Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: Ophthalmol Retina. 2021 Feb 26;5(12):1204–1213. doi: 10.1016/j.oret.2021.02.012

Longitudinal Assessment of Ellipsoid Zone Integrity, Subretinal Hyperreflective Material, and Sub-RPE Disease in Neovascular AMD

Justis P Ehlers 1,2, Robert Zahid 3, Peter K Kaiser 2, Jeffrey S Heier 4, David M Brown 5, Xiangyi Meng 3, Jamie Reese 1,2, Thuy K Le 1,2, Leina Lunasco 1,2, Ming Hu 1,6, Sunil K Srivastava 1,2
PMCID: PMC8387490  NIHMSID: NIHMS1696839  PMID: 33640493

Abstract

Purpose:

To longitudinally assess the effect of anti-vascular endothelial growth factor (VEGF) treatment on ellipsoid zone (EZ) integrity, subretinal hyperreflective material (SHRM), and the sub-retinal pigment epithelium (sub-RPE) compartment in eyes with neovascular age-related macular degeneration (nAMD).

Design:

This study was a post hoc analysis of the OSPREY clinical trial, a prospective, double-masked, phase 2 study comparing brolucizumab 6 mg with aflibercept 2 mg over 56 weeks.

Participants:

Participants with treatment-naïve nAMD at the initiation of the trial were included in the analysis.

Methods:

Eyes were evaluated with spectral domain optical coherence tomography (SD-OCT) at 4-week intervals in the OSPREY trial (n=81). SD-OCT scans collected from each visit were automatically. segmented using a proprietary, machine learning enabled higher-order feature-extraction platform for retinal layer, SHRM, and sub-RPE boundary lines, which were evaluated and corrected as needed by masked trained graders. The current analysis focused only on subjects evaluated with Cirrus (Zeiss) platform, n=28).

Main Outcome Measures:

Outcome measures included change from baseline in EZ-RPE volume, EZ-RPE central subfield thickness (CST), total EZ attenuation, SHRM volume, SHRM CST, and total sub-RPE volume. The correlation between each of these measures and best corrected visual acuity (BCVA) at each visit was evaluated.

Results:

EZ-RPE volume and EZ-RPE CST showed significant increases, and total EZ attenuation, SHRM volume, SHRM CST, and total sub-RPE volume showed significant decreases from baseline at each visit from Week 4–56 (P<.05 at each visit). EZ integrity measures and SHRM volume significantly correlated with BCVA at most visits (P<.05). No significant correlation was found between total sub-RPE volume and BCVA.

Conclusions:

EZ integrity, SHRM, and sub-RPE disease features in eyes with nAMD showed improvement as early as Week 4 of anti-VEGF treatment. EZ integrity measures and SHRM volume were predictors of visual acuity over the first year of treatment.

Précis:

EZ integrity improved, SHRM decreased, and total sub-RPE volume decreased with anti-VEGF treatment in eyes with nAMD. EZ integrity measures and SHRM volume correlated with visual acuity at most visits of the 56-week study.

Treatment of neovascular age-related macular degeneration (nAMD) is critical to preventing and reversing vision loss.1,2 Currently, the standard of care is intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents, which inhibit neovascular activity, exudation, and vascular hyperpermeability.1–3 The effectiveness and durability of anti-VEGF treatment is typically evaluated through spectral-domain (SD) optical coherence tomography (OCT) to assess for the presence of disease activity, such as intraretinal, subretinal, and/or sub-retinal pigment epithelial (sub-RPE) fluid.2 However, analysis of disease features using OCT in clinical trials has been largely limited to global measures of retinal thickness and qualitative evaluation of specific features (eg, presence or absence of fluid) because of limitations in the availability of OCT analysis platforms.4–8 These limited assessments do not adequately capture the complexity of disease progression or treatment response, such as the impact of treatment on ellipsoid zone (EZ) integrity, subretinal hyperreflective material (SHRM), or sub-RPE disease (eg, retinal pigment epithelial detachment [PED]). More recently, advances in image analysis platforms have enabled higher-order OCT assessments, consisting of multilayer segmentation and feature extraction (eg, fluid object and subretinal material).9–14 Detailed and quantitative characterization of specific imaging features over the course of therapy with higher-order OCT analysis techniques may help facilitate personalized and targeted clinical care using such techniques and better inform clinicians regarding biomarkers that may predict functional outcomes.

Although primary indicators of treatment response are reductions in retinal fluid,15 other important retinal features include the integrity of the outer retinal layers, subretinal material, and disease burden of the sub-RPE compartment. Quantitative and automated measures of outer retinal integrity via measures of the EZ on OCT are potentially of significant value, because EZ integrity has been a consistent biomarker of visual acuity.4,5 Assessment of subretinal material may also be valuable because the presence of SHRM on OCT is associated with poorer preservation of the EZ and decreased visual acuity.5,6 In addition, new and emerging therapeutics may have a differential impact on subretinal fibrosis and subretinal material.16 Thus, being able to measure alterations in this feature would be critical to better understanding drug effect. In contrast with outer retinal integrity and subretinal material, disease within the sub-RPE compartment has not consistently been associated with visual acuity.17,18 However, PED has been associated with neovascular reactivations, suggesting that the sub-RPE compartment is an important disease component to monitor.17,19 In addition, the sub-RPE compartment is often more resistant to therapeutic intervention.19 The ability to reliably quantify this compartment could provide significant new insights to emerging therapeutics and their impact on the sub-RPE area, such as drusen resolution and PED response.20 Previous studies have demonstrated that quantification of these features is possible with a variety of higher-order approaches.8–12,21

