Evaluation of Retinal Structure and Visual Function in Blue Cone Monochromacy to Develop Clinical Endpoints for L-opsin Gene Therapy

Abstract

L-cone opsin expression by gene therapy is a promising treatment for blue cone monochromacy (BCM) caused by congenital lack of long- and middle-wavelength-sensitive (L/M) cone function. Eight patients with BCM and confirmed pathogenic variants at the OPN1LW/OPN1MW gene cluster participated. Optical coherence tomography (OCT), chromatic perimetry, chromatic microperimetry, chromatic visual acuity (VA), and chromaticity thresholds were performed with unmodified commercial equipment and/or methods available in the public domain. Adaptive optics scanning laser ophthalmoscope (AOSLO) imaging was performed in a subset of patients. Outer retinal changes were detectable by OCT with an age-related effect on the foveal disease stage. Rod and short-wavelength-sensitive (S) cone functions were relatively retained by perimetry, although likely impacted by age-related increases in the pre-retinal absorption of short-wavelength lights. The central macula showed a large loss of red sensitivity on dark-adapted microperimetry. Chromatic VAs with high-contrast red gratings on a blue background were not detectable. Color vision was severely deficient. AOSLO imaging showed reduced total cone density with majority of the population being non-waveguiding. This study developed and evaluated specialized outcomes that will be needed for the determination of efficacy and safety in human clinical trials. Dark-adapted microperimetry with a red stimulus sampling the central macula would be a key endpoint to evaluate the light sensitivity improvements. VA changes specific to L-opsin can be measured with red gratings on a bright blue background and should also be considered as outcome measures in future interventional trials.

Keywords: color vision; microperimetry; outcome measures; perimetry.

Figure 1

Range of disease severity in BCM patients. OCT along the horizontal meridian (left panels) and NIR-RAFI (right panels) in a representative normal (A) compared with patients with deletion mutations (B) and those with C203R mutations (C). Calibration bars are shown. T, temporal retina; N, nasal retina; OD, right eye; OS, left eye.

Figure 2

Rod and S cone sensitivity distribution. (A) Dark-adapted sensitivity to 500 nm stimuli along the vertical meridian in BCM patients (connected symbols) compared with normal limits (gray band). (B) S cone sensitivity profiles (connected symbols) of the BCM patients using a 440 nm stimulus on a yellow adapting background compared with normal limits (gray band). (C,D) Spectral sensitivity functions (SSF) in a subset of BCM patients measured at 12° eccentricity in superior and inferior fields under dark-adapted (C) and yellow-adapted (D) conditions. Results from each patient (symbols) shifted vertically for visibility, fit by rod (C) and S cone (D) photoreceptor sensitivities adjusted for pre-retinal absorption for each in-dividual, and ordered by age. (E,F) Individualized pre-retinal absorption estimates (symbols) at 500 nm (E) and 440 nm (F) obtained by best ensemble fit functions shown in panels (C) and (D), respectively. Regression lines shown.

Figure 3

Chromatic sensitivities at the central retina. (A,B) NIR-RAFI in representative normal (A) and P8 (B). Retinal locations sampled (black symbols) and the center of fixation (Fix) are shown. (C,D) Microperimetric sensitivity profiles in the normal and P8 along the horizontal and vertical meridians with cyan (C) and red (D) stimuli. Gray region is the normal range. (E) Sensitivities of all BCM patients as a function of eccentricity to cyan (left panel) and red (middle panel) stimuli. Dark gray symbols are from the oldest subject, P3. Cyan minus red sensitivity difference is also shown (right panel). Gray regions show the normal ranges in each panel, the gray arrow indicates the difference expected from rod mediation in the extramacular region.

Figure 4

Chromatic visual acuities. (A,B) Visual acuities in BCM patients recorded with blue on yellow (A) and red on black (B) gratings. Gray region in panel (A) is the normal range. Insets demonstrate the appearance of the gratings to normal trichromats.

Figure 5

Color vision. (A,B) Chromaticity thresholds with blue (A) and red (B) stimuli in BCM patients as a function of LC noise. Hashed region is the color-rendering capability (“phosphor limit”) of the monitor for each color and noise level. The light gray thick line is the normal thresholds. Insets below demonstrate the appearance of the color stimuli embedded in 10%, 50% and 98% LC noise.

Figure 6

Cone cell densities. (A,B) NIR SLO and AO confocal and non-confocal images for P8 (A) and P1 (B). NIR SLO images (left) show the location of the AO images (red squares) and represent the location of peak waveguiding cone density. Confocal (center) and non-confocal split-detection (right) AO images depict the cone mosaic with waveguiding cones marked in blue and non-waveguiding cones marked in purple. Contrast was adjusted on the confocal images for visualization purposes.

Figure 7

Longitudinal changes in structure and function. Long-term serial data showing NIR-RAFI and OCT images of the central retina (left panels) at two ages in seven of the BCM patients. Additionally shown are the quantification of the ONL thickness (right upper panels) and rod sensitivity loss (right lower panels) at the same ages.

Figure 8

Longitudinal AO images. (A,B) NIR SLO (left) and AO (right) images for P6 (A) and P7 (B). NIR images are from the later time point and show the location of the AO images (red squares). AO images are separated by 4.4 and 4.9 years (P6 and P7, respectively). The mosaics are qualitatively stable over this time period. Cones that are waveguiding in the image acquired at the first time point remain waveguiding in the later time point (examples shown by cones labeled with blue arrowheads). Contrast was adjusted on the confocal images for visualization purposes.