International Journal of Molecular Sciences

Article Leber Congenital Amaurosis Due to GUCY2D Mutations: Longitudinal Analysis of Retinal Structure and Visual Function

Samuel G. Jacobson *, Artur V. Cideciyan , Alexander Sumaroka, Alejandro J. Roman , Vivian Wu, Malgorzata Swider, Rebecca Sheplock, Arun K. Krishnan and Alexandra V. Garafalo

Department of , Scheie Eye Institute, Perelman School of , University of Pennsylvania, Philadelphia, PA 19104, USA; [email protected] (A.V.C.); [email protected] (A.S.); [email protected] (A.J.R.); [email protected] (V.W.); [email protected] (M.S.); [email protected] (R.S.); [email protected] (A.K.K.); [email protected] (A.V.G.) * Correspondence: [email protected]; Tel.: +1-215-662-9981

Abstract: augmentation is being planned for GUCY2D-associated Leber congenital amaurosis (LCA). To increase our understanding of the natural history of GUCY2D-LCA, patients were evaluated twice with an interval of 4 to 7 years between visits using safety and efficacy outcome measures previously determined to be useful for monitoring this disorder. In this group of molecularly-identified LCA patients (n = 10; ages 7–37 years at first visit), optical coherence tomography (OCT) was used to measure foveal cone outer nuclear layer (ONL) thickness and rod   ONL at a superior retinal locus. Full-field stimulus testing (FST) with chromatic stimuli in dark- and light-adapted states was used to assay rod and cone vision. Changes in OCT and FST over the Citation: Jacobson, S.G.; Cideciyan, interval were mostly attributable to inter-visit variability. There were no major negative changes A.V.; Sumaroka, A.; Roman, A.J.; Wu, in structure or function across the cohort and over the intervals studied. Variation in severity of V.; Swider, M.; Sheplock, R.; Krishnan, A.K.; Garafalo, A.V. Leber Congenital disease expression between patients occurs; however, despite difficulties in quantifying structure and Amaurosis Due to GUCY2D function in such seriously visually impaired individuals with , the present work supports Mutations: Longitudinal Analysis of the use of OCT as a safety outcome and FST as an efficacy outcome in a clinical trial of GUCY2D-LCA. Retinal Structure and Visual Function. A wide age spectrum for therapy was confirmed, and there was relative stability of structure and Int. J. Mol. Sci. 2021, 22, 2031. function during a typical time interval for clinical trials. https://doi.org/10.3390/ijms22042031 Keywords: cone; clinical trials; rod; optical coherence tomography; outcome measures Academic Editor: Susanne Roosing

Received: 25 January 2021 Accepted: 15 February 2021 1. Introduction Published: 18 February 2021 Inherited retinal degenerations (IRDs) are a heterogeneous group of disorders that share the feature of abnormal visual function at the level of retinal photoreceptors. The Publisher’s Note: MDPI stays neutral source of photoreceptor dysfunction can be due to a variety of pathophysiological mecha- with regard to jurisdictional claims in nisms, and until a decade ago, IRDs were treated mainly with nutrient supplements aimed published maps and institutional affil- iations. to slow the disease [1]. Subsequently, converging information from molecular and retinal biology, animal models, phenotyping, and therapeutic tools reached a critical mass, and now there is great interest in understanding the underlying mechanisms of IRDs and developing novel treatments [2–6]. Among the more severe IRDs are those clinically classified as Leber congenital amau- Copyright: © 2021 by the authors. rosis (LCA), a group of predominantly autosomal recessively inherited diseases with con- Licensee MDPI, Basel, Switzerland. genital , nystagmus (involuntary eye movements), electroretinographic This article is an open access article evidence of photoreceptor dysfunction, and a deceptively benign fundus appearance at distributed under the terms and conditions of the Creative Commons early stages [7,8]. Some forms of LCA are attractive targets for therapy, not only because of a Attribution (CC BY) license (https:// longstanding unmet need but also because of the potential for improvement of vision from creativecommons.org/licenses/by/ treatment. LCA caused by mutations in the GUCY2D gene is theoretically a good candidate 4.0/). for gene augmentation therapy, given the relatively preserved photoreceptor structure de-

Int. J. Mol. Sci. 2021, 22, 2031. https://doi.org/10.3390/ijms22042031 https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2021, 22, 2031 2 of 12

spite disproportionately reduced function [9]. The gene GUCY2D encodes retinal guanylyl (retGC1), which modulates phototransduction in rods and cones [10,11]. Retrospective studies of phenotype in GUCY2D-LCA have shown dysfunction of cones, rods, or both, and patients have different degrees of abnormality in the central retinal structure [12–14]. Considering that this is not a stationary , it is important to know whether there is detectable disease progression over a typical time course of a clinical trial (e.g., several years). To date, there have been no prospective natural history studies of the retinal disease in GUCY2D-LCA. Valuable observations about change in disease expression have been made by studies reviewing clinical notes and examinations as well as grading available imaging and results [14]. To determine outcome measures for future clinical trials of GUCY2D-LCA, we previously studied a cohort of 28 molecularly identified patients [13]. We concluded that in this population, a safety outcome (in addition to ocular and systemic examinations) would be segmented and quantified optical coherence tomography (OCT) scans. To define rod- and cone- mediated visual efficacy, we decided that the full-field stimulus test (FST) under dark- and light-adapted conditions with chromatic stimuli would best capture function [13,15]. In the present study, we sought a greater understanding of the natural history of GUCY2D-LCA by performing follow-up evaluations (using the OCT and FST outcomes) in a subset of 10 patients from our original cohort. Changes in these parameters between the two visits are quantified and reported here.