However, panmacular quantitative assessment of these disease features and their correlation with visual acuity in nAMD over the course of anti-VEGF treatment have not been adequately characterized. Therefore, the present analysis used a proprietary higher-order OCT analysis technique to longitudinally assess OCT imaging features (EZ integrity, SHRM, and the sub-RPE compartment) in nAMD over the course of anti-VEGF treatment. This exploratory post hoc analysis made use of data from the OSPREY study, in which the efficacy of brolucizumab and aflibercept were evaluated over 56 weeks in treatment-naïve patients with nAMD.22

Methods

OSPREY Study Design

The OSPREY trial (ClinicalTrials.gov identifier NCT01796964) has been described previously.22 Briefly, treatment-naïve participants, aged ≥50 years, with active choroidal neovascularization secondary to age-related macular degeneration (AMD) were recruited at 41 investigational centers across the United States. The study eye had to have best corrected visual acuity (BCVA) between 73 and 23 letters inclusive and show leakage on fluorescein angiography as well as subretinal fluid, intraretinal fluid, and sub-RPE fluid on SD-OCT. Ninety participants were randomized 1:1 to receive intravitreal injections of either brolucizumab (6 mg/50 μL) or aflibercept (2 mg/50 μL). Both groups received loading doses at baseline (Week 0), Week 4, and Week 8 and were then treated every 8 weeks (injections at Weeks 16, 24, and 32) with assessment up to Week 40. Following the 8-week dosing cycle, the brolucizumab group was extended to a 12-week dosing cycle (injection at Week 44), while the aflibercept group continued with the 8-week dosing cycle (injections at Weeks 40 and 48), with assessment up to Week 56. Unscheduled treatments were allowed at the investigator’s discretion. Efficacy assessments were conducted at each study visit, which occurred every 4 weeks. Investigational centers obtained SD-OCT scans according to a standardized study protocol and conducted BCVA assessments according to the Early Treatment Diabetic Retinopathy Study visual acuity protocol. The OSPREY trial was approved by institutional review boards, and it complied with the ethical standards defined by the Declaration of Helsinki and Good Clinical Practice. All participants provided written informed consent before participating in the study.

Higher-Order OCT Analysis

The SD-OCT scans conducted for the OSPREY study were transferred to the Cleveland Clinic for post-hoc exploratory analyses. Eighty-one participants (41 brolucizumab patients and 40 aflibercept patients) of the 89 who received treatment were initially analyzed, including those eyes imaged with the Cirrus™ (Carl Zeiss Meditec; Dublin, CA) or the Spectralis® (Heidelberg Engineering; Heidelberg, Germany) system. The imaging protocol for the Spectralis® system in the OSPREY study included a more sparsely sampled macular cube (49 lines) compared to more densely sampled Cirrus™ scans (128 lines). Given the novel nature of this analysis and potential for interpolation errors in the less densely sampled Spectralis® scans, only participants (n = 28) who had SD-OCT scans obtained with the Cirrus™ (Carl Zeiss Meditec; Dublin, CA) SD-OCT system were included in this initial report. Four eyes with scans obtained on the Cirrus system were excluded because of lack of SD-OCT scan availability (n = 3) or poor-quality SD-OCT scans at baseline (n = 1) limiting the ability to segment fluid, SHRM, or layer boundaries. As noted, the macular cube scans were obtained using a 512 × 128 macular cube covering a 6 mm × 6 mm area of the macula centered on the foveal center point.22 SD-OCT macular cube scans from each study visit were imported into the automated machine-learning-augmented segmentation and feature-extraction platform (Cleveland Clinic, Cleveland, OH, USA).13,14,23 The software platform utilized a combination of image processing/filters, machine-learning augmentation, and logic to extract each feature of interest. This process was simultaneously performed for each feature (ie, fluid, retinal layers, and SHRM enclosure boundaries). A higher-order classifier was then utilized to determine the type of fluid (eg, subretinal or intraretinal). Following identification of each individual feature or line of interest, the outputs were transformed into segmentation masks. The masks were then assembled into a fusion editable overlay that was utilized within the user interface. Each feature could be edited for any segmentation errors as needed within the bounds of the logic in place for anatomic feasibility (eg, the internal limiting membrane could not go below the RPE). SHRM was defined as hyperreflective signal above the RPE and below the visualized boundary of the outer retina that was not consistent with the hyporeflective signal of subretinal fluid. Two masked and trained expert readers consecutively checked the automatic segmentations and manually corrected segmentation errors of specific lines and features of interest. All readers received the same training for the SD-OCT analysis, and the same trained readers reviewed and corrected the segmentations for all timepoints for any given participant to minimize inter-timepoint and inter-reader variability. The reading environment was standardized for location, computer configuration, monitor settings, and lighting configuration. The inter-reader intraclass correlation coefficient across the metrics assessed for this analysis range from 0.85 to 0.99. Following initial reads, the project lead reviewed each scan to confirm consistency and accuracy of segmentation.13,24 This same process of automatic segmentation followed by as-needed manual correction was used to segment the features of interest shown in the example B scans of the present study. Using the segmentation boundaries, several measures were exported for evaluation, including EZ-RPE volume, SHRM volume, and total sub-RPE volume across the entire macular cube; the percentage of the macular cube showing total EZ attenuation (ie, 0-μm EZ-RPE thickness); EZ-RPE central subfield thickness (CST); and SHRM CST. For the present analysis, EZ integrity was described based on the EZ-RPE compartment (ie, a surrogate for photoreceptor outer segment length), and it was expressed through the multiple specific EZ metrics, including the panmacular assessments (ie, EZ-RPE volume and percentage total EZ attenuation) and central assessment (ie, EZ-RPE CST). EZ-RPE metrics measured the retinal tissue from the EZ to the RPE band (or the top of the SHRM boundary if present) and excluded intraretinal and subretinal fluid. Total sub-RPE volume was defined as the total volume between the RPE and Bruch’s membrane across the macular cube. EZ-RPE CST and SHRM CST were calculated based on the mean thickness of each parameter on each A scan within the 1-mm diameter area centered on the fovea. The EZ-RPE CST reflected the mean thickness of each EZ-RPE point thickness within the central 1-mm area. The SHRM CST was calculated in a similar manner, as the mean height of any SHRM present on each A scan within the central 1-mm area. If no SRHM was present for a specific A scan, the value was 0. If no SHRM was present within the central subfield, the SHRM CST was 0. Both SHRM volume and SHRM CST excluded subretinal fluid.

