CANVAS Syndrome and RFC1-related Disease

Adult-onset sensory neuronopathy with progressive bilateral vestibular and cerebellar involvement and chronic cough, caused by biallelic intronic pentanucleotide expansions in RFC1 (predominantly AAGGG), with autosomal recessive inheritance.

Literature and regulatory information updated: 9 October 2026.

0. Initial Reference Table

1. Introduction

1.1. From clinical syndrome to molecular disease

The association between bilateral vestibulopathy and cerebellar ataxia was first reported in 1991 and delineated as a syndrome in 2004, according to the review by Szmulewicz and colleagues [18]. That year, Migliaccio and colleagues, in Sydney, described four patients with cerebellar ataxia with bilateral vestibulopathy syndrome (CABV), with onset in the 6th decade, still ambulant 8 to 20 years later, no family history and exclusion of SCA1, SCA2, SCA3, SCA6, SCA7 and Friedreich ataxia; three of the four had peripheral sensory neuropathy [3]. The most enduring contribution of that study was the validation, using a scleral search coil, of the clinical sign that still guides bedside examination: impairment of the visually enhanced vestibulo-ocular reflex (VVOR, visually enhanced vestibulo-ocular reflex, previously called the “doll's eyes” manoeuvre). As the frequency of head rotation increases from 0.1 to 1.0 Hz, eye velocity no longer follows head velocity and corrective saccades appear, because all three compensatory reflexes, the vestibulo-ocular reflex (VOR), the optokinetic reflex and smooth pursuit, are simultaneously impaired [3].

In 2011, a review of 27 patients showed that non-length-dependent sensory impairment, with absent sensory nerve action potentials (SNAP, sensory nerve action potentials), was an integral component rather than an incidental finding, leading to the name “cerebellar ataxia with neuropathy and bilateral vestibular areflexia syndrome”, abbreviated CANVAS [18]. Magnetic resonance imaging (MRI) showed cerebellar atrophy in 22 of the 27 patients, involving the anterior and dorsal vermis and hemispheric crus I; autopsy in one patient revealed loss of Purkinje cells and neurones in the vestibular, trigeminal and facial ganglia, with preservation of the spiral ganglion, and the presence of two sibling pairs suggested a late-onset recessive disease [18]. In 2014, a series of 80 patients added orthostatic hypotension, chronic cough and neuropathic pain to the triad, demonstrated that sensory impairment corresponds to a neuronopathy with marked loss of dorsal root ganglion neurones and identified falls prevention and dysphagia management as clinical priorities [19]. That same year, a New Zealand group described autonomic dysfunction as a major feature of the syndrome [20]. In 2016, clinical diagnostic criteria were proposed, organised into possible, probable, definite and pathologically definite categories, still without a genetic basis and described in section 5 [21].

1.2. Discovery of the gene

In April 2019, Cortese and colleagues (University College London, UCL, Queen Square Institute of Neurology, London) published the molecular cause [7]. Non-parametric linkage analysis under a recessive model identified a single peak at 4p14 (cumulative HLOD, heterogeneity logarithm of the odds, of 5.8) and a shared 1.7 Mb haplotype; exome sequencing in seven patients from three families found no coding variants, and genome sequencing in six patients revealed a biallelic intronic expansion of the AAGGG pentanucleotide in intron 2 of RFC1, confirmed by RP-PCR in all affected individuals from 11 families (23 familial cases) [7]. In the same cohort, the biallelic expansion explained 22% of 150 patients with sporadic late-onset ataxia, 63% of those with both cerebellar ataxia and sensory neuronopathy, and 92% of those with complete CANVAS, showing that the disease is not rare [7].

Independently and almost simultaneously, Rafehi and colleagues (Melbourne) identified the same locus in 35 individuals from 22 Australian families with a clinical diagnosis of CANVAS, combining linkage in five families (combined LOD of 7.04), negative exome sequencing in 27 affected individuals and bioinformatic tools for detecting short tandem repeats in genome data [22]. Before 2019, the only variant associated with CANVAS was a heterozygous missense variant in ELF2 in a family with three affected individuals, according to the same authors [22].

1.3. Nomenclature and reclassification

The discovery broadened the entity beyond the triad. GeneReviews adopted the name “RFC1-Related Disorders”, with a spectrum ranging from complete CANVAS to predominantly or exclusively sensory phenotypes, with or without vestibular and cerebellar involvement [1]. Clinical practice distinguishes complete CANVAS (sensory neuronopathy, bilateral vestibular hypofunction and cerebellar dysfunction); complex sensory ataxia with cerebellar or vestibular involvement; isolated sensory neuronopathy or neuropathy, with or without cough; ataxia with chronic cough; and forms with multisystem features, such as dysautonomia, parkinsonism, motor neurone signs or cognitive impairment [1, 23, 24]. Orphanet retains the clinical name (ORPHA:504476, with the synonyms CANVAS and “CABV syndrome”) and defines the disease as a slowly progressive, late-onset autosomal recessive syndromic cerebellar ataxia with bilateral vestibulopathy and axonal sensory neuropathy, with chronic cough and dysautonomia as variable features [2].

Two reclassifications warrant mention. The 2022 review by Davies and colleagues allowed for pure cerebellar ataxia within the spectrum [25], but a 2024 study found no expansions in any of 54 patients with idiopathic sporadic ataxia without sensory ganglionopathy, and GeneReviews now states that isolated cerebellar ataxia without sensory neuropathy is not suggestive of the disease [1, 26]. Conversely, Australian families previously described as having autosomal dominant hereditary sensory neuropathy type 1 with cough and reflux (HSN1B, linked to 3p22-p24) proved to have RFC1-related disease with pseudodominance, rendering that locus invalid [10]. GeneReviews states that there are no formal diagnostic criteria for RFC1-related disease; diagnosis rests on molecular demonstration of biallelic pathogenic variants in a patient with suggestive findings [1].

1.4. Scope of this article

This article addresses RFC1-related disease across its entire spectrum. GAA expansion ataxia in FGF14 (SCA27B), multiple system atrophy of cerebellar type (MSA-C, multiple system atrophy, cerebellar type), GGC expansion disease in NOTCH2NLC and acquired ganglionopathies are addressed only as differential diagnoses. Autosomal dominant spinocerebellar ataxias and early-onset recessive ataxias enter the differential diagnosis.

2. Molecular Genetics and Pathophysiology

2.1. The RFC1 gene and the replication factor C complex

RFC1 is located at 4p14 and encodes the largest of the five subunits (RFC1 to RFC5) of replication factor C, a ubiquitously expressed complex, including in the central and peripheral nervous systems [1]. The complex acts as a “clamp loader” (clamp loader): it opens the proliferating cell nuclear antigen (PCNA, proliferating cell nuclear antigen) ring and loads it onto DNA, where PCNA maintains high-affinity interaction between the polymerase and template during elongation; it thus participates in DNA replication and repair [1, 22]. The gene is intolerant of loss of function in gnomAD (pLI 0.97; observed/expected ratio 0.18, confidence interval, CI, 0.12–0.31) and shows substantial cerebellar expression in GTEx data [9, 22]. In a zebrafish model, rfc1 is expressed in neural progenitors of the developing cerebellum, where it maintains genomic integrity, and its loss of function causes a severe cerebellar phenotype through impaired Purkinje and granule cell neurogenesis; this is preclinical evidence [27]. Complete loss of RFC1 is incompatible with life in cellular and animal models, and parents heterozygous for a truncating variant remain phenotypically unaffected into old age, indicating that haploinsufficiency is tolerated [9].

2.2. The repeat locus

The expansion lies in the poly(A) tail of an AluSx3 element in intron 2, 2863 bp downstream of exon 2 and 2952 bp upstream of exon 3, where the reference sequence is (AAAAG)11 [7, 22]. Published coordinates are chr4:39350045–39350103 in GRCh37 [7] and chr4:39348425-39348479 in GRCh38, according to the American College of Medical Genetics and Genomics (ACMG) 2026 points-to-consider table [5]. In the MANE Select transcript NM_002913.5, the position corresponds to c.132+2923 [6]. As the repeat is intronic, translation through canonical initiation at an AUG codon is not expected [1].

The locus is highly polymorphic. In 608 chromosomes from 304 healthy British and European controls, the (AAAAG)11 allele accounted for 75.5%, the (AAAAG)exp expansion for 13.0% (15–200 repeats), the (AAAGG)exp expansion for 7.9% (40–1000 repeats, frequently interrupted) and (AAGGG)exp for 0.7% [7]. Eight controls had biallelic expansions of other motifs, including two in whom an AAGGG allele coexisted with an expanded AAAAG or AAAGG allele, showing that disease requires two pathogenic alleles rather than merely two expanded alleles [7]. Tables 7 and 8 of GeneReviews use transcript NM_001204747.1, without HGVS c. notation [1].

2.3. Allele classes and repeat motifs

GeneReviews distinguishes two classes of pathogenic variants: intronic pentanucleotide expansions, present biallelically in more than 99% of patients, and sequence variants or deletions, found in fewer than 1%, always in compound heterozygosity with an expansion [1]. Motif classification has evolved rapidly since 2019, and Table 2.1 summarises the current version.

Table 2.1. Repeat motifs at the RFC1 locus and their classification.

Size thresholds differ between sources. The 2026 GeneReviews text states that pathogenic AAGGG ranges from 250 to more than 2000 repeats and that AAAGG is usually benign between 30 and 110, whereas its Table 8 lists (AAAGG)40-1000 as having variable penetrance, pathogenic above 600, and (AAAAG)12-200 as normal [1]. The 2026 ACMG table defines the pathogenic AAGGG range as 400 to more than 2000 repeats and classifies AAAGG only as “variable penetrance” [5], whereas the 100,000 Genomes Project study described 250 repeats as the smallest known pathogenic AAGGG expansion and pathogenic AAAGG as always exceeding 600 repeats (979 ± 257 in patients versus 238 ± 142 in AAGGG/AAAGG controls; P < 0.0001) [8]. In the Southern blot series of 51 patients, biallelic expansions ranged from 433 to 2750 units, with a median of 989 in the smaller allele and 1802 in the larger [14]. In practice, an AAGGG allele below 400 repeats or an AAAGG allele between 500 and 600 repeats requires cautious interpretation and sizing by an appropriate method.

All pathogenic motifs share a 66 kb haplotype (chr4:39302305–39366034, hg38), and most share a 207 kb block spanning WDR19 and RFC1, with an estimated age of approximately 56,100 years (95% CI 27,680–115,580); reference and expanded AAAAG alleles derive from a different haplotype [8]. The Australian team estimated the age of the most recent common ancestor of the core 0.36 Mb haplotype at approximately 25,880 years (CI 14,080–48,020), probably European, within a linkage disequilibrium block present in Europeans and East Asians but not in Yoruba individuals [22]. All pathogenic motifs are guanine-rich and have a high predicted propensity to form G-quadruplexes, suggesting that size and GC content matter more than the exact motif [8].

2.4. Sequence variants in compound heterozygosity

Among 15 patients from 13 families with typical CANVAS and a single (AAGGG)n allele, exome or genome sequencing found a truncating RFC1 variant in trans in seven patients from five families (38% of families) [9]. In that group's experience, these cases account for approximately 1% (7 versus more than 650 carriers of biallelic expansions) [9]. Table 2.2 presents the variants using standardised nomenclature on the MANE transcript; the variant published as c.2191del corresponds, under the 3' alignment rule of the HGVS (Human Genome Variation Society), to NM_002913.5:c.2192del p.(Gly731GlufsTer6) [4].

Table 2.2. Pathogenic RFC1 sequence variants described in trans with an AAGGG expansion (NM_002913.5; GRCh38).

Phase was confirmed by parental segregation or messenger RNA sequencing, using the synonymous polymorphism rs2066782, which lies on the expansion haplotype. This polymorphism was published as c.2511T>C p.(Ser837=), numbering corresponding to transcript NM_001204747.2; on MANE NM_002913.5 it is c.2508T>C (GRCh38 g.39302305A>G) [4, 9]. Other reports of compound heterozygosity with nonsense, frameshift, splicing variants or deletions are included in the GeneReviews review [1, 32–35]. The authors drew an analogy with Friedreich ataxia, in which 1–4% of patients are compound heterozygotes for the GAA expansion and an intragenic variant [9].

2.5. Repeat nomenclature and database records

The ACMG recommends describing repeats according to HGVS rules for repeated sequences [5]. In ClinVar, the pathogenic expansion appears as NC_000004.12:g.39348425AAGGG[(400_2000)] (VCV000997970), the generic form as g.39348425AARRG[(400_2000)] (VCV000625839), the Māori configuration as g.39348425AAAGG[10_25]AAGGG[n] (VCV000986302), and there are pathogenic records for (ACAGG)n (VCV003370519) and for (AGGGC)n and (AAGGC)n (VCV004531160); the reference allele NM_002913.5(RFC1):c.132+2923= (VCV000996307) is benign [6]. The same records sometimes use transcript NM_001204747.1 (for example c.132+2923_2927AAGGG[(400_2000)]), and almost all have “no assertion criteria provided” [6]. Historical notation with the motif in parentheses, such as (AAGGG)exp, remains the most widely used in the clinical literature and is adopted in this article. In pharmacogenetic literature, “RFC1” has also denoted the reduced folate carrier (SLC19A1), requiring care in bibliographic searches.

2.6. Disease mechanism

The mechanism remains debated, and both loss-of-function and toxic gain-of-function hypotheses have been investigated [1]. In the original article, there was no reduction in messenger RNA or RFC1 protein in patient fibroblasts, lymphoblasts, muscle or brain, no AAGGG RNA foci were detected by fluorescence in situ hybridisation in post-mortem vermis, and RFC1 splicing was normal [7]. Recessive inheritance and compound heterozygosity with truncating variants favour loss of function: in these patients' fibroblasts, the truncating variant triggers nonsense-mediated messenger RNA decay, reducing the transcript and full-length RFC1 protein by approximately 40–50%, which is not observed in patients with biallelic AAGGG [9]. The same authors do not exclude the possibility that hemizygous expression of the repeat-containing allele acts through toxic gain of function of RNA or peptides [9].

The AAGGG motif adopts complex secondary structures, particularly parallel G-quadruplexes and triplex structures, which may interfere with RFC1 transcription or processing [1, 36–38]. RNA foci and repeat-associated peptides have been described in post-mortem brain and dorsal root ganglia, without demonstrated toxicity [1, 39, 40]. In neurones derived from patient induced pluripotent stem cells, expansions did not alter splicing, RFC1 expression or DNA repair; repeat peptides produced in reporter assays were not detected in these neurones, whose reduced synaptic connectivity was corrected by deletion of a single expanded allele but neither reproduced by silencing nor corrected by RFC1 replacement, suggesting a repeat-dependent, protein-independent mechanism (preclinical study) [39]. Partly contrasting with this, a November 2025 preprint, not yet peer reviewed, described tissue-specific reduction in the RFC1 transcript in post-mortem brain and neurones, impaired RFC1 function and increased sensitivity to platinum-induced DNA damage, corrected by excision of the repeat and adjacent AluSx3 [1, 41].

Neuropathology shows severe, diffuse Purkinje cell loss with Bergmann gliosis and a preserved granular layer, without p62 inclusions [7]. The peripheral lesion is a dorsal root ganglion neuronopathy, with degeneration of central (posterior column) and peripheral projections, also affecting the vestibular, trigeminal and facial ganglia [14, 18]. Nerve biopsies show severe chronic loss of large and small myelinated fibres, without active axonal degeneration, regeneration or demyelination (median density of 290 fibres/mm², versus a normal 4000–9000), with relatively preserved unmyelinated fibres [14].