2. Results and Discussion 2.1. Study Population with GUCY2D-LCA Ten patients with GUCY2D-LCA were evaluated on two visits. At the first visit, the patients were part of a larger group to determine outcome measures in GUCY2D-LCA in anticipation of clinical trials of gene augmentation therapy [9]. The biallelic mutations in GUCY2D have been previously reported [13]. Age at first exam ranged from 7.4 to 37.2 years (median, 13.8 years). The second visit occurred after an interval of 4.0 to 7.3 years (median, 4.6 years) (Supplementary Table S1). All patients had nystagmus and reduced visual acuity (range, 20/80 to light perception).

2.2. Serial En Face Near-Infrared Autofluorescence Imaging in GUCY2D-LCA Previous reports of ocular phenotype in GUCY2D-LCA patients have shown en face fundus images, whether color fundus photographs or fluorescence imaging. There were central retinal abnormalities noted in patients in the fifth decade of life but not at earlier ages; rare examples of peripheral retinopathy have also been reported [10]. We used near- infrared reduced illuminance autofluorescence (NIR-RAFI) imaging [16,17] to examine the central of our cohort of patients on two visits to determine if our observations were in keeping with those in other studies that reported en face images in GUCY2D-LCA. There were four patients with macular retinal pigment epithelium (RPE) disturbances visible on NIR-RAFI imaging; the youngest patient was P7 at age 19.4, and there were also changes in P8 at age 28.6, P9 at 30.7, and P10 at 37.2 years (Figure1). The abnormal appearance did not change (by inspection) at the second visit for any of the patients. The results suggest that central retinal RPE abnormalities could be detected in GUCY2D-LCA patients about two decades before previous reports; the changes did not appear to advance in size or retinal extent over the interval in this study. Int.Int. J. J. Mol. Mol. Sci. Sci.2021 2021, 22, 22, 2031, 2031 3 of3 of 12 12

FigureFigure 1. 1.En En face face near-infrared near-infrared reduced reduced illuminance illuminance autofluorescence autofluorescence (NIR-RAFI) (NIR-RAFI) images images from from the the left eye of nine GUCY2D-LCA patients at two visits. Also shown are images of a normal eye ac- left eye of nine GUCY2D-LCA patients at two visits. Also shown are images of a normal eye acquired quired at ages 7 and 14 years. Images from Visit 1 of P2 and both visits of P5 were not available. at ages 7 and 14 years. Images from Visit 1 of P2 and both visits of P5 were not available. The images The images from P1–P4 and P6 are within normal limits, and there is no obvious change on the fromsecond P1–P4 visit. and Retinal P6 are pigment within epithelium normal limits, (RPE) and abno therermalities is no obviousare present change in the on macula the second in P7, visit. P8, RetinalP9, and pigment P10 at Visit epithelium 1; no notable (RPE) changes abnormalities in appearance are present are in evident the macula in the in images P7, P8, at P9, Visit and 2. P10 at Visit 1; no notable changes in appearance are evident in the images at Visit 2. 2.3. Photoreceptor Nuclear Layer Imaging with OCT as a Quantitative Safety Outcome 2.3. Photoreceptor Nuclear Layer Imaging with OCT as a Quantitative Safety Outcome OCT cross-sectional images along the horizontal meridian including the fovea are OCT cross-sectional images along the horizontal meridian including the fovea are shown for a 45-year-old individual with normal vision at the two visits separated by 9 shown for a 45-year-old individual with normal vision at the two visits separated by 9 years years (Figure 2A). Representative horizontal scans through the fovea of one eye for 9 pa- (Figure2A). Representative horizontal scans through the fovea of one eye for 9 patients tients on each of the two visits are shown (Figure 2B). The patients are numbered in order on each of the two visits are shown (Figure2B). The patients are numbered in order of of increasing age at first exam (Supplementary Table S1). Inspection of the scans from Visit increasing age at first exam (Supplementary Table S1). Inspection of the scans from Visit 11 leads leads to to the the impression impression that that there there is is little little difference difference in in foveal foveal ONL ONL thickness thickness in in P1–P4, P1–P4, P6,P6, and and P7, P7, but but P8–P10 P8–P10 appear appear to to have have thinner thinner foveal foveal ONL ONL (Figure (Figure2B). 2B). P8 P8 also also shows shows apparentapparent deposits deposits in in the the central central few few degrees, degrees near, near the the RPE/Bruch’s RPE/Bruch’s membrane membrane region; region; P10 P10 hashas similar similar deposits deposits but but in in the the parafoveal parafoveal area. area. Quantitation Quantitation of theof the foveal foveal ONL ONL thickness thickness at bothat both visits visits and and in bothin both eyes eyes (all (all except except P5) P5) supports supports the the impression impression from from the the qualitative qualitative observationobservation ofof thethe scansscans (Figure(Figure3 3A;A; SupplementarySupplementary TableTable S2).S2). P1–P4 P1–P4 and and P6–P7 P6–P7 had had fo- fovealveal ONL ONL thicknesses thicknesses at at the the first first visit that ranged from 6666 toto 9595 µµmm (median,(median, 8585µ µm);m); normalnormal thickness thickness rangesranges fromfrom 73 to 119 119 µmµm (a (agesges 8–50 8–50 years). years). Thickness Thickness in inP8, P8, P9, P9, and and P10 P10measured measured 46 46µm,µm, 40 40µm,µm, and and 10.6 10.6 µm,µm, respectively. respectively. Although this isis aa relativelyrelatively small small sample,sample, the the data data suggest suggest that that foveal foveal ONL ONL thickness thickness in in patients patients in in the the first first and and second second decades of life can be within or near normal limits, but reduced foveal ONL can occur at

