index for Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay (ARSACS) Identification Based on the Multicenter PROSPAX Study. Mov Disord. 2024;39(8):1343-1351. doi:10.1002/mds.29871. PMID: 38847051.
36. Firth HV, Hurst JA. Ataxic child. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 74–77.
37. Xiromerisiou G, Dadouli K, Marogianni C, Provatas A, Ntellas P, Rikos D, et al. A novel homozygous SACS mutation identified by whole exome sequencing-genotype phenotype correlations of all published cases. J Mol Neurosci. 2020;70(1):131-141. doi:10.1007/s12031-019-01410-z. PMID: 31701440.
38. Nethisinghe S, Abeti R, Kesavan M, Wigley WC, Giunti P. Hsp90 Inhibition: A Promising Therapeutic Approach for ARSACS. Int J Mol Sci. 2021;22(21):11722. doi:10.3390/ijms222111722. PMID: 34769152. PMC8584178.
39. Romano A, Tessa A, Barca A, Fattori F, de Leva MF, Terracciano A, et al. Comparative analysis and functional mapping of SACS mutations reveal novel insights into sacsin repeated architecture. Hum Mutat. 2013;34(3):525-537. doi:10.1002/humu.22269. PMID: 23280630. PMC3629688.
40. Longo L, Mariani I, Cusimano A, Carrotta R, Zacco E, Vazzana R, et al. The isolated HEPN domain of SACSIN exhibits RNA-binding activity. Cell Mol Life Sci. 2026;83(1):378. doi:10.1007/s00018-026-06369-w. PMID: 42873286. PMC13669270.
41. Girard M, Larivière R, Parfitt DA, Deane EC, Gaudet R, Nossova N, et al. Mitochondrial dysfunction and Purkinje cell loss in autosomal recessive spastic ataxia of Charlevoix-Saguenay (ARSACS). Proc Natl Acad Sci U S A. 2012;109(5):1661-1666. doi:10.1073/pnas.1113166109. PMID: 22307627. PMC3277168.
42. Gentil BJ, Lai GT, Menade M, Larivière R, Minotti S, Gehring K, et al. Sacsin, mutated in the ataxia ARSACS, regulates intermediate filament assembly and dynamics. FASEB J. 2019;33(2):2982-2994. doi:10.1096/fj.201801556r. PMID: 30332300.
43. Duncan EJ, Larivière R, Bradshaw TY, Longo F, Sgarioto N, Hayes MJ, et al. Altered organization of the intermediate filament cytoskeleton and relocalization of proteostasis modulators in cells lacking the ataxia protein sacsin. Hum Mol Genet. 2017;26(16):3130-3143. doi:10.1093/hmg/ddx197. PMID: 28535259. PMC5886247.
44. Pilliod J, Moutton S, Lavie J, Maurat E, Hubert C, Bellance N, et al. New practical definitions for the diagnosis of autosomal recessive spastic ataxia of Charlevoix-Saguenay. Ann Neurol. 2015;78(6):871-886. doi:10.1002/ana.24509. PMID: 26288984.
45. Del Bondio A, Longo F, De Ritis D, Spirito E, Podini P, Brais B, et al. Restoring calcium homeostasis in Purkinje cells arrests neurodegeneration and neuroinflammation in the ARSACS mouse model. JCI Insight. 2023;8(12):e163576. doi:10.1172/jci.insight.163576. PMID: 37159335. PMC10371240.
46. Fogel BL, Klopstock T, Lynch DR, Maltecca F, Verma M, Minassian BA, et al. Autosomal Recessive Cerebellar Ataxias: Translating Genes to Therapies. Ann Neurol. 2025;98(3):448-470. doi:10.1002/ana.27271. PMID: 40464291. PMC12392066.
47. Larivière R, Gaudet R, Gentil BJ, Girard M, Conte TC, Minotti S, et al. Sacs knockout mice present pathophysiological defects underlying autosomal recessive spastic ataxia of Charlevoix-Saguenay. Hum Mol Genet. 2015;24(3):727-739. doi:10.1093/hmg/ddu491. PMID: 25260547. PMC4291249.
48. Ndiaye PS, Paul S, Zhao G, Ding C, Pulst SM, Schwarz T. Loss of SARM1 Improves Phenotypes in a Mouse Model of Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay. Neurol Genet. 2026;12(4):e200400. doi:10.1212/nxg.0000000000200400. PMID: 42290675. PMC13262670.