2.7. Penetrance, modifiers and genotype–phenotype correlations

Penetrance of biallelic AAGGG expansions is incomplete. Genomics England and UK Biobank identified, respectively, 56 and 18 biallelic individuals with ataxia or neuropathy and 26 and 207 age-matched biallelic individuals without records of these diseases; a polygenic score constructed from Huntington disease age-at-onset modifiers and DNA repair genes was elevated in affected biallelic individuals (odds ratio, OR, of 1.43 in meta-analysis), with the greatest contribution from polymorphism rs245100 upstream of MSH3 (OR 2.35), and the result was replicated in an Australian cohort of 54 patients [42]. Disease records in biobanks may underestimate symptoms, so these figures do not equate to clinical penetrance. GeneReviews notes that the number of symptomatic individuals may be lower than predicted by genome data, particularly for AAGGG expansions with fewer than 250 repeats [1].

AAGGG expansion size has a mild inverse correlation with age at onset and is associated with greater severity, earlier loss of independent walking and earlier onset of dysarthria and dysphagia, according to the 392-patient series [1, 43]. Earlier studies with fewer patients found no correlation between size and age at onset (n = 34 in the original article; n = 51 in the 100-patient series) [7, 14]; in a French series of 26 patients, a longer smaller allele was associated with greater vestibular and cerebellar impairment [44]. The ACAGG motif is associated with more frequent lower motor neurone involvement (atrophy, cramps, fasciculations), with an OR of 17.9 for muscle atrophy in an individual-data meta-analysis [1, 28, 45, 46]. In compound heterozygosity with a truncating variant, reports suggest a more severe phenotype, with wheelchair dependence at 40–50 years and dysphagia requiring tube feeding, but GeneReviews considers the evidence inconclusive [1, 9]. Some patients with rare motifs (AGGGC, AAGGC) had dysautonomia with syncope or onset at 24 years with rapid progression [8].

2.8. Instability, anticipation and allelic disorders

Germline instability has limited clinical implications: size variation on paternal or maternal transmission is small (less than 10%) and size is relatively stable between siblings [1, 7, 43]. Anticipation is not observed [1, 11]. No other phenotype associated with germline RFC1 variants is known [1].

3. Clinical Spectrum and Manifestations by System

3.1. Overview and main cohorts

RFC1-related disease is an adult-onset neurodegenerative disorder beginning in sensory neurones and involving the vestibular system and cerebellum over subsequent years [1, 14]. Sensory neuropathy appears to be a constant feature, and chronic cough may precede any neurological manifestation by three decades [1]. All published series describe patients without disease-modifying therapy; the data therefore represent natural history under supportive care. The main cohorts differ in recruitment methods, explaining some of the differences in frequency between them (Table 3.1).

Table 3.1. Main molecularly confirmed clinical cohorts.

Table 3.2 reproduces the frequency ranges compiled by GeneReviews from four series published between 2022 and 2024.

Table 3.2. Frequency of selected features (minimum–maximum ranges from Huin 2022, Traschütz 2022, Currò 2024 and Dujardin 2024) [1, 43, 49–51].

3.2. Sensory neuronopathy

Sensory involvement affects all four limbs and may or may not follow a length-dependent pattern. The most common pattern is diffuse reduction in pain sensation, reduced vibration sense in the lower limbs and milder reduction in distal position sense [1]. In the 100-patient series, examined on average 10 years after onset, pinprick sensation was reduced in the face in 22, upper limbs in 55 and lower limbs in 74; the pattern was length-dependent in approximately three-quarters, and distal strength was preserved in all [14]. Symptoms include unsteadiness, sensory loss, paraesthesia, pain and cramps [1]. The feature most likely to mislead the clinician is preserved reflexes: ankle reflexes were absent in 49 of the 100 patients, but knee reflexes were normal in 47 and brisk in 28, and upper limb reflexes normal in 54 and brisk in 26, which the authors attribute to relative preservation of muscle spindle neurones [14]. Sensory conduction is abnormal in all patients studied (95/95), with absent lower limb SNAP in 75 of 90; motor conduction is normal or near-normal, with reduced lower limb motor amplitude in 10 of 93 [14]. The Portuguese series found neuropathy on nerve conduction studies in 15 of 15 patients, even though examination detected it in only 13 [52], and skin biopsies showed very severe epidermal denervation in all patients in the prospective subgroup of Malaquias and colleagues [47]. Nerve ultrasound shows reduced proximal and distal cross-sectional area in upper and lower limbs [1, 53, 54].

3.3. Gait ataxia and balance

Progressive imbalance is the presenting symptom in two-thirds of patients and is often worse in the dark because neuropathy and vestibular loss act together; upper limb coordination and manual dexterity are better preserved than gait [1]. In the European cohort of 70 patients, worsening ataxia without visual control was present in 98%, Romberg sign in 93%, and postural instability with retropulsion in 49% [23]. Gait ataxia began at a median age of 53 years (32–72), earlier in patients with chronic cough (50.3 ± 7.3 versus 61.7 ± 7.4 years; P < 0.001); the first falls occurred a median of 5 years after ataxia onset, in one-quarter before 3 years and sometimes 2 years before ataxia was recognised [23].

3.4. Bilateral vestibular hypofunction

Oscillopsia, particularly with head movements, occurs in approximately one-quarter of patients and may be the presenting manifestation [1]. Vertigo and hearing loss are not part of the disease, although they may coexist because they are common in the population [1]; the 2016 clinical criteria already excluded hearing loss and pyramidal signs from the components of CANVAS, attributing hearing loss, when present, mainly to presbyacusis or noise trauma [21]. The Hereditary Ataxia Overview still lists “severe dizziness or vertigo” as a clue to RFC1 [11]; the more recent dedicated GeneReviews entry states the opposite and is the one followed here [1]. Bilateral vestibular hypofunction was confirmed in 48 of 53 patients tested (91%), including three with a normal bedside head impulse test [14]. In a Swedish series, 47% had preserved caloric responses with abnormal angular VOR, a discordant pattern more common in young patients with mild ataxia, and otolith function was largely preserved [48]. Impaired VVOR indicates coexisting vestibular and cerebellar pathology [1, 3].

3.5. Cerebellar and oculomotor dysfunction

Cerebellar signs include gaze-evoked nystagmus, downbeat nystagmus (downbeat), saccadic pursuit, dysmetric saccades, limb ataxia and dysdiadochokinesia [1]. In the 100-patient series, oculomotor abnormalities were present in 65 (nystagmus in 55, saccadic pursuit in 42, dysmetric saccades in 41) and dysarthria in 40 [14]. Dysarthria and dysphagia occur late: in the European cohort, upper limb ataxia appeared a median of 6.5 years after gait ataxia onset, dysarthria at 8 and dysphagia at 10 [23]. Slow vertical (17%) or horizontal (19%) saccades may occur and are associated with brainstem atrophy [23].

3.6. Chronic cough

The cough is chronic, dry or minimally productive, spasmodic, lasts more than 8 weeks and may be the main presenting symptom; it is accompanied by laryngeal paraesthesia (tickling, irritation, globus), and chest examination is usually normal [1]. It is often diagnosed as asthma or gastro-oesophageal reflux, which may coexist [1]. In the European cohort, cough began at a median age of 35 years (16–69) and preceded ataxia in 80% of cases, by a median of 16 years and up to 36 years [23]. It precedes dysphagia by an average of 26.5 years, making aspiration an unlikely explanation; the authors hypothesise a lesion of peripheral cough reflex pathways through vagal sensory neuronopathy, associated with damage to central networks [23]. Vagus nerve ultrasound in eight patients showed reduced cross-sectional area (left 0.94 ± 0.22 versus 1.86 ± 0.13 mm² in controls), consistent with vagal hypotrophy [55]. The capsaicin cough threshold in patients with CANVAS and cough is similar to that of refractory cough without CANVAS and lower than that of healthy controls, pointing to cough reflex hypersensitivity [16]. Ten patients in the Portuguese series reported an inverse relationship between cough and ataxia severity [47]. Gastro-oesophageal reflux was reported by 31% of European patients, without a statistical association with cough [23].

3.7. Dysautonomia

Autonomic dysfunction includes postural hypotension, erectile dysfunction, chronic constipation or diarrhoea, urinary dysfunction, early satiety, nausea, anhidrosis or hyperhidrosis, and dry mouth and eyes; it is relatively common but rarely disabling [1]. In the European cohort, 62% had autonomic dysfunction (erectile dysfunction in 44% of men, urinary urgency or retention in 39%, postural hypotension in 23%, constipation in 24%, faecal incontinence in 11%), with a median onset 5 years after ataxia [23]. Autonomic testing confirms sympathetic or parasympathetic dysfunction in approximately half of patients undergoing it [1]; in the prospective Portuguese series, cardiovascular dysfunction affected 67%, cardiovagal dysfunction 54% and sudomotor dysfunction 50% [47]. A minority have severe dysautonomia, with symptomatic orthostatic hypotension resembling MSA and cough-induced syncope [1, 56]. In the Swedish series, dysautonomia reached 77% [48].

3.8. Motor involvement and cramps

Brisk reflexes are common; spasticity, Babinski sign, atrophy, fasciculations and weakness are uncommon [1]. The 2026 meta-analysis estimated upper motor neurone signs in 18% (95% CI 6–34%) and lower motor neurone signs in 11% (95% CI 2–24%), with more atrophy in men (OR 2.71) and ACAGG carriers [46]. In an American idiopathic neuropathy cohort, 27.8% of biallelic individuals had weakness on examination and 44.4% motor abnormalities on electromyography, so motor involvement should not exclude testing [57, 58]. Muscle cramps may be the first and only complaint for years, without weakness or atrophy; their prevalence in previous series was 18–42% and electrophysiological motor involvement 4–45% [59].

3.9. Parkinsonism and Parkinson disease

The frequency of extrapyramidal signs varies widely between series: 0–0.5% in some, including the Portuguese series, and 17.1–23.1% in others [1, 23, 43, 47, 60]. In the European cohort, 28% had bradykinesia (one responding to levodopa) and 19% had both dysautonomia and bradykinesia, overlapping with MSA-C [23]. Meta-analysis estimated parkinsonism in 8% (95% CI 1–18%) [46]. In a Brazilian series of 13 patients, dopamine transporter scintigraphy with 99mTc-TRODAT-1 was abnormal in 9, with the putamen more affected than the caudate; three had clinical parkinsonism, all with abnormal scans, and six had reduced uptake without parkinsonism [60]. In case reports and small series with dopamine transporter scintigraphy, totalling 19 individuals, dopaminergic denervation ranged from absent to severe [1, 61–64]. Among 135 patients with familial or early-onset Parkinson disease from Slovenia, Serbia and Croatia without a known monogenic cause, 2 (1.5%) had biallelic AAGGG [65], and the first autopsy of Parkinson disease associated with biallelic RFC1, in a Finnish patient, showed Lewy synucleinopathy (Braak III) without CANVAS neuropathological findings [66]. The causal relationship between the expansion and parkinsonism requires larger cohorts [1].

3.10. Cognition, mood and sleep

Cognitive deficits predominantly affect executive function, attention and working memory; in two systematically assessed series, most patients fulfilled criteria for probable or definite cerebellar cognitive affective syndrome (CCAS), and dementia is uncommon [1, 51, 67]. Meta-analysis estimated cognitive impairment in 31% (95% CI 13–53%), more likely with longer disease duration (OR 1.04 per year) [46]. In a Spanish series of 11 patients, executive dysfunction was followed by abnormalities of social cognition and episodic memory, with frequent depression, anxiety and irritability and isolated cerebellar hypometabolism on fluorodeoxyglucose positron emission tomography (PET) [68]. Polysomnography in 16 Italian patients showed obstructive sleep apnoea in 15 (94%; severe in 4, moderate in 9), restless legs syndrome in 12 (75%), periodic limb movements in 10 (63%) and REM (rapid eye movement) sleep behaviour disorder in one patient [69]. Sleep apnoea was reported in 14% of the European cohort and 6% of the Portuguese cohort, reflecting the difference between systematic investigation and clinical recording [23, 47].

3.11. Neuropathic pain

A meta-analysis of 14 studies with 445 patients estimated neuropathic pain prevalence at 43.2% (95% CI 34.0–52.9%), with high heterogeneity and no validated instruments in most studies [70]. In the 100-patient series, 42 reported neuropathic pain [14].

3.12. Imaging

Brain MRI shows cerebellar atrophy, predominantly of the vermis, in later stages; spinal MRI may show atrophy and T2 hyperintensity of the posterior columns [1]. In the 100-patient series, cerebellar atrophy was present in 57 of 91 (63%), spinal cord atrophy in 19 of 42 (45%) and posterior column abnormalities in 4 of 34 [14]; in the original article, cerebellar atrophy was seen in 35 of 42 (83%) [7]. In the European cohort, atrophy affected the vermis in 87% and hemispheres in 70% of patients with MRI, was mild to moderate in 89% and could remain absent after 20 years of disease; mild brainstem atrophy (13%) was associated with duration and dysphagia [23]. In a Brazilian group (Campinas), 3 T MRI repeated after 2 years showed progressive atrophy of the brainstem, thalamus, hippocampus and cervical spinal cord, with greater sensitivity to change than the SARA scale, and hypothalamic volume was reduced in 19 patients [71, 72].

3.13. Limited phenotypes and progression across systems

The proposed progression is from sensory neurones to the vestibular apparatus and cerebellum: in the 100-patient series, ataxia and sensory complaints began at a median age of 55 years, oscillopsia at 62, dysarthria at 61 and dysphagia at 64; after 10 years, two-thirds had all three systems involved, 16 had neuropathy with cerebellar involvement, 6 with vestibular involvement and 15 isolated sensory neuropathy, and none had an isolated cerebellar or vestibular form [14]. In the 392-patient series, half had complete clinical CANVAS after 10 years, 33% complex sensory ataxia and 14% sensory neuropathy, with or without cough, as the only clinically detectable manifestation [1, 43]. In neuropathy clinics, the picture differs: among patients with idiopathic axonal neuropathy and a biallelic expansion, 42% retained isolated sensory neuropathy after approximately 10 years, two-thirds had no gait or limb ataxia, and vestibular or cerebellar involvement was often subclinical [24]. Axonal sensory neuropathy has been described as the visible tip of a broader subclinical multisystem neurodegenerative process [50].

3.14. Special populations

No paediatric expression of the disease has been described; cough may begin in the 2nd decade, and the earliest neurological onset recorded in large series was at 19 years [1, 14]. The sources consulted do not describe disease evolution during pregnancy. In older people, overlap with common acquired causes (diabetes, monoclonal gammopathy, vitamin B12 deficiency) does not exclude the disease: 25% of biallelic individuals in a Spanish idiopathic axonal neuropathy series had coexisting controlled diabetes or monoclonal gammopathy of undetermined significance (MGUS) [73].

4. Clinical Recognition and Warning Signs

4.1. When to suspect the disease

GeneReviews recommends suspecting the disease in an adult with symptom onset after 25 years and one or more of the following findings, supported by electrophysiological, vestibular and imaging studies and family history: sensory neuropathy or neuronopathy; gait ataxia with positive Romberg sign and dysmetria exacerbated by eye closure, with normal muscle bulk, strength, tone and plantar responses; bilateral vestibulopathy with unsteadiness and oscillopsia; cerebellar dysfunction; refractory chronic cough; and mild dysautonomia [1]. The most useful combination for the clinician is unsteadiness worse in the dark, sensory loss with absent SNAP and preserved or brisk reflexes, bilaterally abnormal head impulse testing and spasmodic dry cough lasting years [3, 14, 23]. In the European cohort, chronic cough, visually dependent sensory ataxia, sensory neuropathy symptoms and axonal sensory neuropathy had a positive predictive value exceeding 90% for the expansion [23]. GeneReviews adds that, in ataxia beginning in middle age or later, asking about chronic cough and requesting nerve conduction studies is highly predictive, and absence of sensory neuropathy on conduction studies makes the disease unlikely; this guidance is based on one author's personal observation [1]. A Swedish series described a symptomatic patient without neuropathy on nerve conduction studies, so the rule is not absolute [48].