Int. J. Mol. Sci. 2021, 22, 2031 4 of 12

later ages. The general tendency of decreasing foveal ONL with age (Figure 3A) was esti- mated using a linear mixed model with age as a fixed effect and the patient nested with eyes as a random effect. The rate of foveal ONL loss accordingly to this model is 0.90 µm/year (p < 0.001). Are there detectable changes in foveal ONL in the 4–7 year intervals sampled (Figure Int. J. Mol. Sci. 2021, 22, 2031 3B)? Eight of the nine patients showed relatively small decreases in foveal ONL thickness4 of 12 (P1: RE, 8.9 µm, LE, 4.8 µm; P2: RE, 9.2 µm, LE, 6.1 µm; P3: RE, 10.5 µm, LE, 8.2 µm; P4: RE, 7.5 µm, LE, 3.1 µm; P6: RE 6.1 µm, LE, 2.0 µm; P7: RE, 4.1 µm, LE, 1.0 µm; P8: RE, 5.3 decadesµm, LE, of6.1 lifeµm; can P9: be RE, within 2.2 µm, or LE, near 1.7 normal µm); and limits, P10 butshowed reduced <1 µm foveal changes ONL in canboth occur eyes. atChanges later ages. in average The general foveal tendency ONL thickness of decreasing in all but foveal the LE ONL of withP3 are age within (Figure the3 A)range was of estimatedtest–retest usingvariability a linear (9.9 mixed µm) which model was with ca agelculated as afixed as 2.77*SD, effect and where the SD patient is the nested mean withwithin-subjects eyes as a random standard effect. deviation The rate[18]. of The foveal observed ONL lossdifference accordingly between to thisright model and left is 0.90eyesµ wasm/year also (withinp < 0.001). test–retest variability.

FigureFigure 2.2.Central Central retinal retinal optical optical coherence coherenc tomographye tomography (OCT) (OCT) scans scans of GUCY2Dof GUCY2D–LCA–LCA patients patients at two at visits.two visits. (A) Horizontal (A) Horizontal scans scans through through the fovea the fovea of a normalof a normal subject subject at visits at visits separated separated by an by interval an of 9 years. The foveal photoreceptor outer nuclear layer (ONL) is labeled. (B) Scans of nine LCA patients (P1–P4, P6–P10) on two visits separated by years, as specified in Supplementary Table S1. T, temporal; N, nasal retina.

Are there detectable changes in foveal ONL in the 4–7 year intervals sampled (Figure3B) ? Eight of the nine patients showed relatively small decreases in foveal ONL thickness (P1: RE, 8.9 µm, LE, 4.8 µm; P2: RE, 9.2 µm, LE, 6.1 µm; P3: RE, 10.5 µm, LE, 8.2 µm; P4: RE, 7.5 µm, LE, 3.1 µm; P6: RE 6.1 µm, LE, 2.0 µm; P7: RE, 4.1 µm, LE, 1.0 µm; P8: RE, 5.3 µm, LE, Int. J. Mol. Sci. 2021, 22, 2031 5 of 12

Int. J. Mol. Sci. 2021, 22, 2031 5 of 12

6.1 µm; P9: RE, 2.2 µm, LE, 1.7 µm); and P10 showed <1 µm changes in both eyes. Changes in averageinterval foveal of 9 years. ONL The thickness foveal photorecep in all buttor the outer LE ofnuclear P3 are layer within (ONL) the is range labeled. of ( test–retestB) Scans of nine variabilityLCA patients (9.9 µm) (P1–P4, which P6–P10) was calculated on two visits as 2.77*SD,separated where by years, SD as is specified the mean in within-subjects Supplementary standardTable deviation S1. T, temporal; [18]. The N, nasal observed retina. difference between right and left eyes was also within test–retest variability.

Figure 3. Foveal outer nuclear layer (ONL) measurements in GUCY2D–LCA patients at two visits. (A) ONL thickness as a Figure 3. Foveal outer nuclear layer (ONL) measurements in GUCY2D–LCA patients at two visits. (A) ONL thickness as functiona function of patient of patient age shows age shows a trend a towardtrend toward reduction reduction of foveal of fo ONLveal withONL increasingwith increasing age. Serial age. Serial data aredata connected are connected by by solidsolid lines. lines. Black Black long dashedlong dashed lines lines represent represent the range the range of normal of normal ONL ONL thickness thickness for comparison for comparison [19]. Gray[19]. Gray dashed dashed line is line is the regressionthe regression line. (line.B) Difference (B) Difference in foveal in foveal ONL ONL thickness thickness from from first visitfirst versusvisit versus follow-up follow-up interval. interval. Short Short dashed dashed lines lines showshow test–retest test–retest variability. variability.

