ATXN7-Related Cone-Rod Dystrophy

Key Points

Question

What are the ophthalmic features of the spinocerebellar ataxia type 7 (SCA7) carriers with preataxia included in this study?

Findings

A cross-sectional study of 15 SCA7 carriers revealed correlations between the number of CAG repeats and visual assessments in 9 carriers with preataxia, including visual acuity, retinal sensitivity, and outer nuclear layer thickness. For carriers with ataxia, cerebellar features progressed substantially together with ophthalmic abnormalities.

Meaning

The correlations between SCA7 disease severity and ophthalmic manifestations found in this study emphasize the potential of visual assessments as biomarkers for disease progression and therapeutic evaluation.

This study describes baseline ophthalmological biomarkers in spinocerebellar ataxia type 7 carriers included in the Integrated Functional Evaluation of the Cerebellum (CERMOI) cross-sectional study.

Abstract

Importance

Reliable biomarkers with diagnostic and prognostic values are needed for upcoming gene therapy trials for spinocerebellar ataxias.

Objective

To identify ophthalmological biomarkers in a sample of spinocerebellar ataxia type 7 (SCA7) carriers.

Design, Setting, and Participants

This article presents baseline data from a cross-sectional natural history study conducted in Paris, France, reference centers for rare diseases from May 2020 to April 2021. Data were analyzed from September to December 2022. Fifteen adult ATXN7 pathogenic expansion carriers (9 with preataxia and 6 with ataxia) were included, all with a Scale for the Assessment and Rating of Ataxia (SARA) score of 15 of 40 or lower. Patients were recruited at the Paris Brain Institute, and all contacted patients accepted to participate in the study.

Main Outcomes and Measures

Three visits (baseline, 6 months, and 12 months) were planned, including neurological examination (SARA and Composite Cerebellar Functional Severity Score), ophthalmological examination (best-corrected visual acuity, microperimetry, full-field electroretinogram, optical coherence tomography, and fundus autofluorescence imaging), and neurofilament light chain (NfL) measurements. Here we report the baseline ophthalmic data from the cohort and determine whether there is a correlation between disease scores and ophthalmic results.

Results

Among the 15 included SCA7 carriers (median [range] age, 38 [18-60] years; 8 women and 7 men), 12 displayed cone or cone-rod dystrophy, with the number of CAG repeats correlating with disease severity (ρ, 0.73, 95% CI, 0.34 to 0.90; P < .001). Two patients with cone-rod dystrophy exhibited higher repeat numbers and greater ataxia scores (median [range] SARA score, 9 [7-15]) compared to those with only cone dystrophy (median [range] SARA score, 2 [0-5]). A correlation emerged for outer nuclear layer thickness with SARA score (ρ, −0.88; 95% CI, −0.96 to −0.59; P < .001) and NfL levels (ρ, −0.87; 95% CI, −0.86 to 0.96; P < .001). Moreover, ataxia severity was correlated with visual acuity (ρ: 0.89; 95% CI, 0.68 to 0.96; P < .001) and retinal sensitivity (ρ, −0.88; 95% CI, −0.96 to 0.59; P < .001).

Conclusions and Relevance

In this cross-sectional study, retinal abnormalities were found at preataxic stages of the disease. Most of the carriers presented with cone dystrophy and preserved rod function. The outer nuclear layer thickness correlated with SARA score and plasma NfL levels suggesting nuclear layer thickness to be a biomarker of disease severity. These findings contribute to understanding the dynamics of SCA7-related retinal dystrophy and may help lay the groundwork for future therapeutic intervention monitoring and clinical trials.