Family history is typically absent or suggests recessive inheritance (affected siblings, rarely consanguinity) [1]. Transmission over two generations does not exclude the disease, because carrier frequency permits pseudodominance [1, 10]. Testing should be considered regardless of family history [74].

4.2. Clinical presentation settings

Presentation depends on the patient's point of entry into the healthcare system, and each specialty has its own clues (Table 4.1).

Table 4.1. Presentation settings and clues for requesting testing.

In the 100-patient series, median age at neurological onset was 52 years and age at diagnosis approximately 72, a delay of approximately 20 years, and only 23 had initially been suspected of having CANVAS; 32 had received another diagnosis (spinocerebellar ataxia in 16, mitochondrial disease in 6, MSA-C in 3) [14, 31]. Previous negative investigations were the rule: Friedreich ataxia in 56, spinocerebellar ataxias in 54 and mitochondrial DNA in 37 [14].

4.3. Warning signs and emergencies

The disease progresses slowly and has no specific acute crises, but some situations require prompt action. Dysphagia with frequent choking indicates a risk of aspiration and malnutrition; it was among the causes of the two premature disease-related deaths in the European cohort (severe dysphagia with cachexia, cough and immobility) [1, 23]. Cough-induced syncope and symptomatic orthostatic hypotension, resembling those in MSA, occur in a minority and are associated with falls [1, 56]. Falls occur early and may precede recognition of ataxia; in the rehabilitation case by Harrell and colleagues, a fall with a wrist fracture and two falls in the preceding year were part of the history [23, 79]. Rapid deterioration, an acute exacerbation or very accelerated progression (phases of up to 5.5 SARA points per year were described between 5 and 15 years of disease) warrant earlier reassessment and investigation for overlapping causes [1, 23]. Acute-onset neuropathy, marked asymmetry or a clear response to immunotherapy suggest another or an overlapping cause.

4.4. Newborn screening

There is no newborn screening for RFC1-related disease, which manifests in adulthood and has no proven preventive intervention [1].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Criteria: from clinical to molecular diagnosis

There are no formal diagnostic criteria for RFC1-related disease; diagnosis is established in a proband with suggestive findings by identifying biallelic pathogenic RFC1 variants [1]. The clinical criteria of Szmulewicz and colleagues (2016), predating gene discovery, remain useful for documenting each component and understanding the limits of clinical diagnosis [21]. The authors emphasise that each of the three cardinal components can independently cause ataxia, so each must be demonstrated independently of the other two: gait ataxia and Romberg sign are non-specific, and the cerebellar component can only be inferred from cerebellum-specific signs (saccadic pursuit, gaze-evoked or direction-changing nystagmus, dysmetric saccades, rebound nystagmus, cerebellar dysarthria, scanning speech, truncal ataxia) [21]. There is no fixed sequence in the appearance of the three components, and a patient may have only two for many years [21]. Coexistence of acquired bilateral vestibulopathy, for example from gentamicin, with cerebellar dysfunction of another cause precludes the diagnosis [21].

Table 5.1. Clinical diagnostic categories of CANVAS proposed in 2016, according to the article text [21].

The article's original table is an image; Table 5.1 reconstructs requirements from the text, which does not detail all combinations of components. The article recommends excluding SCA1, SCA2, SCA3, SCA6 and SCA7 and investigating Friedreich ataxia in all suspected cases, a recommendation now incorporated into the algorithm in section 5.6 [21]. Clinical criteria have known limitations: complete CANVAS is present in only 46–63% of patients with the expansion, depending on the series, and limited phenotypes fall outside them [7, 14, 43]. Conversely, approximately 8–10% of patients with complete CANVAS do not have the biallelic AAGGG expansion [14, 23].

5.2. Demonstrating bilateral vestibulopathy

The 2017 Bárány Society criteria define bilateral vestibulopathy (BVP, bilateral vestibulopathy) and are the standard for documenting the vestibular component (Table 5.2) [80].

Table 5.2. Bárány Society criteria for bilateral vestibulopathy [80].

Bedside testing reliably detects only gains below 0.4, and interpretation is difficult in patients with cerebellar disease; associated sensory neuropathy exacerbates unsteadiness [80]. In CANVAS, bedside head impulse testing was bilaterally abnormal in 87% of patients tested, but vestibular hypofunction was demonstrated instrumentally in 91%, including patients with normal bedside testing [14]. vHIT also has discriminative value: in a Johns Hopkins series, bilateral vestibulopathy (gain <0.6) was present in 80% of patients with CANVAS, 18% with SCA27B and none with idiopathic late-onset cerebellar ataxia (ILOCA); mean horizontal gain was 0.22, 0.85 and 1.11, respectively, and a threshold of 0.50 distinguished CANVAS from SCA27B with 91% specificity and 70% sensitivity [77]. The Swedish series showed that caloric testing may be preserved with pathological angular VOR in nearly half of patients, so vHIT and caloric testing are complementary [48].

5.3. Other investigations

Nerve conduction studies are the highest-yield screening investigation: SNAP are reduced or absent, motor conduction is usually normal, the blink reflex may be abnormal (trigeminal neuronopathy), the H reflex is preserved in a minority and electromyography is usually normal [1, 81]. Once gait ataxia is established, SNAP are often absent in all nerves [31]. In an American idiopathic neuropathy cohort, factors associated with biallelic expansion were abnormal upper limb vibration sense (OR 4.65), sensory-only polyneuropathy on electromyography (OR 7.17) and sensory neuronopathy according to the Camdessanché criteria (OR 3.64), with an area under the curve of 0.80–0.81 for the model [57]. Brain and spinal MRI, nerve ultrasound (reduced cross-sectional area) and autonomic testing (blood pressure response to postural change and to handgrip and sympathetic skin response, for sympathetic function; heart rate variation with the Valsalva manoeuvre, deep breathing and standing, for parasympathetic function) support but do not confirm the diagnosis [1, 31]. Vagus nerve ultrasound and quantitative MRI are investigational biomarkers [55, 71].

In respiratory medicine, cough investigations to exclude treatable causes include spirometry, fractional exhaled nitric oxide (FeNO), full blood count with eosinophils and chest radiography, with a sleep study if apnoea is suspected [1]. A Spanish pilot combining standardised neurological assessment and screening nerve conduction studies in a single chronic cough clinic found marked bilateral reductions in radial and sural SNAP amplitudes in all patients with biallelic expansion, and the authors suggest RFC1 testing in selected patients with unexplained cough, particularly if radial SNAP is reduced [76].

5.4. Molecular testing: methods

The recommended strategy combines targeted repeat analysis to characterise motif and repeat number with gene sequencing when necessary [1]. Table 5.3 summarises the methods and their limitations.

Table 5.3. Methods for detecting RFC1 expansion and limitations.

Table 1 of GeneReviews attributes more than 99% of pathogenic variants to targeted expansion analysis and fewer than 1% to sequencing [1]. This proportion is not clinical sensitivity: it indicates which allele type explains already diagnosed patients, rather than the probability of a test detecting disease in a patient with the phenotype. With classical laboratory diagnosis (positive AAGGG RP-PCR and absence of flanking PCR product), Southern blot confirmed 34 of 34 cases in the original article and 18 of 18 PCR-positive cases in the European cohort, without false positives among 16 PCR-negative individuals or carriers [7, 23]. Accurate sizing is usually unnecessary for the AAGGG motif, but is recommended when there are biallelic expansions with at least one allele of uncertain significance or variable-penetrance size (AAAGG) [1].

Gene sequence testing (exome, genome or a panel including RFC1) is not routinely recommended, but should be performed when only one pathogenic expansion is identified in a patient with a suggestive phenotype: among 13 families in this situation, a truncating variant in trans was found in 5 [1, 9]. In AAGGG/AAAGG compound heterozygotes, the authors of the 100,000 Genomes Project study recommend assessing AAAGG allele size (pathogenic above 500–600 repeats) and AAGGG interruptions, and propose Southern blot, optical mapping or long-read sequencing in patients with a suggestive phenotype and inconclusive screening, for example absence of flanking PCR product with negative AAGGG RP-PCR [8].

5.5. Diagnostic yield by phenotype

Table 5.4. Proportion of patients with biallelic expansion according to phenotype and cohort.

These figures reflect selected cohorts and cannot be directly transferred to another population. Differences between Europe and the Americas or East Asia depend on carrier frequency, recruitment criteria and consanguinity [1, 23]. In all series, monoallelic carriers without a second variant do not explain the phenotype: in the American cohort, 9.1% of patients and 7.2% of controls were monoallelic (non-significant difference), with no second mutation found [58].

5.6. Diagnostic algorithm

The ERN-RND adult ataxia flowchart (revised September 2024) begins by excluding acquired causes when onset is acute or subacute or when history or MRI suggests them (autoimmune, toxic, traumatic, hypoxic, tumour-related, vascular, infectious, deficiency-related and paraneoplastic), and checking for peripheral neuropathy or sensory neuronopathy [17]. It then organises testing by family pattern: in a recessive pattern, test first for Friedreich ataxia and CANVAS/RFC1; in sporadic ataxia, test for Friedreich ataxia, SCA1, SCA2, SCA3, SCA6, SCA7 and SCA17, dentatorubral-pallidoluysian atrophy (DRPLA), CANVAS/RFC1, SCA27B/FGF14 and, if tremor-ataxia begins after 50 years, fragile X-associated tremor/ataxia syndrome (FXTAS), considering MSA-C; in a dominant pattern, the common SCAs, DRPLA and SCA27B [17]. If these tests are negative, the next steps are panel, exome or genome sequencing, biochemistry (alpha-fetoprotein, vitamin E, albumin, cholesterol, lactate, caeruloplasmin, phytanic acid, very-long-chain fatty acids, creatine kinase, organic acids, amino acids, lysosomal enzymes, cholestanol and oxysterols) and investigation of warning signs on MRI and ophthalmological examination [17]. The 2014 EFNS/ENS (European Federation of Neurological Societies/European Neurological Society) consensus, predating gene discovery, recommended considering MSA in sporadic cases beginning after 30 years and testing for the FMR1 premutation after 45 [88].

In practice, for an adult with suspected RFC1-related disease, the sequence is as follows. First, document sensory neuropathy through nerve conduction studies and the vestibular component through vHIT or caloric testing, and actively ask about cough. Second, request flanking PCR and RP-PCR for AAGGG, AAAAG and AAAGG. Third, if the result is biallelic AAGGG, the diagnosis is established; if there is only one AAGGG expansion, request RFC1 sequencing and RP-PCR for alternative motifs; if screening is inconclusive (no flanking product and negative AAGGG RP-PCR) or an AAAGG allele is present, request Southern blot or long-read sequencing. Fourth, if testing is negative with a typical phenotype, consider SCA3, late-onset Friedreich ataxia, SCA27B, SCA4, RNF170, mitochondrial diseases and genome sequencing: in the Australian series, three of four clinical CANVAS families without the expansion received another diagnosis (SCA3, autosomal recessive spastic ataxia of Charlevoix-Saguenay, ARSACS, and probable SCA45) [1, 22].

5.7. Interpretation and report content

Classification follows ACMG/AMP (American College of Medical Genetics and Genomics / Association for Molecular Pathology) rules, under which “pathogenic” and “likely pathogenic” have the same clinical value; biallelic variants of uncertain significance, or a pathogenic variant with a variant of uncertain significance, neither establish nor exclude the diagnosis [1, 89]. For expansions detected by short-read sequencing, the ACMG recommends PCR-free genome sequencing for non-coding expansions, confirmation by an orthogonal method (RP-PCR, Southern blot or long-read sequencing) and, if this is not performed in the laboratory, an explicit recommendation for confirmation; the exact repeat number should only be reported when confirmed or supported by reads spanning the repeat, and otherwise with a confidence interval [5]. The report should indicate software version, interrupted motifs of uncertain significance and HGVS notation for repeated sequences, and should not predict symptoms or age at onset from size [5].

The 2026 ERN-RND/EMQN (European Molecular Genetics Quality Network) consensus uses CANVAS as an example: in a 59-year-old patient with suspected CANVAS and negative exome sequencing in whom the expansion had not been assessed and the report gave no guidance, the report should have indicated targeted testing (for example RP-PCR) in its main conclusions; laboratories should state the limitations of next-generation sequencing (NGS, next-generation sequencing) for expansion disorders, particularly when CANVAS is suspected, and interpret expansions using repeat number, interruptions and locus-specific rules [90]. A complete report specifies the gene, transcript (NM_002913.5), motif of each allele, size when determined and method, zygosity, classification and method limitations. Segregation in parents or siblings helps establish phase when an expansion and a sequence variant are present [9].

A negative test does not exclude the disease when only AAGGG was investigated, only exome or panel sequencing was used, or RP-PCR was not supplemented with sizing of an AAAGG allele [1, 8]. A family may also have one affected individual with biallelic expansion and another with a phenocopy: in the European cohort, a sister of affected twins, with only one expansion, had neuropathy, vestibulopathy and mild ataxia attributed to previous chemotherapy (epirubicin, docetaxel, cyclophosphamide) [23]. There is no ClinGen actionability assessment for RFC1 [31].

5.8. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing are technically possible once biallelic variants have been identified in an affected relative [1]. As this is an adult-onset disease with incomplete penetrance, the ACMG emphasises the need for extensive counselling [5]. Details are provided in section 8.

6. Differential Diagnosis

6.1. Genetic differential diagnosis

Multisystem, asynchronous involvement creates a broad genetic differential: hereditary ataxias, hereditary neuropathies and mitochondrial diseases with ataxia, neuropathy and sometimes bilateral vestibular areflexia [1]. Table 6.1 follows the order of relevance in GeneReviews and adds differential diagnoses described in the other sources.

Table 6.1. Genetic differential diagnosis of RFC1-related disease.

GeneReviews supports the original table with several sources comparing these disorders [1, 97]. SCA27B and MSA-C are the two main differential diagnoses of late-onset cerebellar ataxia [1, 97]. Sensory neuronopathy with bilateral vestibular areflexia and chronic spasmodic cough strongly favour RFC1 over a common SCA [1].

6.2. Multiple system atrophy of cerebellar type

Distinguishing MSA-C has important prognostic implications. Mean survival in MSA is 9.3 years, whereas RFC1-related disease progresses slowly and does not evidently reduce life expectancy [1, 98]. Sensory neuropathy and vestibular dysfunction are not part of MSA, dysautonomia is common and highly disabling in MSA and usually mild in RFC1, and REM sleep behaviour disorder, parkinsonism and MRI showing putaminal, pontine and middle cerebellar peduncle atrophy or the “hot cross bun” sign (hot cross bun) favour MSA [1, 99, 100]. In a French prospective cohort of 418 sporadic late-onset cerebellar ataxias, the 20 RFC1 patients had earlier onset, later appearance of parkinsonism, dysautonomia and pyramidal syndrome, slower progression and median survival exceeding 20 years, compared with MSA-C (65 patients); sensory impairment and chronic cough were discriminative features [101]. Overlap exists: some RFC1 patients with dysautonomia and bradykinesia fulfil possible MSA criteria, although RFC1 does not cause pathologically confirmed MSA [23, 100]. The 2022 Movement Disorder Society criteria distinguish neuropathologically established, clinically established (with MRI markers), clinically probable and possible prodromal MSA [100]. A Chinese study associated biallelic expansion with MSA [85], but the expansion was absent in pathologically confirmed MSA patients [102] and in another Chinese late-onset ataxia and MSA series [103]; GeneReviews cites these studies when identifying MSA as a principal differential diagnosis [1].

