Bilateral vestibulopathy in RFC1-positive CANVAS is distinctly different compared to FGF14-linked spinocerebellar ataxia 27B

The recent discoveries of two new repeat expansion disorders for late-onset cerebellar ataxia (CA), frequently accompanied by vestibulopathy and neuropathy, shed new light on ataxia research. Biallelic intronic repeat expansions in the RFC1 (replication factor C subunit) gene [1, 2] are disease causing in about 90% of Cerebellar ataxia, Neuropathy, and Vestibular Areflexia Syndrome (CANVAS) patients [3, 4], while heterozygous intronic repeat expansions in the Fibro-blast Growth Factor 14 ( FGF14 ) gene explain a relevant number of late-onset CA cases [5–7]. Bilateral vestibulopa-thy (BV) is a characteristic feature of the CANVAS phenotype of RFC1-linked disease, and emerging data suggest that FGF14 repeat expansion carriers may also exhibit vestibular hypofunction [5, 6, 8]. However, differences in the severity of BV, potentially allowing conclusions about the underlying etiology, are yet to be compared between FGF14 and RFC1 repeat expansion carriers. Here, we investigated the role of BV in the differential diagnosis between RFC1-, FGF14-expansion - positive, and repeat-expansion-negative


Dear Sirs,
The recent discoveries of two new repeat expansion disorders for late-onset cerebellar ataxia (CA), frequently accompanied by vestibulopathy and neuropathy, shed new light on ataxia research.Biallelic intronic repeat expansions in the RFC1 (replication factor C subunit) gene [1,2] are disease causing in about 90% of Cerebellar ataxia, Neuropathy, and Vestibular Areflexia Syndrome (CANVAS) patients [3,4], while heterozygous intronic repeat expansions in the Fibroblast Growth Factor 14 (FGF14) gene explain a relevant number of late-onset CA cases [5][6][7].Bilateral vestibulopathy (BV) is a characteristic feature of the CANVAS phenotype of RFC1-linked disease, and emerging data suggest that FGF14 repeat expansion carriers may also exhibit vestibular hypofunction [5,6,8].However, differences in the severity of BV, potentially allowing conclusions about the underlying etiology, are yet to be compared between FGF14 and RFC1 repeat expansion carriers.Here, we investigated the role of BV in the differential diagnosis between RFC1-, FGF14-expansion-positive, and repeat-expansion-negative CA patients.
Patients were recruited through the outpatient clinics at the tertiary referral centers for ataxia and vertigo at the University of Lübeck, Lübeck, Germany.The inclusion criterion was the presence of CA (clinically defined by limb ataxia and/or the presence of cerebellar oculomotor signs), whereas patients with secondary forms (toxic, inflammatory, and paraneoplastic) and known repeat-expansion spinocerebellar ataxias (SCA1, 2, 3, 6, and 17) were excluded.The video head-impulse test (vHIT) [9] was carried out in all patients, and caloric response testing [10] was available in a subset.Genetic and some phenotypic data of several of our RFC1 (19/25, 76%) [4,9] and of all FGF14 [5] repeat expansion carriers have been published before, but here we expand the phenotypic data.
All patients underwent genetic testing for RFC1 and FGF14 repeat expansions.For RFC1, we applied genetic analyses as described [4], including duplex PCR and Sanger sequencing.For FGF14, we performed long-range PCR followed by fragment length analysis as well as repeat-primed PCR and Sanger sequencing.FGF14 repeat expansions were considered disease related with a repeat number > 250 [5].Of note, patients with interrupted GAA repeat expansions/ non-GAA repeat expansions were excluded [11,12].
The sample was subdivided by genetic results in RFC1expansion-positive, FGF14-expansion-positive, and RFC1-and FGF14-expansion-negative individuals.Bilateral vestibulopathy was defined by a mean vestibulo-ocular reflex (VOR) gain < 0.7 assessed by horizontal video head impulse testing [9].In brief, eye and head movements were recorded by a digital video camera (Eye-SeeCam HIT System, Autronics, Hamburg, Germany) at a sampling rate of 220 Hz.At least ten passive and rapid (peak velocity 250°/sec) head movements of small amplitude (10-15°) were performed per side.Head impulses were unpredictable in direction and amplitude.The gain of the horizontal vestibulo-ocular reflex was analyzed at a narrow time interval of 60 ± 10 ms after head movement onset.Only the horizontal vestibulo-ocular reflex was analyzed.Somatosensory impairment (SS) was defined by nerve conduction studies, available for 47/58 patients (81%) or clinically by the presence of a reduced vibrational sense and impaired sense of position at the metatarsophalangeal joint.The presence of BV and SS allowed the clinical differentiation in CANVAS (CA + BV + SS), Cerebellar Ataxia with Bilateral Vestibulopathy (CABV; CA + BV), Cerebellar Ataxia with Somatosensory impairment (CASS; CA + SS), and isolated CA.