To measureTo measure the rod the photoreceptor rod photoreceptor structure structure longitudinally longitudinally in GUCY2D in -LCA,GUCY2D we-LCA, quan- we tifiedquantified ONL in the ONL superior in the retinal superior region retinal of highest region rodof highest density, rod known density, as the known rod hotspot as the rod (RHS)hotspot [20]. Vertical(RHS) [20]. OCT Vertical scans beginningOCT scans atbeginning the fovea at and the crossingfovea and the crossing RHS from the RHS each from visiteach for the visit patients for the are patients shown are (Figure shown4). (Figure The scans 4). ofThe P5 scans were notof P5 available. were not The available. range The of eccentricitiesrange of eccentricities where ONL where measurements ONL measurem were madeents were is marked made (whiteis marked dashed (white lines). dashed Unlikelines). the foveaUnlike ONL the fovea thickness, ONL which thickness, differed which between differed patients, between there patients, was no there obvious was no differenceobvious between difference patients between at the patients RHS. at the RHS. AverageAverage RHS RHS ONL ONL measurements measurements from from individual individual scans scans were were plotted plotted against against age age (Figure(Figure5A); the5A); results the results were were also tabulatedalso tabulated (Supplementary (Supplementary Table Table S2). InS2). five In patients,five patients, µ ONLONL thickness thickness demonstrated demonstrated small small decreases decreases in both in bo eyesth eyes (P1: RE,(P1: 3.3,RE, LE,3.3,3.6 LE, 3.6m; µm; P2: RE, P2: RE, µ µ µ µ 5.5, LE,5.5, 2.9LE, 2.9m; µm; P4: RE,P4: 0.6,RE, LE,0.6, 2.2LE, 2.2m; µm; P6: RE,P6: 0.8,RE, LE,0.8, 6.7LE, 6.7m; µm; P10: P10: RE, 1.9,RE, LE,1.9, 0.3LE, 0.3m), µm), whilewhile P3, P7,P3, andP7, and P8 showed P8 showed a small a small increase increase in average in average ONL ONL thickness thickness for thefor LEthe (4.1, LE (4.1, 4.9, and 2.2 µm, respectively) and a decrease for the RE (1.2, 3.3, and 1.4 µm, respectively). 4.9, and 2.2 µm, respectively) and a decrease for the RE (1.2, 3.3, and 1.4 µm, respectively). P9 showed a small increase in average ONL thickness for the RE (1 µm) and a decrease P9 showed a small increase in average ONL thickness for the RE (1 µm) and a decrease for the LE (0.6 µm). For all nine patients, the change in ONL thickness was within the for the LE (0.6 µm). For all nine patients, the change in ONL thickness was within the range of test–retest variability (7.13 µm), which was calculated in the same way as for the range of test–retest variability (7.13 µm), which was calculated in the same way as for the foveal region. Test–retest variability is shown (Figure5B; horizontal short dashed lines). foveal region. Test–retest variability is shown (Figure 5B; horizontal short dashed lines). The difference in average RHS ONL thickness between RE and LE for both visits was within The difference in average RHS ONL thickness between RE and LE for both visits was test–retest variability for all patients. There appeared to be a weak tendency for decreasing within test–retest variability for all patients. There appeared to be a weak tendency for ONL thickness with age in the RHS region, and this was estimated with the same model as decreasing ONL thickness with age in the RHS region, and this was estimated with the foveal ONL (Figure5A, gray dashed line); the rate of decline was 0.29 µm/year (p = 0.01). same model as foveal ONL (Figure 5A, gray dashed line); the rate of decline was 0.29 µm/year (p = 0.01).

Int. J. Mol. Sci. 2021, 22, 2031 6 of 12 Int. J. Mol. Sci. 2021, 22, 2031 6 of 12

FigureFigure 4. 4. SuperiorSuperior retinal retinal OCT OCT scans scans in inGUCY2DGUCY2D–LCA–LCA patients patients at attwo two visits. visits. (A) ( ANormal) Normal vertical vertical scansscans in in superior superior direction direction starting starting at at the the fovea. fovea. (B (B) )Scans Scans of of the the nine nine LCA LCA patients patients (P1–P4, (P1–P4, P6– P6–P10) P10)on twoon two visits visits separated separated by years,by years, as specified as specified in Supplementary in Supplementary Table Table S1. S1. S, superior S, superior retina. retina. White Whitevertical vertical dashed dashed lines lines outline outline the range the range of eccentricities of eccentricities within within the rodthe hotspotrod hotspot (RHS), (RHS), where where ONL ONLmeasurements measurements were were made. made.

Int. J. Mol. Sci. 2021, 22, 2031 7 of 12 Int. J. Mol. Sci. 2021, 22, 2031 7 of 12

Figure 5. Superior retinal ONL measurements in GUCY2D–LCA patients at two visits. (A) ONL thickness as a function Figure 5. Superior retinal ONL measurements in GUCY2D–LCA patients at two visits. (A) ONL thickness as a function of of patient age. Solid lines connect serial data. Black long dashed lines represent the range of normal ONL thickness for patient age. Solid lines connect serial data. Black long dashed lines represent the range of normal ONL thickness for com- comparison [19]. The gray medium dashed line is the regression line. (B) Differences in average ONL thickness from the parison [19]. The gray medium dashed line is the regression line. (B) Differences in average ONL thickness from the first first visit versus follow-up interval. Short dashed lines show test–retest variability. visit versus follow-up interval. Short dashed lines show test–retest variability.

