Prevalence of intronic repeat expansions in the RFC1 gene in Polish patients with cerebellar syndrome

Abstract

Cerebellar ataxia with neuropathy and vestibular areflexia syndrome (CANVAS) is a recessively inherited neurodegenerative ataxic disorder, which has been associated with intronic biallelic repeat expansions in the RFC1 gene. Our objective was to assess retrospectively the prevalence of CANVAS in Polish population. We screened 2523 Polish patients in whom other repeat expansions were excluded. To determine the repeat expansions in the RFC1 gene in patients, we performed RFC1-flanking PCR and repeat primed PCR (RP-PCR) and to measure the size of the expansion we used Southern blotting and optical genome mapping to compare the results. We have observed the biallelic pathogenic motif/unit AAGGG expansions in 4.6% and expansions of non-pathogenic motifs AAAAG, AAAGG in 25% patients of our studied population. This is the first large-scale cohort study that confirms the relatively frequent occurrence of the CANVAS in Polish population. To increase the current diagnostics of late-onset ataxias within an unexplained molecular background, we suggest involving the RFC1 repeat expansions analysis to the routine diagnostic workflow.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s13353-025-01000-3.

Introduction

Neurodegenerative diseases are a group of disorders that affect the central nervous system, leading to progressive neuronal loss and dysfunction (Behl et al. 2021). Most of them are characterized by the accumulation of abnormal proteins and the degeneration of specific brain regions (Agrawal 2020). Among them hereditary ataxias constitute a group of neurological degenerative disorders that affect coordination and balance.

Cerebellar ataxia with neuropathy and vestibular areflexia syndrome (CANVAS) was first clinically described in 2004 by Szmulewicz et al.(Szmulewicz et al. 2011) and further delineated in 2011 by the same group and suspected to be a disease with a genetic background and a later defined diagnostic criteria (Szmulewicz et al. 2011, 2016). In 2019, an intronic biallelic repeat expansions found in the replication factor subunit 1 gene (RFC1) was linked to the patients presenting with CANVAS (Cortese et al. 2019)—a neurodegenerative disorder characterized by a typical triad of symptoms: cerebellar ataxia, sensory neuropathy and vestibular dysfunction (Davies et al. 2022). Usually, patients present with slowly progressive cerebellar ataxia with neuropathy and complain of sensory impairment such as pins and needles and/or neuropathic pain. The vestibular areflexia is presented with balance impairment and falls. Clinicians also report that chronic or spasmodic cough (Szmulewicz et al. 2016) is very characteristic of CANVAS syndrome and can start up even to three decades before neurological symptoms. Other common features may include an autonomic dysfunction, parkinsonism, and pyramidal signs (Szmulewicz et al. 2016). The microsatellite repeats located within the second intronic region of the RFC1 gene are polymorphic and usually form (AAAAG)₁₁ tracts, whereas a biallelic pentanucleotide repeat expansion (AAGGG)_250–4000 contributes to the symptomatic form of the CANVAS disease (Dominik et al. 2023; Currò et al. 2024). In addition, thanks to advances in sequencing technologies, novel repeat patterns in the RFC1 are being described and they include novel pathogenic motifs, such as ACAGG, AGGGC or mixed expanded alleles of AAAGG/AAGGG as well as rare large uninterrupted AAAGG (Dominik et al. 2023; Akçimen et al. 2019). It has been shown that RFC1 expansion disorder might be a common and possibly underdiagnosed cause of the adult-onset ataxia and its prevalence is still underestimated (Cortese et al. 2019).

In this study, we investigated the prevalence of intronic repeat expansions in the RFC1 gene, including AAGGG, AAAGG and AAAAG expansions, in a cohort of 2523 Polish ataxia patients with excluded spinocerebellar ataxias (SCAs) dynamic mutations. This represents the first large-scale cohort screened in the Polish population.

Material and methods

Patients

This study encompasses a large cohort of patients with progressive cerebellar ataxia syndromes of uncertain genetic origin, evaluated through genetic testing and clinical data collection over multiple years in Poland.

A total of 2523 patients with a clinical suspicion of spinocerebellar ataxias (SCAs) were retrospectively screened for RFC1 repeat expansions, following referral by neurologists from across the country to the Genetic Laboratory at the Institute of Psychiatry and Neurology (IPiN). All individuals were clinically diagnosed with cerebellar ataxia, though detailed medical data were limited, with full clinical context (family history and main neurological signs) available for approximately 2130 cases (84.44%), based on referral forms.