Statistical Analysis

Analyses were conducted with the treatment groups combined. Change from baseline was evaluated using t-tests. Statistical analyses were conducted for hypothesis generation and were not adjusted for multiple comparisons. P values were considered significant at P<0.05. Mean BCVA analyses have been described previously. Analyses were performed using SAS® (Version 9.4).

Results

Patient Characteristics

Participant characteristics at baseline are shown in Table 1. Overall, the mean age was 77 years. Forty-three percent of participants showed predominantly classic lesions, 32% showed minimally classic lesions, and 25% showed occult lesions. The majority of participants showed presence of subretinal fluid, intraretinal fluid, and hyperreflective material at baseline. Mean BCVA was 58 letters at baseline and 65 letters at Week 56 (Table 1).

Table 1.

Participant baseline characteristics and BCVA at baseline and Week 56.

Characteristic All Participants (N=28)
Age, years, mean (SD) 77 (7.6)
Female, n (%) 21 (75.0)
BCVA letters at baseline, mean (SD) 58 (l1.8)
BCVA letters at Week 56, mean (SD) 65 (20.0)
CST, pm, mean (SD) 480 (151)
Lesion type, n (%)
 Predominantly classic 12 (42.9)
 Minimally classic 9 (32.1)
 Occult 7 (25.0)
Presence of hyperreflective material, n (%) 25 (89.3)
Presence of subretinal fluid, n (%) 28 (100)
Presence of intraretinal fluid, n (%) 25 (89.3)
Open in a new tab

BCVA = best corrected visual acuity; CST = central subfield thickness; OCT = optical coherence tomography; SD = standard deviation.

Outer Retinal Integrity and Quantitative Features

EZ-RPE Volume

Baseline mean EZ-RPE volume was 0.78 mm3 (standard deviation [SD] = 0.30). EZ-RPE volume showed a significant increase from baseline beginning at Week 4 (P<0.001), which was maintained through Week 56 (P<0.001 at each visit), suggesting improvement in EZ integrity following anti-VEGF treatment (Figure 1A). Mean increase from baseline ranged from 0.13 mm3 to 0.23 mm3 over Weeks 4–56. The mean time to the maximum EZ-RPE volume was 35.9 weeks (SD = 15.3), which was longer than the mean time to the maximum BCVA (26.0 weeks [SD = 18.1]). When expressed as a percentage change from baseline, the mean percentage change ranged from 28.0% to 44.5%.

Figure 1. Change in EZ Integrity.

Figure 1.

Open in a new tab

(A) Mean change from baseline in EZ-RPE volume. (B) Mean change from baseline in EZ-RPE CST. (C) Representative B-scans (top) and en face EZ-RPE thickness maps (bottom) from the OSPREY study, showing improvement in EZ integrity with treatment. CST = central subfield thickness; EZ = ellipsoid zone; RPE = retinal pigment epithelium.

EZ-RPE Central Subfield Thickness

Baseline mean EZ-RPE CST was 2.81 μm (SD = 6.41). EZ-RPE CST showed a significant increase from baseline at each visit from Weeks 4–56 (P<0.05 at each visit; Figure 1B), indicating improvements in EZ integrity within the central subfield. Mean increase from baseline ranged from 2.72 μm to 12.15 μm over Weeks 4–56. The mean time to the maximum EZ-RPE CST was 38.7 weeks (SD = 15.9).

Percentage Total EZ Attenuation

At baseline, the mean percentage of the macular cube demonstrating total EZ attenuation (0-μm EZ-RPE thickness) was 35.43% (SD = 20.79). The percentage of EZ total attenuation showed a significant decrease from baseline at each visit (P<0.001), suggesting some recovery of the EZ in areas that had shown 0-μm thickness. The mean change from baseline in the percentage of total EZ attenuation ranged from −13.37% to −21.87% across Weeks 4–56. The mean time to the minimum total EZ attenuation was 38.9 weeks (SD = 14.8).