2.4.2.4. Full-Field Full-Field Stimulus Stimulus Testing asas an an Efficacy Efficacy Outcome Outcome in GUCY2D-LCAin GUCY2D-LCA VisualVisual sensitivity sensitivity was was measured measured with with the FSTthe systemFST system we devised we specificallydevised specifically for LCA. for LCA.FST FST has beenhas been used toused quantify to quantify vision in vision many differentin many molecular different forms molecular of LCA forms over the of pastLCA over decade, including as an outcome in our RPE65-LCA trial [9,13,15,21,22]. FST the past decade, including as an outcome in our RPE65-LCA gene therapy trial sensitivities were measured using two color stimuli to determine photoreceptor mediation. [9,13,15,21,22].Analysis of the FST sensitivity sensitivities differences were for measured these chromatic using stimulitwo color indicated stimuli that to all determine sensitivi- pho- toreceptorties were rod-mediatedmediation. Analysis for the dark-adapted of the sensitivity (DA) Blue differences stimulus and for cone-mediated these chromatic for the stimuli indicatedlight-adapted that all (LA) sensitivities Red stimuli were except rod-medi in one patientated for (P6). the Thesedark-adapted stimuli were (DA) previously Blue stimulus andused cone-mediated to define vision for in GUCY2Dthe light-adapted-LCA and to (LA) compare Red structure stimuli withexcept function in one [12 patient,13]. (P6). These Sensitivitystimuli were changes previously between used visits to weredefine graphed vision forin GUCY2D each eye of-LCA the 10and patients to compare structure(Figure6 with) . All function numerical [12,13]. changes reported hereafter correspond to the average of both eyes unlessSensitivity specified. changes For the DAbetween Blue stimulus, visits were P4 hadgrap ahed 7.8 dB for decrease. each eye In of contrast, the 10 P3,patients who (Fig- urewas 6). of All similar numerical age and changes had recordings reported spanning hereafter a comparable correspond interval, to the had average only a of 0.5 both dB eyes decrease. P1, representing a younger age group and a slightly longer interval between unlessvisits, specified. had a 2.2 dBFor increase the DA in Blue sensitivity stimulus, to this P4stimulus. had a 7.8 P10, dB overdecrease. a similar In contrast, interval but P3, who wasin theof similar fourth decadeage and of had life at recordings the first visit spanning and fifth a decade comparable upon return, interval, showed had aonly 1.8 dBa 0.5 dB decrease.decrease P1, in sensitivity. representing The othera younger six patients, age grou P2, P6,p P7,and P5, a P9,slightly and P8, longer showed interval decreases between visits,of 0.4, had 3.1, a 4.2,2.2 dB 2.4, increase 0.42, and in 2.2 sensitivity dB over intervals to this stimulus. of 4.3, 4.2, P10, 7.3, 6.9,over 4.0, a similar and 4.5 interval years, but in respectively.the fourth decade of life at the first visit and fifth decade upon return, showed a 1.8 dB decreaseFor in the sensitivity. DA Blue stimulus, The other what six should patients be considered, P2, P6, P7, a statisticallyP5, P9, and significant P8, showed change? decreases of The0.4, changes3.1, 4.2, in 2.4, mean 0.42, sensitivity and 2.2 versus dB over follow-up intervals time of in 4.3, the patients4.2, 7.3, compared 6.9, 4.0, and with 4.5 our years, respectively.published FST inter-visit variability estimate with 95% and 99% confidence limits [15] are shown (Figure7). The sensitivity change across the follow-up interval did not exceed the For the DA Blue stimulus, what should be considered a statistically significant 95% limits in seven of the ten patients. Decreases in P4 and P7 (both eyes) and P6 (left eye change?only) did The reach changes the limit. in mean sensitivity versus follow-up time in the patients compared with ourFor published the LA Red FST stimulus, inter-visit all patients variability showed estimate changes with in sensitivity 95% and between 99% confidence the visits, limits [15]but are most shown did not (Figure exceed 7). the The inter-visit sensitivity variability change limits across (Figure the7). follow-up Average decreases interval indid not exceedsensitivity the 95% were limits 1.8, 0.84, in seven 1.8, 1.0, of 3.2, the 0.7, ten and patients. 0.6 dB over Decreases follow-up in intervalsP4 and P7 of 4.8,(both 4.4, eyes) 6.9, and P67.2, (left 7.3, eye 4, andonly) 4.5 did years reach for P4,the P3, limit. P1, P10, P7, P9, and P8, respectively. Average increases in sensitivityFor the LA were Red 5.2, stimulus, 1.7, and 0.7all dBpatients over follow-up showed intervalschanges ofin 4.3, sensitivity 4.2, and 6.9-yearsbetween for the vis- its,P2, but P6 most and P5,did respectively. not exceed Thethe inter-visit decrease in variability the right eye limits of P7 (Figure and the 7). increase Average in bothdecreases in sensitivity were 1.8, 0.84, 1.8, 1.0, 3.2, 0.7, and 0.6 dB over follow-up intervals of 4.8, 4.4, 6.9, 7.2, 7.3, 4, and 4.5 years for P4, P3, P1, P10, P7, P9, and P8, respectively. Average in- creases in sensitivity were 5.2, 1.7, and 0.7 dB over follow-up intervals of 4.3, 4.2, and 6.9- years for P2, P6 and P5, respectively. The decrease in the right eye of P7 and the increase in both eyes of P2 exceeded the 95% inter-visit variability limits. The increase in P2 may

Int. J. Mol. Sci. 2021, 22, 2031 8 of 12

reflect the maturation of the patient over this interval and an ability to perform the test

Int. J. Mol. Sci. 2021, 22, 2031 more reliably at the second visit. 8 of 12 An overall analysis (combining all data from all eyes) for sensitivity change versus time after baseline was performed separately for LA Red and DA Blue stimuli. In the LA Red results, the effect of time was not significant (p = 0.26). In the DA Blue results, there eyes of P2 exceeded the 95% inter-visit variability limits. The increase in P2 may reflect the was a small decreasing trend of 0.28 log units per decade (p ≤ 0.001). This tendency resem- maturation of the patient over this interval and an ability to perform the test more reliably blesat the data second found visit. in another form of LCA with seriously impaired vision [23].