49. Larivière R, Sgarioto N, Márquez BT, Gaudet R, Choquet K, McKinney RA, et al. Sacs R272C missense homozygous mice develop an ataxia phenotype. Mol Brain. 2019;12(1):19. doi:10.1186/s13041-019-0438-3. PMID: 30866998. PMC6416858.
50. Salem IH, Blais M, Zuluaga-Sánchez VM, Rouleau L, Becker EBE, Dupré N. ARSACS: Clinical Features, Pathophysiology and iPS-Derived Models. Cerebellum. 2025;24(1):24. doi:10.1007/s12311-024-01777-9. PMID: 39753868.
51. Thiffault I, Dicaire MJ, Tetreault M, Huang KN, Demers-Lamarche J, Bernard G, et al. Diversity of ARSACS mutations in French-Canadians. Can J Neurol Sci. 2013;40(1):61-66. doi:10.1017/s0317167100012968. PMID: 23250129.
52. Longo F, De Ritis D, Miluzio A, Fraticelli D, Baets J, Scarlato M, et al. Assessment of Sacsin Turnover in Patients With ARSACS: Implications for Molecular Diagnosis and Pathogenesis. Neurology. 2021;97(23):e2315-e2327. doi:10.1212/wnl.0000000000012962. PMID: 34649874. PMC8665432.
53. De Ritis D, Ferrè L, De Winter J, Tremblay-Desbiens C, Blais M, Bassi MT, et al. Reduction of sacsin levels in peripheral blood mononuclear cells as a diagnostic tool for spastic ataxia of Charlevoix-Saguenay. Brain Commun. 2024;6(4):fcae243. doi:10.1093/braincomms/fcae243. PMID: 39091421. PMC11291951.
54. Breckpot J, Takiyama Y, Thienpont B, Van Vooren S, Vermeesch JR, Ortibus E, et al. A novel genomic disorder: a deletion of the SACS gene leading to spastic ataxia of Charlevoix-Saguenay. Eur J Hum Genet. 2008;16(9):1050-1054. doi:10.1038/ejhg.2008.58. PMID: 18398442.
55. McMillan HJ, Carter MT, Jacob PJ, Laffan EE, O'Connor MD, Boycott KM. Homozygous contiguous gene deletion of 13q12 causing LGMD2C and ARSACS in the same patient. Muscle Nerve. 2009;39(3):396-399. doi:10.1002/mus.21222. PMID: 19208398.
56. Yeow D, Katz M, Rodgers J, Davis M, Robertson T, Gardner RJM, et al. Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay in Two Half-Siblings. Ann Clin Transl Neurol. 2026;13(2):413-417. doi:10.1002/acn3.70236. PMID: 41353788. PMC12883688.
57. Kim D, Ryoo N, Park YH, Bagyinszky E, An SSA, Kim S. A Novel Mutation in Sacsin, p.Val1335IIe, May Cause Late-Onset Sacsinopathy Due to Haploinsufficiency. Curr Issues Mol Biol. 2023;45(12):9917-9925. doi:10.3390/cimb45120619. PMID: 38132465. PMC10741900.
58. National Center for Biotechnology Information. ClinVar: SACS[gene]. Available at: https://www.ncbi.nlm.nih.gov/clinvar/?term=SACS%5Bgene%5D (accessed on 2026-11-06).
59. Sortino V, Sapuppo A, Bernabini A, Pappalardo XG, Falsaperla R, Di Pasquale G, et al. The recurrent SACS gene p.Arg272His hotspot expands the neurodevelopmental and cognitive spectrum of ARSACS. Neurogenetics. 2026;27(1):62. doi:10.1007/s10048-026-00935-x. PMID: 42671486.
60. Srikajon J, Pitakpatapee Y, Limwongse C, Chirapapaisan N, Srivanitchapoom P. Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay (ARSACS) in a Thai Patient: The Classic Clinical Manifestations, Funduscopic Feature, and Brain Imaging Findings with a Novel Mutation in the SACS Gene. Tremor Other Hyperkinet Mov (N Y). 2020;10:1. doi:10.5334/tohm.68. PMID: 32775015. PMC7394210.
61. Laaraje A, Alaoui B, Radi A, Hasani A, Agadr A, Abilkassem R. Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay with Early-Onset Epilepsy and Novel Clinical Features: A rare entity from Morocco in the Middle East and North Africa region. Sultan Qaboos Univ Med J. 2025;25(1):756-761. doi:10.18295/2075-0528.2906. PMID: 40979611. PMC12445317.
