Peculiar Clinical Findings in Young Choroideremia Patients: A Retrospective Case Review

Purpose: To report peculiar clinical findings in young choroideremia (CHM) patients. Methods: We retrospectively reviewed young (age <20 years at the first evaluation) CHM patients examined at the Regional Reference Center for Hereditary Retinal Degenerations at the Eye Clinic in Florence between 2012 and 2018. We took into consideration patients with ophthalmological examinations, fundus color photographs, fundus autofluorescence (FAF) images, optical coherence tomography (OCT) scans, full-field electroretinograms, and Goldmann visual fields. Results: In our series, we studied 8 young CHM patients (average age 13.8 years, median age 12.5, range 10–20) for a total of 16 eyes. Visual acuity (VA) was 20/20 in 7 patients and 20/25 in both eyes of 1 patient. We identified a peculiar central FAF pattern (detectable in 3 patients), characterized by reduced central hypo-autofluorescence. Long OCT scans showed different forms of parapapillary retinal involvement from the mildest to the most severe form when the macula is still preserved. In 3 patients, at the time of atrophic changes at the posterior pole, it was possible to detect a progressive reduction of foveal pigmentation during follow-up. We found mutations of the CHM gene in all 6 patients who had been screened. Conclusions: CHM is a progressive retinal disorder which involves both the peripheral and the central retina. Using a multimodal imaging approach, we described peculiar central abnormalities underlying the early involvement of the central retina in young CHM patients with a good VA.

Keywords: Choroideremia; Young age; Fundus autofluorescence; Retinal dystrophy

Choroideremia (CHM) (Online Mendelian Inheritance in Man [OMIM] identifier #303100) is an X-linked chorioretinal dystrophy caused by defects in the CHM gene, which encodes Rab escort protein 1 (REP1), a key mediator of membrane trafficking in the retina and retinal pigment epithelium (RPE). CHM is estimated to affect approximately 1 in 50,000 individuals. Its clinical course is characterized by progressive degeneration of the RPE and by photoreceptor loss. Most patients retain good central acuity into the fifth decade of life [[1]-[3]] despite progressive structural and functional alterations to the central retina in the early stages of the disease [[3]-[8]]. In fact, since the CHM gene is also expressed in cones and these cells are dependent on surrounding retinal cells that also express REP1 as part of the cone visual cycle, it would not be unexpected to find subtle functional cone abnormalities before the onset of degeneration.

The mechanism of retinal degeneration is poorly understood, and there is still uncertainty regarding which cell type(s) is/are primarily affected [[1]]. In order to better characterize the pathogenic processes which are involved in CHM, we studied the clinical features of young CHM patients. Recently, Khan et al. [[3]] studied young CHM patients, demonstrating a reduction in subfoveal choroidal thickness with disease progression and a simultaneous increase in foveal retinal thickness, both of which occurred while visual acuity (VA) was maintained. In our study, we reviewed clinical, optical coherence tomography (OCT), and fundus autofluorescence (FAF) examinations of young CHM patients.

The patients included in this retrospective single-center study were identified at the Regional Reference Center for Hereditary Retinal Degenerations at the Eye Clinic in Florence. We included CHM patients younger than 20 years at the initial examination. All patients underwent a comprehensive ophthalmological examination including best corrected VA (BCVA) measurement, intraocular pressure evaluation, biomicroscopy of the anterior segment, and fundoscopy after dilation.

We considered the following diagnostic examinations: color fundus photography, OCT, electrophysiological tests, and Goldmann visual field testing. Color fundus photographs of the posterior pole and the surrounding areas of both eyes were taken using a fundus camera with a 45° field of vision centered on the macula (FF450 Retinograph; Carl Zeiss Meditec, Jena, Germany). The OCT examination was performed using a Triton Swept Source OCT device (Topcon Medical Systems Inc., Oakland, NJ, USA) and a SPECTRALIS confocal scanning laser ophthalmoscope (Heidelberg Engineering, Heidelberg, Germany). The electrophysiological examination was performed using the electrophysiological diagnostic unit RETIMAX (Roland Consult, Brandenburg an der Havel, Germany) according to the International Society for Clinical Electrophysiology of Vision (ISCEV) protocols. FAF images were obtained using the SPECTRALIS device (Heidelberg Engineering; blue autofluorescence [BAF], λexc = 488 nm) and ultrawide-field digital scanning laser technology (Daytona; Optos, Düsseldorf, Germany; green autofluorescence [GAF], 532-nm FAF).