FigureFigure 6. 6. Dark- andand light-adaptedlight-adapted chromatic chromatic full-field full-field stimulus stimulus test (FST) test (FST) results results in GUCY2D in GUCY2D–LCA –LCA patientspatients at twotwo visits. visits. Upper Upper panels: panels: Sensitivity Sensitivity for the for blue the stimulus blue stimulus in the dark-adapted in the dark-adapted state, which state, −2 whichwas detected was detected by rods inby all rods cases. in Lowerall cases. panels: Lower Sensitivity panels: forSensitivity the red stimulus for the on red a 10stimulus phot cd.m on a 10 phot cd.mwhite-2 background,white background, which was which detected was bydetected cones inby all cones patients in all except patients the RE except of P6. the Photoreceptor RE of P6. Photo- receptormediation mediation was determined was determined based on chromatic based on differences: chromatic dark-adapted differences: (DA) dark-adapted Blue minus (DA) DA Red Blue mi- nus(not DA shown); Red (not and shown); light-adapted and light-adapted (LA) blue (not shown)(LA) blue minus (not LA shown) Red. Theminus 0 dB LA sensitivity Red. The level 0 dB sensi- corresponds to −0.57 log 1/(phot cd.m−2). tivity level corresponds 10to −0.57 log10 1/(phot cd.m−2).

Int. J. Mol. Sci. 2021, 22, 2031 9 of 12 Int. J. Mol. Sci. 2021, 22, 2031 9 of 12

Figure 7. FST sensitivity changes from first visit in GUCY2D–LCA patients as a function of follow-up interval. Symbols, testing conditions, and stimulus levels correspond to those in Figure6. Dashed

lines indicate 95 and 99% confidence intervals for test–retest variability. Figure 7. FST sensitivity changes from first visit in GUCY2D–LCA patients as a function of follow- An overall analysis (combining all data from all eyes) for sensitivity change versus up interval. Symbols, testing conditions, and stimulus levels correspond to those in Figure 6. Dashedtime lines after indicate baseline 95 and was 99% performed confidence separately intervals for for test–retest LA Red variability. and DA Blue stimuli. In the LA Red results, the effect of time was not significant (p = 0.26). In the DA Blue results, 3. Summarythere was and a small Conclusions decreasing trend of 0.28 log units per decade (p ≤ 0.001). This tendency resembles data found in another form of LCA with seriously impaired vision [23]. Patients with GUCY2D-LCA were assessed twice using structural and functional ex- aminations3. Summary with andan interval Conclusions of 4 to 7 years between visits. Conclusions about disease pro- gressionPatients are as follows: with GUCY2D -LCA were assessed twice using structural and functional examinations(1) En face imaging with an performed interval of with 4 to NIR-RAFI 7 years between methods visits. revealed Conclusions no abnormalities about disease in the progressionfirst two decades are as follows:of life, but there were central RPE changes in the older patients of this cohort.(1) EnProgression face imaging of these performed changes with on NIR-RAFIthe second methods visit was revealed not evident. no abnormalities in the(2) first Cross-sectional two decades ofretinal life, butimaging there wereat the central fovea RPEand changesin the superior in the older retina patients was per- of this formedcohort. with Progression OCT. Changes of these were changes measurable, on the given second carefully-acquired visit was not evident. high quality scans (despite (2)the Cross-sectional nystagmus), in retinal9 of 10 imagingpatients. atAt thethe foveafovea, and only in one the patient superior in one retina eye was had per- a decreaseformed in with thickness OCT. Changes that would were be measurable, considered givenoutside carefully-acquired the range of inter-visit high quality variabil- scans ity.(despite Changes the in nystagmus),the superior inretinal 9 of10 ONL patients. were all At attributable the fovea, only to inter-visit one patient variability in one eye and had a notdecrease to structural in thickness changes. that would be considered outside the range of inter-visit variability. Changes(3) Visual in thesensitivity superior was retinal quantified ONL were separately all attributable for rods to and inter-visit cones using variability dark- andand not light-adaptedto structural chromatic changes. FST. Analyzed individually, most patients did not show signifi- cant changes(3) Visual in rod sensitivity function. wasExceptions quantified were separately the decreases for rods in rod and function cones using in both dark- eyes and of twolight-adapted patients (1.0 chromatic and 0.4 log FST. unit Analyzed over 4.8 individually, and 7.3 years) most and patients in one eye did notof a showthird significantpatient changes in rod function. Exceptions were the decreases in rod function in both eyes of two

Int. J. Mol. Sci. 2021, 22, 2031 10 of 12

patients (1.0 and 0.4 log unit over 4.8 and 7.3 years) and in one eye of a third patient (0.3 log unit over 4.2 years). One of these patients also showed a decrease of similar magnitude in cone function, but only in one eye (0.4 log unit in 7.3 years). There was a fourth patient showing an increase in cone function over a 4.3-year interval. A trend analysis combining all available data showed a small decrease in rod function (0.28 l.u. per decade for DA Blue), and a rate for LA Red that was not statistically significant. In summary, given the appropriate expertise in recording, analyzing, and interpreting relevant data from these seriously visually impaired GUCY2D-LCA patients, change could be detected with the quantitative methods used. The detected changes in this group of patients, however, were mostly attributable to inter-visit variability. There were no major negative changes in structure and/or function in this cohort and over the intervals studied.

4. Materials and Methods 4.1. Subjects Ten patients with GUCY2D-LCA were evaluated at two visits. At the first visit, the patients ranged in age from 7 to 37 years, and they were re-evaluated after intervals of approximately 4–7 years. All patients had molecular testing in a CLIA (Clinical Laboratory Improvement Amendments)-approved laboratory. Procedures followed the Declaration of Helsinki, and the study was approved by the relevant institutional review board. Informed consent, assent, and parental permission were obtained, and the work was HIPAA (Health Insurance Portability and Accountability Act)-compliant.