Among these, a subset of 1125 (44.58%) patients underwent in-depth clinical and imaging evaluation in a tertiary neurological center. This subset was distinguished by the following:

• Clinical signs suggesting inherited or neurodegenerative ataxia,

• Negative results for dynamic mutations in common SCA subtypes (1, 2, 3, 6) and DRPLA,

• Additional negative testing for Huntington’s disease (HD) and Friedreich’s ataxia (FRDA) in selected individuals.

The mean age at examination for the entire screened population was 44.95 years (± 19.45), with 1220 (48.3%) female patients. The majority presented with cerebellar syndrome, including gait disturbances, dysarthria, and imbalance. Cerebellar atrophy was a frequent MRI finding. All studied individuals were index cases. Among the 2523, 560 patients (22.2%) had a positive family history, typically involving balance issues and tremor, while the remaining 1963 (77.8%) were sporadic cases. All patients were referred between 1997 and 2020.

The deeply phenotyped subset was further characterized by the following:

• Cerebellar syndrome in over 46.73%,

• Cerebellar atrophy confirmed on MRI in 37.19%,

• Frequent pyramidal and extrapyramidal signs, spastic paraparesis or tetraparesis, and speech disorders,

• Less commonly, cognitive decline, involuntary movements, seizures, and sensory neuropathy. For more details, see Table 1.

Table 1.

Clinical description of the study group

	No. of patients	% of study group
Neurological symptoms
Cerebellar symptoms
Cerebellar syndrome	520	20.60
Ataxia	177	7.04
Dysarthria	228	9.05
Intention tremor	101	4.02
Head tremor	51	2.01
Resting tremor	25	1.01
Nystagmus	76	3.02
Movement disorders
Balance disturbances	165	6.53
Gait disturbances	127	5.03
Pyramidal syndrome	152	6.03
Spastic tetraparesis	76	3.02
Spastic paraparesis	13	0.50
Extrapyramidal syndrome	63	2.51
Muscular symptoms and neuropathies
Muscle atrophy	25	1.01
Polyneuropathy	63	2.51
Cognitive and psychiatric symptoms
Dementia syndrome/intellectual regression	76	3.02
Seizure and other symptoms
Epileptic seizures, myoclonus	25	1.01
Urinary sphincter control issues	13	0.50
Imaging findings:
Cerebellar atrophy	799	31.66
Cortico-subcortical atrophy	139	5.53
Excluded genetic disorders
SCA (Spinocerebellar ataxias) types 1, 2, 3, 6; DRPLA	1876	74.37
HD (Huntington’s disease)	279	11.06
FRDA (Friedreich’s ataxia)	114	4.52

Control group description

The control group consisted of 537 unrelated, neurologically healthy individuals of Polish origin, recruited from the general population. Women accounted for 50.84% of the cohort. The mean age was 36 years (± 17.82). All participants had no history of neurological disorders, based on self-report and/or clinical documentation. Peripheral blood samples were collected for genetic analysis to assess the frequency of RFC1 repeat motifs in the general population.

The study protocol, including the use of control samples, was approved by the local institutional ethics committee. Informed consent was obtained from all participants prior to sample collection, in accordance with the Declaration of Helsinki.

Methods

To study the prevalence of the pathogenic (AAGGG_n) biallelic repeat expansion in undiagnosed late onset ataxia patients from Poland, we used the following workflow illustrated in Fig. 1.

Fig. 1.

Diagram showing the workflow of patients’ analysis towards repeat expansion in the RFC1 gene. The study started with one group of undefined late-onset ataxia and after flanking and repeat primed PCR (Polymerase Chain Reaction), several patients were selected for southern blot and optical genome mapping analysis to determine their expansion ranges

Flanking PCR, RP-PCR

Flanking PCR and RP-PCR (Repeat-Primed PCR) were performed as previously described by Cortese et al. (Cortese et al. 2019) on genomic DNA (for primers sequences and PCRs conditions, see supplementary Table 1). Flanking PCR enabled to amplify the intronic RFC1 repeat region within the normal range and expansion carriers are identified when only one or none of the products were visible on the capillary electrophoresis. For those individuals RP-PCR was performed to screen for three different potentially expanded motifs. The first group of primers refers to expanded benign motif (AAAAG)_15–200. The second set was used to determine expanded the other benign (AAAGG)_40–1000 repeat expansions. Initially considered non-pathogenic, the AAAGG motif can be recognized as pathogenic if its expansion is sufficiently extensive (> 500 repeats) and occurs together with the AAGGG expansion on the second allele (Dominik et al. 2023). The last set of primers was used for the pathogenic expansion (AAGGG)_250–4000 detection. The CANVAS syndrome is considered to be confirmed if the biallelic expansion of the pathogenic motif AAGGG exceeds 250 repeats (> 6.5 Kb in Southern blot analysis). An ABI 3130 Genetic Analyzer (Applied Biosystems) was used to analyze the flanking PCR and RP-PCR fragment lengths assessment. The results were analyzed using GeneMapper software 4.0 (Applied Biosystems). Flanking PCR method is subject to a size determination error of ± 1 repeat within the accurate range.