Trial Registration

ClinicalTrials.gov Identifier: NCT04288128

Introduction

Spinocerebellar ataxias (SCAs) are a group of inherited neurological disorders characterized by a progressive degeneration of the cerebellum, brainstem, and spinal cord. To date, 50 subtypes of SCA and 39 associated genes have been identified and are transmitted in an autosomal dominant manner. However, clinical manifestations exhibit significant variability across these disorders, reflecting their genetic heterogeneity. The most frequent and severe subtypes are caused by pathological CAG repeat expansion encoding for a polyglutamine stretch in the corresponding protein. Oculomotor abnormalities are often present in patients with SCA even at preataxic and early disease stage and correlate with brainstem atrophy and CAG repeat size. In some SCA subtypes, the optic nerve and the retina may be primarily involved. Spinocerebellar ataxia type 7 (SCA7) is the most known SCA form associated with retinal degeneration. The toxic effect of the polyglutamine stretch in ataxin-7 (coded by ATXN7, OMIM*607640), a protein highly expressed in the nuclei and the inner segments of photoreceptors, is responsible for the dysregulation of several cellular functions leading to their death. Visual symptoms usually include decreased visual acuity, photophobia, and dyschromatopsia. Fundus examination may range from normal appearance (ie, no retinopathy or occult maculopathy) to a cone-rod dystrophy. There is no current treatment for these patients; however, advancing technologies on gene silencing techniques (eg, RNA interference and antisense oligonucleotides) have shown encouraging results in mouse models, opening the path for future clinical trials. Of course, this demands a better knowledge of the natural history of SCA7 and the identification of biomarkers that could be used to monitor the efficacy of an experimental therapy. The Integrated Functional Evaluation of the Cerebellum (Evaluation fonctionnelle intégrée du cervelet [CERMOI]) study (NCT04288128) aims to meet this objective, providing a translational and integrated overview of cerebellar dysfunction in SCA7 carriers over 1 year. In this study, we report the ophthalmic characteristics at baseline of SCA7 carriers enrolled in the CERMOI study.

Methods

ATXN7 pathogenic expansion carriers from reference centers for rare diseases in Paris, France, were included in the study. This study followed the tenets of the Declaration of Helsinki and was approved by the French ethics committee. The study was conducted from May 2020 to April 2021, and data were analyzed from September to December 2022. All participants provided written informed consents. In this study, participants did not receive any additional incentives, stipends, or compensation for their participation. Visits included a neurological examination and plasma neurofilament light chain (NfL) measurement at Paris Brain Institute (Institut du Cerveau [ICM]) and an ophthalmological examination at the National Reference Center for Rare Retinal Diseases REFERET of the Quinze-Vingts Hospital, Paris, France. Reporting for this article followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Neurological Examination

A complete neurological examination was performed to stage the disease. Scale for the Assessment and Rating of Ataxia (SARA) and the Composite Cerebellar Functional Severity (CCFS) scores were measured as previously reported. Only those with a SARA score of 15 of 40 or less were included in the study. Patients were classified as having preataxia if their SARA score was less than 3 of 40.

NfL Measurement

We measured plasma NfL on fasting blood. All biosamples were frozen at –80 °C after collection and stored in the local biobank. NfL levels were measured in duplicate using an ultrasensitive single-molecule array on the Simoa HD-1 Analyzer (Quanterix), as previously described.