4.2. Near-Infrared Autofluorescence A confocal scanning laser ophthalmoscope (Spectralis HRA; Heidelberg Engineer- ing, Heidelberg, Germany) was used to record en face images and estimate RPE health. NIR-RAFI was performed using methods previously described [16,17]. For NIR-RAFI, excitation at 790 nm was used at 100% laser power setting and 105% detector sensitiv- ity. All images were acquired at high speed mode (30◦ × 30◦ square field sampled onto 768 × 768 pixels) and with the automatic normalization feature turned off. The manufac- turer’s ART (Automatic Real Time-function) feature was used whenever possible with a 21-frame average.

4.3. Optical Coherence Tomography Cross-sectional images of the retina were obtained with a spectral-domain OCT in- strument (RTVue-100; Optovue Inc., Fremont, CA, USA). The scans were 30◦ in length along the horizontal and vertical meridians to study the retinal changes at the fovea and at the rod-rich region 10–15◦ in the superior retina (rod hotspot, RHS). To measure ONL thickness at the fovea, three horizontal scans from each visit for each patient were selected based on the shape of the foveal depression (criteria for selection: similar contour of the foveal depression in all three scans, deepest pit, no inner retinal layers or thinnest ones, and the same position of blood vessels). The foveal ONL thickness was defined as the distance between the internal limiting membrane and outer limiting membrane, and it was measured with custom programs (MATLAB R2018a, MathWork, Natick, MA, USA). Unlike at the fovea, where one measurement was performed, ONL thickness at RHS represents the average of five measurements along the vertical scan at 0.5-degree increments within a range of 10–12◦ superiorly. These measurements were also performed on three scans except for P1, LE, and P10, LE, where only two scans were available.

4.4. Visual Function FST was performed to estimate perceptual light sensitivity as previously described [11,17]. Stimuli were 200 ms photopically-matched blue and red flashes (peak 465 nm and 637 nm, respectively) in the dark-adapted (DA) and light-adapted (LA, steady white 10 phot cd.m−2 background) states [15,21] using a ganzfeld stimulator (Colordome; Diagnosys LLC, Little- ton, MA, USA). Participants responded when they perceived the lights presented. Several Int. J. Mol. Sci. 2021, 22, 2031 11 of 12

threshold estimates for each color were obtained within a session and averaged. Photore- ceptor mediation was determined by sensitivity differences between the chromatic stimuli. Differences higher than 3.2 dB indicate rod-mediation of the blue stimuli, and those lower than 22.2 dB indicate cone mediation of the red stimuli.

Supplementary Materials: Supplementary materials are available online at https://www.mdpi. com/1422-0067/22/4/2031/s1. Author Contributions: Conceptualization, S.G.J. and A.V.C.; data curation, S.G.J., A.V.C., A.S., A.J.R., A.K.K.; formal analysis, A.S., A.J.R., V.W., M.S.; funding acquisition, S.G.J.; project administration, R.S., A.V.G.; writing—original draft, S.G.J.; writing—review and editing, all authors. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Brynn’s Fight for Sight, Brier Fund for LCA Research, Sanofi US Services, Inc., and Research to Prevent Blindness (Departmental Grant), and was encouraged by the tireless efforts of Lea Bramnick. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of the University of Pennsylvania. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Sacchetti, M.; Mantelli, F.; Merlo, D.; Lambiase, A. Systematic eview of Randomized Clinical Trials on Safety and Efficacy of Pharmacological and Nonpharmacological Treatments for Pigmentosa. J. Ophthalmol. 2015, 2015, 1–11. [CrossRef] 2. Thompson, D.A.; Iannaccone, A.; Ali, R.R.; Arshavsky, V.Y.; Audo, I.; Bainbridge, J.W.B.; Besirli, C.G.; Birch, D.G.; Branham, K.E.; Cideciyan, A.V.; et al. Advancing Clinical Trials for Inherited Retinal Diseases: Recommendations from the Second Monaciano Symposium. Transl. Vis. Sci. Technol. 2020, 9, 2. [CrossRef] 3. Donato, L.; Scimone, C.; Alibrandi, S.; Rinaldi, C.; Sidoti, A.; D’Angelo, R. Transcriptome Analyses of lncRNAs in A2E-Stressed Retinal Epithelial Cells Unveil Advanced Links between Metabolic Impairments Related to Oxidative Stress and . Antioxidants 2020, 9, 318. [CrossRef] 4. Scimone, C.; Alibrandi, S.; Scalinci, S.Z.; Battagliola, E.T.; D’Angelo, R.; Sidoti, A.; Donato, L. Expression of Pro-Angiogenic Markers Is Enhanced by Blue Light in Human RPE Cells. Antioxidants 2020, 9, 1154. [CrossRef][PubMed] 5. Donato, L.; Scimone, C.; Alibrandi, S.; Pitruzzella, A.; Scalia, F.; D’Angelo, R.; Sidoti, A. Possible A2E Mutagenic Effects on RPE Mitochondrial DNA from Innovative RNA-Seq Bioinformatics Pipeline. Antioxidants 2020, 9, 1158. [CrossRef] 6. Donato, L.; Scimone, C.; Alibrandi, S.; Abdalla, E.M.; Nabil, K.M.; D’Angelo, R.; Sidoti, A. New Omics–Derived Perspectives on Retinal Dystrophies: Could Ion Channels-Encoding or Related Act as Modifier of Pathological Phenotype? Int. J. Mol. Sci. 2020, 22, 70. [CrossRef] 7. Hollander, A.I.D.; Roepman, R.; Koenekoop, R.K.; Cremers, F.P. Leber congenital amaurosis: Genes, proteins and disease mechanisms. Prog. Retin. Eye Res. 2008, 27, 391–419. [CrossRef][PubMed] 8. Kondkar, A.A.; Abu-Amero, K.K. Leber congenital amaurosis: Current genetic basis, scope for genetic testing and person-alized medicine. Exp. Eye Res. 2019, 189, 107834. [CrossRef] 9. Cideciyan, A.V.; Jacobson, S.G. Proctor Award Lecture: Leber congenital amaurosis (LCA): Potential for improvement of vision. Investig. Ophthalmol. Vis. Sci. 2019, 60, 1680–1695. [CrossRef] 10. Boye, S.E. Insights gained from gene therapy in animal models of retGC1 deficiency. Front. Mol. Neurosci. 2014, 7, 43. [CrossRef] [PubMed] 11. Boye, S.E. Leber Congenital Amaurosis Caused by Mutations in GUCY2D. Cold Spring Harb. Perspect. Med. 2014, 5, a017350. [CrossRef] 12. Jacobson, S.G.; Cideciyan, A.V.; Peshenko, I.V.; Sumaroka, A.; Olshevskaya, E.V.; Cao, L.; Schwartz, S.B.; Roman, A.J.; Olivares, M.B.; Sadigh, S.; et al. Determining consequences of retinal membrane guanylyl cyclase (RetGC1) deficiency in human Leber congenital amaurosis en route to therapy: Residual cone-photoreceptor vision correlates with biochemical properties of the mutants. Hum. Mol. Genet. 2012, 22, 168–183. [CrossRef] 13. Jacobson, S.G.; Cideciyan, A.V.; Sumaroka, A.; Roman, A.J.; Charng, J.; Lu, M.; Choudhury, S.; Schwartz, S.B.; Heon, E.; Fishman, G.A.; et al. Defining Outcomes for Clinical Trials of Leber Congenital Amaurosis Caused by GUCY2D Mutations. Am. J. Ophthalmol. 2017, 177, 44–57. [CrossRef] Int. J. Mol. Sci. 2021, 22, 2031 12 of 12