Southern blotting

Southern blotting (SB) was performed to estimate the potential size of the expansions as previously described by Cortese et al. (Cortese et al. 2019). Briefly, enzymatic digestion of 5 µg DNA was performed with EcoRI for 3 h at 37 °C. The products were electrophoresed on a 1.2% agarose gel overnight (~ 15 h) and the gel was subsequently washed in depurinating, denaturing, and neutralizing solutions. The blot was then assembled with the DNA transferring overnight (~ 15 h) to a positively charged membrane (Roche Applied Science) using the upward transfer method. The membrane was then subjected to UV transillumination for 3 min to crosslink the DNA. Hybridization with a mixture of salmon sperm and RFC1 specific probe in digoxigenin granules solution (DIG, Roche Diagnostics) followed overnight (~ 15 h). The membrane was washed, blocked, and incubated with anti-DIG antibody. Detection buffer and chemiluminescent substrate (CDP-STAR) were used to visualize the hybridized fragments. The pentanucleotide repeat size was estimated against DIG labelled DNA molecular weight marker II (Roche Diagnostics).

Optical genome mapping

Optical Genome Mapping (OGM; Bionano Genomics, San Diego, CA, USA) was performed to determine the size of the repeat expansions within the RFC1 gene. Ultra-high molecular weight (UHMW) gDNA was isolated from the fresh blood samples frozen at − 80 °C using Bionano Prep SP Frozen Human Blood DNA Isolation kit according to the manufacturer’s instructions (Bionano Genomics). Isolated gDNA was quantitated using BR (broad range) dsDNA assay (Qubit, Thermofisher). Subsequently 750 ng of UHMW gDNA was labeled with Bionano Prep Direct Label and Stain (DLS) kit following manufacturer’s instructions (Bionano Genomics). The labeled and stained gDNA was quantitated using Qubit dsDNA HS (high specificity) assay kit (Thermofisher). Labeled gDNA with a concentration of 4 − 12 ng/µl was loaded onto the Saphyr chips and run on the Saphyr instrument. The de novo assembly pipeline, serving as the reference for this reaction, was applied. Each sample achieved an effective coverage of approximately 100 × with a throughput target of 500 Gbp. The data were analyzed using Bionano Access (Bionano Genomics, San Diego, CA, USA), aligned to the human reference genome GRCh38, and then visualized on the Bionano Access 1.7.1.1 genome browser. The analysis focused on the presence of repeat expansions in intron 2 of the RFC1 gene (hg38: 4:39,287,456–39,366,362).

Results

Microsatellite repeats analysis in patients group

In 1751/2523 individuals, at least one amplified product was observed in the flanking PCR reaction. The remaining 772 patients were further investigated with RP-PCR. In this group, 455 patients showed following expansions: the biallelic pathogenic expansion AAGGG/AAGGG was observed in 116 patients (4.6%), non-pathogenic AAAGG/AAAGG in 20 and AAAAG/AAAAG in 253 patients. We also observed 52 carriers of the pathogenic expansion. Four patients showed AAGGG/AAAGG expansions and 48 showed AAGGG/AAAAG expansions. Additionally, the expansion of an AAAAG/AAAGG repeats was observed in 14 patients (see Table 2.).

Table 2.

Distribution of expanded motifs in the studied population

	Expanded motif	Number of patients	~ % in studied population	Number of healthy individuals	~ % in healthy individuals
Expandedalleles frequency	AAGGG	168	6.7%	26	4.84%
	AAAGG	38	1.5%	22	4.10%
	AAAAG	315	12.5%	46	8.57%
Genotypes	AAGGG/AAGGG	116	4.6%	22	4.10%
	AAGGG/AAAGG	4	0.16%	4	0.74%
	AAGGG/AAAAG	48	1.9%	-	-
	AAAAG/AAAGG	14	0.6%	6	1.12%
	AAAAG/AAAAG	253	10%	40	7.45%
	AAAGG/AAAGG	20	0.8%	12	2.23%

The Southern blotting was performed in 12 patients with > 5 µg of DNA available (Fig. 2.). They all have been selected after RP-PCR specific for AAGGG motif showing a characteristic ladder. The Southern blotting was used to indicate if the pathogenic motif appears as bi- or monoallelic expansion and to measure possible size. Their clinical symptoms included progressive cerebellar syndrome including imbalance, dysarthria and gait problems and showed cerebellar atrophy on MRI. The presence of expansion was confirmed in 8 patients with 3 cases of biallelic expansion in the pathogenic range and 5 cases of heterogenic (pathogenic/non-pathogenic range). In 4 patients, the SB failed to reveal the results due to the poor DNA quality. Following these results, we have applied Optical Genome Mapping (OGM, Bionano Sciences).