A family history was taken and a pedigree was created. After genetic counseling, a molecular genetic analysis was performed at the Genetics Department of Careggi Teaching Hospital in Florence.

The patients were clinically diagnosed by a characteristic fundus appearance, visual field defects, and electroretinography, supported in some cases by a positive family history. The clinical and genetic features of the patients are shown in Table 1. In our series, we identified 8 young CHM patients (average age 13.8 years, median age 12.5, range 10–20) for a total of 16 eyes. Most of the patients were asymptomatic at disease discovery, with 37.5% (3/8) reporting difficulty seeing in the dark as their major concern. VA was 20/20 in 7 patients (7/8) and 20/25 in both eyes of 1 patient (1/8). The spherical equivalents are reported in Table 1. The anterior segment was unremarkable. The ophthalmoscopic examination revealed fundus abnormalities in all patients. At the final follow-up (possible for 6 patients), the average age was 18.2 years (range 13–26). VA remained stable in all patients. The mean follow-up was 3.8 ± 1.2 years (range 2–8).

Graph: Table 1. The clinical and genetic features of the CHM patients

Genotyping results were available for 6 patients (6/8). The pedigrees of the families are summarized in Figure 1. We examined 10 female CHM carriers who displayed the typical fundus features of a CHM carrier. More specifically, we examined 2 female carriers in the CHM-1 family (IV-3 and IV-5), 2 female carriers in the CHM-2 family (I-2 and II-2), 1 female carrier in the CHM-3 family (I-2), 4 female carriers in the CHM-4 family (II-5, III-3, III-4, and III-11), and 1 female carrier in the CHM-5 family (I-2).

Graph: Fig. 1. Pedigrees of the choroideremia (CHM) patients. CHM-1 family (patient P1; V-3); CHM-2 family (P2, III-1; P3, III-2); CHM-3 family (P4, II-1); CHM-4 family (P5, IV-6; P6, IV-2); CHM-5 family (P7, II-1; P8, II-2).

The fundus phenotypes are shown in Figure 2. At the first evaluation, the mildest phenotype was present in 3 patients (patient P1, P3, and P5; n = 3/8); it was characterized by significant pigmentary changes predominantly at the mid-periphery (Fig. 2a–c). Three patients (P4, P7, and P8; n = 3/8) had areas of atrophy, particularly well defined, between the vascular arcades and the equator; interspersed between these areas of atrophy there were regions that retained pigmentation. Particularly in P8, areas of chorioretinal atrophy were detectable at the posterior pole (Fig. 2d, g, h). Two patients (P2 and P6; n = 2/8) had widespread chorioretinal atrophy and large peripheral areas of pigmentary changes (Fig. 2e, f); in these patients, atrophic changes were also evident at the posterior pole.

