Chronic cough (CC), defined as cough lasting >8 weeks,1 affects 3–12% of the general population2,3 and accounts for around 10% of pulmonology consultations.4 When the cause is not identified after a targeted study, it is classified as unexplained CC (UCC), and when the CC not respond to treatment of possible underlying cause, it is classified as refractory CC (RCC).5 Recently, the European Respiratory Society (ERS) Task Force proposed a new term to encompass both CC types, namely, cough hypersensitivity syndrome,5,6 clinically reflecting CC as an irritative cough that is frequently triggered by stimuli that do not normally produce cough.

A number of neurological diseases have been associated with CC, including Holmes-Adie syndrome7 and sensory neuropathy type 1B,8 although the underlying mechanism is still unknown. Recently, a rare disease called cerebellar ataxia with neuropathy and bilateral vestibular areflexia syndrome (CANVAS) has been postulated as a possible cause of CC, given that CC is highly prevalent (30–97%) in patients with CANVAS.9–12 The most widely accepted hypothesis regarding the underlying mechanism of CC is an alteration of the sensory innervation of the C fibres in both the upper respiratory tract and oesophagus, leading to the hypersensitivity that favours the development of CC.9,13,14 Various studies of patients with CANVAS also report gastroesophageal reflux (GER),9,13,15–17 suggesting that early alteration of esophageal motility may be the possible cause of GER and, in turn, of CC.

The relevance of CC for patients with CANVAS is that CC seems to precede the onset of neurological symptoms by years, and even decades.13,15,18 Therefore, given the possibility of early diagnosis of a progressively disabling neurological syndrome, genetic screening and counselling is important. The genetic cause of CANVAS, an hereditary autosomal recessive disease is, in most cases, the presence of biallelic expansion of the AAGGG repeat motif (AAGGG)exp in intron 2 of replication factor C subunit 1 (RFC1).10 It has also been reported that CANVAS may exceptionally be due to (AAGGG)exp and a truncating variant (nonsense or frameshift),19–21 or (AAAGG)exp when repeats are >500.22 Further alleles involved have also recently been identified (AGGGC, AAGGC and AGAGG).22 Early diagnosis is clearly important in terms of improving quality of life (QoL) for the patient and reducing the CC burden for the health system.23,24

While the precise prevalence of CANVAS is unknown, allelic frequency in (AAGGG)exp in RFC1 is estimated at 0.7% of the healthy population.10 Prevalence of biallelic carriers is therefore estimated as approximately 1 in 20000 births,10 and, since CC is a common symptom of CANVAS, a higher prevalence would be expected in those patients. In fact, that was what was observed in the study by Guilleminault et al.,25 which is, to our knowledge, the only study to date that has analyzed repeat expansion in RFC1 in patients with RCC; those authors found that 16.2% and 8.8% of patients had biallelic and monoallelic expansion, respectively. Subsequent publications raise the possibility of including genetic screening as part of RCC and UCC studies12,26; however, more research is needed to replicate and expand on available results, including, for example, for patients with respiratory pathologies26 and patients with CC undergoing a protocolized study.

This study is one of the first screens for RFC1 mutations in patients with CC describing patient clinical and functional characteristics and evaluate their QoL. Furthermore, it is the first to evaluate use of the Rydel-Seiffer fork in patients with CC to detect possible peripheral vibratory sensitivity impairment, which could be a warning sign to suspect CANVAS.

Ours is one of the first studies that screens for RFC1 mutations in patients undergoing CC studies. The fact that we did not differentiate at study outset between patients with RCC, UCC, or CC resolved after treatment has made it possible to identify which patients should undergo genetic screening, namely, those in the RCC and UCC subgroups. Our study is also the first to demonstrate the usefulness of Rydel-Seiffer fork results for patient selection for genetic screening, suggesting that this test could be used as criterion for onward referral when genetic screening is not available in all specialized CC consultations (and specially for patients with an altered Rydel-Seiffer fort test), whether in the centre itself or through a referral circuit.

Detecting RFC1 mutations as a means of early diagnosis of CANVAS is important to determine a possible underlying cause of CC that have been classified as UCC. As for future treatments for CANVAS, while there is as yet no cure, some of the symptoms can be treated, using for instance, gabapentin. Genetic diagnosis is important to avoid unnecessary examinations in individuals at risk. Identifying the genetic cause in a family's index case makes it possible to identify other family members at risk of developing or transmitting the disease to their offspring. Furthermore, in the case of CANVAS, the high frequency of carriers in the general population (0.7%)10 would indicate that genetic study is especially important in order to provide genetic and reproductive counselling to family members.

We report prevalence rates for biallelic and monoallelic (AAGGG)exp in RFC1 of 6.1% and 18.2%, respectively, higher than estimated for the general population. Our values are largely in agreement with those reported by Guilleminault et al.,25 with the difference that their biallelic expansion rate was lower, and monoallelic expansion rate was higher, than our rates. Interestingly, that same study described only female heterozygous carriers of (AAGGG)exp in RFC1, whereas one of the monoallelic mutation carriers in our study was a man; a possible explanation is that we studied a broader sample of patients, including a subgroup of patients whose CC resolved after treatment, to which this male patient belonged. Regarding candidates for genetic screening, mutation prevalence was greater in patients with RCC, and was also greater in patients with UCC, but only those with monoallelic expansion (50%; n=2). The higher than expected prevalence of heterozygous (AAGGG)exp carriers is somewhat suggestive of a relationship with CC, although a more robust conclusions would require further studies.