Fig. 2.

Southern blotting performed in 12 patient samples. Samples 2, 4, and 7 (red) showed the presence of biallelic expansions in pathogenic range > 6.5 Kb. Patients 3, 6, 9, and 11 (blue) were the carriers of the expansion on one allele (1 allele < 6.5 Kb, 2 allele > 6.5 Kb). Patient 10 (green) had expansion of pathogenic motif (AAGGG) on 2 alleles (AAGGG/AAGGG) but in non-pathogenic range (< 6.5 Kb), while samples 1, 5, 8, and 12 were unsuccessful

Due to the specific requirements of the DNA isolation procedure, OGM was performed on 3 freshly collected blood samples and frozen at − 80 °C from clinical CANVAS syndrome patients. Unfortunately, we could not obtain fresh blood samples necessary for OGM from patients previously tested by the Southern blot method. The three patients were chosen to perform the OGM based on the RP-PCR results, which indicated the presence of biallelic expansion in the first patient and heterogenic expansions (AAAAG/AAGGG) in two others. In the first patient who showed typical CANVAS syndrome (cerebellar ataxia with neuropathy and vestibular areflexia) (Fig. 3A), we identified homozygous pathogenic repeat expansion in the RFC1 gene consisting of 1060 repeats (5302 bp) of AAGGG on one allele and 1188 repeats (5942 bp) on the second allele (Fig. 3A). In the second (Fig. 3B) patient, the expansion of 929 AAGGG repeats (4644 bp) and 443 repeats (2214 bp) expansion of non-pathogenic motif AAAAG was observed. The third (Fig. 3C) patient showed an expansion of AAGGG 825 repeats (4125 bp) and 547 repeats (2733 bp) on AAAAG. Both of them turned out to be the carriers of the pathogenic motif but showed a strong clinical triad of symptoms in line with CANVAS syndrome. Moreover, particularly in Fig. 3C patient, the expansion of the non-pathogenic motif on the other allele was quite broad > 500 repeats.

Fig. 3.

Graphical diagram of the RFC1gene analysis using OGM. A Homozygous repeat expansion in the RFC1 gene in patient 1. B and C Heterozygous repeat expansion of the RFC1 gene in patient 2 and 3, respectively. Green bars represent the reference genome, blue bars indicate de novo genome maps of the studied samples. Black vertical lines mark labeled specific sequence motifs on each bar, and the black lines between the reference and assembled map indicate alignment

Both SB and OGM underscore the potential diagnostic value of sizing expansions; however, these findings are not statistically significant. SB results offer an estimation of potential expansion sizes, though in this study, only a select few expansions were measured.

Microsatellite repeats analysis in control group

In our control group of 537 healthy individuals (1074 chromosomes), flanking PCR analysis revealed 88 heterozygotes with allele sizes of 335–633 bp/345–685 bp, 256 homozygotes 335–633 bp and 193 individuals without product for further RP-PCR. Out of these, 22 individuals (44 chromosomes) showed presence of the pathogenic AAGGG motif on both alleles, yielding a frequency of 4.10% (for more details, see Table 2).

Discussion

In one of the largest studied ataxia cohorts consisted of 2523 patients from Poland in whom the unequivocal genetic cause was not determined, the biallelic pathogenic (AAGGG) repeat expansions in the RFC1 were observed in 4.6% of patients. These results were based on PCR analysis – RP-PCR, a method which cannot estimate the size of the expansions, however, which can indicate the presence or absence of known pathogenic configurations and flanking PCR which can indicate a presence or absence of normal and to some extend -pathogenic alleles. All patients included in the study showed clinical picture of cerebellar ataxia, but they were not divided into subgroups of presented clinical symptoms due to insufficient medical data and not fully determined clinical phenomenology, considering particular sensory neuropathy and vestibulopathy.