6.3. Acquired sensory neuronopathies

Acquired causes of sensory neuronopathy include paraneoplastic disease, Sjögren syndrome, acute and chronic immune-mediated neuropathies (including anti-MAG neuropathy), metabolic causes (diabetes, vitamin B12 deficiency, copper deficiency) and toxic causes (alcohol, pyridoxine intoxication, platinum derivatives) [1, 104, 105]. Coexistence is common and misleading: of 71 patients with CANVAS-RFC1 and sensory ganglionopathy, 6 women had a diagnosis of primary Sjögren syndrome, all with sicca syndrome, but only 4 fulfilled the 2016 ACR-EULAR (American College of Rheumatology/European League Against Rheumatism) criteria; four were treated with immunosuppression without response [78]. In an idiopathic axonal neuropathy series, three RFC1 patients had received immunotherapy for diagnoses of Sjögren syndrome, sensory CIDP or paraneoplastic neuropathy [24], and 15% of 27 patients with ataxia and ganglionopathy attributed to coeliac disease or gluten sensitivity had the expansion [26]. The recommendation is to test RFC1 in ataxia or sensory neuropathy attributed to acquired causes when phenotype, evolution or treatment response is atypical [1]. Paraneoplastic syndrome does not appear more frequent: cancer occurred in 12% of RFC1 patients and in 14% of 100 ataxia controls without the expansion [14]. Antigliadin antibodies are positive in 10–15% of patients with ataxia, with uncertain causality [96].

6.4. Bilateral vestibulopathy from other causes

In the 255-patient bilateral vestibulopathy series cited in the Bárány Society criteria, an aetiology was identified in approximately 30%: ototoxicity in 13%, bilateral Ménière disease in 7% and meningitis in 5% [80]. Ototoxicity includes gentamicin and other aminoglycosides, chemotherapy, loop diuretics, very high-dose salicylates and immune checkpoint inhibitors [1, 80]. Other causes are bilateral vestibular schwannomas in NF2-related schwannomatosis, meningeal carcinomatosis and systemic autoimmune and inflammatory diseases [1, 80]. Bilateral vestibular hypofunction also occurs in SCA3 (and SCA1, SCA2 and SCA6), Friedreich ataxia, Gaucher disease and Charcot-Marie-Tooth disease, and the anatomical proximity of vestibular and cerebellar nuclei makes the two components difficult to separate [1]. The 2016 criteria give the example of a patient with posterior fossa decompression for Chiari malformation, sepsis treated with gentamicin and bilateral peripheral vestibulopathy, who should not be classified as CANVAS [21].

6.5. Other causes of progressive cerebellar dysfunction

Multifactorial and acquired cerebellar diseases are 4–5 times more common than hereditary ones, and up to 40% of adults with late-onset cerebellar ataxia without family history remain without a genetic cause despite comprehensive assessment [11]. Toxic, nutritional, vascular, neoplastic, inflammatory and paraneoplastic causes should be considered [1]. Among treatable adult ataxias, ataxia with vitamin E deficiency, cerebrotendinous xanthomatosis and Refsum disease have specific biochemical markers and are included in the European algorithm [17, 96].

7. Treatments and Risk Management Strategies

7.1. Principles and evidence base

There is no FDA- or EMA-approved disease-modifying treatment for RFC1-related disease, and the ClinicalTrials.gov search on 9 October 2026 identified no disease-modifying drug trials [15, 106]. Nor are there published disease-specific clinical practice guidelines: GeneReviews recommendations are based on the authors' personal experience, that is, expert opinion [1]. Treatment is supportive and directed at each component, aiming to maximise function and reduce complications, through a multidisciplinary team including neurology, physical and rehabilitation medicine, physiotherapy, occupational therapy and, as needed, speech and language therapy, respiratory physiotherapy, nutrition, gastroenterology, respiratory medicine and otorhinolaryngology [1, 96]. Rehabilitation recommendations draw on the general degenerative ataxia literature rather than disease-specific studies [31]. For problems shared with other conditions, such as refractory chronic cough and neuropathic pain, the corresponding general guidelines apply, all off-label for this disease.

7.2. Initial assessment after diagnosis

Table 7.1 reproduces the assessments recommended by GeneReviews to establish disease extent and the patient's needs.

Table 7.1. Recommended assessments after diagnosis [1].

Given the high frequency of obstructive sleep apnoea in a series with systematic polysomnography (15 of 16 patients), active enquiry about sleep disorder symptoms is justified, although GeneReviews recommends it only on clinical suspicion [1, 69].

7.3. Rehabilitation of ataxia, balance and vestibular function

Rehabilitation is the foundation of treatment [1, 15]. GeneReviews recommends neurorehabilitation, physiotherapy with balance exercises, gait training and strengthening, occupational therapy for activities of daily living, mobility aids (walking sticks, walking frames, ramps for powered wheelchairs), feeding aids (weighted cutlery) and dressing aids (hooks), home adaptations to prevent falls (grab rails, raised toilet seats), exercise within individual capacity and weight management, and considers vestibular rehabilitation [1, 108]. The 2014 EFNS/ENS consensus assigned class II evidence to physiotherapy and class III to occupational therapy in chronic ataxias [88]. A 2018 AAN (American Academy of Neurology) review, since retired by the society and therefore without value as a current recommendation, considered that 4 weeks of inpatient rehabilitation probably improved ataxia and function in degenerative ataxias [109]. A 2025 review reports that 4 weeks of intensive coordinative training improved SARA, gait and dexterity in degenerative ataxias, with partial benefit at 3 months [15].

Vestibular rehabilitation does not restore VOR but may improve function. In a 65-year-old man with confirmed CANVAS, a programme of 7 sessions over 4 months, with gaze stabilisation exercises, balance in low light, reduced base of support and gait training, monitored by an app, increased gait speed from 1.02 to 1.13 m/s and Functional Gait Assessment (FGA) from 20 to 27 out of 30, without a change in VOR gain or Dizziness Handicap Inventory, and with a loss of 4 lines in dynamic visual acuity [79]. This is single-case evidence. The patient association Ataxia UK, in a leaflet written with two of the gene's discoverers, recommends physiotherapy, speech and language therapy, occupational therapy for aids and home adaptation, and regular neurological follow-up [110].

Non-invasive neuromodulation is under investigation. Repetitive cerebellar transcranial magnetic stimulation (rTMS) improved SARA by 1.56 points versus sham in a meta-analysis of 8 trials with 237 patients, predominantly with SCA3 [15], and the AAN review reported possible benefit at 21 days [11, 109]. Two registered studies of cerebellar transcranial alternating current stimulation (tACS) and cerebello-spinal direct current stimulation (tDCS) in ataxias including CANVAS ended in 2021 and 2022 without registry results (NCT04153110 and NCT05621200) [106]. None of these data is specific to RFC1.

7.4. Chronic cough

Cough is often the most troublesome and longest-standing symptom. Investigation should exclude and treat coexisting causes, such as asthma, gastro-oesophageal reflux or ineffective oesophageal motility and rhinitis, and management involves respiratory medicine, otorhinolaryngology and gastroenterology [1]. GeneReviews proposes considering a cough neuromodulator (gabapentin, amitriptyline, pregabalin), a trial of low-dose morphine, speech and language therapy or physiotherapy for cough control, and a proton pump inhibitor if reflux is present [1]. All these drugs are used off-label.

General chronic cough guidelines differ. The 2020 ERS (European Respiratory Society) guideline recommends a trial of low-dose prolonged-release (PR) morphine, 5–10 mg twice daily, in adults with refractory chronic cough (strong recommendation, moderate evidence), based on a randomised trial in which severity decreased by 1.96 points on a 0 to 9 scale and the Leicester Cough Questionnaire (LCQ) increased by 2 points; approximately half of patients respond, usually within a week, and the drug should be stopped if there is no response within 1–2 weeks; codeine is not recommended because of CYP2D6 variability [111]. The same guideline suggests a trial of gabapentin or pregabalin (conditional recommendation, low evidence): in the gabapentin trial, at a maximum tolerated dose of 1800 mg/day, the LCQ improved by 1.8 points and cough frequency decreased by 27%; in the trial of pregabalin 300 mg/day combined with speech and language therapy, in adults with refractory chronic cough, the LCQ improved by 3.5 points, without a reduction in frequency; the risk of withdrawal due to adverse effects was 2.3 times that of placebo [111]. The ERS also suggests cough control therapy delivered by physiotherapy or speech and language therapy (conditional, moderate evidence), does not recommend routine chest computed tomography (CT) when radiography and examination are normal or routine antacids without reflux symptoms, and advises sequential trials of one drug at a time, discontinuing in the absence of response and, if effective, continuing for several months followed by withdrawal to assess remission [111]. The 2016 CHEST (American College of Chest Physicians) guideline, based on 11 trials with 570 participants, suggests gabapentin and multimodal speech and language therapy, but does not support morphine, although both improve cough-related quality of life, and considers inhaled corticosteroids and esomeprazole ineffective in the absence of asthma or acid reflux [112]. Neuromodulators are not used in children [111].

Disease-specific data come from an uncontrolled prospective French series, published in 2026, of 17 patients with unexplained refractory chronic cough and biallelic expansion, whose cough had lasted 236 ± 128 months [16]. Previous pregabalin or gabapentin did not alter cough in 4 of 4 patients, and amitriptyline had no effect in 3 of 8, produced a partial response with disabling residual cough in 4 and a good response in 1. With oral prolonged-release morphine 10 mg twice daily or equivalent for 6 months, 4 patients (23.5%) stopped coughing, 11 (64.7%) had a good response with non-disabling residual cough, 1 had a partial response and 1 discontinued because of nausea [16]. Stated limitations are absence of a control group, subjective assessment using a non-validated scale, and failure to repeat the visual analogue scale and LCQ at 6 months [16]. The authors attribute the effect to morphine action on central and peripheral µ receptors at doses below analgesic doses [16]. A 2025 review describes treating cough as refractory cough with low-dose morphine or gabapentin or pregabalin, and states that baclofen may help, although experts consider other drugs more effective; the evidence is anecdotal [15].

Morphine requires particular care in this population, in whom dysphagia, falls and obstructive sleep apnoea are frequent [23, 69]. The American prolonged-release morphine label contraindicates it in significant respiratory depression, acute or severe asthma without monitoring, monoamine oxidase inhibitor (MAOI) use within the preceding 14 days and gastrointestinal obstruction, and warns of dependence, abuse, overdose and death at any dose [113]. The ERS guideline lists constipation and drowsiness as common adverse effects [111].

A randomised placebo-controlled crossover trial of pregabalin for CANVAS-associated cough was registered in New Zealand in December 2023 (ACTRN12623001356651; 75 mg twice daily for 12 weeks per phase, adjusted between 75 and 225 mg/day; 18 planned patients; primary outcome cough frequency measured by a mobile phone app) [114]. The registry status remains “not yet recruiting”, and no published results were found [114].

7.5. Neuropathic pain

Neuropathic pain affects approximately 43% of patients, but GeneReviews considers treatment rarely necessary and primarily recommends advice to avoid injuries in limbs with sensory loss [1, 70]. There are no disease-specific trials. The 2025 systematic review and meta-analysis by NeuPSIG (Neuropathic Pain Special Interest Group of the IASP), updating the 2015 recommendations, proposes as first-line treatment, with a strong recommendation and moderate certainty, tricyclic antidepressants (number needed to treat, NNT, of 4.6), α2δ ligands, gabapentin and pregabalin (NNT 8.9), and serotonin and noradrenaline reuptake inhibitors (SNRIs; NNT 7.4); as second line, 8% capsaicin patches and 5% lidocaine patches; and as third line, botulinum toxin A, rTMS and opioids [115]. When cough and pain coexist, an α2δ ligand may serve both purposes, although in the French series its efficacy against cough was nil in 4 patients [16, 111]. In the European Union, pregabalin is indicated for peripheral and central neuropathic pain in adults, encompassing pain in this disease; in the United States, pregabalin and gabapentin indications are aetiology-specific and do not include it [116–118].

7.6. Ataxia, nystagmus and oscillopsia

4-Aminopyridine (4-AP; fampridine, or dalfampridine in American nomenclature), useful in episodic ataxia type 2 and SCA27B and used for downbeat nystagmus, may be considered according to GeneReviews, but there are no data supporting its use in RFC1-related disease [1, 11]. A retrospective series from 4 university centres in Germany, Switzerland and France, published in 2026, described 30 patients treated off-label with 4-aminopyridine, acetyl-DL-leucine (ADLL) or both [119]. The daily 4-aminopyridine dose was usually 20 mg (5–30 mg), and acetyl-DL-leucine 5000 mg (500 mg to 4000–8000 mg) [119]. Subjective improvement occurred in 3 of 22 patients on 4-aminopyridine (13.6%) and 3 of 15 on acetyl-DL-leucine (20.0%), objective gait improvement in 2 of 13 and 2 of 10, no effect on downbeat nystagmus, and one improvement was not maintained at 8 and 11 weeks [119]. Three patients had participated in the randomised crossover ALCAT trial of acetyl-DL-leucine versus placebo for 6 weeks, which was negative [119]. Adverse effects were non-serious: 43.5% with 4-aminopyridine (mainly worsening dizziness, agitation, nausea) and 16.7% with acetyl-DL-leucine; treatments were discontinued mainly for lack of benefit [119]. The authors conclude that responses are rare and limited, unlike in SCA27B [119].

Riluzole 100 mg/day (no maximum dose defined in the source) is, according to the Hereditary Ataxia Overview, the only drug with demonstrated symptomatic improvement in ataxias of mixed aetiology, and requires liver enzyme monitoring [11]; the EFNS/ENS consensus assigned it level B and amantadine level C, regardless of aetiology [88]. There are no data on riluzole in RFC1-related disease, and a 2025 review reports perceived benefit in only a minority of patients with hereditary ataxias [15]. Baclofen may be used for spontaneous nystagmus and spasticity, with fatigue as a limiting factor, and levodopa for overlapping parkinsonism; one of the 70 patients in the European cohort responded to levodopa [15, 23]. Spasticity, when present, may be treated with baclofen, tizanidine or dantrolene, according to general hereditary ataxia guidance [11]. Omaveloxolone, approved by the FDA in 2023 for Friedreich ataxia from 16 years, has neither an indication nor data in this disease [15]. A review recommends, for any off-label drug in adult ataxias, a therapeutic trial of at least 3 months and gradual withdrawal if there is no benefit [15].

7.7. Dysphagia, dysarthria, dysautonomia, cognition and sleep

Dysarthria is treated with speech and language therapy and, rarely, alternative communication methods (writing pads, digital devices) [1]. In dysphagia, modifying food consistency reduces aspiration, and a video-oesophagram helps define appropriate consistency [1]. Severe dysphagia may require tube feeding, described in a compound heterozygote [9]. For dysautonomia, GeneReviews recommends considering treatment of erectile dysfunction, urinary incontinence or retention, constipation or diarrhoea, and dry eyes and mouth, without specifying drugs [1]. Cognitive dysfunction warrants assessment with the CCAS scale, and psychiatric symptoms standard treatment; cognitive behavioural therapy, including Goal Management Training, is proposed for hereditary ataxias generally [1, 11]. The sources consulted do not describe studies treating sleep apnoea or restless legs syndrome in this disease. Social work should coordinate appointments, equipment and medication, and the need for palliative care or home nursing should be reassessed over time [1].