62. Burguêz D, Oliveira CM, Rockenbach MABC, Fussiger H, Vedolin LM, Winckler PB, et al. Autosomal recessive spastic ataxia of Charlevoix-Saguenay: a family report from South Brazil. Arq Neuropsiquiatr. 2017;75(6):339-344. doi:10.1590/0004-282x20170044. PMID: 28658401.
63. Santos M, Damásio J, Carmona S, Neto JL, Dehghani N, Guedes LC, et al. Molecular Characterization of Portuguese Patients with Hereditary Cerebellar Ataxia. Cells. 2022;11(6):981. doi:10.3390/cells11060981. PMID: 35326432. PMC8946949.
64. Duquette A, Brais B, Bouchard JP, Mathieu J. Clinical presentation and early evolution of spastic ataxia of Charlevoix-Saguenay. Mov Disord. 2013;28(14):2011-2014. doi:10.1002/mds.25604. PMID: 23913799.
65. Hendrickx N, Mentré F, Traschütz A, Gagnon C, Schüle R, Synofzik M, et al. Prediction of Individual Disease Progression Including Parameter Uncertainty in Rare Neurodegenerative Diseases: The Example of Autosomal-Recessive Spastic Ataxia Charlevoix Saguenay (ARSACS). AAPS J. 2024;26(3):57. doi:10.1208/s12248-024-00925-7. PMID: 38689016.
66. Bouhlal Y, Zouari M, Kefi M, Ben Hamida C, Hentati F, Amouri R. Autosomal recessive ataxia caused by three distinct gene defects in a single consanguineous family. J Neurogenet. 2008;22(2):139-148. doi:10.1080/01677060802025233. PMID: 18569450.
67. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 10, Disorders of Oxidative Phosphorylation, secção 10.1 (Tabela 10.2).
68. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, Clinical Approach to Inborn Errors of Metabolism in Paediatrics, secção 1.5 (Tabela 1.19).
69. Rezende Filho FM, Bremner F, Pedroso JL, de Andrade JBC, Marianelli BF, Lourenço CM, et al. Retinal Architecture in Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay (ARSACS): Insights into Disease Pathogenesis and Biomarkers. Mov Disord. 2021;36(9):2027-2035. doi:10.1002/mds.28612. PMID: 33893680.
70. Vogel AP, Rommel N, Oettinger A, Stoll LH, Kraus EM, Gagnon C, et al. Coordination and timing deficits in speech and swallowing in autosomal recessive spastic ataxia of Charlevoix-Saguenay (ARSACS). J Neurol. 2018;265(9):2060-2070. doi:10.1007/s00415-018-8950-4. PMID: 29968200.
71. Gagnon C, Lessard I, Lavoie C, Côté I, St-Gelais R, Mathieu J, et al. An exploratory natural history of ataxia of Charlevoix-Saguenay: A 2-year follow-up. Neurology. 2018;91(14):e1307-e1311. doi:10.1212/wnl.0000000000006290. PMID: 30158165. PMC6177270.
72. Lessard I, St-Gelais R, Hébert LJ, Côté I, Mathieu J, Brais B, et al. Functional mobility in walking adult population with ataxia of Charlevoix-Saguenay. Orphanet J Rare Dis. 2021;16(1):432. doi:10.1186/s13023-021-02054-2. PMID: 34649570. PMC8515729.
73. Shimazaki H, Takiyama Y, Sakoe K, Ando Y, Nakano I. A phenotype without spasticity in sacsin-related ataxia. Neurology. 2005;64(12):2129-2131. doi:10.1212/01.wnl.0000166031.91514.b3. PMID: 15985586.
74. Kneer K, Straub S, Wittlinger J, Stahl JH, Winter N, Timmann D, et al. Neuropathy in ARSACS is demyelinating but without typical nerve enlargement in nerve ultrasound. J Neurol. 2024;271(5):2494-2502. doi:10.1007/s00415-023-12159-2. PMID: 38261029. PMC11055797.
75. Kuchay RAH, Mir YR, Zeng X, Hassan A, Musarrat J, Parwez I, et al. ARSACS as a Worldwide Disease: Novel SACS Mutations Identified in a Consanguineous Family from the Remote Tribal Jammu and Kashmir Region in India. Cerebellum. 2019;18(4):807-812. doi:10.1007/s12311-019-01028-2. PMID: 30963395.