Southern blotting was performed on selected patients DNA to confirm RP-PCR results and gain insight into the size of the expanded alleles. The OGM was used in 3 clinically suspected CANVAS patients. The highly restricted procedure of DNA extraction and staining yielded reliable results. The overall bench working time is comparable to the Southern blotting, with the assurance of consistently good-quality results. Despite its expenses, it is worthwhile to apply the procedure for sizing expansions in particular cases. Unfortunately, none of both methods gives the detailed structure of the motif. OGM provides very accurate sizing and SB only approximate size of the expansion. Recent advances in long-read sequencing (LRS) technologies, such as those offered by Oxford Nanopore and PacBio platforms, have introduced highly accurate methods for identifying and characterizing repetitive sequences, including intronic AAGGG expansions in the RFC1 gene (Nakamura et al. 2020). These technologies enable direct measurement of repeat number and structure across long genomic regions, overcoming many of the limitations inherent to traditional PCR-based approaches and SB. Although our current study relied on RP-PCR combined with confirmatory SB and OGM, in selected cases, we acknowledge that LRS represents a comprehensive analysis of pathogenic repeat expansions. Future studies incorporating LRS are likely to enhance diagnostic precision and may uncover additional variability in repeat structures that remain undetectable by conventional methods.

It is increasingly evident that CANVAS is a heterogeneous disorder with different configuration motifs and sizes contributing to the phenotypic spectrum. This is a very important issue in the context of diagnostics and genetic counseling for people from families affected by CANVAS and disease spectrum. The biallelic expansions of the non-pathogenic AAAAG/AAAAG motif is also quite frequent in our studied group, observed in 253 patients (10%), and the frequency of such genotype seems to be too high to have a serious effect on neurogenerative mechanisms (patient 3B—929 AAGGG repeats/443 repeats AAAAG, and patients 3 C AAGGG 825 repeats and 547 AAAAG repeats). However, we cannot exclude that such expansions > 500 repeats may reveal other functional effects within the nervous system. Current knowledge does not allow for precise determination of the pathogenicity of these expansions. Despite suggestions that expansions of other motifs are significant, their role has not been confirmed. Outcomes of our research corresponds with other European populations where the frequency of pathogenic biallelic AAGGG expansion estimates between 0.7% up to 22% of undiagnosed ataxia patients which shows that CANVAS syndrome has been underestimated in clinical diagnosis so far (Cortese et al. 2019; Dominik et al. 2021; Rafehi et al. 2019). Patients were genetically diagnosed – PCR, RP-PCR, to determine the motif, and selected samples were used in SB and OGM to determine their size. Table 3 presents studies on the genetics of CANVAS syndrome conducted across various populations (Cortese et al. 2019, 2020; Akçimen et al. 2019; Aboud Syriani et al. 2020; Fan et al. 2020; Tsuchiya et al. 2020; Beecroft et al. 2020; Kontogeorgiou et al. 2021; Tagliapietra et al. 2021; Montaut et al. 2021; Ghorbani et al. 2022).

Table 3.