7.8. Doses of symptomatic drugs

No drug has an approved indication for RFC1-related disease; all uses in Table 7.2 are off-label, and doses derive from the drug label, the guideline recommending it for the symptom generally or the published disease-specific series, as indicated in each row.

Table 7.2. Symptomatic drugs used in RFC1-related disease (adults).

7.9. Comparative regulatory status

Table 7.3. Regulatory status of drugs used off-label, USA and European Union.

The sources consulted did not include INFARMED (Portuguese National Authority of Medicines and Health Products) or ANVISA (Brazilian Health Regulatory Agency) data on access to these drugs in Portugal or Brazil for this disease.

7.10. Interventional studies and registered studies

Table 7.4. Interventional studies in RFC1-related disease and general studies supporting recommendations.

These data have limitations preventing firm conclusions. Both disease-specific series lack a control group, use non-standardised doses and subjective assessments, and cough and ataxia fluctuate, favouring regression to the mean and placebo effects [16, 119]. The randomised trials cited by cough guidelines were conducted in refractory cough generally and cannot be directly compared with each other or extrapolated without reservation to the vagal neuronopathy of this disease; the mean cough severity reduction with morphine corresponds to just under 2 points on a 9-point scale, and only approximately half of patients respond [111]. The French series is the only one assessing morphine in the disease and has no comparator group [16]. Negative studies, such as ALCAT for acetyl-DL-leucine and absence of gabapentinoid effects on cough in 4 patients, form part of the evidence [16, 119]. A calculation from the European cohort showed the sample size needed for future trials: detecting a 50% reduction in SARA progression with 80% power requires 330 patients in a 1-year trial or 132 in a 2-year trial [23].

Studies registered on ClinicalTrials.gov are observational or concern devices and biomarkers: a natural history study (NCT05177809, Tübingen, 150 patients, recruiting until December 2026); the spinocerebellar ataxia natural history and genetic modifiers study, including RFC1 and SCA27B patients (NCT01060371, 1400 patients, until 2030); oxidative stress biomarkers (NCT07156214, Rome); urinary dysfunction (NCT06817707, Nice); chronic cough and CANVAS (NCT04703595, Barcelona, completed without registry results); video-oculography (NCT05278091) and nerve ultrasound (NCT05095870), in Nîmes; Nanopore sequencing of expansion disorders (NCT04621422, Bordeaux); and a multichannel vestibular implant for bilateral vestibulopathy, not specific to CANVAS (NCT06500975, Johns Hopkins, until 2029) [106].

7.11. Developments in the past 12 months

Between October 2025 and October 2026, no authorisations, regulatory applications, phase 1 to 3 disease-modifying therapy trials or company announcements concerning the disease appeared in the sources searched (Europe PMC, ClinicalTrials.gov and general searches); absence from these sources does not prove that none exist [106]. Table 7.5 summarises developments with their evidence level.

Table 7.5. Developments from October 2025 to October 2026.

7.12. Agents to avoid

GeneReviews considers it reasonable to assume that drugs with known toxicity to peripheral nerves, the cerebellum or vestibular system may exacerbate the disease, but states that there is no evidence-based guidance to avoid specific drugs and that decisions should follow clinical need; alcohol, which worsens ataxia, should be consumed only in moderation [1]. The 2020 version, reflected in GeneGuidance, gave as examples neurotoxic chemotherapy and pyridoxine (peripheral nerve), phenytoin (cerebellum), aminoglycosides (vestibular system) and chronic alcohol consumption [31]. These examples are drugs with known neurotoxicity rather than formal contraindications. The Charcot-Marie-Tooth Association maintains a list of potentially toxic drugs in hereditary neuropathies [1]. A preprint suggests that heterozygous AAGGG carriers have increased risk and severity of platinum-induced neuropathy [1, 41], and a CANVAS phenocopy was attributed to epirubicin, docetaxel and cyclophosphamide chemotherapy in an affected family [23]. Elevated vitamin B6 should be investigated in neuropathy patients taking supplements; supplements were stopped in two patients with cramps [59].

7.13. Pregnancy and lactation

Typical disease onset is beyond reproductive age, and the sources consulted do not describe pregnancies in patients. Considerations mainly concern symptomatic drugs, whose American labels state that: pregabalin may cause fetal harm based on animal data and breastfeeding is not recommended; gabapentin, according to published cohort studies, did not increase major malformation or miscarriage risk despite adverse animal data; dalfampridine has no adequate pregnancy data and reduced offspring viability in animals; and prolonged morphine use during pregnancy may cause neonatal withdrawal syndrome and is not recommended during breastfeeding [113, 117, 118, 120]. The New Zealand pregabalin trial excluded pregnant and breastfeeding women [114]. Cough neuromodulators are not used in children [111].

7.14. Anaesthesia and the perioperative period

There are no specific guidelines. A report of a 69-year-old woman with CANVAS (without described genetic confirmation), frail and wheelchair-dependent, undergoing pancreaticoduodenectomy illustrates the risks [122]. She received 4 weeks of prehabilitation with daily physiotherapy, incentive spirometry and nutrition; anaesthesia combined spinal anaesthesia (hyperbaric bupivacaine 0.5% 15 mg, morphine 200 µg and clonidine 75 µg), thoracic epidural and opioid-sparing general anaesthesia, with arterial, central venous and advanced haemodynamic monitoring; she required noradrenaline 2–10 µg/min to maintain mean arterial pressure above 65 mmHg during the 12 hours of surgery and until day 3; non-depolarising neuromuscular blockade was reversed with sugammadex 200 mg; she was extubated at the end of surgery and discharged on day 12 [122]. Extrapolating from MSA, Friedreich ataxia and SCAs, the authors recommend euvolaemia and continuous monitoring because of baroreflex dysfunction, noradrenaline as the preferred vasopressor, consideration of isoprenaline or pacing for atropine-resistant bradycardia, caution with fludrocortisone because of supine hypertension, attention to aspiration risk from gastroparesis, central respiratory depression, obstructive apnoea and vocal cord paresis, avoidance of suxamethonium (hyperkalaemia, excessive paralysis), cautious use of non-depolarising relaxants and early mobilisation [122]. Neuraxial blockade is possible but may exacerbate dysautonomia, with scarce evidence [122]. The frequency of obstructive sleep apnoea warrants attention during post-anaesthetic recovery [69].

7.15. Transition and long-term care

Transition from paediatric to adult medicine does not apply because the disease begins in adulthood. The relevant transition is from independence to dependence: the introduction of a walking stick, walking frame and wheelchair, dysphagia and increasing caregiver burden require regular reassessment of aids, social work and, in advanced stages, palliative care and respite care [1].

7.16. Guidelines and their differences

Table 7.6. Guidelines and consensus documents used in this article.

8. Reproductive Options and Family Counselling

8.1. Inheritance and recurrence risks

The disease has autosomal recessive inheritance [1]. A patient's parents are presumed heterozygous for an intronic expansion, almost always AAGGG, or more rarely a sequence variant or deletion [1]. Parental testing confirms heterozygosity and makes risk calculation more reliable, but with average onset in the 6th decade, parents have often already died [1]. If only one parent is a carrier, “Hereditary Ataxia Overview” advises considering a de novo variant or postzygotic mosaicism, and, with apparent homozygosity, an undetected deletion or uniparental isodisomy [11]. Heterozygotes have no disease described to date and are presumed not to be at risk, although confirmation requires long-term follow-up; the preprint on platinum neuropathy in carriers is the only suggested exception and has not been peer reviewed [1, 41]. In segregation analysis of 38 first-degree relatives in the European cohort, 9 of 10 affected individuals had two pathogenic expansions and the 28 healthy individuals had zero or one; the exception was the chemotherapy phenocopy [23]. A healthy 52-year-old brother with (AAGGG)1000 and (AAAGG)60 had normal examination and nerve conduction studies, showing that an AAGGG allele with a short AAAGG allele does not cause disease [9].

Table 8.1. Risks for relatives of a patient with biallelic expansions.

8.2. Pseudodominance and carrier frequency

AAGGG carrier frequency is high: approximately 8% in populations of northern European ancestry, 5.3% in American populations, 4.7% in East Asia, 2.9% in South Asia and 0.7% in African populations [1, 12]. In Australian controls it was approximately 1 in 16 [13]. Consequently, a patient may have affected children with a carrier partner, and disease appears in consecutive generations without consanguinity; this is how Australian families with “dominant hereditary sensory neuropathy with cough” proved to have RFC1-related disease [1, 10]. Probable consanguinity was present in 27% of RFC1-positive families in the European cohort and multiplex history in 32% [23]. The parallel with other recessive ataxias in populations with high carrier frequency, such as ARSACS in Saguenay–Lac-Saint-Jean, is useful in counselling [11].

8.3. Penetrance and predictive testing

Incomplete penetrance, suggested by hundreds of biallelic UK Biobank individuals without recorded ataxia or neuropathy, and the role of polygenic modifiers require caution when informing an asymptomatic biallelic sibling [42]. The report should not predict symptoms or age at onset from expansion size [5]. GeneReviews does not identify presymptomatic screening with therapeutic benefit and refers assessment of at-risk relatives to genetic counselling [1]. The value of diagnosis, even without disease-modifying therapy, includes ending the diagnostic odyssey, informing prognosis, explaining cough, avoiding unnecessary immunotherapies and enabling reproductive counselling [31, 78].

8.4. Cascade testing

Carrier detection in relatives requires identification of the familial variants [1]. In compound heterozygotes with an expansion and a truncating variant or a rare motif, cascade testing should investigate both alleles, combining RFC1 sequencing with RP-PCR for alternative motifs [8, 9]. When the family has an uncommon motif (ACAGG, AGGGC, Māori configuration), the laboratory should use a motif-specific assay [1].

8.5. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing are possible once biallelic variants have been identified in an affected relative, and views on their use differ among professionals and families; the decision is personal [1]. For adult-onset expansion disorders, the ACMG requires extensive counselling before prenatal diagnosis [5]. The ideal time for counselling and partner carrier testing is before pregnancy, and it should be offered to affected young adults, carriers or those at risk [1].

8.6. Minors and families

The disease begins in adulthood and has no proven preventive intervention, so predictive testing in minors does not alter surveillance or treatment during childhood [1]. The sources consulted contain no RFC1-specific recommendation on testing minors. Counselling should be renewed at each appointment, for example when family planning arises, and should include social work support [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1. What each figure measures

Frequency figures for this disease measure different things and should not be interchanged. Table 9.1 separates carrier frequency, genotypic prevalence, recorded clinical prevalence and yield in selected cohorts.

Table 9.1. Epidemiological indicators and what they measure.

The discrepancy between the Orphanet category and genomic data is not resolved by choosing one figure: the former counts patients labelled as CANVAS before testing existed, the latter count genotypes, and incomplete penetrance and underdiagnosis lie between them [1, 2, 42]. The number of symptomatic people may be lower than predicted from genome data, particularly for short AAGGG expansions [1]. The 2020 study authors considered the disease probably underdiagnosed [14], and more than a thousand individuals with pathogenic expansions had been identified by 2026 [1]. Ataxia UK states that there are no data on how many people have CANVAS [110].

9.2. Europe

In European cohorts, biallelic expansion explains 15–22% of late-onset ataxias, 62% of patients with sensory and cerebellar involvement and almost 100% of patients with ataxia and chronic cough [1, 7, 50, 124, 125]. It also explains approximately one-third of pure or predominantly sensory idiopathic axonal neuropathies [1, 24, 126]. In an unselected Turkish late-onset ataxia cohort, with 50% consanguinity, the proportion was 14% of families, and the authors estimate 1–4% in other populations depending on region and consanguinity [23]. In Poland, among 2523 patients with cerebellar syndrome, 4.6% had biallelic AAGGG and 25% non-pathogenic AAAAG or AAAGG expansions [83]. In Sweden, 8 of 30 patients (27%) were from Norrbotten, suggesting a regional founder effect [48]. Patients with rare motifs were predominantly European, except for AAGGC (South Asian family) and ACAGG (East Asia) [8].

9.3. Portugal and Brazil

In Portugal, the multicentre series by Malaquias and colleagues included 67 patients, with multisystem features in 82% [47], and the series from the São João Local Health Unit (ULS), Porto, included 15 patients, with the complete triad in two-thirds [52]. A Portuguese family with 4 affected individuals among 15 members was cited in the 2024 Turkish report [74]. The editorial accompanying the Portuguese series notes that no consistent statistics compare disease impact between countries, that the disease occurs in people of Portuguese descent and that it is under-recognised [127]. In Brazil, the original article included patients from Ribeirão Preto [7]; in a Brazilian cohort of 23 adult ataxias negative for common ataxias, RP-PCR suggested biallelic AAGGG in 2, but sequencing showed biallelic AAAGG in one, and biallelic AAGGG was confirmed in a family with two affected siblings [82]. A Brazilian centre described nigrostriatal dysfunction in 9 of 13 patients [60], and the Campinas group studied progressive brain atrophy and the hypothalamus [71, 72]. In southern Brazil, a 20-year cohort of ataxias without vertical family history had no access to RFC1 expansion testing at its institution, and many public ataxia clinics in the country lack Friedreich ataxia testing and SCA panels, necessitating private testing and creating social disparity [128]. The sources consulted contain no population prevalence data for Portugal or Brazil.

9.4. The Americas, Oceania and Asia

In ataxia cohorts from North and South America and East Asia, the proportion was lower, at 0.6% to 4.3% [1, 82, 84–86]. In the United States, in an idiopathic peripheral neuropathy cohort, biallelic expansion was present in 2.3% of patients (6.9% of pure sensory neuropathies) and 1 of 778 controls [57, 58]. In Australia, more than one-third of adults with ataxia had biallelic variants, predominantly AAGGG, but also ACAGG, long AAAGG and the Māori motif [13]. The (AAAGG)10-25(AAGGG)exp(AAAGG)4-6 configuration in patients of Māori and Cook Islands ancestry probably reflects a founder allele [30]. ACAGG is almost exclusive to East Asia and the Pacific [1, 28, 29]. In South Korea, the first chronic cough study in a non-Caucasian population found 2.3% biallelic individuals [75]. The expansion has also been described in Lebanese and Indonesian families and in carriers with admixed Native American ancestry [22, 23].

9.5. European networks and patient associations

The relevant European reference network is ERN-RND (rare neurological diseases), which publishes the adult ataxia flowchart, and, for sensory neuropathy, EURO-NMD (neuromuscular diseases) [17]. Patient associations include Ataxia UK (United Kingdom), with a dedicated CANVAS leaflet; Euro-ATAXIA (European federation); APAHE (Portuguese Association of Hereditary Ataxias); ABAHE (Brazilian Association of Hereditary and Acquired Ataxias); and the National Ataxia Foundation (USA) [1, 110, 129]. GeneReviews also lists Charcot-Marie-Tooth associations (CMTA in the USA, CMT UK, ACMT-Rete in Italy), Ménière disease associations, the Multiple System Atrophy Trust and Muscular Dystrophy UK [1].

10. Illustrative Clinical Case

10.1. Turkish siblings with cough decades before ataxia

Gökçay and colleagues described in 2024 the first molecularly confirmed Turkish CANVAS family, followed at Ege University in Izmir [74]. There were seven siblings, born to healthy non-consanguineous parents who had emigrated from the same village in Macedonia; five were affected (one woman and four men, aged 76, 72, 67, 62 and 57 years) and two were healthy, with no affected individuals in other generations (pedigree in Figure 1 of the article: https://pmc.ncbi.nlm.nih.gov/articles/PMC11438076/figure/Fig1/) [74].