Correlation Between EZ Integrity and BCVA

There was a significant positive correlation between EZ-RPE volume and BCVA at each visit from Week 4–56 (P<0.05) and between EZ-RPE CST and BCVA at each visit from Week 8–56 (P<0.05; Figure 2). The significant correlations ranged from weak to moderate (0.40≤ r ≤0.62) for EZ-RPE volume and were typically moderate for EZ-RPE CST (0.54≤ r ≤0.66, except at Week 24 r=0.43). There was a significant negative correlation between percentage of total EZ attenuation and BCVA at most visits from Week 4–56 (P<0.05, excluding Week 12, 16, 40; Figure 2). These correlations were generally weaker (0.38 ≤ r ≤0.55) at corresponding weeks than those for EZ-RPE volume or EZ-RPE CST. The correlation with BCVA was not significant at baseline for any of the 3 EZ-integrity metrics.

Figure 2. Correlations with BCVA.

Figure 2.

Open in a new tab

Pearson correlation coefficients between OCT outcome measures and BCVA (ETDRS letters) at each study visit. Red indicates a positive correlation, and blue indicates a negative correlation. Lighter tints correspond to weaker correlations. *P≤.05; **P≤.01; ***P≤.001. BCVA = best corrected visual acuity; CST = central subfield thickness; ETDRS = Early Treatment Diabetic Retinopathy Study; EZ = ellipsoid zone; OCT = optical coherence tomography; RPE = retinal pigment epithelium; SHRM = subretinal hyperreflective material.

Change in BCVA from baseline to Week 56 showed a significant and weak correlation with change in EZ-RPE CST from baseline to Week 56 (r=0.39, P=0.04) and change in percentage of total EZ attenuation (r=0.46, P=0.02), indicating that greater improvement in EZ integrity was weakly associated with greater increase in BCVA. Change in BCVA from baseline to Week 56 was not significantly correlated with change in EZ-RPE volume from baseline to Week 56 (r=−0.20, P=0.32). Change in BCVA from baseline to Week 56 was also significantly correlated with early change in percentage of total EZ attenuation, quantified as change from baseline to Week 4 (r=0.49, P=0.009). Change in BCVA from baseline to Week 56 was not significantly correlated with early change in EZ-RPE volume (r=−0.28, P=0.16) or early change in EZ-RPE CST (r=0.07, P=0.74).

Subretinal Material

SHRM Volume

Baseline mean SHRM volume was 0.37 mm3 (SD = 0.32). SHRM volume showed a significant decrease from baseline at each visit (P<0.001; Figure 3A). Mean change from baseline in SHRM volume ranged from −0.25 mm3 to −0.32 mm3 across Weeks 4–56. The mean time to the minimum SHRM volume was 25.6 weeks (SD = 16.5), which was similar to the mean time to the maximum BCVA (26.0 weeks [SD = 18.1]). The mean percentage change ranged from −70.0% to −87.3% over Weeks 4–56.

Figure 3. Change in SHRM.

Figure 3.

Open in a new tab

(A) Mean change from baseline in SHRM volume. (B) Mean change from baseline in SHRM CST. (C) Representative B-scans for 2 subjects (first subject in rows 1 and 2; and second subject in rows 3 and 4 from the OPSREY study, showing reduction in SHRM with treatment. Rows 2 and 4 are the same as rows 1 and 3, respectively, but with SHRM highlighted in yellow. CST = central subfield thickness; SHRM = subretinal hyperreflective material.

SHRM Central Subfield Thickness

Baseline mean SHRM CST was 57.39 μm (SD = 49.55). There was a significant decrease from baseline in SHRM CST at each visit (P<0.001; Figure 3B). Mean change from baseline in SHRM CST ranged from −33.48 to −46.93 μm from Weeks 4–56. The mean time to the minimum SHRM CST was 15.9 weeks (SD = 14.9), which was shorter than the mean time to maximum BCVA (26.0 weeks [SD = 18.1]).

Correlation Between SHRM Metrics and BCVA

The correlations between each of the SHRM metrics and BCVA for each study visit are shown in Figure 2. There was a significant negative correlation between SHRM volume and BCVA at baseline and most subsequent study visits (P<0.05; excluding Week 20, 24, 40, and 56). There was also a significant negative correlation between SHRM CST and BCVA at baseline and Week 4, 8, 28, and 32 (P<0.05). The significant correlations were weak to moderate for both SHRM volume (0.40≤ r ≤0.62) and SHRM CST (0.38≤ r ≤0.59).

Change in BCVA from baseline to Week 56 showed a significant weak correlation with change in SHRM volume from baseline to Week 56 (r=0.44, P=0.02) and from baseline to Week 4 (r=0.41, P=0.04), suggesting that greater decrease in SHRM volume was weakly associated with greater increase in BCVA. Change in BCVA from baseline to Week 56 was not significantly correlated with change in SHRM CST from baseline to Week 56 (r=0.32, P=0.10) or from baseline to Week 4 (r=0.36, P=0.07).