Different populations study of CANVAS syndrome genetics

Researchers	Number of studied cases	Summary of the results	Ref
Cortese A. et al. 2019	173	The biallelic expansions of an intronic repeat in the RFC1 gene were a common cause of late-onset ataxia. The researchers analyzed a large cohort of patients with ataxia and identified the repeat expansion in the RFC1 gene in a significant proportion of cases, within the CANVAS familial cohort 23/23 pathogenic expansion and in sporadic cases 33/150 (22%)	Cortese et al. 2019)
Akcimen F. et al. 2019	177	The study investigated the RFC1 repeat expansion in Canadian (1/154, 0.6%) and Brazilian (2/23, 8.7%) ataxia cohorts and identified novel conformations AAGAG and AGAGG in heterozygous states. They found that the expansion frequency in both cohorts was lower than previously reported in other populations	Akçimen et al. 2019)
Cortese A. et al. 2020	363	The study found that the presence of the RFC1 expansion was associated with a distinctive clinical phenotype characterized by sensory ataxia, prominent neuropathy, and vestibular dysfunction which was CANVAS cases 63/70 (90%) and adult-onset ataxia 42/293 (14%). In addition, they identified a novel expansion pattern of (AAGGG)n/(GGGAA)n in one patient	Cortese et al. 2020)
Aboud S,D. et al. 2020	911	In this study, the authors screened a North American cohort with adult-onset ataxia for RFC1 repeat expansions 29/911 (3,2%) and reported that affected individuals had a variable age of onset and clinical presentation, and that the majority had a slowly progressive ataxia phenotype with additional neurological features such as peripheral neuropathy and oculomotor abnormalities	Aboud Syriani et al. 2020)
Fan Y. et al. 2020	193	The study found no evidence of biallelic intronic AAGGG repeat expansion in RFC1 among patients with late-onset ataxia and multiple system atrophy (MSA) among Chinese population	Fan et al. 2020)
Tsuchiya M. et al. 2020	49	The authors identified 3/49 (6.1%) patients with the expansion, all of whom had CANVAS syndrome. They also found novel biallelic ACAGG repeat expansion. The study suggests that RFC1 repeat expansion is a rare but possible cause of late-onset cerebellar ataxia in Japanese populations	Tsuchiya et al. 2020)
Beecroft S. et al. 2020	15	The study identified a specific pathogenic repeat configuration in the RFC1 gene in a group of Māori patients with CANVAS syndrome. The repeat configuration was found to be specific to the Māori population and likely due to a founder allele	Beecroft et al. 2020)
Kontogeorgiou Z. et al. 2021	77	In this study, the researchers identified biallelic pentanucleotide repeat expansions in RFC1 in 3/77 (3.9%) of Greek patients with late-onset ataxia	Kontogeorgiou et al. 2021)
Tagliapietra M. et al. 2021	234	The article describes a patients diagnosed with a RFC1 AAGGG repeat expansion. That presented chronic idiopathic axonal polyneuropathy: pure sensory neuropathy 21/40 (53%), predominantly sensory 10/56 (18%), sensorimotor 3/138 (2%). All together the pathogenic expansion has been found in 14.1% of studied patients	Tagliapietra et al. 2021)
Montaut S. et al. 2021	100	The study identified biallelic RFC1 repeat expansions in a French idiopathic sporadic late-onset ataxia 15/100 (15%), suggesting that this expansion could be an underrecognized cause of sporadic ataxia	Montaut et al. 2021)
Ghorbani F. et al. 2022	626	The study aimed to determine the prevalence of intronic repeat expansions in RFC1 in Dutch patients with CANVAS and adult-onset ataxia. The intronic repeat expansions in RFC1 were present in 55% of the patients presented CANVAS and 2.7% of the patients with adult-onset ataxia. The study suggested that intronic repeat expansions in RFC1 are a relatively rare cause of adult-onset ataxia in the Dutch population	Ghorbani et al. 2022)
Tyagi N. et al. 2023	1998	The aim of the study was to determine prevalence of bi-allelic (AAGGG)exp in Indian ataxia and other neurological disorders. The study group included 1998 uncharacterized ataxia samples. They have found different expanded motifs at RFC1 locus, like AAAAG, AAAGG, AAAGGG, AAAAGG, AAGAG, AACGG, AAGGC, AGAGG in Indian cohort	Tyagi et al. 2023)
Curro R. et al. 2024	2334	The study investigated the role of the repeat size in influencing clinical variables in RFC1 disease. Of 2334 patients, 556 (24%) carried biallelic AAGGG expansions at PCR screening. They confirmed the biallelic pathogenic expansions in 392 patients by SB	Currò et al. 2024)

To date, CANVAS patients are mainly of European origins with a few studies of other populations with different expanded motifs (Tsuchiya et al. 2020; Tyagi et al. 2023). Analysis of the AAGGG core haplotype suggests that the pathogenic repeat expansion arose in Europe, approximately 50,000 years ago (Dominik et al. 2023; Rafehi et al. 2019). The study of Rafehi et al.(Rafehi et al. 2019), it was indicated that in a North American/Australian cohort, the majority of individuals carrying AAGGG expansion were of European ancestry. However, it was also present in non-European individuals i.e., East Asians with admixed Native American ancestry, despite the disorder being highly overrepresented in European populations (Rafehi et al. 2019; Tsuchiya et al. 2020). Also, the pathogenic AAGGG expansion carrier frequency ranges from ~ 1 to 5% in the healthy population as it was previously reported (Cortese et al. 2019; Akçimen et al. 2019; Sułek-Piątkowska et al. 2010). Compared to the previously reported control cohort by Cortese et al.(Cortese et al. 2019), our data suggest a higher prevalence of pathogenic AAGGG motifs in the general population. While Cortese’s study, which included 304 healthy individuals, identified four pathogenic-length expansions (0.7% of 608 chromosomes), our cohort of 537 individuals yielded 48 AAGGG-positive chromosomes via RP-PCR (4.47%). Although we could not directly confirm expansion size in our control group, prior SB analysis in patients showed that 3 out of 8 RP-PCR-positive individuals carried truly pathogenic-length expansions. Extrapolating this proportion to our control group suggests that around 1.67% of alleles might be pathogenic in length (over twice the frequency reported by Cortese et al. (Cortese et al. 2019)). It remains unclear whether the Cortese cohort included heterozygous or homozygous carriers, and whether shorter, non-pathogenic expansions were present, which limits direct comparability but further emphasizes the variability in population data. Nevertheless, it might be beneficial for patients with ataxic disorders but no clear genetic diagnosis to be tested for the RFC1 repeat expansions, particularly if the patients present with sensory neuropathy or/and unexplained cough.