76. Fortin J, Synofzik M, Pedneault-Tremblay ÉA, Hermle D, Thieme A, Timmann D, et al. The Cerebellar Cognitive-Affective Syndrome Scale Reveals Consistent, Early, and Progressive Neuropsychological Deficits in Autosomal-Recessive Spastic Ataxia of Charlevoix-Saguenay: A Large International Cross-Sectional Study. Mov Disord. 2026;41(4):1034-1040. doi:10.1002/mds.70201. PMID: 41669957. PMC13067325.
77. Muona M, Berkovic SF, Dibbens LM, Oliver KL, Maljevic S, Bayly MA, et al. A recurrent de novo mutation in KCNC1 causes progressive myoclonus epilepsy. Nat Genet. 2015;47(1):39-46. doi:10.1038/ng.3144. PMID: 25401298. PMC4281260.
78. R Lal R, Iype M, Sreedharan M, Ahamed S, O Surendran M, R S S, et al. Rare and Atypical Manifestations of ARSACS-A Report of Two Cases. Cerebellum. 2026;25(2):47. doi:10.1007/s12311-026-01989-1. PMID: 41941018.
79. Beauchesne W, Ouellet-Dupuis F, Frigon MA, Savard C, Gagné-Ouellet V, Gagnon C, et al. Cannabis use in patients with Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay. J Clin Neurosci. 2022;103:44-48. doi:10.1016/j.jocn.2022.06.019. PMID: 35809457.
80. Lessard I, Côté I, St-Gelais R, Hébert LJ, Brais B, Mathieu J, et al. Natural History of Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay: a 4-Year Longitudinal Study. Cerebellum. 2024;23(2):489-501. doi:10.1007/s12311-023-01558-w. PMID: 37101017.
81. Lessard I, Hébert LJ, Brais B, Duchesne E, Rodrigue X, Brisson JD, et al. Selection of Clinical Outcome Assessments for Trial Readiness in ARSACS - 2-year Progression and Responsiveness to Change Part 1: Disease Severity, Swallowing, Upper Limb Function, and Participation. Cerebellum. 2025;24(4):106. doi:10.1007/s12311-025-01858-3. PMID: 40450178.
82. Lessard I, Duchesne E, Hébert LJ, Brais B, Rodrigue X, Routhier F, et al. Selection of Clinical Outcome Assessments for Trial Readiness in ARSACS - 2-year Progression and Responsiveness to Change Part 2: Mobility, Balance, and Lower Limb Coordination. Cerebellum. 2025;24(4):95. doi:10.1007/s12311-025-01849-4. PMID: 40332679.
83. Firth HV, Hurst JA. Ataxic adult. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 70–73.
84. Ribeiro EM, coord. Ataxia de Friedreich. In: Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 90–92.
85. Gouvernement du Québec. Tests de porteur pour quatre maladies héréditaires récessives. Available at: https://www.sante.gouv.qc.ca/tests4maladies
86. Mercier J, Prévost C, Engert JC, Bouchard JP, Mathieu J, Richter A. Rapid detection of the sacsin mutations causing autosomal recessive spastic ataxia of Charlevoix-Saguenay. Genet Test. 2001;5(3):255-259. doi:10.1089/10906570152742326. PMID: 11788093.
87. Tunca C, İşlek Camadan EE, Smolina N, Palvadeau RJ, Öztop Çakmak Ö, Vural A, et al. Sacsin levels in PBMCs: A diagnostic assay for SACS variants in peripheral blood cells - A PROSPAX study. Mov Disord. 2024;39(12):2291-2297. doi:10.1002/mds.30012. PMID: 39314081.
88. Firth HV, Hurst JA. Hereditary spastic paraplegias (HSP). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 448–451.
89. Klebe S, Azzedine H, Durr A, Bastien P, Bouslam N, Elleuch N, et al. Autosomal recessive spastic paraplegia (SPG30) with mild ataxia and sensory neuropathy maps to chromosome 2q37.3. Brain. 2006;129(Pt 6):1456-1462. doi:10.1093/brain/awl012. PMID: 16434418.
90. de Silva R, Greenfield J, Cook A, Bonney H, Vallortigara J, Hunt B, et al. Guidelines on the diagnosis and management of the progressive ataxias. Orphanet J Rare Dis. 2019;14(1):51. doi:10.1186/s13023-019-1013-9. PMID: 30786918. PMC6381619.
91. Vogel AP, Stoll LH, Oettinger A, Rommel N, Kraus EM, Timmann D, et al. Speech treatment improves dysarthria in multisystemic ataxia: a rater-blinded, controlled pilot-study in ARSACS. J Neurol. 2019;266(5):1260-1266. doi:10.1007/s00415-019-09258-4. PMID: 30840144.
