Advertisement

Spectral-Domain OCT Analysis of Risk Factors for Macular Atrophy Development in the HARBOR Study for Neovascular Age-Related Macular Degeneration

Open AccessPublished:April 02, 2020DOI:https://doi.org/10.1016/j.ophtha.2020.03.031

      Purpose

      To identify baseline risk factors for macular atrophy (MA) development in HARBOR via a longitudinal assessment of monthly spectral-domain (SD)-OCT scans. Previous analyses of MA in HARBOR examined data from color fundus photography (CFP) and fluorescein angiography (FA).

      Design

      Retrospective, post hoc analysis of SD-OCT images from HARBOR (ClinicalTrials.gov identifier, NCT00891735), a phase 3, multicenter, prospective, randomized, double-blind, active treatment–controlled clinical trial.

      Participants

      Patients (N = 1097) with subfoveal choroidal neovascularization secondary to neovascular age-related macular degeneration (nAMD) treated with intravitreal ranibizumab 0.5 mg monthly (n = 275), 0.5 mg pro re nata (PRN) after 3 loading doses (n = 275), 2.0 mg monthly (n = 274), or 2.0 mg PRN (n = 273).

      Methods

      Evaluable SD-OCT macular cube scans from patients with 24 months of follow-up (N = 941) were examined monthly from baseline to month 24 by masked reading center–trained graders. Atrophy diagnosis criteria were consistent with those proposed by the Classification of Atrophy Meetings (CAM) group: hypertransmission of light into the choroid, loss of retinal pigment epithelium, and loss of outer retinal layers. Multivariable proportional hazards regression was performed for time to atrophy development.

      Main Outcome Measures

      Risk factors for MA as determined by time to MA development over 24 months of treatment.

      Results

      Baseline risk factors for MA were confirmed from prior analyses that used CFP and FA data: absence of subretinal fluid, presence of intraretinal cysts, presence of Type 3 neovascularization, and presence of atrophy in the fellow eye. This analysis of SD-OCT data identified new baseline risk factors for MA: higher central drusen volume, lower choroidal thickness, presence of nascent atrophy, presence of reticular pseudodrusen, and increased central foveal thickness. Ranibizumab treatment regimen and dose level were not found to be risk factors for MA development.

      Conclusions

      In this analysis of a major nAMD trial using CAM atrophy criteria, new baseline risk factors for MA development were identified using an SD-OCT dataset. Risk factors for MA development identified by prior analyses were confirmed. Monthly treatment with ranibizumab 0.5 mg was not found to be a risk factor for MA development over 24 months.

      Abbreviations and Acronyms:

      AMD (age-related macular degeneration), BCVA (best-corrected visual acuity), CAM (Classification of Atrophy Meetings), CATT (Comparison of Age-Related Macular Degeneration Treatments Trials), CFP (color fundus photography), CFT (central foveal thickness), CI (confidence interval), CNV (choroidal neovascularization), cRORA (complete RPE and outer retinal atrophy), FA (fluorescein angiography), GA (geographic atrophy), HR (hazard ratio), IVAN (Inhibition of VEGF in Age-related choroidal Neovascularisation), MA (macular atrophy), nAMD (neovascular age-related macular degeneration), PED (pigment epithelial detachment), PRN (pro re nata), RPD (reticular pseudodrusen), RPE (retinal pigment epithelium), SD (spectral-domain), SHRM (subretinal hyperreflective material), SRF (subretinal fluid), VEGF (vascular endothelial growth factor)
      Age-related macular degeneration (AMD) is clinically classified into early, intermediate, and late—or advanced—subtypes.
      • Ferris 3rd, F.L.
      • Wilkinson C.P.
      • Bird A.
      • et al.
      Clinical classification of age-related macular degeneration.
      Advanced AMD is characterized by the development of atrophy or choroidal neovascularization (CNV), defined by the growth of new vessels in the intraretinal, subretinal, or sub–retinal pigment epithelium (RPE) space.
      • Wong T.Y.
      • Chakravarthy U.
      • Klein R.
      • et al.
      The natural history and prognosis of neovascular age-related macular degeneration: a systematic review of the literature and meta-analysis.
      The term macular neovascularization has recently been used by some authors as an alternative to CNV
      • Spaide R.F.
      Improving the age-related macular degeneration construct: a new classification system.
      • Rosenfeld P.J.
      Optical coherence tomography and the development of antiangiogenic therapies in neovascular age-related macular degeneration.
      • de Oliveira Dias J.R.
      • Zhang Q.
      • Garcia J.M.B.
      • et al.
      Natural history of subclinical neovascularization in nonexudative age-related macular degeneration using swept-source OCT angiography.
      ; however, the term CNV will be used herein. Anti–vascular endothelial growth factor (anti-VEGF) therapy is the current standard of care for neovascular AMD (nAMD), and several large phase 3 clinical trials have reported anti-VEGF treatment to be effective in reducing CNV exudation and improving and maintaining vision.
      • Brown D.M.
      • Kaiser P.K.
      • Michels M.
      • et al.
      Ranibizumab versus verteporfin for neovascular age-related macular degeneration.
      • Rosenfeld P.J.
      • Brown D.M.
      • Heier J.S.
      • et al.
      Ranibizumab for neovascular age-related macular degeneration.
      • Busbee B.G.
      • Ho A.C.
      • Brown D.M.
      • et al.
      Twelve-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration.
      • Chakravarthy U.
      • Harding S.P.
      • Rogers C.A.
      • et al.
      Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial.
      • Martin D.F.
      • Maguire M.G.
      • Ying G.S.
      • et al.
      Ranibizumab and bevacizumab for neovascular age-related macular degeneration.
      • Kodjikian L.
      • Souied E.H.
      • Mimoun G.
      • et al.
      Ranibizumab versus bevacizumab for neovascular age-related macular degeneration: results from the GEFAL noninferiority randomized trial.
      • Schmidt-Erfurth U.
      • Kaiser P.K.
      • Korobelnik J.F.
      • et al.
      Intravitreal aflibercept injection for neovascular age-related macular degeneration: ninety-six-week results of the VIEW studies.
      In the Inhibition of VEGF in Age-related choroidal Neovascularisation (IVAN), Comparison of Age-Related Macular Degeneration Treatments Trials (CATT), and HARBOR trials, a subset of eyes undergoing treatment for nAMD with anti-VEGF agents have been reported to demonstrate atrophy, termed macular atrophy (MA).
      • Chakravarthy U.
      • Harding S.P.
      • Rogers C.A.
      • et al.
      Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial.
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      • Grunwald J.E.
      • Pistilli M.
      • Ying G-s
      • et al.
      Growth of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT).
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      • Bailey C.
      • Scott L.J.
      • Rogers C.A.
      • et al.
      Intralesional macular atrophy in anti–vascular endothelial growth factor therapy for age-related macular degeneration in the IVAN trial.
      The etiology of MA remains unknown, but different hypotheses have implicated natural progression of underlying or impending atrophy, photoreceptor damage resulting from CNV, and treatment with anti-VEGF therapy.
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      Recently, the Classification of Atrophy Meetings (CAM) group proposed a consensus on the definition of atrophy using spectral-domain (SD)-OCT as a more sensitive detection method than color fundus photography (CFP) or fluorescein angiography (FA) assessments.
      • Sadda S.R.
      • Guymer R.
      • Holz F.G.
      • et al.
      Consensus definition for atrophy associated with age-related macular degeneration on OCT: Classification of Atrophy Report 3.
      The SD-OCT–based detection of MA prevalence and progression in the HARBOR population found no significant differences in the incidence or progression rates of MA among study arms, ranibizumab doses, or treatment regimens.
      • Sadda S.R.
      • Morgenthien E.
      • Gune S.
      A longitudinal SD-OCT analysis of macular atrophy in the HARBOR trial.
      Because MA has been noted to occur frequently in eyes with nAMD, an interest exists in understanding why atrophy occurs in this setting. Previous studies have suggested that continuous or monthly therapy
      • Chakravarthy U.
      • Harding S.P.
      • Rogers C.A.
      • et al.
      Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial.
      ,
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      or the specific anti-VEGF agent
      • Grunwald J.E.
      • Pistilli M.
      • Ying G-s
      • et al.
      Growth of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT).
      may be risk factors for development of MA. In a prior analysis of HARBOR that used CFP and FA for atrophy assessment, baseline subretinal fluid (SRF) was associated with a lower risk of atrophy, whereas baseline atrophy in the fellow eye and baseline presence of intraretinal cysts were associated with a higher risk of MA development.
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      Previous reports from CATT support the findings of baseline SRF and fellow eye atrophy, whereas an analysis of IVAN also reported an increased risk of MA associated with fellow eye atrophy at baseline.
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Bailey C.
      • Scott L.J.
      • Rogers C.A.
      • et al.
      Intralesional macular atrophy in anti–vascular endothelial growth factor therapy for age-related macular degeneration in the IVAN trial.
      However, an OCT-based analysis provides a potentially more accurate and robust assessment of MA, because the 3-dimensional attributes of dense volume SD-OCT allow specific retinal layers to be visualized, compared with the 2-dimensional assessment provided by CFP and FA. Moreover, because OCT data are often obtained more frequently than data from other imaging methods, OCT-based analysis allows for a more precise longitudinal assessment of the evolution of atrophy.
      The primary objective of this analysis was to take advantage of the monthly SD-OCT data to determine whether patient characteristics (demographic or ocular, as assessed by SD-OCT at baseline) or ranibizumab treatment (dose level or regimen) were risk factors for MA development over 2 years in the HARBOR trial population. Because SD-OCT is the most sensitive imaging method for detecting atrophy, and HARBOR SD-OCT images were obtained monthly, this analysis may provide the most accurate assessment of atrophy thus far.