Ophthalmic Examination

Ophthalmic history, including previous surgical procedures and concomitant or past ocular diseases, was collected. A complete ophthalmic examination was then performed, including best-corrected visual acuity (BCVA) using the Early Treatment Diabetic Retinopathy Study (ETDRS) charts, slitlamp examination, color vision tested by the Cambridge color test (Metropsis; Cambridge Research System), mesopic microperimetry (Macular Integrity Assessment device; MAIA, CenterVue Spa), full-field electroretinography (Espion E2; Diagnosys) performed according to standards of the International Society for Clinical Electrophysiology of Vision, spectral-domain optical coherence tomography (SD-OCT; Spectralis Heidelberg Retina Angiograph and OCT; Heidelberg Engineering), retinal fundus photographs, and short-wavelength and near-infrared fundus autofluorescence (Heidelberg Retina Angiograph II; Heidelberg Engineering) imaging. All imaging modalities were evaluated by 2 independent operators (N.K. and B.B.) to check for any abnormalities. In case of disagreement, an open adjudication was performed in the presence of a retinal imaging expert (I.A.) who was responsible for the final decision. If a central area of hypoautofluorescence was present on short-wavelength or near-infrared fundus autofluorescence, the surface was calculated separately by 2 operators (N.K. and B.B.). SD-OCT images were used to automatically obtain the thickness of the outer nuclear layer at the foveal center (1-mm diameter) after careful manual correction of any segmentation error, when present (Heidelberg Eye Explorer version 1.9.10.0; Heidelberg Engineering). For mesopic microperimetry, testing parameters included a Goldmann size III stimulus, 200-millisecond presentation time; 4-2 strategy threshold, and 68 stimulus points distributed within a circular area approximately 20° in diameter, centered on the fovea. The mean sensitivity in the overall pattern was extracted. Examinations with a high fixation loss rate (>30%) or a duration longer than 20 minutes were excluded from the analysis. For full-field electroretinography, patients’ data were compared against a data set of healthy individuals. Normal ranges were defined for all electroretinogram components as the mean value (±2 SDs), as previously reported.

Statistical Analysis

Data analysis was performed using SPSS version 21.0 (IBM) and R version 4.2.2 (R Foundation). First, the agreement between eyes was tested for all quantitative variables; since there were no differences (eTables 1 and 2 in Supplement 1), all further analyses were performed using the data from the worse eye. The Spearman correlation coefficient was calculated between the SARA score, the CCFS score, and the NfL levels with variables from the ophthalmic examination using the Holm-Bonferroni correction.

Results

We included 15 SCA7 carriers (9 with preataxia and 6 with ataxia) (Table 1). The median (range) age was 38 (18-60) years; there were 8 women and 7 men. CAG repeats in the ATXN7 gene ranged between 38 and 62 (median = 41). Overall, the median (range) SARA score at baseline was 2 (0-15), and the median (range) CCFS score was 0.85 (0.77-0.12). The median (range) plasma NfL levels were 15.53 (6.31-51.57) pg/mL.

Table 1. Clinical and Genetic Characteristics at Baseline.

Characteristic	SCA7 allele (N = 15)
Sex, No. (%)
Female	8 (53)
Male	7 (47)
CAG repeat length of ATXN7 short alleles, median (IQR) [range]	10 (10-10) [10-14]
CAG repeat length of ATXN7 expanded alleles, median (IQR) [range]a	41 (39-47) [38-62]
Age at baseline, median (IQR) [range], y	38 (29-40) [18-60]
Age at first ophthalmic symptoms, y
No.	9
Median (IQR) [range]	36 (17-48) [6-60]
Ophthalmic disease duration
No.	9
Median (IQR) [range], y	8 (5-11) [0-12]
SARA score, median (IQR)b	2 (0-11.5)
CCFSc	0.85 (0.84-1.03)
Plasma NfL, median (IQR), pg/mL	15.53 (13.27-23.23)

Abbreviations: CCFS, composite cerebellar functional severity score; NfL, neurofilament light chain; SARA, Scale for the Assessment and Rating of Ataxia; SCA7, spinocerebellar ataxias type 7.

^a Normal value < 36 CAG repeats.

^b Maximum value = 40.

^c Normal mean (SD) = 0.85 (0.05).