Graph: Fig. 2. Color fundus photographs of the choroideremia patients. a Patient P1 (10 years; BCVA 20/20 OU). The posterior pole and optic disc are normal. Very mild peripapillary atrophy is present. Peripheral retinal involvement is evident as a pigmentary disturbance. The changes appear as granular clumps of pigmentation (arrow). b P3 (12 years; BCVA 20/20 OU). The fundus images show retinal pigment epithelium (RPE) dystrophy involving the mid-periphery (arrow). No macular abnormalities are present. Peripapillary atrophy is evident. c P4 (13 years; BCVA 20/20 OU). The fundus images show RPE dystrophy involving the mid-periphery (arrow). No macular abnormalities are present. Peripapillary atrophy is detectable. d P5 (14 years; BCVA 20/20 OU). The fundus images display areas of atrophy, particularly well defined in the mid-peripheral retina, between the vascular arcades and the equator (arrowheads). Interspersed between these areas of atrophy there are regions that retain pigmentation (arrow). e P2 (11 years; BCVA 20/20 OU). The fundus photographs show a preserved macular region, widespread chorioretinal atrophy, and large peripheral pigmentary rearrangements. f P6 (20 years; BCVA 20/25 OU). The fundus photographs show widespread chorioretinal atrophy and large peripheral pigmentary rearrangements with atrophic changes also at the posterior pole. g P7 (12 years; BCVA 20/20 OU). The fundus images display areas of atrophy, particularly well defined in the mid-peripheral retina, between the vascular arcades and the equator (arrowheads). Interspersed between these areas of atrophy there are regions that retain pigmentation (arrow). h P8 (18 years; BCVA 20/20 OU). The fundus images display large areas of atrophy at the periphery with evident peripapillary chorioretinal atrophy.

In all patients, peripapillary atrophy was present. In particular, we report an important intra-patient symmetry. During follow-up, we observed fundus changes both at the peripheral retina and at the posterior pole. The area of peripheral retina covered by pigmentary changes evolved into areas of atrophy, particularly well defined at the mid-peripheral retina, between the vascular arcades and the equator (Fig. 3a–c). In 1 patient (P5), the fundus changes were particularly evident during follow-up (4 years) (Fig. 3d, e). The peripapillary atrophy was progressive and advanced in a centrifugal manner toward the macula (Fig. 3). At the time of the peripheral changes to the RPE, macular RPE abnormalities occurred. In 3 patients (P2, P4, and P5) we observed a different pigmentation between the fovea and the posterior pole (the foveal region was darker than the surrounding area). Among these patients, in P2 and P4 we were able to detect progressive foveal depigmentation at the time of the atrophic changes at the posterior pole during follow-up (5 and 2 years, respectively) (Fig. 3f–i).

Graph: Fig. 3. Fundus photographs at follow-up. a Ultra-wide-field fundus photograph of patient P1 at the last fundus photograph examination (in 2017). b, c The images show the same area as image a but magnified (green dashed circle) during follow-up. Pigmentary changes were present in 2015 (b) and evolved in the areas of retinal atrophy by 2017 (c). d, e Color fundus photographs of P5. d At the first evaluation (in 2013). Retinal dystrophy with pigmentary changes was evident at the mid-periphery. The yellow dashed circle delineates 2 small circular atrophic areas in the dystrophic retina. e The wide-field image shows the severe atrophic changes in pigmentary rearrangements during follow-up (4 years later, in 2017). The yellow dashed circle delineates the corresponding areas in d and e. f, g Color fundus photographs of the posterior pole of P2. The reduction in macular retinal pigment epithelium pigmentation is evident (green arrowheads). f At the first evaluation (in 2012). g Three years later (in 2015). h, i Color fundus photographs of the posterior pole of P4. The arrowheads show a reduction in macular pigmentation. h At the first evaluation (in 2015). i One year later (in 2016).

The FAF patterns are shown in Figure 4. GAF and BAF were available for all patients. In 2 patients (P1 and P3), the examination was characterized by a speckled FAF at the mid-periphery, a rather normal FAF at the extreme periphery, and a normal macular autofluorescence (Fig. 4a–d). In P3, the speckled FAF pattern involved the vascular arcades too (Fig. 4c, d).