The index case was a 67-year-old man with 10 years of unsteady gait, worse in the dark, oscillopsia with head movements, distal numbness and tingling since 55 years, erectile dysfunction for 5 years and chronic cough for more than 30 years, resistant to antitussives [74]. On examination, eye movements were normal without nystagmus, head impulse testing showed bilateral horizontal corrective saccades, ankle reflexes were absent with the remaining reflexes normal, there was glove-and-stocking hypoaesthesia and dysarthria without dysmetria or dysdiadochokinesia, and gait was broad-based with inability to tandem walk; SARA score was 13 [74]. Nerve conduction studies recorded no sensory potentials, and MRI showed vermian atrophy, particularly of lobules VI and VII, with widening of the primary and prepyramidal fissures (Figure 2 of the article: https://pmc.ncbi.nlm.nih.gov/articles/PMC11438076/figure/Fig2/) [74].

In all five affected siblings, chronic cough was the first symptom, beginning between 33 and 45 years, but was reported only when specifically asked about; paraesthesia (48–55 years), oscillopsia (51–60) and imbalance (51–57) followed, and two had erectile dysfunction [74]. The two oldest siblings had SARA scores of 32 and 28, with dysmetria, dysdiadochokinesia and dysarthria, and required walking assistance at 65 and 63 years; the youngest had SARA scores of 10 and 5 [74]. vHIT, performed in three affected individuals, showed very low VOR gains in all three canals on both sides (0.04–0.30), versus 0.72–1.09 in healthy siblings [74]. Before genetic testing, acquired causes had been excluded through biochemistry, immunofixation, antinuclear, anti-double-stranded DNA, anti-SSA and anti-SSB antibodies, paraneoplastic antibodies, coeliac disease antibodies, cerebrospinal fluid studies and whole-body CT, and an SCA panel was negative [74]. Flanking PCR produced no band in the five affected individuals but did in the two healthy siblings, and RP-PCR confirmed homozygous AAGGG in all five, without size determination [74]. The report does not describe treatment.

10.2. Lessons from the case

The case brings together features enabling bedside diagnosis: a dry cough lasting decades that emerges only when specifically asked about, unsteadiness worse in the dark, oscillopsia, bilateral corrective saccades on head impulse testing, absent SNAP with largely preserved reflexes and vermian atrophy [1, 74]. The siblings also illustrate variability in severity with the same genotype, from assisted walking to mild ataxia, and recessive inheritance without affected individuals in other generations [74]. The authors conclude that testing should be requested in late-onset ataxia with sensory and vestibular symptoms, particularly with chronic cough, irrespective of family history [74]. The extensive acquired-cause investigation performed before genetic testing corresponds to the first step of the European algorithm; today, with RFC1 in first-tier testing for sporadic or recessive ataxia, diagnostic testing can be requested in parallel [17]. Absence of expansion sizing, acceptable with a typical biallelic AAGGG pattern, is consistent with guidance that sizing is usually unnecessary for this motif [1].

11. Prognosis and Long-term Follow-up

11.1. Prognosis

The disease progresses slowly, with early sensory neurone involvement and subsequent appearance of vestibular and cerebellar dysfunction [1]. Disability data are consistent between large series (Table 11.1), all without disease-modifying therapy.

Table 11.1. Disability milestones and progression.

Life expectancy appears not to be reduced [1]. This prognosis contrasts with hereditary ataxias generally, in which survival is usually up to 25 years after manifestations, and MSA, with mean survival of 9.3 years [1, 11]. SARA progression is almost twice that described by the authors for Friedreich ataxia, and acceleration phases resembling MSA-C warrant reassessment [23]. Poorer prognostic factors are larger AAGGG expansions, associated with earlier loss of walking and earlier dysarthria and dysphagia, and, in isolated reports, compound heterozygosity with a truncating variant and some rare motifs [1, 8, 9, 43]. Quantitative brainstem MRI showed greater sensitivity to change over 2 years than SARA and is a candidate progression biomarker [71].

11.2. Surveillance

GeneReviews recommends the surveillance in Table 11.2; there are no other guidelines with specific intervals [1].

Table 11.2. Recommended surveillance.

Clinical reports prioritise falls prevention and dysphagia management [19], and vestibular rehabilitation may improve gait without restoring VOR [79].

12. Summary and Final Considerations

In seven years, RFC1-related disease has evolved from a rare syndrome with a long name into one of the most frequent genetic causes of adult-onset ataxia and sensory neuropathy. The cause is almost always a biallelic AAGGG pentanucleotide expansion in the poly(A) tail of an AluSx3 element in intron 2 of the gene, an allele carried by approximately 8% of people of northern European ancestry. Alternative motifs (ACAGG in East Asia, a Māori founder configuration, AGGGC and very long AAAGG) and, in approximately 1% of cases, a truncating variant in trans complete the molecular spectrum. The mechanism remains debated, between loss of function suggested by compound heterozygosity and a protein-independent repeat effect.

Clinical disease begins in sensory neurones and spreads over years to the vestibular apparatus and cerebellum. Spasmodic dry cough, often preceding unsteadiness by decades and labelled as asthma or reflux, neuronopathy with preserved reflexes and bilateral vestibular hypofunction are the keys to recognition. Isolated cerebellar ataxia without sensory neuropathy does not suggest the disease. Limited phenotypes, such as idiopathic sensory neuropathy, refractory cough or parkinsonism, have widened the range of specialties in which diagnosis can be made.

Diagnosis requires targeted testing because panels and exome sequencing do not detect the expansion; RP-PCR with flanking PCR is the basis, Southern blot and long-read sequencing size and characterise uncertain alleles, and gene sequencing is required when there is only one expansion. Reports should state these limitations, and a negative result in a typical phenotype requires investigation of alternative motifs and differential diagnoses such as SCA27B, SCA3 and late-onset Friedreich ataxia. Distinction from MSA-C and immune ganglionopathies prevents incorrect prognoses and ineffective immunotherapies.

There is no approved disease-modifying treatment, and the searches performed found no disease-modifying drug trials. Management involves rehabilitation and symptomatic control, with weak and predominantly extrapolated evidence: low-dose morphine for cough, recommended by ERS but not CHEST, showed a response in an uncontrolled series, and 4-aminopyridine and acetyl-DL-leucine rarely help. Progression is slow, with a walking stick after approximately 10 years and apparently preserved life expectancy, and surveillance is annual. High carrier frequency gives genetic counselling particular features, such as pseudodominance and partner carrier testing, and incomplete penetrance calls for caution when informing asymptomatic relatives.