Functional Examinations

The mean (SD) BCVA was 74.53 (28.23) ETDRS letter score (about 20/32 Snellen equivalent). Color vision was impaired in 5 carriers (33%), with no preferential axis. Thirteen carriers could complete the microperimetry, while 2 were excluded for the long examination duration due to low vision (20/200 and 20/800) and unstable fixation. The mean (SD) retinal sensitivity was 23.87 (3.92) dB (reference database range: 24-36 dB). Full-field electroretinography showed an impairment of the cone system in 9 carriers (60%); 2 of them (13%) had both photopic and scotopic functions altered, with undetectable traces in 1 patient (7%). The number of CAG repeats correlated with disease severity (ρ, 0.73; 95% CI, 0.34-0.90; P < .001) (Figure 1). In 6 carriers with no photoreceptor impairment (median age: 43 years), CAG repeats were 40 or less and SARA scores were less than 3 of 40 (Figure 1). The 2 patients with rod and cone dysfunction (aged 18 and 38 years) had 62 and 46 CAG repeats with a SARA score of 15 and 12, respectively (Figure 1). Finally, the number of CAG repeats in carriers with only cone dysfunction (n = 7; median age: 42 years) ranged between 38 and 49, with a SARA score ranging between 0 and 15 (Figure 1).

Figure 1. Distribution of Age, Number of CAG Repeats, and Scale for the Assessment and Rating of Ataxia (SARA) Score Based on Full-Field Electroretinography Alterations.

Imaging Analysis

Three carriers (20%) had no retinal abnormality on any imaging modality. They all had a SARA score of 1 or lower and a CCFS score less than 0.85. All other carriers presented with cone dystrophy or cone-rod dystrophy at variable stages (eFigure in Supplement 1). Alterations included a central area of hypoautofluorescence on short-wavelength and near-infrared fundus autofluorescence, a foveal disruption of the photoreceptor ellipsoid zone, and/or a central outer retinal atrophy on SD-OCT. Specifically, hypoautofluorescence was detectable and measurable in the short-wavelength fundus autofluorescence of 7 carriers (47%; mean [SD] surface, 0.54 [0.72] mm²) and in the near-infrared fundus autofluorescence of 12 carriers (80%; mean [SD] surface, 0.42 [0.78] mm²), half of them in the preataxic stage. The ellipsoid zone was disrupted and measurable in 10 carriers (67%; mean [SD] horizontal diameter, 937.42 [877.01] μm; mean [SD] vertical diameter, 1329.89 [2600.8] μm). The overall cohort had a mean (SD) central outer nuclear layer thickness of 69.97 (31.25) μm.

Correlations

SARA score showed a correlation with BCVA (ρ, 0.89; 95% CI, 0.68 to 0.96; P < .001) and mean retinal sensitivity (ρ, −0.88; 95% CI, −0.96 to −0.59; P < .001) (Figure 2). Among all imaging parameters, central outer nuclear layer thickness correlated with SARA score (ρ, −0.88; 95% CI, −0.96 to −0.59; P < .001) and plasma NfL levels (ρ, −0.87; 95% CI, −0.86 to −0.96; P < .001) (Table 2).

Figure 2. Correlations Between the Scale for the Assessment and Rating of Ataxia (SARA) Score and Best-Corrected Visual Acuity (BCVA), Retinal Sensitivity on Microperimetry, and Outer Nuclear Layer Thickness on Optical Coherence Tomography.

ETDRS indicates Early Treatment Diabetic Retinopathy Study.

Table 2. Correlations Between Ophthalmic Data and Pathological CAG Repeats, Scale for the Assessment and Rating of Ataxia (SARA) Score, and Plasma Neurofilament Light Chain (NfL).