Graph: Fig. 4. Fundus autofluorescence (FAF) images. a–d Normal macular FAF. Wide-field FAF (green FAF [GAF] and blue FAF [BAF]) in patient P1 (a, b) and P3 (c, d). GAF (a) is characterized by a normal macular autofluorescence, a speckled FAF at the mid-periphery (arrow), and a fairly normal FAF at the extreme periphery. BAF (b) shows a normal macular autofluorescence. A very mild speckled FAF can be seen along the vascular arcades (arrowheads) in P1 (a, b). Peripapillary hypo-autofluorescence is evident. e–j Hyper-autofluorescent parafovea. GAF and BAF in P7 (e–h) and P4 (i, j). The GAF images (e, g, i) show well-demarcated hypo-autofluorescent areas at the mid-periphery. Regions that are characterized by speckled hypo-autofluorescence are interspersed between these areas of atrophy, also involving the posterior pole. The macular area is characterized by a particular FAF pattern: it is possible to clearly see a central hyper-autofluorescent area (orange arrowhead); using BAF (f, h, j), we were able to discern the hyper-autofluorescent parafovea (orange dashed circle) with preserved foveolar hypo-autofluorescence (green arrowhead). o Hyper-autofluorescent parafovea (on the left side, orange box, left eye of P7) in comparison with a normal parafovea (on the right side, light-blue box, left eye of P1). Peripapillary hypo-autofluorescence is evident. k–n Homogeneous FAF pattern. The GAF and BAF images of P8 (k, l) and P2 (m, n) show a homogeneous FAF at the posterior pole, inside the vascular arcades without a hyper-autofluorescent parafovea. For P2, the GAF image (m) shows peripheral hyper-autofluorescence due to the autofluorescent properties of the sclera in the atrophic areas. Peripapillary hypo-autofluorescence is evident.

In 3 patients (P4, P5, and P7), well-demarcated hypo-autofluorescent areas were evident at the mid-periphery. Regions with speckled hypo-autofluorescence were interspersed between areas of atrophy, also involving the posterior pole. FAF was quite well preserved at the extreme periphery (Fig. 4e–j). The macular area of these patients (P4, P5, and P7) was characterized by a peculiar FAF pattern: using BAF, we were able to discern a very small hypo-autofluorescent area only at the foveola (the para-foveal region appeared almost hyper-autofluorescent) in comparison with P1 and P3 (mildest phenotype). In fact, the retinal vessels at the boundaries of the foveal avascular zone in P4, P5, and P7 were well identified due to the lighter background in comparison to P1 and P3 (Fig. 4f, h, j); using GAF, we did not easily observe the physiological central hypo-autofluorescence (Fig. 4e, g, i) in comparison with P1 and P3 (mildest phenotype).

In P2, P6, and P8, the FAF images revealed a homogeneous residual FAF at the posterior pole (Fig. 4k–n). In all patients, it was possible to see peripapillary hypo-autofluorescence.

The OCT features are summarized in Figure 5. On OCT, retinal abnormalities involving the ellipsoid zone (EZ), interdigitation zone (IZ), and RPE were detected in all patients of our series (n = 8/8). More specifically, IZ and EZ abnormalities occurred in the areas of preserved RPE (assessed using FAF).

Graph: Fig. 5. Optical coherence tomography (OCT) abnormalities. a, b P1. The OCT scans show a normal macular profile. Very mildly hyper-reflective outer retinal layers (ellipsoid zone [EZ] and interdigitation zone [IZ]) and retinal pigment epithelium (RPE) abnormalities ("undulated/disrupted" profile) are detectable in the peripapillary area (yellow arrowhead). The long OCT scan (b; 3D Line 12 × 5 mm) shows the extension of the RPE, EZ, and IZ abnormalities. In particular, the peripapillary area is displayed. c–g P3. A normal macular profile can be seen (c). Very mildly hyper-reflective outer retinal layers (EZ and IZ) and RPE abnormalities (undulated/disrupted profile) are detectable in the peripapillary area (yellow arrowhead). These alterations are detectable at the posterior pole and inside the vascular arcades (d; yellow arrowhead). The long OCT scan (e) shows hyper-reflective material at the posterior pole above the RPE (subretinal drusenoid deposits), which appear hyper-autofluorescent in f (blue fundus autofluorescence) and g (green fundus autofluorescence) (yellow dashed circle). h–j P5 (h), P4 (i), and P2 (j). The IZ is not detectable. Irregularities of the EZ and RPE are widespread in the macular area. Between the blue arrowheads, it is possible to distinguish the RPE, EZ, and ELM. From the blue to the red arrowheads, the RPE, EZ, and ELM appear disrupted. In the peripapillary area, atrophy of the outer retinal layers is evident. The inner retinal layers and outer retinal tubulations (ORTs) are detectable (asterisks). k Long OCT scan of P6. The image shows more advanced alterations. The IZ is not detectable, and there is severe disruption of the EZ layer with RPE atrophy. In the transition zone, ORTs are present, and in particular it is possible to see the rolling of the ELM with ORT development. In the peripapillary area, it is possible to detect atrophic changes (outer and inner retinal layer atrophy). The figure in the small box shows intraretinal cysts which are detectable in the transition zone at the level of the inner nuclear layer.