13. References

1. Cortese A, Curro R, Kaski D, Birring S, Reilly MM, Houlden H. RFC1-Related Disorders. 2020 Nov 25 [updated 2026 Sep 22]. In: Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK564656/
2. Orphanet. Cerebellar ataxia with neuropathy and bilateral vestibular areflexia syndrome (ORPHA:504476) [Internet]. Paris: INSERM; [updated 2026 Jul 2; cited 2026 Oct 9]. Available from: https://www.orpha.net/en/disease/detail/504476
3. Migliaccio AA, Halmagyi GM, McGarvie LA, Cremer PD. Cerebellar ataxia with bilateral vestibulopathy: description of a syndrome and its characteristic clinical sign. Brain. 2004 Feb;127(Pt 2):280-93. doi:10.1093/brain/awh030 PMID: 14607788.
4. 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.
5. Guha S, Rajan-Babu IS, Kesari A, Garg S, Rose NC, Seifert BA, et al. Detection of repeat expansion variants using next generation sequencing: A points to consider statement of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2026 Jul;28(7):102520. doi:10.1016/j.gim.2026.102520 PMID: 42287278.
6. National Center for Biotechnology Information. ClinVar: RFC1 repeat expansion records (VCV000997970, VCV001333160, VCV000625839, VCV000986302, VCV003370519, VCV004531160, VCV000996307) and VCV001703247 [Internet]. Bethesda (MD): NCBI [cited 2026 Oct 9]. Available from: https://www.ncbi.nlm.nih.gov/clinvar/variation/997970/
7. Cortese A, Simone R, Sullivan R, Vandrovcova J, Tariq H, Yau WY, et al. Biallelic expansion of an intronic repeat in RFC1 is a common cause of late-onset ataxia. Nat Genet. 2019 Apr;51(4):649-658. doi:10.1038/s41588-019-0372-4 PMID: 30926972.
8. Dominik N, Magri S, Currò R, Abati E, Facchini S, Corbetta M, et al. Normal and pathogenic variation of RFC1 repeat expansions: implications for clinical diagnosis. Brain. 2023 Dec 1;146(12):5060-5069. doi:10.1093/brain/awad240 PMID: 37450567.
9. Ronco R, Perini C, Currò R, Dominik N, Facchini S, Gennari A, et al. Truncating Variants in RFC1 in Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia Syndrome. Neurology. 2023 Jan 31;100(5):e543-e554. doi:10.1212/WNL.0000000000201486 PMID: 36289003.
10. Grosz BR, Ellis M, Trivedi S, Scriba C, Stoll M, Zhu D, et al. Pseudodominant Inheritance of Biallelic RFC1 Expansions-Revisiting the 3p22-p24 HSN1B Locus. J Peripher Nerv Syst. 2025 Dec;30(4):e70070. doi:10.1111/jns.70070 PMID: 41084404.
11. Perlman S. Hereditary Ataxia Overview. 1998 Oct 28 [updated 2022 Jun 16; revised 2026 Jul 9]. In: Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1138/
12. Ibañez K, Jadhav B, Zanovello M, Gagliardi D, Clarkson C, Facchini S, et al. Increased frequency of repeat expansion mutations across different populations. Nat Med. 2024;30:3357-68. PMID: 39354197. doi:10.1038/s41591-024-03190-5.
13. Davies KC, Rafehi H, Fearnley LG, Snell P, Gillies G, Field TA, et al. Comprehensive Characterisation of the RFC1 Repeat in an Australian Cohort. Cerebellum. 2025 Jun 7;24(4):111. doi:10.1007/s12311-025-01867-2 PMID: 40481300.
14. Cortese A, Tozza S, Yau WY, Rossi S, Beecroft SJ, Jaunmuktane Z, et al. Cerebellar ataxia, neuropathy, vestibular areflexia syndrome due to RFC1 repeat expansion. Brain. 2020 Feb 1;143(2):480-490. doi:10.1093/brain/awz418 PMID: 32040566.
15. Ariello LE, Rastall DPW, Rosenthal LS. Treatment of primary adult-onset neurodegenerative cerebellar ataxias. Neurotherapeutics. 2026 Jan;23(1):e00805. doi:10.1016/j.neurot.2025.e00805 PMID: 41387161.
16. Guilleminault L, Chazelas P, Magdelaine C, Villeneuve T, Lia AS, Magy L. Efficacy of morphine on cough in patients with repeat expansions of RFC1 and refractory chronic cough. ERJ Open Res. 2026 Jan;12(1):. doi:10.1183/23120541.00840-2025 PMID: 41532091.
17. European Reference Network for Rare Neurological Diseases (ERN-RND). Diagnostic flowchart for adult ataxias [Internet]. Published 2019 Feb 11; last revision 2024 Sep 16 [cited 2026 Oct 9]. Available from: https://www.ern-rnd.eu/wp-content/uploads/2024/09/Final-Flowchart-adult-ataxias_en-New.pdf
18. Szmulewicz DJ, Waterston JA, MacDougall HG, Mossman S, Chancellor AM, McLean CA, et al. Cerebellar ataxia, neuropathy, vestibular areflexia syndrome (CANVAS): a review of the clinical features and video-oculographic diagnosis. Ann N Y Acad Sci. 2011 Sep;1233:139-47. doi:10.1111/j.1749-6632.2011.06158.x PMID: 21950986.
19. Szmulewicz DJ, McLean CA, MacDougall HG, Roberts L, Storey E, Halmagyi GM. CANVAS an update: clinical presentation, investigation and management. J Vestib Res. 2014;24(5-6):465-74. doi:10.3233/VES-140536 PMID: 25564090.
20. Wu TY, Taylor JM, Kilfoyle DH, Smith AD, McGuinness BJ, Simpson MP, et al. Autonomic dysfunction is a major feature of cerebellar ataxia, neuropathy, vestibular areflexia 'CANVAS' syndrome. Brain. 2014;137:2649–56. PMID: 25070514. doi:10.1093/brain/awu196.
21. Szmulewicz DJ, Roberts L, McLean CA, MacDougall HG, Halmagyi GM, Storey E. Proposed diagnostic criteria for cerebellar ataxia with neuropathy and vestibular areflexia syndrome (CANVAS). Neurol Clin Pract. 2016 Feb;6(1):61-68. doi:10.1212/CPJ.0000000000000215 PMID: 26918204.
22. Rafehi H, Szmulewicz DJ, Bennett MF, Sobreira NLM, Pope K, Smith KR, et al. Bioinformatics-Based Identification of Expanded Repeats: A Non-reference Intronic Pentamer Expansion in RFC1 Causes CANVAS. Am J Hum Genet. 2019 Jul 3;105(1):151-165. doi:10.1016/j.ajhg.2019.05.016 PMID: 31230722.
23. Traschütz A, Cortese A, Reich S, Dominik N, Faber J, Jacobi H, et al. Natural History, Phenotypic Spectrum, and Discriminative Features of Multisystemic RFC1 Disease. Neurology. 2021 Mar 2;96(9):e1369-e1382. doi:10.1212/WNL.0000000000011528 PMID: 33495376.
24. Currò R, Salvalaggio A, Tozza S, Gemelli C, Dominik N, Galassi Deforie V, et al. RFC1 expansions are a common cause of idiopathic sensory neuropathy. Brain. 2021 Jun 22;144(5):1542-1550. doi:10.1093/brain/awab072 PMID: 33969391.
25. Davies K, Szmulewicz DJ, Corben LA, Delatycki M, Lockhart PJ. RFC1-Related Disease: Molecular and Clinical Insights. Neurol Genet. 2022 Oct;8(5):e200016. doi:10.1212/NXG.0000000000200016 PMID: 36046423.
26. Hadjivassiliou M, Currò R, Beauchamp N, Dominik N, Grunewald RA, Shanmugarajah P, et al. Can CANVAS due to RFC1 biallelic expansions present with pure ataxia? J Neurol Neurosurg Psychiatry. 2024 Jan 11;95(2):171-174. doi:10.1136/jnnp-2023-331381 PMID: 37414537.
27. Nobilleau F, Audet S, da Silva Babinet A, Tork S, Zaouter C, Liao M, et al. RFC1 regulates the expansion of neural progenitors in the developing zebrafish cerebellum. Nat Commun. 2025 Jul 1;16(1):6019. doi:10.1038/s41467-025-60775-5 PMID: 40595562.
28. Scriba CK, Beecroft SJ, Clayton JS, Cortese A, Sullivan R, Yau WY, et al. A novel RFC1 repeat motif (ACAGG) in two Asia-Pacific CANVAS families. Brain. 2020;143:2904–10. PMID: 33103729. doi:10.1093/brain/awaa263.
29. Tsuchiya M, Nan H, Koh K, Ichinose Y, Gao L, Shimozono K, et al. RFC1 repeat expansion in Japanese patients with late-onset cerebellar ataxia. J Hum Genet. 2020;65:1143–7. PMID: 32694621. doi:10.1038/s10038-020-0807-x.
30. Beecroft SJ, Cortese A, Sullivan R, Yau WY, Dyer Z, Wu TY, et al. A Māori-specific RFC1 pathogenic repeat configuration in CANVAS, likely due to a founder allele. Brain. 2020;143:2673–80. PMID: 32851396. doi:10.1093/brain/awaa203.
31. GeneGuidance. RFC1 — gene record and clinical curation. GeneGuidance database, version of 2026-09-26.
32. Benkirane M, Da Cunha D, Marelli C, Larrieu L, Renaud M, Varilh J, et al. RFC1 nonsense and frameshift variants cause CANVAS: clues for an unsolved pathophysiology. Brain. 2022;145:3770-5. PMID: 35883251. doi:10.1093/brain/awac280.
33. Arteche-Lopez A, Avila-Fernandez A, Damian A, Soengas-Gonda E, de la Fuente RP, Gomez PR, et al. New cerebellar ataxia, neuropathy, vestibular areflexia syndrome cases are caused by the presence of a nonsense variant in compound heterozygosity with the pathogenic repeat expansion in the RFC1 gene. Clin Genet. 2023;103:236-41. PMID: 36250766. doi:10.1111/cge.14249.
34. Weber S, Coarelli G, Heinzmann A, Monin ML, Richard N, Gerard M, et al. Two RFC1 splicing variants in CANVAS. Brain. 2023;146:e14-e6. PMID: 36478048. doi:10.1093/brain/awac466.
35. King KA, Wegner DJ, Bucelli RC, Shapiro J, Paul AJ, Dickson PI, et al. Whole-genome and long-read sequencing identify a novel mechanism in RFC1 resulting in CANVAS syndrome. Neurol Genet. 2022;8:e200036. PMID: 36524104. doi:10.1212/NXG.0000000000200036.
36. Abdi MH, Zamiri B, Pazuki G, Sardari S, Pearson CE. Pathogenic CANVAS-causing but not nonpathogenic RFC1 DNA/RNA repeat motifs form quadruplex or triplex structures. J Biol Chem. 2023;299:105202. PMID: 37660923. doi:10.1016/j.jbc.2023.105202.
37. Kudo K, Hori K, Asamitsu S, Maeda K, Aida Y, Hokimoto M, et al. Structural polymorphism of the nucleic acids in pentanucleotide repeats associated with the neurological disorder CANVAS. J Biol Chem. 2024;300:107138. PMID: 38447794. doi:10.1016/j.jbc.2024.107138.
38. Wang J, Qiu D, Zhou J, Mergny JL, Alberti P. Pentanucleotide guanine-rich WGGGW repeats, including CANVAS AGGGA repeats, form a variety of noncanonical structures. Nucleic Acids Res. 2026;54:gkag051. PMID: 41614301. doi:10.1093/nar/gkag051.
39. Maltby CJ, Krans A, Grudzien SJ, Palacios Y, Muiños J, Suárez A, et al. AAGGG repeat expansions trigger RFC1-independent synaptic dysregulation in human CANVAS neurons. Sci Adv. 2024 Sep 6;10(36):eadn2321. doi:10.1126/sciadv.adn2321 PMID: 39231235.
40. Wada T, Doi H, Okubo M, Tada M, Ueda N, Suzuki H, et al. RNA foci in two bi-allelic RFC1 expansion carriers. Ann Neurol. 2024;95:607-13. PMID: 38062616.
41. Currò R, Dominik N, Facchini S, Schnekenberg RP, Perini C, Ronco R, et al. CANVAS causing AAGGG repeat expansions cause tissue-specific reduction in RFC1 expression and increase sensitivity to DNA damage [preprint]. bioRxiv. 2025 Nov 18. doi:10.1101/2025.11.18.688292.
42. Wang X, Fearnley LG, Davies KC, Snell P, Lee S, Jackson VE, et al. DNA Repair Pathway Variants Are Enriched in Individuals with Biallelic AAGGG CANVAS and RFC1-Related Disease. Mov Disord. 2026 Jul 19;:. doi:10.1002/mds.70428 PMID: 42473260.
43. Curro R, Dominik N, Facchini S, Vegezzi E, Sullivan R, Galassi Deforie V, et al. Role of the repeat expansion size in predicting age of onset and severity in RFC1 disease. Brain. 2024;147:1887-98. PMID: 38193360. doi:10.1093/brain/awad436.
44. Dupré M, Hermann R, Vidoni L, Quadrio I, Latour P, Subtil F, et al. Impact of the intronic RFC1 expansion size in CANVAS phenotype: an oculomotor study. J Neurol. 2025 Jun 3;272(6):442. doi:10.1007/s00415-025-13150-9 PMID: 40461673.
45. Miyatake S, Yoshida K, Koshimizu E, Doi H, Yamada M, Miyaji Y, et al. Repeat conformation heterogeneity in cerebellar ataxia, neuropathy, vestibular areflexia syndrome. Brain. 2022;145:1139-50. PMID: 35355059.
46. Massucco S, Hamedani M, Ponzano M, Manca R, Roveta E, Pichiecchio A, et al. Motor, Extrapyramidal, and Cognitive Involvement in RFC1 Disease: A Systematic Review and Meta-Analysis. Neurol Genet. 2026 Jun;12(3):e200393. doi:10.1212/NXG.0000000000200393 PMID: 42109533.
47. Malaquias MJ, Braz L, Santos Silva C, Damásio J, Jorge A, Lemos JM, et al. Multisystemic RFC1-Related Disorder: Expanding the Phenotype Beyond Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia Syndrome. Neurol Clin Pract. 2023 Oct;13(5):e200190. doi:10.1212/CPJ.0000000000200190 PMID: 37674869.
48. Alm V, Säll L, Samuelsson K, Press R, Arnardottir S, Lindstrand A, et al. Novel Clinical Insights From a Swedish RFC1 Spectrum Disorder Cohort. Eur J Neurol. 2026 Feb;33(2):e70488. doi:10.1111/ene.70488 PMID: 41603480.
49. Huin V, Coarelli G, Guemy C, Boluda S, Debs R, Mochel F, et al. Motor neuron pathology in CANVAS due to RFC1 expansions. Brain. 2022;145:2121-32. PMID: 34927205. doi:10.1093/brain/awab449.
50. Traschütz A, Wilke C, Haack TB, Bender B, Synofzik M; RFC1 Study Group. Sensory axonal neuropathy in RFC1-disease: tip of the iceberg of broad subclinical multisystemic neurodegeneration. Brain. 2022;145:e6-e9. PMID: 35230382. doi:10.1093/brain/awac003.
51. Dujardin K, Tard C, Digle E, Herlin V, Mutez E, Davion JB, et al. Cognitive impairment is part of the phenotype of cerebellar ataxia, neuropathy, vestibular areflexia syndrome (CANVAS). Mov Disord. 2024;39:892-7. PMID: 38480525. doi:10.1002/mds.29750.
52. Fernandes AA, Alexandre PL, Vedor S, Figueiredo R, Marques P, Braz L. Comprehensive phenotyping of RFC1-related disorder: integrating electrophysiological, brain imaging, and otoneurological data in deep phenotyping. Arq Neuropsiquiatr. 2025 Nov;83(11):1-7. doi:10.1055/s-0045-1811723 PMID: 41145152.
53. Salvalaggio A, Cacciavillani M, Tierro B, Coraci D, Curro R, Ferrarini M, et al. Nerve ultrasound in CANVAS-spectrum disease: reduced nerve size distinguishes genetically confirmed CANVAS from other axonal polyneuropathies. J Peripher Nerv Syst. 2024;29:464-71. PMID: 39219417. doi:10.1111/jns.12655.
54. Garvey A, Melville IZ, Scriba CK, Yong V, Rodrigues M, Kao J, et al. Nerve ultrasound, neuronopathy and cough predict sensory neuropathy patients with RFC1 expansions. Brain Commun. 2025;7:fcaf434. PMID: 41322202. doi:10.1093/braincomms/fcaf434.
55. Di Giovanni J, Silvestri G, Dalla Zanna G, Tomasello F, Funcis A, Rossi S, et al. Vagus nerve hypotrophy in genetically confirmed cerebellar ataxia, neuropathy and vestibular areflexia syndrome (CANVAS) patients. Clin Neurophysiol. 2026 Jul;187:2111865. doi:10.1016/j.clinph.2026.2111865 PMID: 41930793.
56. Record CJ, Alsukhni RA, Curro R, Kaski D, Rubin JS, Morris HR, et al. Severe distinct dysautonomia in RFC1-related disease associated with Parkinsonism. J Peripher Nerv Syst. 2022;27:311-5. PMID: 36177974. doi:10.1111/jns.12515.
57. Stino AM, Muthukumar L, Reynolds EL, Todd P, Ovunc S, Jin SC, et al. Bi- and Mono-Allelic RFC1 Expansion in a North American Cohort With Idiopathic Axonal Neuropathy. Ann Clin Transl Neurol. 2026 Jul 5;:. doi:10.1002/acn3.70462 PMID: 42402724.
58. Tang Z, Ovunc SS, Iwase R, Mehinovic E, Thomas S, Ulibarri J, et al. Homozygous RFC1 AAGGG Repeat Expansions Are Common in Idiopathic Peripheral Neuropathy. Ann Neurol. 2026 Jul;100(1):95-108. doi:10.1002/ana.78226 PMID: 41964406.
59. Machado J, Zwicker J, Bourque PR, Lochmüller H, Warman-Chardon J, Aleman A, et al. Cramps as the initial presentation of CANVAS/RFC1-related disorder. J Neuromuscul Dis. 2026 Sep 18;:22143602261485644. doi:10.1177/22143602261485644 PMID: 42758088.
60. Azevedo MPC, Lobo CC, Schmitt GS, Matos PCAAP, Barsottini OGP, Pedroso JL, et al. Nigrostriatal dysfunction in RFC1-related disorder/CANVAS. Parkinsonism Relat Disord. 2023 Oct;115:105854. doi:10.1016/j.parkreldis.2023.105854 PMID: 37729670.
61. Nakamura H, Doi H, Mitsuhashi S, Miyatake S, Katoh K, Frith MC, et al. Long-read sequencing identifies the pathogenic nucleotide repeat expansion in RFC1 in a Japanese case of CANVAS. J Hum Genet. 2020;65:475–80. PMID: 32066831.
62. Barghigiani M, De Michele G, Tessa A, Fico T, Natale G, Sacca F, et al. Screening for RFC-1 pathological expansion in late-onset ataxias: a contribution to the differential diagnosis. J Neurol. 2022;269:5431-5. PMID: 35633373. doi:10.1007/s00415-022-11192-x.
63. Kytövuori L, Sipilä J, Doi H, Hurme-Niiranen A, Siitonen A, Koshimizu E, et al. Biallelic expansion in RFC1 as a rare cause of Parkinson's disease. NPJ Parkinsons Dis. 2022;8:6. PMID: 35013364.
64. Horowitz T, Guedj E, Eusebio A, Fluchere F, Azulay JP, Delmont E, et al. Molecular imaging in CANVAS: a contribution for differential diagnosis? Mov Disord Clin Pract. 2024;11:879-85. PMID: 38576115. doi:10.1002/mdc3.14041.
65. Kovanda A, Šušmelj L, Jaklič H, Lukežič T, Maver A, Petrovic I, et al. Biallelic RFC1 Expansions Are a Rare Cause of Early-Onset and Familial Parkinson's Disease. Clin Genet. 2026 Mar;109(3):604-606. doi:10.1111/cge.70095 PMID: 41177915.
