PURPOSE
To describe the natural history of optical coherence tomography (OCT) and mesopic microperimetry (MP) parameters in the Rate of Progression of USH2A -related Retinal Degeneration (RUSH2A) study.
DESIGN
Multicenter, international, prospective, longitudinal natural history study.
METHODS
Study Population: 105 participants with biallelic disease-causing variants in USH2A with a clinical diagnosis of Usher Syndrome or autosomal recessive retinitis pigmentosa who underwent annual testing over 4 years.
MAIN OUTCOME MEASURE(S)
Ellipsoid zone (EZ) area and central sub-field thickness (CST) on OCT. Mean sensitivity (MS) and mean pointwise sensitivity at predefined functional transition points (MS FTP ) on MP.
RESULTS
A total of 101 and 94 participants met criteria for OCT and MP analysis, respectively. The average EZ area was 3.9 ± 5.3 mm 2 at baseline which decreased to 3.3 ± 4.1 mm 2 at 4 years (rate of change [ROC] [−0.18 mm 2 /year]). The average CST decreased from 249.7 ± 35.6 to 244.1 ± 39.3 microns over 4 years (−2.14 microns/year). The average MP MS declined from 6.0 ± 5.1 dB (baseline) to 4.5 ± 4.2 (4 years) (−0.39 dB/year). The MS FTP significantly decreased over time (17.9 dB [baseline] to 10.8 dB [4 years]). Notably, at 4 years, there was a meaningful decrease (≥ 7dB) in MS FTP in 46% of study eyes.
CONCLUSIONS
This study establishes the natural history of key structural and functional parameters in USH2A -retinopathy. The MS FTP is a novel, robust parameter that showed clinically meaningful change over time and is a promising tool to monitor treatment efficacy in clinical trials.
U sher syndrome (USH) encompasses a heterogenous group of autosomal recessive (AR) disorders that result in dual sensory impairment (deafness and retinitis pigmentosa [RP]) with or without vestibular dysfunction. Based on clinical symptoms and severity, four sub-types of USH (Types 1, 2, 3 and atypical USH [Usher 4]) involving at least 15 different genes, have been described. Patients affected with USH Type 2 typically have mild to moderate congenital hearing impairment and retinal degeneration that begins in the second or third decade of life. Three genes have been implicated in USH Type 2 ( USH2A, ADGRV1 , and WHRN ). Disease causing variants in Usherin ( USH2A ) cause USH Type 2A (USH2) and account for over 50% of cases of Usher syndrome. Intriguingly, some genes implicated in USH also cause ARRP without hearing loss, also termed nonsyndromic ARRP, and USH2A is the commonest cause of nonsyndromic ARRP (12%-25%). , Biallelic truncating variants in USH2A almost always lead to USH2, whereas biallelic missense variants in the USH2A interconnection domain lead to nonsyndromic ARRP. Eyes affected with USH2A- related nonsyndromic ARRP have progressive rod-cone degeneration, the onset and severity of which is delayed in comparison to USH2.
The visual acuity, visual field and electroretinography (ERG) in USH2A -related disease have been described in some large cohorts. However there is a paucity of literature describing the longitudinal natural history of the retinal phenotype using quantitative structural imaging (spectral-domain optical coherence tomography [SD-OCT]) and functional (mesopic fundus-guided microperimetry [MP]) modalities in USH2 and ARRP. These modern tools enable better understanding of visual function through higher resolution structure-function correlation in eyes with inherited retinal dystrophies (IRDs) than older modalities, and provide an opportunity to develop clinically meaningful outcome measures for therapeutic clinical trials. For instance, the ellipsoid zone (EZ) width and area measurements from SD-OCT are known to have higher reliability than visual acuity, visual field, and ERG in IRDs. , Further, fundus-guided MP, a tool that provides precise topographic evaluation of macular function with registration of stimuli to macular locations, is correlated with macular OCT measures, thus providing in-depth structure-function correlation. ,
The Rate of Progression of USH2A -related Retinal Degeneration (RUSH2A) study is an international multicenter investigation designed to characterize the natural history of USH2A -related retinal degeneration. , , Baseline macular SD-OCT and MP parameters of the RUSH2A cohort showed similar EZ area and MP mean sensitivity characteristics in USH2 and ARRP phenotypes. The objectives of this report are to describe the 4-year progression of OCT and mesopic MP parameters in participants and assess macular structure-function associations observed during the study. Further, as newer treatments are being developed for USH2A -related IRD (NCT04355689, NCT05085964), this report will also examine the usefulness of MP parameters as potential clinical trial outcome measures in USH2A -related IRD.