Further evaluation of use of the Rydel-Seiffer fork for early detection of altered peripheral vibratory sensitivity seems appropriate, since 2 of the 6 patients with altered Rydel-Seiffer fork results had a positive result for early-stage CANVAS (biallelic (AAGGG)exp in RFC1, without neurological symptoms). The result of this simple and rapid test could serve as an early sign of CANVAS and enable patient selection for genetic screening, or, where a genetic study was not available, for referral to neurology.

Because no statistically significant differences were found in the clinical or functional characteristics of patients with and without (AAGGG)exp in RFC1, we could not define in further detail which specific patients should undergo genetic testing. In the complementary tests, we only detected a statistically significant difference in relation to chest CT results: patients with no alteration more frequently presented with at least one (AAGGG)exp in RFC1 (p=0.03). Although differences in the GER tests were not statistically significant, the 2 patients with biallelic (AAGGG)exp had impaired esophageal motility, suggesting that further study is needed in this regard to understand the mechanisms underlaying CC.

The questionnaires, which confirmed a poorer QoL that especially affected the psychosocial sphere, corroborate the results of previously published studies.23

The main strength of our study is that it was supported by a multidisciplinary team of neurologists, geneticists, gastroenterologists, and pulmonologists. Limitations include the small sample size – due to the high cost associated with complementary testing and especially genetic studies of patients with CC – and the fact that the study was conducted in a single centre. Another limitation was not having access to long-read sequencing to determine if the heterozygous carrier of (AAGGG)exp/(AAAGG)exp had >500 repeats in the (AAAGG)exp allele. Likewise, we were unable to analyze the possible presence of recently described exceptionally pathogenic motifs associated with CANVAS (AGGGC, AAGGC y AGAGG), although no case has been described to date in Spain. Furthermore, patients with CC showing neurological symptoms compatible with CANVAS were not analyzed because this was a reason for exclusion, and a more exhaustive neurological study would possibly have identified other neurological causes of CC.

Genetic screening for (AAGGG)exp in RFC1 should be considered in specialized CC consultations, especially of patients with RCC and UCC. Also meriting consideration in these consultations is use of the Rydel-Seiffer fork, as a low-cost, rapid, and simple complementary test that could identify which patients with CC should be referred for neurological and/or genetic studies. Finally, all CC studies it is important to alert to possible symptoms indicative of CANVAS, in the interest of early diagnosis and of benefit for both the patient and the health system.

Substantial contributions to study conception and design of the study: EP, VP, LGQ, PG, LQ, FB, MR, ASC, ACL. Acquisition, analysis, and interpretation of data: EP. Drafting the article: EP, ACL. Revising the article critically for important intellectual content and final approval of the version to be submitted: all co-authors.

This research was supported by a grant from the Spanish Society of Pulmonology and Thoracic Surgery (SEPAR) [1144, 2021]. Also for this research, EP was supported by the Catalan Society of Pulmonology (SOCAP) [IIBSP-TOS-2020-143, 2022], LGQ and PG by the Instituto de Salud Carlos III and co-funded by EDRF/FEDER, ‘Una manera de hacer Europa’ under grant from CIBERER [ACCI ER21P1AC705, 2020], and ASC by a grant from the Spanish Ministry of Universities [FPU20/06692, 2020].

EP has received, in the last 3 years, conference travel and attendance expenses from AstraZeneca, Sanofi, Gebro Pharma, Chiesi, FAES Farma, and GSK, and for talks at meetings sponsored by AstraZeneca and GSK, and has received funds/grants for research projects from state agencies, non-profit foundations, and GSK.

VP has received, in the last 3 years, honoraria for speaking at sponsored meetings from Astrazeneca, Boehringer-Ingelheim, Chiesi, Gebro, GSK, Luminova-Medwell, and Sanofi, has received assistance with travel expenses from Astrazeneca and Chiesi, and has acted as a consultant for Astrazeneca, Chiesi, GSK, and Menarini.

LGQ has received conference travel and attendance expenses and fees for talks at meetings sponsored by PTC Therapeutics, and has received funds/grants for research projects from state agencies.

PG, LQ, FB, MR and ASC declare no conflicts of interest.

ACL has received, in the last 3 years, fees for talks at meetings sponsored by AstraZeneca, Zambón, Boehringer Ingelheim, Chiesi, SANOFI, GSK, MSD, Novartis, Orion Pharma, and Sanofi, has received travel and attendance expenses for conferences from Gebro, GSK, Novartis, AstraZeneca, and Sanofi, and has received funds/grants for research projects from several state agencies, non-profit foundations, and AstraZeneca and GSK.