92. Lessard I, Masterman V, Côté I, Gagnon C, Duchesne E. A rehabilitation program to increase balance and mobility in ataxia of Charlevoix-Saguenay: An exploratory study. PLoS One. 2022;17(12):e0279406. doi:10.1371/journal.pone.0279406. PMID: 36576926. PMC9797069.
93. ClinicalTrials.gov. Studies registered for “ARSACS”, “sacsin”, “Charlevoix-Saguenay” and “spastic ataxia” (NCT05479656, NCT05768750, NCT06261424, NCT06596850; ERN-RND European registry NCT04319796). Bethesda (MD): National Library of Medicine. Available at: https://clinicaltrials.gov (API v2 accessed on 2026-11-06).
94. Groupe de recherche interdisciplinaire sur les maladies neuromusculaires (GRIMN). Programme PACE-ARSACS. Available at: https://www.grimn.ca/programme-pace-arscs
95. Schatton C, Synofzik M, Fleszar Z, Giese MA, Schöls L, Ilg W. Individualized exergame training improves postural control in advanced degenerative spinocerebellar ataxia: A rater-blinded, intra-individually controlled trial. Parkinsonism Relat Disord. 2017;39:80-84. doi:10.1016/j.parkreldis.2017.03.016. PMID: 28365204.
96. Niyomwungere E, Routhier F, Gagnon C, Kirby RL, Rodrigue X, Lessard I, et al. Efficacy of Manual Wheelchair Skills Training for Improving Skills and Confidence in People With Hereditary Degenerative Disorders: Protocol for a Sequential Multimethods Study. JMIR Res Protoc. 2025;14:e66974. doi:10.2196/66974. PMID: 40742628. PMC12355141.
97. INFARMED. Lioresal 10 mg and 25 mg tablets (baclofen) — Summary of Product Characteristics. Novartis Farma. Infomed, med_id 5074. Available at: https://extranet.infarmed.pt/INFOMED-fo/
98. Agência Nacional de Vigilância Sanitária (ANVISA). Open data — registered medicines (DADOS_ABERTOS_MEDICAMENTOS.csv). Available at: https://dados.anvisa.gov.br/dados/DADOS_ABERTOS_MEDICAMENTOS.csv (accessed on 2026-11-06).
99. U.S. Food and Drug Administration. Baclofen tablets, USP — prescribing information (DailyMed set id 2278954a-7605-426a-ae85-1bf0a3117420, revision of 2026-04-22). Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=2278954a-7605-426a-ae85-1bf0a3117420
100. Allergan/AbbVie. BOTOX (onabotulinumtoxinA) for injection — prescribing information (DailyMed set id 33d066a9-34ff-4a1a-b38b-d10983df3300). Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=33d066a9-34ff-4a1a-b38b-d10983df3300
101. INFARMED. Botox 100 Units Allergan, powder for solution for injection (botulinum toxin A) — Summary of Product Characteristics. AbbVie. Infomed, med_id 29406. Available at: https://extranet.infarmed.pt/INFOMED-fo/
102. Viatris Specialty. DETROL (tolterodine tartrate) tablets — prescribing information (set id 41eb00d2-64e3-4ec9-8732-4bfc81b04180, revision of 2023-02-15). Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=41eb00d2-64e3-4ec9-8732-4bfc81b04180
103. U.S. Food and Drug Administration. Oxybutynin chloride tablets (set id 0697b3cf-b146-47c0-b5a5-5905ae8d1c97, revision of 2025-11-18) and oxybutynin chloride extended-release tablets (set id 02ff3be7-fc5c-4b91-8c67-ecdf6e19c42a) — prescribing information. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0697b3cf-b146-47c0-b5a5-5905ae8d1c97
104. INFARMED. ADT 10, 25 and 75 mg film-coated tablets (amitriptyline hydrochloride) — Summary of Product Characteristics. Generis. Infomed, med_id 136. Available at: https://extranet.infarmed.pt/INFOMED-fo/
105. U.S. Food and Drug Administration. Amitriptyline hydrochloride tablets — prescribing information (set id 0067b698-ee1d-4524-b1cc-af1028d13736, revision of 2024-06-12). Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0067b698-ee1d-4524-b1cc-af1028d13736
106. ClinicalTrials.gov. NCT06261424. IMPACT, a Supervised Rehabilitation Program for Spastic Ataxias: A Rater-blinded, Randomized Controlled Trial. Available at: https://clinicaltrials.gov/study/NCT06261424
107. ClinicalTrials.gov. NCT04297891. Phenotypes, Biomarkers and Pathophysiology in Spastic Ataxias (PROSPAX). Available at: https://clinicaltrials.gov/study/NCT04297891
108. ClinicalTrials.gov. NCT07798674. TreatHSP/SPAX Master Protocol: Adaptive Platform for Longitudinal Progression, Biomarkers and Pathophysiology in Ataxias, Hereditary Spastic Paraplegias and Spastic Ataxias. Available at: https://clinicaltrials.gov/study/NCT07798674
109. Di Folco C, Dubec-Fleury C, Traschütz A, Kessler C, Reich S, Gagnon C, et al. Spastic Ataxia Composite (SPAXCOM): A Scale to Evaluate the Progression of Subjects with Spasticity and Ataxia. Mov Disord. 2025;40(11):2476-2485. doi:10.1002/mds.70006. PMID: 40832806. PMC12661626.