Mean ± standard deviation (SD) and the frequency of individuals with percentages are shown.Differences between the groups were analyzed dependent on the data distribution by ANOVA or Kruskal Wallis tests, and, if significant and relevant, post hoc tests were applied.We analyzed the utility of the VOR gain to discriminate between RFC1-expansion-positive and RFC1-expansion-negative CABV patients using a receiver operating characteristic (ROC) curve.p values < 0.05 were considered significant.Analyses were performed using GraphPad Prism 9 and jamovi Version 2.3.
Investigating the capacity of the VOR gain to distinguish RFC1-expansion-positive from FGF14-expansion-positive and repeat-expansion-negative cerebellar ataxia patients with BV, we found a high accuracy (Area under the Receiver Operator Curve (ROC): 0.97, p < 0.0001) in predicting RFC1-expansion positivity (Fig. 1B).Illustrating vHIT results of one RFC1expansion-positive individual and one FGF14 expansion carrier are shown in Fig. 2.
Our study revealed a mildly reduced angular VOR gain in two-thirds of FGF14 repeat expansion carriers, which is in line with the two initial studies, reporting BV in a subset of FGF14 repeat expansion carriers, and a recent study on an independent cohort, where 75% of FGF14-expansion-positive patients had BV [8].Furthermore, BV was present in all RFC1 repeat expansion carriers where the severity of BV was significantly greater compared to FGF14-expansion-positive individuals as well as in repeat-expansion-negative CA patients.This finding is important for at least two reasons: First, our data suggest that the severity of BV aids in clinically distinguishing RFC1-expansion-positive from RFC1-expansionnegative ataxic individuals (including FGF14 repeat expansion carriers) with high accuracy.Second, this difference in the magnitude of VOR gain reduction intimates a potentially different pathophysiological underpinning for the VOR gain reduction in RFC1-and FGF14-linked disease.One option is that the markedly reduced angular VOR gain (< 0.5) reflects peripheral vestibular organ (neuropathy/ganglionopathy) lesions and is hence associated with caloric hyporesponsiveness.In contrast, a mildly pathological vHIT result may indicate cerebellar (floccular) pathology which may not be present on caloric testing [14].We do however note that experimental floccular damage may increase, decrease, or have no effect on VOR gain [15].Degeneration of the vestibular ganglia is well established for CANVAS [16].Although histological analyses in RFC1 repeat expansion carriers are yet to be performed, we found evidence for genuine vestibular dysfunction in RFC1expansion-positive individuals by caloric hyporesponsiveness in all RFC1 repeat expansion carriers (who underwent caloric testing).The distinct underpinnings of BV in FGF14-linked disease warrant further in-depth investigation, including histopathology.In addition, our study confirmed further relevant features of FGF14-linked different from RFC1-linked disease, such as no relevant occurrence of chronic cough and less frequent dysarthria compared to other forms of late-onset cerebellar ataxia [8].
Besides a relatively small sample size, the present study's limitations include a potential underestimation of RFC1expansion positivity, as we did not investigate sequence variants, which were recently highlighted as an additional, albeit rare, cause of RFC1-linked disease [17].Moreover, the retrospective analysis results in partly incomplete data, particularly regarding somatosensory deficits, as nerve conduction studies were not available in 20% of patients.Finally, we reported only data on the horizontal VOR but not on additional vestibulo-ocular investigations, such as the  This work suggests that BV in CA patients with RFC1 repeat expansions is more severe as compared to CA patients with FGF14 repeat expansions, which may find clinical utility as the severity of BV (i.e., VOR gain) may be readily measured by the vHIT and, hence, may facilitate the identification of patients with RFC1-repeat-expansion positivity.Additionally, our data raise the possibility that BV in FGF14 repeat expansion carriers may be caused by cerebellar rather than vestibular dysfunction, although we note the aforementioned variation in the effect of floccular pathology on VOR gain; hence, this hypothesis requires further investigation.

