High-resolution images of retinal structure in patients with choroideremia

Reema Syed, Sanna M Sundquist, Kavitha Ratnam, Shiri Zayit-Soudry, Yuhua Zhang, J Brooks Crawford, Ian M MacDonald, Pooja Godara, Jungtae Rha, Joseph Carroll, Austin Roorda, Kimberly E Stepien, Jacque L Duncan, Reema Syed, Sanna M Sundquist, Kavitha Ratnam, Shiri Zayit-Soudry, Yuhua Zhang, J Brooks Crawford, Ian M MacDonald, Pooja Godara, Jungtae Rha, Joseph Carroll, Austin Roorda, Kimberly E Stepien, Jacque L Duncan

Abstract

Purpose: To study retinal structure in choroideremia patients and carriers using high-resolution imaging techniques.

Methods: Subjects from four families (six female carriers and five affected males) with choroideremia (CHM) were characterized with best-corrected visual acuity (BCVA), kinetic and static perimetry, full-field electroretinography, and fundus autofluorescence (FAF). High-resolution macular images were obtained with adaptive optics scanning laser ophthalmoscopy (AOSLO) and spectral domain optical coherence tomography (SD-OCT). Coding regions of the CHM gene were sequenced.

Results: Molecular analysis of the CHM gene identified a deletion of exons 9 to 15 in family A, a splice site mutation at position 79+1 of exon 1 in family B, deletion of exons 6 to 8 in family C, and a substitution at position 106 causing a premature stop in family D. BCVA ranged from 20/16 to 20/63 in carriers and from 20/25 to 5/63 in affected males. FAF showed abnormalities in all subjects. SD-OCT showed outer retinal layer loss, outer retinal tubulations at the margin of outer retinal loss, and inner retinal microcysts. Patchy cone loss was present in two symptomatic carriers. In two affected males, cone mosaics were disrupted with increased cone spacing near the fovea but more normal cone spacing near the edge of atrophy.

Conclusions: High-resolution retinal images in CHM carriers and affected males demonstrated RPE and photoreceptor cell degeneration. As both RPE and photoreceptor cells were affected, these cell types may degenerate simultaneously in CHM. These findings provide insight into the effect of CHM mutations on macular retinal structure, with implications for the development of treatments for CHM. (ClinicalTrials.gov number, NCT00254605.).

Conflict of interest statement

Disclosure: R. Syed, None; S.M. Sundquist, None; K. Ratnam, None; S. Zayit-Soudry, None; Y. Zhang, None; J.B. Crawford, None; I.M. MacDonald, None; P. Godara, None; J. Rha, None; J. Carroll, None; A. Roorda, P; K.E. Stepien, None; J.L. Duncan, None

Figures

Figure 1.
Figure 1.
Pedigrees for the 4 families with choroideremia. Squares: males. Circles: females. Diamond: sex unknown. Filled symbols: males with CHM. Circles with dark centers: female carriers. Small filled circles: spontaneous abortions (sex unknown). Diagonal line: deceased. Transmission was consistent with X-linked inheritance in each family and the four families were unrelated.
Figure 2.
Figure 2.
Color fundus photo (top), FAF (middle), and SD-OCT images (bottom two rows) in female CHM carriers. Outlines on color fundus photo and FAF images: location of AOSLO images. Black lines: horizontal SD-OCT scan locations; the span of the lower magnified SD-OCT scans is indicated by the solid black lines coming from the images above them. Yellow ellipses: examples of hypopigmentation on color fundus photo that correlate with hypoautofluorescence on FAF. Blue squares: areas of hypoautofluorescence that correlate with areas of attenuated SD-OCT bands 3, 4a, and 4b; yellow arrows indicate the border at which band 3 becomes attenuated or disappears completely. In (D) (B-V-3), bands 3 and 4 were attenuated throughout the macula. Red arrow: bridging hyperreflective structures or interlaminar bridge seen in A-III-1. All OCT images were acquired using the Spectralis system with the exception of C-II-3, whose images were acquired with the Bioptigen system. White horizontal and vertical scale bars on the SD-OCT images indicate 200 μm.
Figure 3.
Figure 3.
Color fundus photo (top), FAF (middle), and SD-OCT images in male CHM patients. In the most severely affected patient (B-IV-2, [D]), AOSLO images did not reveal unambiguous cone mosaics, although an island of hyperautofluorescent RPE and outer retinal structures was present at the fovea; the SD-OCT scans in this patient demonstrate outer retinal tubulations (asterisk) and inner retinal microcysts. Outlines on color fundus photo and FAF images: locations of AOSLO images. Black lines: horizontal SDOCT scan locations; the span of the lower magnified SDOCT scans is indicated by the solid black lines coming from the images above them. Black asterisks: outer retinal tabulations. Red arrows: interlaminar bridges. Black arrow: edge of region where all outer retinal bands are present. Yellow arrow: intact band 2, but attenuated band 3. Blue arrow: only bands 1, 2, and 4 are visualized. All OCT images were acquired using the Spectralis system with the exception of C-III-2, whose images were acquired with the Bioptigen system. White horizontal and vertical scale bars on the SD-OCT images indicate 200 μm.
Figure 4.
Figure 4.
High-resolution AOSLO images for two carriers (left panels) and two affected males with CHM (right panels). Blue line: SD-OCT scan location. White boxes: cone spacing Z-scores within 2 SD of the normal mean. Black boxes: cone spacing Z-scores greater than 2 SD of the normal mean. Black arrow: area where bands 2, 3, and 4 are intact (magnified AOSLO, inset [A2]). Yellow arrows: area where bands 3 and 4 are attenuated but band 2 is intact; black scale bar: 1°. Large black squares show locations of magnified AOSLO insets; (A1, B1) show increased cone spacing in regions with disruption of band 3 but preservation of band 2; (A2, B2) show normal cone spacing in regions with intact outer retinal bands; (A3) shows a region where unambiguous cones were not visible and cone spacing measures were not possible with band 3 loss and band 2 preservation, and (B3) shows a region where cone spacing Z-scores were normal but the cones appeared abnormal with irregular packing; band 2 is irregular and band 3 is absent. (C2) shows normal cone spacing and an intact mosaic with intact outer retinal bands, while (C1) shows disrupted cones that were not arranged in mosaics, precluding quantitative measurement of cone spacing, in a region where band 2 is preserved but band 3 is attenuated or absent. Inset (D2) shows increased cone spacing near the fovea in a region where band 2 is intact but band 3 is irregular, while inset (D1) highlights an area at the edge of atrophy where cone mosaics appear abnormal although cone spacing is within 2 SD of the normal mean, band 2 is attenuated and band 3 is absent. All OCT images were acquired using the Spectralis system with the exception of (A, C), showing images from C-II-3 and C-III-2, whose images were acquired with the Bioptigen. White horizontal and vertical scale bars on the SD-OCT images indicate 200 μm.
Figure 5.
Figure 5.
Low-frequency structures observed in AO images at the fovea in a CHM patient, C-III-2. Top panels: comparison of foveal montages from flood illuminated AO and AOSLO images. The flood-illuminated AO image (left) shows low-frequency structures, not visible on AOSLO images (right) which show cone photoreceptors. White scale bar: 100 μm. Bottom panels, magnified images of the region delimited by the white box in the left-upper panel show a comparison of AO flood images of the fovea in the same patient over 6 months; the low-frequency structures remain stable. White circles: three corresponding retinal locations. White scale bar: 50 μm.

Source: PubMed

3
Sottoscrivi