66. Sipilä J, Rauramaa T, Kytövuori L, Majamaa K. Neuropathology of a patient with Parkinson's disease and biallelic expansion in RFC1. Brain Pathol. 2026 Sep 18;:e70152. doi:10.1111/bpa.70152 PMID: 42757740.
67. Albajar I, Iruzubieta P, Barandiaran M, Alves P, Pujol A, Verdura E, et al. Further evidence of cerebellar cognitive affective/Schmahmann syndrome in RFC1-related syndrome. Mov Disord. 2024;39:1246-8. PMID: 39031430. doi:10.1002/mds.29885.
68. Delgado-Álvarez A, Gajate-García V, Cabrera-Martín MN, Fenollar-Cortés M, Horga A, Matias-Guiu JA. Comprehensive neuropsychological assessment and FDG-PET imaging in patients with RFC1-related disorder. Parkinsonism Relat Disord. 2026 Jul;148:108360. doi:10.1016/j.parkreldis.2026.108360 PMID: 42172733.
69. Funcis A, Brunetti V, Rossi S, Zanna GD, de Scisciolo M, Maneri G, et al. Prevalence and Clinical Correlates of Sleep Disorders in RFC1-Spectrum Disorders: A Cross-Sectional Study. Mov Disord. 2025 Sep;40(9):1990-1995. doi:10.1002/mds.30279 PMID: 40879541.
70. Neophytou P, Kalampokini S, Avraam C, Anagiotos A, Deltas C, Ioannou C, et al. Uncovering Neuropathic Pain in CANVAS: Lessons from a Case-Based Literature Review. Pain Ther. 2026 Oct;15(5):1209-1219. doi:10.1007/s40122-026-00867-2 PMID: 42525371.
71. Lobo CC, Rezende TJR, Pimentel-Silva LR, Jarola GM, Santos NBS, da Silva Schmitt G, et al. Longitudinal Evaluation of Ataxia and Brain Structural Changes in RFC1-Related Disorder. Mov Disord Clin Pract. 2025 Dec;12(12):2311-2316. doi:10.1002/mdc3.70344 PMID: 40908706.
72. Souza VM, Lobo CC, Rezende TJR, Schmitt GS, de Assis Pereira Matos PCA, de Lima FD, et al. Hypothalamic atrophy in CANVAS/RFC1. J Neurol. 2025 Jun 17;272(7):465. doi:10.1007/s00415-025-13194-x PMID: 40526300.
73. Gajate-García V, Fenollar-Cortés M, Galán L, Oancea-Ionescu R, Guerrero-Solá A, Horga A. RFC1 Repeat Expansions in Chronic Idiopathic Axonal Polyneuropathy: Prevalence, Phenotype, and Diagnostic Implications. J Peripher Nerv Syst. 2026 Sep;31(3):e70155. doi:10.1111/jns.70155 PMID: 42584421.
74. Gökçay F, Baskan GN, Şahbaz I, Koç MK, Başak AN, Celebisoy N. Cerebellar ataxia, neuropathy and vestibular areflexia syndrome (CANVAS): a family with five affected sibs from Turkey. BMC Neurol. 2024 Sep 28;24(1):356. doi:10.1186/s12883-024-03782-1 PMID: 39342186.
75. Park KE, Lee J, Choi JP, Oh JY, Won HK, Lee HY, et al. RFC1 Repeat Expansion in Chronic Cough: Findings From the Korean Chronic Cough Registry. Allergy Asthma Immunol Res. 2026 Jan;18(1):55-66. doi:10.4168/aair.2026.18.1.55 PMID: 41592538.
76. Gajate-García V, Fenollar-Cortés M, Rodríguez-Hermosa JL, Lara-González M, Herrero-Forte C, Muerte-Moreno I, et al. Targeted neurological screening for RFC1-related disease in unexplained chronic cough. J Neurol. 2026 Mar 16;273(4):. doi:10.1007/s00415-026-13751-y PMID: 41840142.
77. Ariello LE, Gold DR, Millar JL, Schubert MC, Mu W, Rosenthal L, et al. Video head impulse test differentiates RFC1-CANVAS, FGF14-SCA27B, and idiopathic late-onset cerebellar ataxias. J Neurol. 2025 Oct 7;272(10):679. doi:10.1007/s00415-025-13433-1 PMID: 41055766.
78. Fernández-Eulate G, Debs R, Maisonobe T, Latour P, Cohen-Aubart F, Saadoun D, et al. Sjögren syndrome and RFC1-CANVAS sensory ganglionopathy: co-occurrence or misdiagnosis? J Neurol. 2023 Jan;270(1):460-465. doi:10.1007/s00415-022-11382-7 PMID: 36155842.
79. Harrell RG, Cassidy AR, Klatt BN, Hovareshti P, Whitney SL. Vestibular rehabilitation in cerebellar ataxia with neuropathy and vestibular areflexia syndrome (CANVAS)- A case report. J Otol. 2023 Oct;18(4):199-207. doi:10.1016/j.joto.2023.06.004 PMID: 37877066.
80. Strupp M, Kim JS, Murofushi T, Straumann D, Jen JC, Rosengren SM, et al. Bilateral vestibulopathy: Diagnostic criteria Consensus document of the Classification Committee of the Bárány Society. J Vestib Res. 2017;27(4):177-189. doi:10.3233/VES-170619 PMID: 29081426.
81. Infante J, García A, Serrano-Cárdenas KM, González-Aguado R, Gazulla J, de Lucas EM, et al. Cerebellar ataxia, neuropathy, vestibular areflexia syndrome (CANVAS) with chronic cough and preserved muscle stretch reflexes: evidence for selective sparing of afferent Ia fibres. J Neurol. 2018;265:1454–62. PMID: 29696497. doi:10.1007/s00415-018-8872-1.
82. Akçimen F, Ross JP, Bourassa CV, Liao C, Rochefort D, Gama MTD, et al. Investigation of the RFC1 Repeat Expansion in a Canadian and a Brazilian Ataxia Cohort: Identification of Novel Conformations. Front Genet. 2019;10:1219. doi:10.3389/fgene.2019.01219 PMID: 31824583.
83. Tomczuk F, Ziora-Jakutowicz K, Dominik N, Houlden H, Cortese A, Rutkowska K, et al. Prevalence of intronic repeat expansions in the RFC1 gene in Polish patients with cerebellar syndrome. J Appl Genet. 2026;67(3):731-742. doi:10.1007/s13353-025-01000-3. PMID: 40802152.
84. Aboud Syriani D, Wong D, Andani S, De Gusmao CM, Mao Y, Sanyoura M, et al. Prevalence of RFC1-mediated spinocerebellar ataxia in a North American ataxia cohort. Neurol Genet. 2020;6:e440. PMID: 32582864. doi:10.1212/NXG.0000000000000440.
85. Wan L, Chen Z, Wan N, Liu M, Xue J, Chen H, et al. Biallelic intronic AAGGG expansion of RFC1 is related to multiple system atrophy. Ann Neurol. 2020;88:1132–43. PMID: 32939785. doi:10.1002/ana.25902.
86. Ando M, Higuchi Y, Yuan JH, Yoshimura A, Higashi S, Takeuchi M, et al. Genetic and clinical features of cerebellar ataxia with RFC1 biallelic repeat expansions in Japan. Front Neurol. 2022;13:952493. PMID: 36034314. doi:10.3389/fneur.2022.952493.
87. Zhang XY, Yu H, Liu GL, Wu ZY, Dong HL. Non-coding repeat expansions within NOTCH2NLC and RFC1 genes contribute to unsolved inherited peripheral neuropathies. J Hum Genet. 2026 Sep 10;:. doi:10.1038/s10038-026-01509-3 PMID: 42716992.
88. van de Warrenburg BP, van Gaalen J, Boesch S, Burgunder JM, Dürr A, Giunti P, et al. EFNS/ENS Consensus on the diagnosis and management of chronic ataxias in adulthood. Eur J Neurol. 2014 Apr;21(4):552-62. doi:10.1111/ene.12341 PMID: 24418350.
89. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17:405-24. PMID: 25741868. doi:10.1038/gim.2015.30.
90. Maver A, Lohmann K, Urbanczyk LM, Arnesen A, Barić I, Bauer P, et al. Consensus recommendations for next-generation sequencing-based genetic testing in Rare Neurological diseases. Eur J Hum Genet. 2026 Oct;34(10):1503-1512. doi:10.1038/s41431-026-02198-4 PMID: 42521809.
91. Borsche M, Thomsen M, Szmulewicz DJ, Lubbers B, Hinrichs F, Lockhart PJ, et al. Bilateral vestibulopathy in RFC1-positive CANVAS is distinctly different compared to FGF14-linked spinocerebellar ataxia 27B. J Neurol. 2024;271:1023-7. PMID: 37861706. doi:10.1007/s00415-023-12050-0.
92. Figueroa KP, Gross C, Buena-Atienza E, Paul S, Gandelman M, Kakar N, et al. A GGC-repeat expansion in ZFHX3 encoding polyglycine causes spinocerebellar ataxia type 4 and impairs autophagy. Nat Genet. 2024;56:1080-9. PMID: 38684900. doi:10.1038/s41588-024-01719-5.
93. Cortese A, Callegari I, Currò R, Vegezzi E, Colnaghi S, Versino M, et al. Mutation in RNF170 causes sensory ataxic neuropathy with vestibular areflexia: a CANVAS mimic. J Neurol Neurosurg Psychiatry. 2020a;91:1237–8. PMID: 32943585. doi:10.1136/jnnp-2020-323719.
94. Van Daele SH, Moisse M, Race V, Van Eesbeeck A, Keldermans L, Vermeer S, et al. RNF170 mutation causes autosomal dominant sensory ataxia with variable pyramidal involvement. Eur J Neurol. 2022;29:345-9. PMID: 34469621. doi:10.1111/ene.15091.
95. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 2, Inborn Errors of Metabolism in Adults — A Diagnostic Approach to Neurological and Psychiatric Presentations, Table 2.1.
96. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 70–72.
97. Pellerin D, Wilke C, Traschütz A, Nagy S, Currò R, Dicaire MJ, et al. Intronic FGF14 GAA repeat expansions are a common cause of ataxia syndromes with neuropathy and bilateral vestibulopathy. J Neurol Neurosurg Psychiatry. 2024;95:175-79. PMID: 37399286. doi:10.1136/jnnp-2023-331490.
98. Fanciulli A, Wenning GK. Multiple-system atrophy. N Engl J Med. 2015;372:249–63. PMID: 25587949. doi:10.1056/NEJMra1311488.
99. Chelban V, Bocchetta M, Hassanein S, Haridy NA, Houlden H, Rohrer JD. An update on advances in magnetic resonance imaging of multiple system atrophy. J Neurol. 2019;266:1036–45. PMID: 30460448. doi:10.1007/s00415-018-9121-3.
100. Wenning GK, Stankovic I, Vignatelli L, Fanciulli A, Calandra-Buonaura G, Seppi K, et al. The Movement Disorder Society Criteria for the Diagnosis of Multiple System Atrophy. Mov Disord. 2022 Jun;37(6):1131-1148. doi:10.1002/mds.29005 PMID: 35445419.
101. Pei I, Storck A, Bogdan T, Andries-Ros AC, Tranchant C, Iosif A, et al. Does CANVAS mimic MSA-C? A prospective multimodal study. J Neurol. 2026 Aug 19;273(9):. doi:10.1007/s00415-026-14005-7 PMID: 42616133.
102. Sullivan R, Yau WY, Chelban V, Rossi S, O'Connor E, Wood NW, et al. RFC1 intronic repeat expansions absent in pathologically confirmed multiple systems atrophy. Mov Disord. 2020;35:1277–9. PMID: 32333430. doi:10.1002/mds.28074.
103. Fan Y, Zhang S, Yang J, Mao CY, Yang ZH, Hu ZW, et al. No biallelic intronic AAGGG repeat expansion in RFC1 was found in patients with late-onset ataxia and MSA. Parkinsonism Relat Disord. 2020;73:1–2. PMID: 32151945. doi:10.1016/j.parkreldis.2020.02.017.
104. Kumar KR, Cortese A, Tomlinson SE, Efthymiou S, Ellis M, Zhu D, et al. RFC1 expansions can mimic hereditary sensory neuropathy with cough and Sjögren syndrome. Brain. 2020;143:e82-e.
105. Johnson RM, Curro R, Cortese A, Lunn MPT, Carr AS. Late-onset sensory ataxic neuropathy: important clinical clues to correct diagnosis. Pract Neurol. 2026. Epub ahead of print.
106. ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine. Search for “RFC1”, “CANVAS ataxia”, “cerebellar ataxia neuropathy vestibular areflexia” and “RFC1 repeat expansion” (API v2) [cited 2026 Oct 9]. Available from: https://clinicaltrials.gov/
107. Schmitz-Hübsch T, du Montcel ST, Baliko L, Berciano J, Boesch S, Depondt C, et al. Scale for the assessment and rating of ataxia: development of a new clinical scale. Neurology. 2006;66:1717-20. PMID: 16769946. doi:10.1212/01.wnl.0000219042.60538.92.
108. Martineau L, Noreau A, Dupré N. Therapies for ataxias. Curr Treat Options Neurol. 2014;16:300. PMID: 24832479. doi:10.1007/s11940-014-0300-y.
109. Zesiewicz TA, Wilmot G, Kuo SH, Perlman S, Greenstein PE, Ying SH, et al. Comprehensive systematic review summary: Treatment of cerebellar motor dysfunction and ataxia [RETIRED]: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology. Neurology. 2018 Mar 6;90(10):464-471. doi:10.1212/WNL.0000000000005055 PMID: 29440566.
110. Ataxia UK. CANVAS (Cerebellar ataxia, neuropathy and vestibular areflexia syndrome) [information leaflet, Internet]. London: Ataxia UK; 2021 Nov [cited 2026 Oct 9]. Available from: https://www.ataxia.org.uk/wp-content/uploads/2022/06/CANVAS-information-leaflet-2.pdf
111. Morice AH, Millqvist E, Bieksiene K, Birring SS, Dicpinigaitis P, Domingo Ribas C, et al. ERS guidelines on the diagnosis and treatment of chronic cough in adults and children. Eur Respir J. 2020 Jan;55(1):. doi:10.1183/13993003.01136-2019 PMID: 31515408.
112. Gibson P, Wang G, McGarvey L, Vertigan AE, Altman KW, Birring SS, et al. Treatment of Unexplained Chronic Cough: CHEST Guideline and Expert Panel Report. Chest. 2016 Jan;149(1):27-44. doi:10.1378/chest.15-1496 PMID: 26426314.
113. Rhodes Pharmaceuticals. MS CONTIN (morphine sulfate) extended-release tablets: prescribing information. NDA 019516 [Internet]. U.S. Food and Drug Administration; label effective 2026 Aug 14 [cited 2026 Oct 9]. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22MS%20CONTIN%22
114. Roxburgh R (sponsor). Feasibility of Pregabalin for the treatment of CANVAS associated chronic cough. Australian New Zealand Clinical Trials Registry ACTRN12623001356651 [Internet]. Registered 2023 Dec 21 [updated 2023 Dec 21; cited 2026 Oct 9]. Available from: https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?ACTRN=12623001356651
115. Soliman N, Moisset X, Ferraro MC, de Andrade DC, Baron R, Belton J, et al. Pharmacotherapy and non-invasive neuromodulation for neuropathic pain: a systematic review and meta-analysis. Lancet Neurol. 2025 May;24(5):413-428. doi:10.1016/S1474-4422(25)00068-7 PMID: 40252663.
116. European Medicines Agency. Lyrica (pregabalin): EPAR — medicine overview [Internet]. Amsterdam: EMA; authorised 2004 Jul 5; last EC decision 2026 Mar 25 [cited 2026 Oct 9]. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/lyrica
117. Pfizer (Parke-Davis). LYRICA (pregabalin): prescribing information. NDA 021446 [Internet]. U.S. Food and Drug Administration; label effective 2020 Jun 15 [cited 2026 Oct 9]. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22LYRICA%22
118. Viatris. NEURONTIN (gabapentin): prescribing information. NDA 020235 [Internet]. U.S. Food and Drug Administration; label effective 2026 May 12 [cited 2026 Oct 9]. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22NEURONTIN%22
119. Heindl F, Traschütz A, Synofzik M, Wirth T, Anheim M, Tarnutzer AA, et al. RFC1-related disorders: A case series of 4-aminopyridine and acetyl-DL-leucine treatment. Cerebellum. 2026 Apr 10;25(2):. doi:10.1007/s12311-026-01982-8 PMID: 41961150.
120. Merz Pharmaceuticals. AMPYRA (dalfampridine) extended-release tablets: prescribing information. NDA 022250 [Internet]. Silver Spring (MD): U.S. Food and Drug Administration; label effective 2026 Jan 16 [cited 2026 Oct 9]. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22AMPYRA%22
121. European Medicines Agency. Fampyra (fampridine): EPAR — medicine overview [Internet]. Amsterdam: EMA; authorised 2011 Jul 20; last EC decision 2025 Aug 6 [cited 2026 Oct 9]. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/fampyra
122. Weinberg L, Hungenahally A, Meyerov J, Miles LF, Cox DRA, Muralidharan V. Anaesthesia techniques and advanced monitoring in CANVAS patients - Implications for postoperative morbidity and patient recovery: A case report. Int J Surg Case Rep. 2021 Jun;83:106058. doi:10.1016/j.ijscr.2021.106058 PMID: 34098187.
123. Erdmann H, Schaub A, Lucas MC, Scholz V, Benet-Pages A, Becker K, et al. Repeat-associated ataxias in a German patient cohort analysed by targeted parallel long-read sequencing. Brain. 2026;149:993-1006. PMID: 40898875. doi:10.1093/brain/awaf318.
124. Montaut S, Diedhiou N, Fahrer P, Marelli C, Lhermitte B, Robelin L, et al. Biallelic RFC1-expansion in a French multicentric sporadic ataxia cohort. J Neurol. 2021;268:3337-43. PMID: 33666721. doi:10.1007/s00415-021-10499-5.
125. Erdmann H, Schoberl F, Giurgiu M, Leal Silva RM, Scholz V, Scharf F, et al. Parallel in-depth analysis of repeat expansions in ataxia patients by long-read sequencing. Brain. 2023;146:1831-43. PMID: 36227727. doi:10.1093/brain/awac377.
126. Tagliapietra M, Cardellini D, Ferrarini M, Testi S, Ferrari S, Monaco S, et al. RFC1 AAGGG repeat expansion masquerading as chronic idiopathic axonal polyneuropathy. J Neurol. 2021;268:4280-90. PMID: 33884451. doi:10.1007/s00415-021-10552-3.
127. Bronstein AM. Cerebellar ataxia, neuropathy and vestibular areflexia syndrome (CANVAS). Arq Neuropsiquiatr. 2025 Nov;83(11):1-2. doi:10.1055/s-0045-1813234 PMID: 41330564.
128. Aschoff CAM, Silva TO, Hasan A, Migliorini E, Donis KC, Rare Genomes Project Consortium, et al. The Real Life of Ataxia Patients Without a Vertical Family History: a Twenty-year Experience in South Brazil. Cerebellum. 2026 Aug 19;25(5):. doi:10.1007/s12311-026-02067-2 PMID: 42616279.
129. Ataxia patient associations: Euro-ATAXIA (https://euro-ataxia.eu/), APAHE — Portuguese Association of Hereditary Ataxias (https://www.apahe.pt/), ABAHE — Brazilian Association of Hereditary and Acquired Ataxias (https://www.abahe.org.br/), National Ataxia Foundation (https://www.ataxia.org/) [Internet; cited 2026 Oct 9].