MATERIALS AND METHODS
STUDY DESIGN
Participants were enrolled in the RUSH2A study (NCT03146078) at 16 clinical sites in Europe and North America. The study was approved by the ethics boards at each site and adhered to the tenets of the Declaration of Helsinki. All participants provided written consent. The study design, inclusion and exclusion criteria, procedures, and findings from MP and OCT at baseline have been published previously. Briefly, participants were at least 8 years of age with rod-cone degeneration associated with biallelic pathogenic or likely pathogenic sequence variants in USH2A (as per published guidelines) . Participants with a baseline best corrected visual acuity (BCVA) with Early Treatment of Diabetic Retinopathy Study (ETDRS) letter score of 54 or greater (Snellen equivalent 20/80 or better) in the study eye, kinetic visual field of at least 10 degrees diameter in all meridians using the III4e target (Octopus 900 Pro, Haag Streit, Mason, Ohio), and stable fixation were enrolled in the primary cohort (N = 105). The study eye was defined as the eye with better visual acuity at baseline, and participants were examined annually. Patients reported sex as male or female. Tests of visual function and imaging were conducted by study-certified clinical center staff using standardized protocols. Testing was in the study eye only except that OCT and BCVA were assessed in both eyes. The analysis cohorts for this report consist of study eyes that have test results available for at least 2 of the 5 time points, including baseline and 1 to 4 years.
OCT IMAGING AND GRADING
High resolution OCT images centered at the fovea [30° x 25° volume scan (121 B-scans) and 30° x 5° 7-line raster scan] were obtained using a Heidelberg Spectralis HRA+OCT unit (Heidelberg Engineering GmbH, Heidelberg, Germany) as previously described.
Duke Reading Center readers assessed the OCT scans; the grayscale was used for additional contrast. Retinal cystic changes were identified within the retinal layers as well-defined, black or dark round or oval shapes and were differentiated from cavitations and outer retinal tubulations, as previously described. Central subfield thickness (CST) was measured semi-automatically using HEYEX software (Heidelberg Engineering GmBH, Heidelberg, Germany). Readers adjusted image centration and inner and outer segmentation boundaries as needed. Readers used the Duke Optical Coherence Tomography Retinal Analysis Program (DOCTRAP) to manually annotate A-scans with the intact EZ on each B-scan obtained from OCT macular volumes. , The annotated A-scans were used to measure EZ area. After the first reader completed the B-scan gradings, a second senior reader reviewed them and corrected them when needed.
MICROPERIMETRY TESTING
Fundus-guided mesopic (standard) MP was performed at baseline (3 tests per participant) and annually (1 test per participant) using a Macular Integrity Assessment (MAIA-2) unit (iCare, Raleigh, NC) with software version 1.7 or higher as described earlier. Each participant completed the 3 baseline tests within the same day. Briefly, following pupil dilation and ten-minute mesopic adaptation, a custom circular grid consisting of 89 points that covered the retinal area within the arcades was tested. Testing points were arranged in concentric rings located at 2°, 4°, 6.5°, 9°, 12° and 15° from the foveal center ( Figure 1 A). The foveal area was determined based on the red-free fundus image as described previously.
The follow-up MP was performed with baseline test 1 set as the reference test per protocol . As fatigue is known to negatively affect threshold sensitivity of perimetry (including MP), , , and the fact that the first test during the day often shows higher sensitivity measures, using baseline reference test 1 for follow-up analysis was considered more robust. However, during the study, some of the follow-up tests were incorrectly referenced to baseline test 2 or 3; in these cases, functional transition points (FTP) selection employed the reference test 2 or 3 instead of test 1 (n = 5 and 18, respectively).
MICROPERIMETRY ANALYSIS
The US Food and Drug Administration (FDA) has provided guidance on clinically meaningful change in MP as a mean change of sensitivity of ≥ 7dB across ≥ 5 prespecified points. , Hence, multiple analyses were performed on MP data to identify the most robust outcome measure that is likely to show ≥ 7dB mean change over the study duration in our cohort. Mean sensitivity (MS) values from points across the entire MP grid and from the points located at various eccentricities (ring analysis: rings 1 to 5) were ascertained.