Sub-RPE Compartment

Baseline mean total sub-RPE volume was 0.80 mm3 (SD = 0.65). Total sub-RPE volume showed a significant decrease from baseline at each visit (P<0.05 at each visit; Figure 4A). Mean change from baseline ranged from −0.14 mm3 to −0.20 mm3 over Weeks 4–56, while mean percentage change from baseline ranged from −10.1% to −15.5%. The mean time to the minimum sub-RPE volume was 29.6 weeks (SD = 18.3), which was somewhat longer than the mean time to maximum BCVA (26.0 weeks [SD = 18.1]).

Figure 4. Change in total sub-RPE volume.

Figure 4.

Open in a new tab

(A) Mean change from baseline in total sub-RPE volume. (B) Representative B-scans (top) and en face sub-RPE thickness maps (bottom) for 1 subject from the OSPREY study, showing reduction in sub-RPE volume with treatment. RPE = retinal pigment epithelium.

The correlation between total sub-RPE volume and BCVA was not significant at any visit (Figure 2). The correlation between change in BCVA from baseline to Week 56 was not significantly correlated with change in total sub-RPE volume from baseline to Week 56 (r=0.23, P=0.24) or from baseline to Week 4 (r=0.17, P=0.40).

Discussion

The present post-hoc exploratory analysis of the OSPREY study examined longitudinal changes in SD-OCT imaging features of patients with treatment-naïve nAMD from the initiation of anti-VEGF therapy through 56 weeks of treatment. Using higher-order OCT analysis, we observed that anti-VEGF treatment improved EZ integrity, reduced SHRM, and reduced disease within the sub-RPE compartment. Correlations between anatomic features and visual acuity were observed for EZ-RPE and SHRM metrics but not for sub-RPE volume.

Improvement in EZ integrity compared with baseline was evident as early as Week 4 of treatment and throughout the 56-week treatment period (Figure 1). These findings are consistent with the finding of improvement in EZ integrity as early as Month 3 of treatment.4,25 Analysis of the percentage of the macular cube with total EZ attenuation indicated improvement in EZ integrity, even in macular areas that had shown EZ-RPE thickness of 0 μm. Findings that EZ integrity improves with treatment suggest that some of the disruptions are due to misplacement or attenuation of outer retinal structures as opposed to permanent damage.4 Although not in nAMD, a similar analysis was completed in the VISTA study in patients with diabetic macular edema. Using higher-order OCT, this analysis quantitatively demonstrated improvement in EZ integrity with anti-VEGF treatment (aflibercept).13 Previous studies have shown an association between EZ integrity and visual acuity in patients with nAMD before treatment or at the end of treatment.18,25–28 Although the present study did not find an association at baseline, the present study demonstrated a consistent correlation with visual acuity at each visit beginning at Week 4 or 8 of treatment for EZ-RPE volume and EZ-RPE CST, respectively; and at most visits beginning at Week 4 for percentage total EZ attenuation (Figure 2). The finding of a correlation after initiation of treatment is likely because of a high incidence of complete EZ attenuation in the foveal area before the onset of treatment. In addition, increase in EZ-RPE CST and decrease in percentage of total EZ attenuation over the course of treatment were associated with increase in visual acuity, which is consistent with previous findings of EZ features in the central subfield.28 The present study also found that early decrease (from baseline to Week 4) in percentage total EZ attenuation was associated with increase in visual acuity over the course of treatment (from baseline to Week 56), suggesting that this early biomarker is a potential predictor of future outcomes. Reliable measurement of the outer retinal layers has been the major limiting factor in using the outer retina as an imaging biomarker.17 However, higher-order OCT analysis techniques, including machine-learning augmentation and advanced image-analysis methods, enable efficient evaluation of the outer retinal layers.9–12

In addition to improvements in EZ integrity, analyses of OCT imaging features showed reduction from baseline in SHRM with anti-VEGF treatment, which was observed throughout the 56-week treatment period (Figure 3). SHRM is thought to consist of fibrovascular tissue and other material such as blood, fluid, and lipids in the subretinal space.17,29 In the present study, SHRM volume decreased by an average of 72.2% from baseline to Week 4, consistent with the hypothesis that SHRM is largely composed of fluid, which can readily resolve with anti-VEGF therapy.6 The observed reduction in SHRM volume and SHRM CST in the present analysis is consistent with the previous finding that the proportion of patients who show SHRM decreases with anti-VEGF therapy.6 SHRM has been shown to be negatively correlated with visual acuity.6,29 The proximity of SHRM to the photoreceptors and the barrier created by the SHRM between the photoreceptors and the RPE have been hypothesized to play roles in the association between SHRM and poor visual acuity.6,29,30 In the present study, SHRM volume was found to negatively correlate with visual acuity at baseline and most subsequent study visits (Figure 2). The correlations between SHRM volume across the entire macular cube with visual acuity were more consistently found than the correlations between SHRM CST and visual acuity. Decrease in SHRM volume was also correlated with increase in BCVA, which is consistent with previous findings in the central subfield.28 Interestingly, the present study found that early decrease in SHRM volume (from baseline to Week 4) was a predictor of increase in BCVA over the duration of treatment (from baseline to Week 56).