In 2 patients (P1 and P3), the central macular OCT examination result was normal. The OCT scans, performed on speckled FAF areas, showed RPE, EZ, and IZ abnormalities: a reduction in reflectivity of the IZ (on some OCT scans, the IZ was not detectable), an alteration of the RPE, and the EZ profile resembling an "undulated" and "disrupted" pattern. In these abnormal areas, it was possible to detect hyper-reflective deposits above the RPE (this lesion appeared hypo-reflective on infrared reflectance imaging and partially hyper autofluorescent using FAF) resembling a small subretinal drusenoid deposit (Fig. 5e–g).

In 3 patients (P4, P5, and P7), the OCT examination showed more advanced retinal alterations. In the foveal region of P4 and P5, it was possible to see only the EZ and the RPE layer. There was widespread disruption of the EZ in the macular area.

Two patients (P2 and P6) had a very disrupted EZ that was only detectable in the foveal region. Outer retinal tubulation (ORT) was well defined in the transition zone. In these patients, the atrophy of the outer retinal layers and the choroid was evident on the macular OCT scans as well. In P6, small intraretinal cysts at the level of the inner nuclear layer were detectable in the transition zone. P8 had persistence of the inner retinal layers (foveal hypoplasia) on macular line scans through the fovea.

In our series, we reviewed long OCT scans performed in order to study the peripapillary areas in CHM patients (Swept Source OCT 3D+LINE 12 × 9 mm). All examinations were performed at the last follow-up. The patients with the milder phenotype (P1 and P3) presented very mild signs of peripapillary OCT alterations: EZ-RPE irregularities, as well as back-scatter due to RPE atrophy (especially in P3).

In P4, P6, and P7, back-scatter due to RPE atrophy, as well as absence of the EZ, IZ, external limiting membrane (ELM), and outer nuclear layer, was evident; the internal retinal layers were preserved, and on some scans it was possible to discern a residual hyper-reflective band attributable to the outer plexiform layer above Bruch's membrane. In the transition zone, the ELM was rolled up, forming ORTs (detectable in 5 patients). On a single OCT scan of P6, many ORTs were detectable.

In the last 3 patients (with a more severe phenotype), it was possible to see atrophy of the internal retinal layers as well. In P6, intraretinal cysts were present in the transition zone, particularly at the level of the inner nuclear layers.

The different functional and anatomical features of CHM in young people were described in a previous work [[3]]. On fundoscopic examination, the earliest changes were seen in the peripheral retina as pigmentary changes. These appeared as granular clumps of pigmentation. During follow-up, the areas of the peripheral retina characterized by pigmentary changes evolved into areas of atrophy, well defined in the mid-peripheral retina between the vascular arcades and the equator.

Retinal imaging was useful to identify different stages of retinal degeneration: mild RPE changes, patchy RPE changes, and confluent chorioretinal atrophy. The same alterations characterize retinal dystrophy in female CHM carriers [[3], [9]]. The OCT examination performed on dystrophic areas showed irregular profiles of the RPE, EZ, IZ, and ELM [[11]]. In particular, hyper-reflective lesions above the RPE were described. The same alterations on OCT were described in female CHM carriers [[9]] as subretinal drusenoid deposits [[9], [12]]. In advanced cases, OCT showed atrophy of the outer retinal layers and, later, of the internal retinal layers.