Defining functional transition points (FTP)
The MP loci located on the inner border of a functional transition area were identified using a method similar to that described previously. , ‘Pathways’ of loci, commencing at the central locus and ending at the most peripheral ring of loci, were created ( Figure 1 B). A large decrease in sensitivity (≥7db) between adjacent loci along any of these pathways indicated a likely transition from within the EZ to outside the EZ . These inner loci would be expected to have a greater decrease in sensitivity as the EZ contracted over time, and all such inner loci with a baseline value of ≥ 8dB qualified as a candidate FTP. The percentage of qualifying adjacent candidate FTPs, ranging from 100% to 25%, were used to rank the the candidate FTPs . All candidate FTPs that were qualified by 100% of adjacent candidate FTPs were selected as a FTP. After these points had been selected, if the selected number of FTPs was less than 5, then candidate FTPs with the next highest percentage of adjacent qualifying candidate FTPs were included as FTPs until 5 points were selected. Eyes with less than 5 FTPs at baseline were excluded from FTP analysis. Mean sensitivity of FTPs (MS FTP ) was calculated for each eye at each visit.
ADDITIONAL TESTING
BCVA was evaluated using ETDRS charts , or the electronic version of the test. Static perimetry (SP) was performed using the Octopus 900 device (Haag-Streit, Mason, OH). Full-field stimulus threshold (FST) measurements were obtained using the Espion E 3 system (Diagnosys LCC, Lowell, MA).
STATISTICAL METHODS
The distributions of OCT and MP measures for the study eye at each visit were summarized by means, standard deviations (SDs), medians, interquartile ranges (IQRs) and ranges. Spaghetti plots were used to display the trajectories for each eye by the reported duration of disease (time since onset of symptoms). To address a potential floor effect, a cohort with the capacity to exhibit further progression was identified for each measure. Preserved cohorts of eyes were defined by determining a threshold baseline value above which the slopes of the measure over time were greater than 0. Because the values of CST are affected by the presence of cystoid macular edema (CME), CST values at visits when the eye had CME were excluded from analysis.
The annual rate of change of EZ area, CST, and the mean sensitivity across all 89 MP loci (MP MS) was ascertained in the entire cohort and in the clinical sub-groups (USH2 and ARRP). Mixed effects longitudinal regression models with a random intercept were employed to estimate the annual rates of change, accompanied by 95% confidence intervals (95% CI). Time was calculated by dividing the number of days from baseline by 365.25. Additionally, an outlier down-weighted model was utilized to account for exceptional rates of change. For the outlier down-weighted model, the rate of decline for each eye was first estimated using a simple linear regression model. Next, a robust regression model using M-estimation with a Huber weighting function was used to determine the weights for eyes in the mixed effect model.
The changes in mean pointwise sensitivity (PWS) at various eccentricities ( Figure 1 A) (Ring 1 [points 1-5], Ring 2 [points 6-13], Ring 3 [points 14-25], Ring 4 [points 26-41], Ring 5 [points 42-89]) were calculated for the preserved cohort. The annual rate of change for these measures was calculated using the longitudinal regression models described above. The FTPs of each participant eye that met our inclusion criteria at baseline were used to calculate the percentage of eyes that met the FDA guidance of ≥ 7dB mean change in MP FTP . , , The coefficient of repeatability (CoR) for the MP MS is 2.2 dB and for PWS (individual points) is 8.9 dB, as previously published. ,
To evaluate the correlation between changes in OCT and MP measures with changes in other functional measures (BCVA, FST, and SP) from baseline to 4 years, Spearman correlation coefficients (r s ) were calculated. Inter-eye symmetry (right vs left) of OCT measures and changes was assessed with intraclass coefficients (ICCs).
All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA), and reported P-values are two-sided.