Variable	CAG repeats (95% CI)a	P value	SARA score (95% CI)b	P value	Plasma NfL (95% CI)	P value
Functional parameters
BCVA	0.70 (0.26 to 0.89)	.004	0.89 (0.68 to 0.96)	<.001	0.71 (0.29 to 0.89)	.003
Color vision scores
Protan	0.77 (0.40 to 0.92)	.001	0.81 (0.49 to 0.93)	<.001	0.64 (0.16 to 0.86)	.01
Deutan	0.71 (0.29 to 0.89)	.003	0.87 (0.61 to 0.95)	<.001	0.66 (0.21 to 0.87)	.009
Tritan	0.75 (0.35 to 0.91)	.001	0.91 (0.74 to 0.97)	<.001	0.68 (0.23 to 0.88)	.007
Microperimetry
63% BCEA area	0.34 (−0.27 to 0.74)	.26	0.67 (0.17 to 0.89)	.01	0.44 (−0.17 to 0.76)	.13
95% BCEA area	0.34 (−0.27 to 0.74)	.26	0.67 (0.17 to 0.89)	.01	0.44 (−0.17 to 0.76)	.13
Average threshold	−0.36 (−0.77 to −0.28)	.25	−0.88 (−0.96 to −0.59)	<.001	−0.62 (−0.87 to −0.04)	.04
Full-field electroretinogram
Dark 0.01-amplitude	−0.61 (−0.85 to −0.12)	.02	−0.28 (−0.69 to −0.29)	.32	−0.19 (−0.63 to 0.37)	.50
Dark 0.01-culmination time	0.11 (−0.43 to 0.59)	.70	0.26 (−0.30 to 0.68)	.35	0.42 (−0.13 to 0.76)	.12
Dark 3.0-amplitude	−0.78 (−0.92 to −0.42)	.001	−0.47 (−0.79 to 0.07)	.08	−0.42 (−0.76 to 0.14)	.12
Dark 3.0-culmination time	0.22 (−0.34 to 0.65)	.44	0.45 (−0.10 to 077)	.10	0.26 (−0.30 to 0.68)	.35
Dark 10.0-amplitude	−0.75 (−0.91 to −0.37)	.001	−0.63 (−0.86 to −0.15)	.01	−0.36 (−0.73 to 0.19)	.18
Dark 10.0-culmination time	0.24 (−0.32 to 0.66)	.40	0.21 (−0.30 to 0.65)	.46	0.03 (−0.49 to 0.53)	.92
Light-adapted flicker amplitude	−0.75 (−0.91 to −0.36)	.001	−0.65 (−0.87 to −0.19)	.008	−0.59 (−0.84 to −0.09)	.02
Light-adapted flicker period	−0.10 (−0.58 to 0.44)	.73	0.07 (−0.46 to 0.56)	.81	−0.13 (−0.60 to 0.42)	.65
Structural parameters
SW-FAF (n = 8)
Surface	0.84 (0.267 to 0.97)	.01	0.59 (−0.23 to 0.91)	.12	0.66 (−0.14 to 0.92)	.08
Horizontal diameter	0.55 (−0.29 to 0.90)	.17	0.17 (−0.61 to 0.78)	.68	0.31 (−0.52 to 0.83)	.46
Vertical diameter	0.38 (−0.46 to 0.85)	.36	0.10 (−0.66 to 0.75)	.82	0.14 (−0.63 to 0.77)	.75
NIR-FAF (n = 12)
Surface	0.36 (−0.29 to 0.77)	.25	0.18 (−0.45 to 0.68)	.58	0.36 (−0.29 to 0.77)	.25
Horizontal diameter	0.67 (0.13 to 0.89)	.02	0.53 (−0.08 to 0.84)	.07	0.59 (0.00 to 0.86)	.05
Vertical diameter	0.31 (−0.34 to 0.74)	.33	0.20 (−0.43 to 0.69)	.52	0.40 (−0.24 to 0.79)	.19
Optical coherence tomography
Horizontal diameter (n = 9)	0.37 (−0.41 to 0.82)	.33	0.25 (−0.50 to 0.78)	.51	0.18 (−0.56 to 0.75)	.64
Vertical diameter (n = 9)	0.49 (−0.28 to 0.86)	.18	0.43 (−0.35 to 0.84)	.25	0.37 (−0.41 to 0.82)	.34
Outer nuclear layer thickness (n = 12)	−0.75 (−0.92 to −0.27)	.003	−0.88 (−0.96 to −0.59)	<.001	−0.87 (−0.86 to −0.96)	<.001

Abbreviations: BCVA, best-corrected visual acuity; BCEA, bivariate contour ellipse area; NIR-FAF, near-infrared fundus autofluorescence; SW- FAF, short-wavelength fundus autofluorescence.