14. Bouzia, Z.; Georgiou, M.; Hull, S.; Robson, A.G.; Fujinami, K.; Rotsos, T.; Pontikos, N.; Arno, G.; Webster, A.R.; Hardcastle, A.J.; et al. GUCY2D-Associated Leber Congenital Amaurosis: A Retrospective Natural History Study in Preparation for Trials of Novel . Am. J. Ophthalmol. 2020, 210, 59–70. [CrossRef] 15. Roman, A.J.; Cideciyan, A.V.; Aleman, T.S.; Jacobson, S.G. Full-field stimulus testing (FST) to quantify visual perception in severely blind candidates for treatment trials. Physiol. Meas. 2007, 28, N51–N56. [CrossRef][PubMed] 16. Cideciyan, A.V.; Swider, M.; Aleman, T.S.; Roman, M.I.; Sumaroka, A.; Schwartz, S.B.; Stone, E.M.; Jacobson, S.G. Reduced- illuminance autofluorescence imaging in ABCA4-associated retinal degenerations. J. Opt. Soc. Am. A Opt. Image Sci. Vis. 2007, 24, 1457–1467. [CrossRef][PubMed] 17. Cideciyan, A.V.; Swider, M.; Jacobson, S.G. Autofluorescence imaging with near-infrared excitation: Normalization by reflectance to reduce signal from choroidal fluorophores. Invest. Ophthalmol. Vis. Sci. 2015, 56, 3393–3406. [CrossRef][PubMed] 18. Bland, J.M.; Altman, D.G. Measurement error. BMJ 1996, 312, 1654. [CrossRef][PubMed] 19. Sumaroka, A.; Garafalo, A.V.; Cideciyan, A.V.; Charng, J.; Roman, A.J.; Choi, W.; Saxena, S.; Aksianiuk, V.; Kohl, S.; Wissinger, B.; et al. Blue Cone Caused by the C203R Missense Mutation or Large Deletion Mutations. Investig. Opthalmology Vis. Sci. 2018, 59, 5762–5772. [CrossRef][PubMed] 20. Curcio, C.A.; Sloan, K.R.; Kalina, R.E.; Hendrickson, A.E. Human photoreceptor topography. J. Comp. Neurol. 1990, 292, 497–523. [CrossRef][PubMed] 21. Roman, A.J.; Schwartz, S.B.; Aleman, T.S.; Cideciyan, A.V.; Chico, J.D.; Windsor, E.A.; Gardner, L.M.; Ying, G.-S.; Smilko, E.E.; Maguire, M.G.; et al. Quantifying rod photoreceptor-mediated vision in retinal degenerations: Dark-adapted thresholds as outcome measures. Exp. Eye Res. 2005, 80, 259–272. [CrossRef][PubMed] 22. Jacobson, S.G.; Cideciyan, A.V.; Ratnakaram, R.; Heon, E.; Schwartz, S.B.; Roman, A.J.; Peden, M.C.; Aleman, T.S.; Boye, S.L.; Sumaroka, A.; et al. Gene therapy for Leber congenital amaurosis caused by RPE65 mutations: Safety and efficacy in 15 children and adults followed up to 3 years. Arch. Ophthalmol. 2012, 130, 9–24. [CrossRef][PubMed] 23. Jacobson, S.G.; Cideciyan, A.V.; Sumaroka, A.; Roman, A.J.; Charng, J.; Lu, M.; Choi, W.; Sheplock, R.; Swider, M.; Kosyk, M.S.; et al. Outcome Measures for Clinical Trials of Leber Congenital Amaurosis Caused by the Intronic Mutation in the CEP290 Gene. Investig. Opthalmology Vis. Sci. 2017, 58, 2609–2622. [CrossRef][PubMed]