110. Ryeznik Y, Hilgers RD, Heussen N, Comets E, Mentré F, Hendrickx N, et al. A Pharmacometrics-Informed Trial Simulation Framework for Optimizing Study Designs for Disease-Modifying Treatments in Rare Neurological Disorders. CPT Pharmacometrics Syst Pharmacol. 2025;14(10):1705-1714. doi:10.1002/psp4.70082. PMID: 40763305. PMC12521073.
111. Santangelo R. Developing gene therapy for Ataxia of Charlevoix-Saguenay. McGill Reporter. 2024 Jul 31. Available at: https://reporter.mcgill.ca/developing-gene-therapy-for-ataxia-of-charlevoix-saguenay/
112. Márquez BT, Leung TCS, Hui J, Charron F, McKinney RA, Watt AJ. A mitochondrial-targeted antioxidant (MitoQ) improves motor coordination and reduces Purkinje cell death in a mouse model of ARSACS. Neurobiol Dis. 2023;183:106157. doi:10.1016/j.nbd.2023.106157. PMID: 37209925.
113. Naef V, Vecchia SD, Giacich M, Licitra R, Bachetti T, Vargas GC, et al. Long-term benefits of TUDCA supplement in ARSACS zebrafish model. Sci Rep. 2025;15(1):25429. doi:10.1038/s41598-025-10850-0. PMID: 40659819. PMC12260040.
114. Paré AM, Larochelle N, Dvorak RA, Minotti S, Atkinson CS, Cheng-Boivin Z, et al. HDAC inhibitor SAHA resolved intermediate filament bundling and defective autophagy caused by the lack of sacsin in ARSACS [preprint]. Research Square. 2024. doi:10.21203/rs.3.rs-4745938/v1.
115. Anheim M, Fleury M, Monga B, Laugel V, Chaigne D, Rodier G, et al. Epidemiological, clinical, paraclinical and molecular study of a cohort of 102 patients affected with autosomal recessive progressive cerebellar ataxia from Alsace, Eastern France: implications for clinical management. Neurogenetics. 2010;11(1):1-12. doi:10.1007/s10048-009-0196-y. PMID: 19440741.
116. Dougherty SC, Harper A, Al Saif H, Vorona G, Haines SR. A Chromosomal Deletion and New Frameshift Mutation Cause ARSACS in an African-American. Front Neurol. 2018;9:956. doi:10.3389/fneur.2018.00956. PMID: 30498468. PMC6249318.
117. Pedroso JL, Braga-Neto P, Abrahão A, Rivero RL, Abdalla C, Abdala N, et al. Autosomal recessive spastic ataxia of Charlevoix-Saguenay (ARSACS): typical clinical and neuroimaging features in a Brazilian family. Arq Neuropsiquiatr. 2011;69(2B):288-291. doi:10.1590/s0004-282x2011000300004. PMID: 21625752.
118. Dupré N, Bouchard JP, Brais B, Rouleau GA. Hereditary ataxia, spastic paraparesis and neuropathy in the French-Canadian population. Can J Neurol Sci. 2006;33(2):149-157. doi:10.1017/s031716710000490x. PMID: 16736723.
119. Gagnon C, Brais B, Lessard I, Lavoie C, Côté I, Mathieu J. Development and validation of a disease severity index for ataxia of Charlevoix-Saguenay. Neurology. 2019;93(16):e1543-e1549. doi:10.1212/wnl.0000000000008313. PMID: 31534027. PMC6815205.