RESULTS
STUDY POPULATION
Of the 105 participants in the primary cohort, 101 met the inclusion criterion for OCT analysis, with a minimum of two available data points during the study period. Since MP measurements were conducted only in selected clinical centers, the MP analysis included 94 study eyes. Among the participants in the OCT analysis cohort, 63 (62%) were diagnosed with USH2 and 38 (38%) were diagnosed with ARRP. The mean age of the participants at baseline was 37 years (SD, 13), and 57 (56%) were female. The median duration of disease at enrollment was 13 years (IQR, 7 to 20). Ninety-five (92%) of 103 living participants completed OCT testing at 4 years. Of those, two were ungradable.
In the MP analysis cohort of 94 participants, 56 (60%) were diagnosed with USH2 and 38 (40%) were diagnosed with ARRP. The mean age of the MP cohort at baseline was 38 years (SD, 13), and 53 (56%) were female. The median duration of disease at enrollment was 13 years (IQR, 7-20). Eighty-seven (84%) of 103 living participants completed MP testing at 4 years. Of those, two were excluded as their testing were unreliable.
OCT OUTCOMES
The EZ area for all study eyes in the OCT cohort was 3.9 ± 5.3 mm 2 at baseline, which decreased to 3.3 ± 4.1 mm 2 at Year 4 visit ( Table 1 ). Figure 2 A displays a spaghetti plot of EZ area values for each study eye plotted against the duration of disease at each visit, demonstrating an overall progressive reduction in EZ area over time. Sixty-eight (67%) of eyes with baseline EZ area ≤ 3 mm 2 exhibited a floor effect; hence, eyes with baseline EZ area > 3 mm 2 comprised the preserved cohort. In the preserved cohort, EZ area was 9.2 ± 6.5 mm 2 at baseline, which decreased to 7.4 ± 4.9 mm 2 at Year 4 ( Table 1 ). In the entire cohort, the average change of EZ area was −0.18 (95% CI, −0.24, −0.12) mm 2 /year, and in the preserved cohort, it was −0.53 (95%CI, −0.70, −0.37) mm 2 /year ( Table 2 ). The estimated decline based on the outliers down-weighted model was −0.34 (−0.47, −0.22) mm 2 /year ( Table 2 ). Notably, the annual rates of change of EZ area were higher in USH2 (vs ARRP) in the preserved OCT cohorts (difference in slope = 0.38; p = 0.01; e-Table 1). The annual rate of change for EZ area was 1.22 mm 2 /year greater with higher baseline value (≥11 mm 2 ) than lower baseline value (<5 mm 2 ) and 0.34 mm 2 /year greater with younger enrollment age (< 30 years) than older enrollment age (≥50 years) (e-Table 1).
| Outcomes | Baseline a | Year 1 | Year 2 | Year 3 | Year 4 b |
|---|---|---|---|---|---|
| Entire Cohort | |||||
| EZ Area (mm 2 ) | |||||
| N | 101 | 100 | 86 | 96 | 93 |
| Mean ± SD | 3.9 ± 5.3 | 3.9 ± 5.1 | 3.7 ± 4.7 | 3.5 ± 4.4 | 3.3 ± 4.1 |
| Range | 0.0 to 28.3 | 0.0 to 28.1 | 0.1 to 24.6 | 0.0 to 20.8 | 0.0 to 22.5 |
| CST w/o CME (microns) | |||||
| N | 72 | 62 | 52 | 56 | 49 |
| Mean ± SD | 249.7 ± 35.6 | 245.9 ± 35.7 | 244.9 ± 34.5 | 240.9 ± 34.5 | 244.1 ± 39.3 |
| Range | 159.0 to 323.0 | 155.0 to 323.0 | 150.0 to 320.0 | 145.0 to 324.0 | 172.0 to 322.0 |
| MP Mean Sensitivity (dB) | |||||
| N b | 94 | 84 | 79 | 81 | 77 |