^a Pathologic allele.

^b Maximum value 40.

Discussion

In this cross-sectional study, we reported ophthalmological abnormalities in SCA7 carriers with preataxia and ataxia assessed by a multimodal functional assessment and imaging approaches as part of the baseline CERMOI study. We found retinal abnormalities in the preataxic stage of the disease. Outer nuclear layer thickness was found to be a potential surrogate biomarker with correlations with SARA score and plasma NfL levels.

In SCA7, the retinal phenotype was initially reported as a macular dystrophy and subsequently better defined as a cone or cone-rod dystrophy. Interestingly, in our cohort, most of the carriers presented with a cone dystrophy, demonstrating preserved rod function. However, the occurrence of generalized cone-rod dysfunction is correlated with the CAG repeat size and the severity of the cerebellar ataxia scored by SARA (Figure 1). Therefore, patients with cone-rod dystrophy were younger than the others since they carried larger CAG expansion. Previous reports have consistently shown that age and disease duration were not different when comparing groups of patients in various stages of the retinal dystrophy, suggesting that the pathological allele is more important in determining the retinal phenotype and its progression. Longitudinal studies will be crucial for confirming these hypotheses and subsequently establishing a potential intervention timeframe. Indeed, it is not clear whether cone dystrophy inevitably evolves to cone-rod dystrophy or if rods may be indefinitely preserved in certain cases. This is not trivial, as it may shed further light on the pathogenesis of SCA7-related retinal dystrophy, which is still largely unknown. On this matter, one of the proposed mechanisms of photoreceptor degeneration is related to the interaction between ataxin-7 and CRX, a nuclear transcription factor that is mostly expressed in photoreceptors, and variants of which are associated with an autosomal dominant cone-rod dystrophy. Other mechanisms include interaction with neural retina leucine zipper protein (NRL) and nuclear receptor subfamily 2, group E, member 3 (NR2E3) which alters the differentiation of photoreceptors. Understanding the molecular mechanisms of SCA7 may facilitate the discovery of new therapeutic approaches. Recently, RNA interference–based therapy and antisense oligonucleotides showed encouraging results in SCA mouse models. In this context, natural history studies with genotype-phenotype correlations and deep phenotyping are warranted as they could reveal novel biomarkers for systemic assessment and disease progression. Specifically, ocular biomarkers may be particularly useful as they can be monitored through noninvasive in vivo examinations. Previous genotype-phenotype correlation studies have found a significant association between clinical scores and the number of CAG repeats. However, the correlation between clinical scores and ophthalmological findings has been only partially explored. Recently, Marianelli et al proposed a novel classification of the retinal degeneration based on the fundus examination and OCT imaging. Briefly, the authors defined 4 stages of the disease, ranging from no abnormalities to generalized dystrophy with nummular confluent atrophic lesions. SARA score correlated with ophthalmological findings of this classification. While the latter may particularly be useful for clinicians in assessing the disease, it may be less suitable to evaluate the short-term (ie, 1- or 2-year) progression of the disease, which is required in most therapeutic clinical trials.