Fig. 1
Fig. 1 BV (box plot of the mean horizontal vestibulo-ocular reflex (VOR) gain of both sides, assessed by video-head-impulse test) in RFC1-expansion-positive, FGF14-expansion-positive, and repeatexpansion-negative late-onset cerebellar ataxia patients.Only individuals with a confirmed mean VOR gain of < 0.7 were included in the analysis.A VOR gain is lower in RFC1 repeat expansion carriers (n = 25) compared to FGF14-positive patients (n = 6) and repeatexpansion-negative cerebellar ataxia patients (n = 12).p < 0.001 (ANOVA), p values of post-hoc tests are shown in the figure.B

Fig.
Fig. 2 Illustrating video headimpulse test (vHIT) results of one RFC1 and one FGF14 repeat expansion carrier with similar age at onset and age at examination.A vHIT examination of a 66-year-old RFC1repeat expansion-positive female with five years disease duration until vHIT was performed (age at onset 61 years).At the time of examination, she had downbeat nystagmus, saccadic pursuit, severe dysarthria, as well as limb, stance, and gait ataxia.Nerve conduction studies confirmed sensory neuropathy corresponding to a CANVAS (Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia Syndrome) phenotype.In keeping with the diagnosis of RFC1-linked disease, she had a chronic cough.A mean VOR gain of 0.16 at 60 ms showed severe VOR impairment.B vHIT examination of a 68-year-old male FGF14 repeat expansion carrier with likewise five years disease duration until the vHIT was performed (age at onset 63 years).During examination, he had mild dysarthria, gait disturbance, limb and gait ataxia, gaze-evoked horizontal nystagmus, and impaired VOR suppression.Neuropathy was excluded by nerve conduction studies, leading to a CABV (Cerebellar Ataxia with Bilateral Vestibulopathy) phenotype.A mean VOR gain of 0.66 at 60 ms revealed mild VOR impairment upon vHIT testing

Table 1
Demographics of the study sample, clinical, and genetic findings Cerebellar oculomotor signs were defined as the presence of saccadic pursuit, gaze-evoked horizontal or downbeat nystagmus Number of individuals and frequencies in percent are displayed.Mean VOR is shown for all individuals as well as for individuals with confirmed BV (mean VOR gain of < 0.7) only.For AAE, AAO, and DD, mean ± standard deviation is shown RFC1 gene encoding replication factor C subunit 1, FGF14 gene encoding Fibroblast Growth Factor 14, AAE age at examination, SD standard deviation, AAO age at onset, DD disease duration, DBN downbeat nystagmus, CANVAS cerebellar Ataxia, neuropathy, and vestibular areflexia syndrome, CABV cerebellar ataxia with bilateral vestibulopathy, CASS cerebellar ataxia with somatosensory impairment, CA isolated cerebellar ataxia, VOR vestibulo-ocular reflex, BV bilateral vestibulopathy