Reduction in total sub-RPE volume compared with baseline was also evident throughout the 56-week treatment period (Figure 4). Total sub-RPE volume in the present study referred to the entire disease burden in the sub-RPE compartment, including fluid, fibrovascular material, and drusen.17 The reduction in total sub-RPE volume observed is consistent with previous studies that showed that anti-VEGF therapy reduces the proportion of patients with PED and that sub-RPE volume is reduced after 1 month of therapy.17,18,31 In contrast, the sub-RPE area has been found to increase from Month 1 to Month 2 following the initiation of treatment in a subgroup of patients with nAMD showing eventual recurrence of neovascular activity.31 PED volume has also been found to increase at clinical visits preceding those at which anti-VEGF injection was administered in patients with vascularized PEDs receiving treatment as needed, suggesting volumetric changes in PED may be a useful predictor of disease progression and retreatment needs.32,33 These findings highlight the value in monitoring for changes in sub-RPE area and volume. No correlation between total sub-RPE volume and visual acuity was observed in the present study at baseline or during treatment. The lack of correlation is consistent with other studies that have shown no association between PED volume and visual acuity before treatment.29,30 In addition, no association was found between change in total sub-RPE volume and change in BCVA.

Limitations of the present analysis include the small sample and the exploratory post hoc nature of the analysis. Additional limitations relate to the measurements of the outer retina. The measurement approach to EZ integrity has been described in multiple ways by various groups, including reflectivity of the EZ band, en face defects in EZ visibility, and EZ-RPE compartment measures (ie, photoreceptor outer segment length).10–12,14 An important limitation of the present analysis is that the EZ integrity metrics did not measure EZ-band reflectivity but rather focused on the EZ-RPE compartment. EZ reflectivity is an area for future analysis, particularly the integrative assessment of the EZ-RPE compartment with EZ-band reflectivity. Notably, multiple reports have demonstrated the functional association of EZ integrity metrics as measured based on EZ-RPE compartment features.13,34,35 An additional limitation of the present study is that the outer retinal assessment focused on the integrity of a single band, the EZ. The other outer retinal bands, such as the external limiting membrane and interdigitation zone, have also been linked to visual acuity but were not segmented in this analysis.28,36 With the automated segmentation system used in the present analysis, the external limiting membrane and the interdigitation zone were not segmented in part because of their fainter signal on OCT, which makes consistent segmentation more challenging. For this specific analysis, an additional limitation is that only eyes imaged with the Cirrus™ system were included based on the notable differences in macular-cube scan sampling density between the Spectralis® and Cirrus™ platforms. Importantly, the Spectralis® data and combined system data (not shown) demonstrated similar findings to the Cirrus™ data presented in this report.37 However, in order to minimize the risk of errors based on larger interpolation distances, the Spectralis® data were not included. This highlights the importance of the prespecified imaging protocols during clinical trial design.

The results of this study serve as an important proof of concept and will be used to guide future OCT imaging analyses in nAMD using larger data sets. The present analysis demonstrated an improvement in EZ integrity, reduction in SHRM volume, and reduction in sub-RPE disease burden as early as Week 4 of treatment. Improvements from baseline were maintained throughout the 56-week treatment period. Measures of EZ integrity and SHRM volume appear to be important biomarkers of functional outcomes.

Acknowledgments

The authors would like to thank all the OSPREY investigators. Editorial support was provided by IMPRINT Science, New York, NY, USA, and was funded by Novartis Pharmaceuticals.

Financial Support: This work was funded by a research grant provided by Novartis Pharmaceuticals, East Hanover, NJ, USA, which participated in the design of the study, conducting the study, data collection, data management, and data analysis, interpreted the data, and prepared, reviewed, and approved the manuscript. This work was also supported through the NIH/NEI K23-EY022947–01A1.

Conflict of Interest: Justis P. Ehlers reports grants NIH/NEI K23-EY022947–01A1 and NIH/NEI R34-EY029308, and grants from Novartis, Alcon, Oxurion, Aerpio, Regeneron, Genentech, the Ohio Department of Development TECH-13–059, The Tom and Maryanne Wagner Advanced Imaging Research Fund, The Betty Powers Optical Coherence Tomography Research Fund, The Norman C. and Donna L. Harbert Endowed Chair Fund, The Tony and Leona Campane Image-Guided Surgery and Advanced Image Analysis Research Fund; personal fees from Novartis, Leica/Bioptigen, Alcon, Allergan, Alimera, Thrombogenics, Stanten, Aerpio; personal fees and other from Zeiss; and patents/intellectual property/licensing with Bioptigen/Leica outside the submitted work. Robert Zahid and Xiangyi Meng are employees of Novartis. Peter K. Kaiser reports personal fees from Novartis, Allergan, Regeneron, Bayer, Kanghong, Kodiak, and RegenexBio outside the submitted work. Jeffrey S. Heier reports personal fees from 4DMT, Adverum, Aerie, Aerpio, Aldeyra, Allegro, Alzheon, Annexon, Apellis, Asclepix, Beaver-Visitec, Galimedix, Genentech, Gyroscope, iRenix, jCyte, Kala, Kanghong, NGM, Notal Vision, Novartis, Ocugenix, Oculis, Ocunexus, Ocular Therapeutix, Palatin, Pfizer, Regeneron, Regenxbio, Santen, Scifluor, Shire, Stealth, Tyrogenex, Voyant; and grants from Aerie, Aerpio, Apellis, Genentech, Graybug, Gyroscope, Hemera, Janssen R&D, KalVista, Kanghong, Novartis, Ophthotech, Optovue, Regeneron, Regenxbio, and Stealth outside the submitted work. David M. Brown reports grants and personal fees from Novartis, Heidelberg Engineering, and Carl Zeiss Meditec outside the submitted work. Sunil K. Srivastava reports personal fees from Novartis, Bausch and Lomb, Abbvie, Clearside, Zeiss, and Regenerxbio; and personal fees and other from Allergan, Eyepoint, and Regeneron outside the submitted work. Jamie Reese, Thuy K. Le, Leina Lunasco, and Ming Hu have nothing to disclose.