| Mean ± SD | 6.0 ± 5.1 | 5.5 ± 5.0 | 4.7 ± 4.3 | 4.5 ± 3.9 | 4.5 ± 4.2 |
| Range | 0.0 to 22.2 | 0.0 to 22.1 | 0.3 to 19.4 | 0.0 to 18.6 | 0.0 to 19.1 |
| MP Average sensitivity within Intact EZ Area | |||||
| N | 81 | 77 | 72 | 77 | 80 |
| Median (Q1, Q3) | 23.0 (21.0, 25.3) | 23.0 (21.0, 25.1) | 22.8 (21.0, 25.0) | 23.6 (20.8, 25.4) | 23.1 (20.0, 25.2) |
| Range | 13.3 to 29.7 | 1.7 to 29.5 | 14.0 to 29.2 | 2.0 to 28.9 | −1.0 to 27.8 |
| MP Average sensitivity outside intact EZ area | |||||
| N | 83 | 79 | 74 | 80 | 82 |
| Median (Q1, Q3) | 2.4 (1.6, 5.5) | 2.4 (0.9, 5.4) | 1.7 (0.8, 4.3) | 1.9 (0.9, 3.9) | 1.7 (0.6, 3.7) |
| Range | −0.5 to 16.7 | −0.5 to 17.3 | −0.2 to 17.6 | −0.3 to 15.8 | −0.6 to 14.7 |
| EZ Functional Transition points | |||||
| Median (Q1, Q3) | 18.0 (13.0, 22.0) | 16.0 (9.0, 21.0) | 14.0 (7.0, 20.0) | 12.0 (3.0, 20.0) | 11.0 (1.5, 19.0) |
| Range | 8.0 to 36.0 | −1.0 to 36.0 | −1.0 to 32.0 | −1.0 to 33.0 | −1 to 32.0 |
| Preserved Cohort | |||||
|---|---|---|---|---|---|
| EZ Area (mm 2 ) | |||||
| N | 33 | 33 | 29 | 32 | 31 |
| Mean ± SD | 9.2 ± 6.5 | 8.9 ± 6.3 | 8.3 ± 5.6 | 7.9 ± 5.2 | 7.4 ± 4.9 |
| Range | 3.2 to 28.3 | 2.4 to 28.1 | 3.2 to 24.6 | 2.2 to 20.8 | 2.4 to 22.5 |
| MP Mean Sensitivity (dB) | |||||
| N | 86 | 77 | 71 | 73 | 71 |
| Mean ± SD | 6.5 ± 5.1 | 6.0 ± 4.9 | 5.2 ± 4.3 | 4.9 ± 3.9 | 4.8 ± 4.2 |
| Range | 1.1 to 22.2 | 0.8 to 22.1 | 0.7 to 19.4 | 0.6 to 18.6 | 0.0 to 19.1 |
a At the baseline visit, 2 study eyes had ungradable EZ, and 39 of the study eyes had missed CST due to CME. MP MS were missed for various reasons: no MAIA (11), test not done (1), and ungradable (6).
b At the 48-month visit, OCT was missed for reasons such as death (2), missing visit (5), virtual visit (2), test not done (1), and 2 ungradable cases. For study eyes, 46 had missed CST due to death (2), missing visit (5), virtual visit (2), test not done (1), and CME (36). MP MS were missed due to death (2), missing visit (5), virtual visit (2), no MAIA (1), test not done (8), and ungradable (2).SD is standard deviation; IQR is interquartile range.
| —–Preserved Cohort b —– | |||
|---|---|---|---|
| Outcomes | Entire cohort a | All | Outliers Down-Weighted c |
| EZ Area (mm 2 /year) | N = 101 | N = 33 | N = 33 |
| Annual Change d | −0.18 | −0.53 | −0.34 |
| 95% CI | (−0.24, −0.12) | (−0.70, −0.37) | (−0.47, −0.22) |
| CST w/o CME (Microns/Year) | N = 72 | N = 72 | N = 72 |
| Annual Change d | −2.14 | −2.14 | −2.05 |
| 95% CI | (−2.66, −1.63) | (−2.66, −1.63) | (−2.50, −1.59) |
| MP Mean Sensitivity (dB) | N = 94 | N = 86 | N = 86 |
| Annual Change | −0.39 | −0.42 | −0.35 |
| 95% CI | (−0.47, −0.31) | (−0.51, −0.33) | (−0.42, −0.27) |
| MP Average Sensitivity Within Intact EZ Area | N = 103 | ||
| Annual Change | −0.07 | ||
| 95% CI | (−0.25, 0.10) | ||
| MP Average Sensitivity Outside Intact EZ Area | N = 103 | ||
| Annual Change | −0.34 | ||
| 95% CI | (−0.43, −0.26) | ||
| EZ Functional Transition Points | N = 87 | ||
| Annual Change | −1.54 | ||
| 95% CI | (−1.72, −1.36) | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