Our study attempts to find more precise correlations between neurological scores and retinal dystrophy through a thorough multimodal examination. Specifically, we found a correlation between the SARA score and both functional (ie, BCVA and microperimetry) and imaging parameters (outer nuclear layer thickness). Moreover, outer nuclear layer thickness was correlated with NfL levels, which is a valid biomarker in SCAs able to discriminate preataxic from both control individuals and ataxic carriers and to predict clinical and radiological progression. Previous studies on SCA7-associated retinal dystrophy defined some biomarkers of progression of the disease, including the OCT macular thickness, the automated perimetry, or electrophysiology. Compared to automated perimetry, microperimetry has the advantage of testing specific areas of the retina through an eye-tracking system that guarantees precision and repeatability. OCT parameters demonstrated potential as surrogate biomarkers for the progression of inherited retinal diseases, although regulatory agencies, such as the US Food and Drug Administration or the European Medicines Agency, have not yet accepted them as clinically meaningful for the approval of novel therapies. Ellipsoid zone disruption usually represents the first sign of photoreceptor dysfunction and degeneration which usually precede the loss of the cell bodies. In rod-cone dystrophies, the width of the intact ellipsoid zone is used to monitor the progression of the disease, as its narrowing represents the first sign of the ongoing photoreceptor degeneration. In cone and cone-rod dystrophies, disruption of the central ellipsoid zone occurs from the earliest stages of the disease. Furthermore, although the reliability of preserved ellipsoid zone measurements in rod-cone dystrophies has been demonstrated, no such studies have been conducted for cone or cone-rod dystrophies. In such instances, the automated measurement of the thinning of the outer nuclear layer, signifying the loss of photoreceptor cell bodies, could potentially offer a more robust correlation with the progressive cone degeneration, thereby demonstrating greater sensitivity in disease assessment. Nonetheless, to validate this hypothesis and confirm the clinical meaning of OCT changes in cone-rod dystrophies, additional longitudinal and comparative studies are warranted.

In fundus autofluorescence imaging, SCA7-related dystrophies display the distinctive presentation observed in cone or cone-rod dystrophies. Specifically, short-wavelength fundus autofluorescence reveals an initial phase of hyperautofluorescence during the early stages, associated with the window defect arising from ellipsoid zone disruption. This is followed by subsequent central hypoautofluorescence due to retinal pigment epithelial atrophy. In contrast, near-infrared fundus autofluorescence usually reveals hypoautofluorescence even on initial ellipsoid zone disruption, showcasing heightened sensitivity in detecting subtle early changes in the disease. Notably, lesions on near-infrared fundus autofluorescence generally appear larger than in short-wavelength fundus autofluorescence, exhibiting hypoautofluorescence even in areas where the ellipsoid zone is preserved on OCT. This was also observed in other inherited retinal dystrophies. It is plausible that near-infrared fundus autofluorescence might unveil a process in the disease progression that precedes the disruption of the ellipsoid zone and subsequent retinal pigment epithelial degeneration, which ultimately leads to the loss of the short-wavelength fundus autofluorescence signal.

Strengths and Limitations

Our study has some limitations, including the small sample size and the cross-sectional analysis. However, SCA7 is a rare disease and we included carriers at preataxic or early ataxic phases. The number of enrolled carriers is comparable to or higher than to most previous studies focusing on SCA7-related retinal dystrophy. Strengths of the study include the setting of the study in reference expert centers for rare diseases, the use of a complete multimodal imaging and functional approach for the ophthalmic assessment, and the inclusion of preataxic and early ataxic carriers who will be the target for future gene silencing therapies.

Conclusions

In conclusion, SCA7-related retinal dystrophy is characterized by a cone dystrophy or a cone-rod dystrophy phenotype that was correlated with neurological findings and NfL levels through ocular functional and imaging parameters in this study. Outer nuclear layer assessed by OCT, a simple and noninvasive examination, may be a promising surrogate biomarker to monitor a potential intravitreal treatment. The longitudinal assessment of these parameters through prospective natural history studies will further assess the potential value of these biomarkers to track the disease progression and to monitor therapeutic interventions in clinical trials.

Supplement 1.

eTable 1. Functional data from both eyes of patients with SCA7

eTable 2. Imaging data from both eyes of patients with SCA7

eFigure. Short-wavelength fundus autofluorescence, near-infrared fundus autofluorescence, optical coherence tomography and full field electroretinogram responses of three SCA7 carriers and one healthy control

Supplement 2.

Data sharing statement