Thanks to the long OCT scans, we were able to detect peripapillary alterations in the young CHM patients. It was possible to see different forms of retinal involvement from the mildest to the most severe form. In particular, in some cases we were able to see alterations on OCT when the macula was still preserved. The atrophic changes involved first the RPE/outer retinal layers and subsequently the internal retinal layers. Moreover, in the peripapillary area, it was possible to see the formation of ORTs.

The exact pathogenetic mechanism of ORTs is still unclear, but there is evidence that these lesions may represent an attempt to preserve photoreceptors. Müller cells may be activated by the progressive damage to photoreceptors, and this may promote the expression of glial fibrillary acidic protein, facilitating ORT formation. This fact may explain the significant association of ORTs with the presence of geographic atrophy, suggesting that a dysmorphic or absent RPE facilitates ORT formation [[13]-[17]]. In our series, 5 patients (n = 5/8) presented with ORTs in the peripapillary area. We know that significant parapapillary and peripapillary atrophy characterizes male CHM patients, also younger ones [[3]]. For this reason, this area is probably the first site where ORTs originate in CHM.

Regarding FAF alterations, we know that the remaining islands, characterized by a preserved RPE, appear as areas with a normal or increased autofluorescence intensity due to lipofuscin accumulation, whereas areas with RPE atrophy are characterized by hypoautofluorescence. The previous study on FAF in young CHM patients [[3]] showed that a normal FAF area appeared to correlate with age, although there was significant variation between individuals. A high degree of inter-eye symmetry was detected in previous works [[3], [18]-[20]]. Greater inter-eye differences (especially in BCVA and microperimetry) are seen in later stages of the disease, and they are likely to arise from differences in foveal involvement as the centripetal degeneration encroaches on the fovea earlier in one eye than in the other, leading to foveal cone reduction [[19], [21]].

FAF follow-up data on CHM patients showed that RPE degeneration in CHM occurred in a centripetal fashion, with early loss of peripheral FAF leading to the formation of a residual "island" of preserved FAF with sharply demarcated, scalloped edges, usually centered around the fovea. This island was shrinking over time and eventually encroaching on the fovea, which is associated with a sharp decline in VA generally during the fourth-to-fifth decade of life [[18], [22]]. Altogether, the rate of FAF area reduction seemed to depend on the size of the area at baseline, with larger areas progressing more quickly. Therefore, the slowing of degeneration in later years was due to a small area at baseline [[20]].

Khan et al. [[3]] demonstrated a reduction in retained macular autofluorescence, and they described three FAF patterns at the posterior pole: normal, speckled, and absent (hypo-autofluorescence) [[3], [23]]. We are in agreement with the previous works, but in our series, we revealed an important finding that has not previously been reported about the central FAF pattern in young CHM patients. In particular, we described FAF findings in young CHM patients using two FAF devices in order to show the different FAF features of this peculiar central FAF pattern. Our study is the first that compares BAF and GAF imaging in young CHM patients. We identified a peculiar central FAF pattern (detectable in 3 patients), characterized by a reduction in foveal hypo-autofluorescence. In fact, using BAF it was possible to see a very small hypo-autofluorescent area only at the foveola (with the reduction of the hypo-autofluorescent foveal signal, the parafoveal region appeared hyper-autofluorescent). Using GAF, central hypo-autofluorescence was not detectable at all.

In fact, we know that in contrast to BAF, GAF is not significantly affected by macular pigments (lutein and zeaxanthin) due to a lack of absorption [[24]]. For this reason, a reduction in central hypo-autofluorescence is more evident using GAF, thanks to the already weak central hypo-autofluorescence in physiological conditions. This FAF pattern was probably not related to the macular pigments (lutein and zeaxanthin); in fact, no differences were found between CHM patients and normal subjects concerning macular pigments [[5]]. Moreover, in our patients, foveolar hypo-autofluorescence was preserved using BAF.