In our study, we have found that some patients with cerebellar ataxia may have (1) pathogenic AAGGG motif only in one allele, (2) expansion of non-pathogenic genotypes AAAGG or AAAAG, (3) heterogenous arrangements of both alleles with pathogenic expansion of AAGGG on one allele and non-pathogenic AAAGG or AAAAG expansion on the other allele; (4) pathogenic genotype AAGGG with normal nucleotide repeats < 250. Our findings showed that the biallelic AAGGG expansion in the RFC1 gene was found in 4.6% of patients with late-onset cerebellar ataxia of unknown cause. Whitmore, if we hypothesize that the presence of only one AAGGG_exp allele (AAGGG_exp/AAAAG_exp or AAAGG_exp) within genotype is sufficient to cause CANVAS, the prevalence of this disorder among undiagnosed ataxias would be even higher (6.7%). However, the significance of such findings in relation to CANVAS phenomenology is unknown. We have only few patients with precise neurological diagnostic work-up so far and one patient with severe clinical symptomatology of CANVAS had AAGGG expansion with 825 repeats in only one allele and the expansion of 547 AAAAG repeats on the other allele, as measured by OGM. We cannot exclude that the abnormalities pointed 1 to 4 are genetic risk factors for the developing of the disease, are the only cause of the disease or are the cause of other, non-CANVAS ataxia. So, it is possible that minority of the cases that not fully blown symptomatology of CANVAS (e.g., ataxia with the absence of vestibulopathy and sensory neuropathy) may be caused by expansion of other than AAGGG allele or even normal repeat length of pathogenic AAGGG motif on both alleles. They may also have other genetic cause of the disease. It is worth mentioning that two patients, who show to be a carrier of the pathogenic motif, were initially suspected of CANVAS syndrome based on clinical observation. These patients underwent reexamination by professionals, confirming clinical presentation of CANVAS syndrome. This gives a further notice to conclude that this is very fragile matter. Determination of genotypes and presented symptoms are not that obvious.

Following other issues of CANVAS syndrome, that had been reported lately (Dominik et al. 2023; Akçimen et al. 2019; Nakamura et al. 2020; Tyagi et al. 2023; Scriba et al. 2020). Tyagi et al. have investigated the RFC1 locus in diverse neurodegenerative outcomes in an India cohort (Tyagi et al. 2023). They have confirmed the presence of the pathogenic motif AAGGG and reported cases with new pathogenic expansion of (AAAGG)_exp/(AAGGG)_exp in the same locus (for different repeat motifs see Table 3.). Another group has identified novel conformations AAGAG and AGAGG (Akçimen et al. 2019). Perhaps, as in the case of other recessive diseases caused by dynamic mutations such as Friedreich's ataxia or Unverricht-Lundborg myoclonic epilepsy, a complex mutation including a repeat expansion on one allele and a point mutation on the other may cause CANVAS syndrome (Benkirane et al. 2022; Ronco et al. 2023). Several other (Cortese et al. 2019, 2020; Akçimen et al. 2019; Aboud Syriani et al. 2020; Fan et al. 2020; Tsuchiya et al. 2020; Beecroft et al. 2020; Kontogeorgiou et al. 2021; Tagliapietra et al. 2021; Montaut et al. 2021; Ghorbani et al. 2022; Tyagi et al. 2023) researchers have been reporting pathogenic motifs in their cohorts and assumptions that different may also appear and that further investigation is required. These findings could correspond to our study, where some of the patients demonstrated no product in flanking and RP-PCR, suggesting the possible presence of different motifs. This raises a question on the exact method of testing CANVAS syndrome patients. In our study, we have limited the examination of the conformation of repeat motifs to RP-PCR. The SB and OGM was used to see whether sizing are appropriate as other groups have reported (Cortese et al. 2019, 2020; Akçimen et al. 2019; Aboud Syriani et al. 2020; Fan et al. 2020; Tsuchiya et al. 2020; Beecroft et al. 2020; Kontogeorgiou et al. 2021; Tagliapietra et al. 2021; Montaut et al. 2021; Ghorbani et al. 2022; Tyagi et al. 2023). Our study confirms accurate sizing is crucial for characterizing patients’ genotypes. Dominik et al. had given implications for genetic testing and clinical diagnosis of variants of the RFC1 repeat expansions and agreed on sizing the microsatellite repeats in clinical diagnosis (Dominik et al. 2023).