Abbreviations and Acronyms:

BCVA

best corrected visual acuity

CST

central subfield thickness

ETDRS

Early Treatment Diabetic Retinopathy Study

EZ

ellipsoid zone

nAMD

neovascular age-related macular degeneration

OCT

optical coherence tomography

PED

pigment epithelium detachment

RPE

retinal pigment epithelium

SD

standard deviation

SD-OCT

spectral-domain optical coherence tomography

SHRM

subretinal hyperreflective material

VEGF

vascular endothelial growth factor

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Schmidt-Erfurth U, Chong V, Loewenstein A, et al. Guidelines for the management of neovascular age-related macular degeneration by the European Society of Retina Specialists (EURETINA). Br J Ophthalmol. 2014;98(9):1144–1167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wykoff CC, Clark WL, Nielsen JS, Brill JV, Greene LS, Heggen CL . Optimizing anti-VEGF treatment outcomes for patients with neovascular age-related macular degeneration. J Manag Care Spec Pharm. 2018;24(2-a Suppl):S3–S15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006;355(14):1432–1444. [DOI] [PubMed] [Google Scholar]
  • 4.Riedl S, Cooney L, Grechenig C, et al. Topographic analysis of photoreceptor loss correlated with disease morphology in neovascular age-related macular degeneration. Retina. 2020;40(11):2148–2157. [DOI] [PubMed] [Google Scholar]
  • 5.Ryu CL, Al-Humaid S, Rampakakis E, Galic IJ, Chen JC. Correlation of visual acuity with fibrotic scar location in treated neovascular age-related macular degeneration eyes. Retina. 2016;36(7):1324–1330. [DOI] [PubMed] [Google Scholar]
  • 6.Willoughby AS, Ying GS, Toth CA, et al. Subretinal hyperreflective material in the Comparison of Age-related Macular Degeneration Treatments Trials. Ophthalmology. 2015;122(9): 1846–1853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Dugel PU, Koh A, Ogura Y, et al. HAWK and HARRIER: phase 3, multicenter, randomized, double-masked trials of brolucizumab for neovascular age-related macular degeneration. Ophthalmology. 2020;127(l):72–84. [DOI] [PubMed] [Google Scholar]
  • 8.Lee H, Kang KE, Chung H, Kim HC. Automated segmentation of lesions including subretinal hyperreflective material in neovascular age-related macular degeneration. Am J Ophthalmol. 2018;191:64–75. [DOI] [PubMed] [Google Scholar]
  • 9.Chiu SJ, Li XT, Nicholas P, Toth CA, Izatt JA, Farsiu S. Automatic segmentation of seven retinal layers in SDOCT images congruent with expert manual segmentation. Opt Express. 2010;18(18): 19413–19428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ross DH, Clark ME, Godara P, et al. RefMoB, a reflectivity feature model-based automated method for measuring four outer retinal hyperreflective bands in optical coherence tomography. Invest Ophthalmol Vis Sci. 2015;56(8):4166–4176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.de Sisternes L, Hu J, Rubin DL, Leng T. Visual prognosis of eyes recovering from macular hole surgery through automated quantitative analysis of spectral-domain optical coherence tomography (SD-OCT) scans. Invest Ophthalmol Vis Sci. 2015;56(8):4631–4643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Loo J, Fang L, Cunefare D, Jaffe GJ, Farsiu S. Deep longitudinal transfer learning-based automatic segmentation of photoreceptor ellipsoid zone defects on optical coherence tomography images of macular telangiectasia type 2. Biomed Opt Express. 2018;9(6):2681–2698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ehlers JP, Uchida A, Hu M, et al. Higher-order assessment of OCT in diabetic macular edema from the VISTA study: ellipsoid zone dynamics and the retinal fluid index. Ophthalmol Retina. 2019;3(12):1056–1066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Itoh Y, Vasanji A, Ehlers JP. Volumetric ellipsoid zone mapping for enhanced visualisation of outer retinal integrity with optical coherence tomography. Br J Ophthalmol. 2016;100(3):295–299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Brown DM, Tuomi L, Shapiro H, Pier Study Group. Anatomical measures as predictors of visual outcomes in ranibizumab-treated eyes with neovascular age-related macular degeneration. Retina. 2013;33(1):23–34. [DOI] [PubMed] [Google Scholar]
  • 16.Zhang R, Liu Z, Zhang H, Zhang Y, Lin D. The COX-2-selective antagonist (NS-398) inhibits choroidal neovascularization and subretinal fibrosis. PLoS One. 2016;11(1):e0146808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Schmidt-Erfurth U, Waldstein SM. A paradigm shift in imaging biomarkers in neovascular age-related macular degeneration. Prog Retin Eye Res. 2016;50:1–24. [DOI] [PubMed] [Google Scholar]