Furthermore, this central paradoxical hyper-autofluorescence pattern is different from the central hyper-autofluorescence FAF pattern found in retinitis pigmentosa [[25]]; in retinitis pigmentosa patients with central hyper-autofluorescence (type 3), VA is very low, the structure of the outer retinal layers on OCT is more damaged, and the FAF pattern is superimposable between GAF and BAF. In our series, we reported progressive central retinal RPE depigmentation during follow-up. The lack of central RPE pigment may be responsible for the reduction in central hypo-autofluorescence. Melanin absorbs the short-wave-length excitation beam, decreasing the overall auto-fluorescent signal [[26]]. Probably after a centripetal loss of pigment granules, atrophy of the RPE occurs. This mechanism could be responsible for the peculiar FAF pattern described.

Obviously, we cannot exclude the possibility that more pathogenetic mechanisms are involved in this particular FAF pattern; for example, an increased amount of lipofuscin in the RPE (due to RPE/parafoveal photoreceptor impairment) could be responsible for a primary hyper-autofluorescent parafovea. For these reasons, we suggest that this intermediate stage (reduction in central hypo-autofluorescence with parafoveal hyper-autofluorescence) is characterized by initial alterations to the para-foveal photoreceptors (cones) or initial impairment of the parafoveal RPE with consequent involvement of the photoreceptors.

In fact, Nabholz et al. [[7]] reported a decrease in cone density in the parafovea of young CHM patients (<25 years) using adaptive optics-based infrared fundus ophthalmoscopy. This cone involvement could be the consequence of a generalized alteration of the metabolism in photoreceptors and/or in the RPE caused by a CHM mutation. There is still uncertainty as to whether the neuroretina, RPE, and choroid are all primarily affected, or whether one or more of these layers is secondarily affected [[7], [27]]. However, there is some evidence that RPE cells degenerate first [[8], [29]-[32]]. An RPE-dependent photoreceptor death model has been supported by cell-specific Chm ablation experiments in the conditional knockout mouse model [[33]] and in zebrafish rep1 mutants [[34]], which suggests that photoreceptor death is not autonomous and requires contact with a mutant RPE. Photoreceptors appear to lose outer segments following the loss of the underlying RPE and form tubulations at the edges of degeneration [[3], [31]].

This finding was also evident from our OCT results. The reduction in central hypo-autofluorescence was not associated with reduced VA. This is in agreement with previous works which described central retinal abnormalities with preserved VA in the natural history of CHM [[3], [8]]. Aleman et al. [[4]] described alterations in all their patients, including their youngest ones (6 years). The earliest OCT scan abnormality described was approximation of the EZ to the RPE and/or loss of the IZ signal, probably representing shortening or loss of the photoreceptors' outer segments. These abnormalities were observed in areas with normal RPE pigmentation, and also coincided with outer nuclear layer preservation in most individuals; moreover, a reduction in subfoveal choroidal thickness with disease progression and a simultaneous increase in foveal retinal thickness were reported [[3], [6], [32]].

In addition to progressive alterations, congenital changes have been described, such as persistence of the inner retinal layers [[3]]. The involvement of the parafovea was also reported by Jacobson et al. [[6]]; they described a retinal thickening as the earliest finding that anticipated or coincided with photoreceptor abnormalities. They proposed that this thickening was a surrogate marker for the earliest stages of remodeling, and that the pathobiology underlying this disease expression involves the Müller cell, the main glial cell of the retina. Unfortunately, in our study, microperimetry was not performed, and this is a limitation of our work, but we believe that CHM patients characterized by this peculiar FAF pattern (reduction in central hypo-autofluorescence) may not have a detectable localized dysfunction; in fact, the full-field light sensitivity thresholds remained comparable with those of age-matched controls throughout a wide range of disease stages, in contrast to rod sensitivity [[35]].

In conclusion, our clinical, OCT, and FAF results are further evidence of the possible early involvement of the central retina in the course of CHM disease. The reduction in central hypo-autofluorescence and the phase of a hyper-autofluorescent parafovea could be further markers of metabolic dysfunction at the level of the central retina in CHM disease.

This study was approved by the local research ethics committee, and all investigations were conducted in accordance with the principles of the Declaration of Helsinki.

The authors have no conflicts of interest and no funds have been received for this study.