From the research conducted so far, the molecular mechanism of pathogenic expansion is not fully elucidated to assess the role of currently known as non-pathogenic or smaller pathogenic expansions. Perhaps, they also influence the development of the disease, posing an open question that needs to be addressed in the future to characterized the genetic content of CANVAS syndrome. Moreover, detection of the pathogenic expansion in patients with other neurological conditions such as MSA or parkinsonism may expand the spectrum of disorders related to mutations in the RFC1 gene.

Conclusion and future directions

Our study confirms that the most prevalent genetic cause of CANVAS is biallelic expansion of the AAGGG repeat in the RFC1 gene. Additionally, we identified atypical pathogenic motifs and non-expanded AAGGG alleles, underscoring the genetic heterogeneity of this disorder and the limitations of single-method diagnostics.

CANVAS is a clinically variable multisystem syndrome involving cerebellar ataxia, sensory neuropathy, and vestibular areflexia. The broad phenotypic spectrum often complicates clinical diagnosis, highlighting the need to refine diagnostic criteria and raise awareness among neurologists of incomplete or atypical presentations.

To improve diagnostic accuracy, comprehensive molecular testing strategies are essential. Screening should include all known RFC1 repeat motifs, with confirmation by multiple orthogonal methods such as RP-PCR, SB, and LRS. This approach minimizes both false-negative and false-positive results, particularly for complex or borderline repeat structures.

Our findings also support the broader inclusion of repeat expansion analysis in the diagnostic workflow of unresolved ataxia syndromes. Systematic screening may uncover new pathogenic mechanisms and enable better classification of neurogenetic diseases.

Integrating RFC1 testing into routine diagnostic panels will not only enhance clinical accuracy and genetic counseling but also support future research into the pathophysiology, modifier factors, and potential therapeutic targets of RFC1-related disorders.

Limitations of a study

This study has several important limitations, primarily related to its retrospective and multicenter nature. The 2523 patients included were referred for genetic testing from clinical centers across Poland and examined by different neurologists, which likely introduced variability in clinical assessment and diagnostic consistency.

• Clinical data were available only in a limited form—typically as brief descriptions on referral forms submitted with blood samples for genetic analysis. As a result, comprehensive clinical phenotyping, including standardized neurological examination, was not available for most cases. Furthermore, the inclusion criteria were based solely on the exclusion of the most common dynamic SCA mutations prevalent in the Polish population (Nakamura et al. 2020), and additional NGS or whole-exome/genome sequencing and LRS was not performed.

• SB was not carried out for RFC1 AAGGG/AAGGG genotypes due to the lack of sufficient high-quality DNA, which is a typical limitation in retrospective studies. OGM was also not performed, as it requires fresh, high-molecular-weight DNA, which was not available for the archived samples. We limited the study to flanking PCR and RP-PCR. Consequently, we were unable to determine the precise size of the detected expansions in many cases.

• Moreover, the clinical presentation of patients (beyond the presence of cerebellar ataxia) was not reviewed or confirmed by an independent neurologist prior to genetic testing. This further limits the ability to correlate specific phenotypes with genotypes in this cohort.

While the study provides valuable insight into the genetic landscape of RFC1 expansions in a large Polish cohort, the absence of standardized clinical data and advanced genomic techniques limits the ability to draw firm conclusions about genotype–phenotype correlations or pathogenic repeat sizes in many cases.

Supplementary Information

Below is the link to the electronic supplementary material.

ESM 1 ^{(14.6KB, docx)}

(DOCX 14.6 KB)

Abbreviations

CANVAS

Cerebellar ataxia with neuropathy and vestibular areflexia syndrome

SCA

Spinocerebellar ataxia

DRPLA

Dentatorubral-pallidoluysian atrophy

Author contribution

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Tomczuk Filip, Ziora-Jakutowicz Karolina, Dominik Natalia, Houlden Henry, Cortese Andrea, Rutkowska Karolina, Pollak Agnieszka, Ploski Rafal, Janik Piotr, Elert-Dobkowska Ewelina, and Sulek Anna. The first draft of the manuscript was written by Tomczuk Filip and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by the Ministry of Science and Higher Education and Institute of Psychiatry and Neurology statutory grant “Retrospective molecular study of patients with ataxia for cerebellar ataxia, neuropathy and vestibular areflexia syndrome – CANVAS”.

Data availability

Anonymized data from this study will be shared only by request from qualified investigator.

Declarations

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Institute of Psychiatry and Neurology Ethics Committee no. 30/2021.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Competing interests

The authors declare no competing interests.