  • 18.Lai TT, Hsieh YT, Yang CM, Ho TC, Yang CH. Biomarkers of optical coherence tomography in evaluating the treatment outcomes of neovascular age-related macular degeneration: a real-world study. Sci Rep. 2019;9(1):529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Schmidt-Erfurth U, Waldstein SM, Deak GG, Kundi M, Simader C. Pigment epithelial detachment followed by retinal cystoid degeneration leads to vision loss in treatment of neovascular age-related macular degeneration. Ophthalmology. 2015;122(4): 822–832. [DOI] [PubMed] [Google Scholar]
  • 20.Querques G, Georges A, Ben Moussa N, Sterkers M, Souied EH. Appearance of regressing drusen on optical coherence tomography in age-related macular degeneration. Ophthalmology. 2014; 121(1):173–179. [DOI] [PubMed] [Google Scholar]
  • 21.Moraes G, Fu DJ, Wilson M, et al. Quantitative analysis of OCT for neovascular age-related macular degeneration using deep learning. Ophthalmology. 2020. Published online. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Dugel PU, Jaffe GJ, Sallstig P, et al. Brolucizumab versus aflibercept in participants with neovascular age-related macular degeneration: a randomized trial. Ophthalmology. 2017;124(9):1296–1304. [DOI] [PubMed] [Google Scholar]
  • 23.Xu D, Yuan A, Kaiser PK, et al. A novel segmentation algorithm for volumetric analysis of macular hole boundaries identified with optical coherence tomography. Invest Ophthalmol Vis Sci. 2013;54(1): 163–169. [DOI] [PubMed] [Google Scholar]
  • 24.Arepalli S, Srivastava SK, Hu M, et al. Assessment of inner and outer retinal layer metrics on the Cirrus HD-OCT platform in normal eyes. PLoS One. 2018;13(10):e0203324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Shin HJ, Chung H, Kim HC. Association between foveal microstructure and visual outcome in age-related macular degeneration. Retina. 2011;31(8):1627–1636. [DOI] [PubMed] [Google Scholar]
  • 26.El-Emam S, Chhablani J, Barteselli G, et al. Correlation of spectral domain optical coherence tomography characteristics with visual acuity in eyes with subfoveal scarring after treatment for wet age-related macular degeneration. Retina. 2013;33(6): 1249–1257. [DOI] [PubMed] [Google Scholar]
  • 27.Hayashi H, Yamashiro K, Tsujikawa A, Ota M, Otani A, Yoshimura N. Association between foveal photoreceptor integrity and visual outcome in neovascular age-related macular degeneration. Am J Ophthalmol. 2009;148(1):83–89. [DOI] [PubMed] [Google Scholar]
  • 28.Lee H, Jo A, Kim HC. Three-dimensional analysis of morphologic changes and visual outcomes in neovascular age-related macular degeneration. Invest Ophthalmol Vis Sci. 2017;58(2):1337–1345. [DOI] [PubMed] [Google Scholar]
  • 29.Keane PA, Liakopoulos S, Chang KT, et al. Relationship between optical coherence tomography retinal parameters and visual acuity in neovascular age-related macular degeneration. Ophthalmology. 2008;115(12):2206–2214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Keane PA, Patel PJ, Ouyang Y, et al. Effects of retinal morphology on contrast sensitivity and reading ability in neovascular age-related macular degeneration. Invest Ophthalmol Vis Sci. 2010;51(11):5431–5437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Sasaki M, Kato Y, Fujinami K, et al. Advanced quantitative analysis of the sub-retinal pigment epithelial space in recurrent neovascular age-related macular degeneration. PLoS One. 2017;12(11):e0186955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Penha FM, Gregori G, Garcia Filho CA, Yehoshua Z, Feuer WJ, Rosenfeld PJ. Quantitative changes in retinal pigment epithelial detachments as a predictor for retreatment with anti-VEGF therapy. Retina. 2013;33(3):459–466. [DOI] [PubMed] [Google Scholar]
  • 33.Penha FM, Rosenfeld PJ, Gregori G, et al. Quantitative imaging of retinal pigment epithelial detachments using spectral-domain optical coherence tomography. Am J Ophthalmol. 2012;153(3):515–523. [DOI] [PubMed] [Google Scholar]
  • 34.Runkle AP, Kaiser PK, Srivastava SK, Schachat AP, Reese JL, Ehlers JP. OCT Angiography and Ellipsoid Zone Mapping of Macular Telangiectasia Type 2 From the AVATAR Study. Invest Ophthalmol Vis Sci. 2017;58(9):3683–3689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Banaee T, Singh RP, Champ K, et al. Ellipsoid zone mapping parameters in retinal venous occlusive disease with associated macular edema. Ophthalmol Retina. 2018;2(8):836–841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hirota K, Itoh Y, Rii T, Inoue M, Hirakata A. Correlation between foveal interdigitation zone band defect and visual acuity after surgery for macular pseudohole. Retina. 2015;35(5):908–914. [DOI] [PubMed] [Google Scholar]
  • 37.Ehlers JP, Clark J, Reese J, Meng X, Srivastava SK. Longitudinal dynamics in outer retinal integrity in nAMD: higher-order OCT feature extraction and insights from the OSPREY trial. The Macula Society 43rd Annual Meeting; February 19–22, 2020; San Diego, CA. [Google Scholar]

RESOURCES