      Methods

      This post hoc study was a retrospective analysis of SD-OCT images from the HARBOR trial (ClinicalTrials.gov identifier, NCT00891735), a phase 3, multicenter, prospective, randomized, double-blind, active treatment–controlled clinical trial whose study design and primary 12- and 24-month outcomes were reported previously.
      • Busbee B.G.
      • Ho A.C.
      • Brown D.M.
      • et al.
      Twelve-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration.
      ,
      • Ho A.C.
      • Busbee B.G.
      • Regillo C.D.
      • et al.
      Twenty-four-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration.
      All participants provided written informed consent, and the study protocol was approved by institutional review boards before the study start. The study adhered to the tenets of the Declaration of Helsinki and was conducted in accordance with Good Clinical Practice (International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use E6), applicable United States Food and Drug Administration regulations, and the Health Insurance Portability and Accountability Act.
      In brief, HARBOR evaluated the efficacy and safety of 2 doses and 2 regimens of ranibizumab in 1097 patients 50 years of age or older with subfoveal nAMD randomized 1:1:1:1 to receive ranibizumab 0.5 mg monthly (n = 275), 0.5 mg pro re nata (PRN; n = 275), 2.0 mg monthly (n = 274), or 2.0 mg PRN (n = 273) for 24 months. Study eye inclusion criteria included best-corrected visual acuity (BCVA) of 20/40 to 20/320 (Snellen equivalent) using Early Treatment Diabetic Retinopathy Study charts; active subfoveal lesions with classic CNV, some classic CNV component, or with purely occult CNV; total area of lesion less than 12 disc areas or 30.48 mm2; and total CNV area constituting 50% or more of total lesion area based on FA. Exclusion criteria relevant to the present analysis included subfoveal fibrosis or subfoveal atrophy in the study eye. As part of the screening process, a central reading center evaluated the patient’s fundus photographs and fluorescein angiograms to provide an objective masked assessment of the patient’s eligibility regarding classification of the CNV lesion and assessments of CNV area and leakage. All patients were evaluated monthly; patients in the PRN arms received 3 monthly loading doses initially, followed by re-treatment if the study eye met the prespecified re-treatment criteria: a 5-letter or more decrease in BCVA from the previous visit, or any evidence of disease activity on SD-OCT, such as intraretinal fluid, SRF, or pigment epithelial detachment (PED).
      • Busbee B.G.
      • Ho A.C.
      • Brown D.M.
      • et al.
      Twelve-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration.

       Study Objectives

      The primary objective of this analysis was to determine the SD-OCT characteristics of patients with nAMD that may be risk factors for MA development. This post hoc assessment of SD-OCT images was conducted on HARBOR study patients who completed the month 24 visit and had evaluable SD-OCT scans (N = 941). The statistical analysis population of this report consisted of patients who did not have MA at baseline (N = 761).

       Grading of Macular Atrophy

       OCT Imaging Protocol

      Patients in HARBOR underwent SD-OCT imaging by certified OCT imaging technicians every month for the 24-month duration of the study. OCT imaging was performed in both eyes using the Cirrus HD-OCT instrument (Carl Zeiss Meditec, Inc, Dublin, CA), with the macular cube scan comprising 512 A-scans × 128 B-scans over a 6 × 6-mm square centered on the fovea. Raw OCT data (.ib and associated Carl Zeiss raw data file format) for each patient were exported in a masked deidentified fashion and transferred for analysis to the Doheny Image Reading Center/Doheny Image Reading & Research Lab.

       Grading Overview

      Images were assessed using Cirrus OCT Advanced RPE Analysis software version 6.02.81 (Carl Zeiss Meditec, Inc).
      • Abdelfattah N.S.
      • Zhang H.
      • Boyer D.S.
      • et al.
      Progression of macular atrophy in patients with neovascular age-related macular degeneration undergoing antivascular endothelial growth factor therapy.
      All OCT scans were reviewed by certified Doheny Image Reading Center/Doheny Image Reading & Research Lab OCT graders in accordance with a centralized grading protocol. A total of 44 preselected variables were assessed, including drusen volume within the central 3- and 5-mm circles, presence of choroidal hypertransmission, RPE loss or thinning, loss of outer retinal layers, presence of SRF, subretinal hyperreflective material (SHRM), intraretinal cysts, outer retinal tubulations, subretinal drusenoid deposits (the OCT correlate of reticular pseudodrusen [RPD]), intraretinal hyperreflective foci, PED, type of PED, severity of cysts, density of SHRM, status of the RPE under cysts, location of lesions in relation to the fovea, SRF volume, SHRM thickness, PED volume, central foveal thickness (CFT), and subfoveal choroidal thickness.
      All cases were reviewed by 1 or more graders (N.S.A., J.L., Y.S., and S.B.); select cases were reviewed by a second independent grader to assess repeatability. In the instance of discrepancy between 2 graders, the graders met in open adjudication to resolve the discrepancy and generate a single response for each case. If graders were unable to resolve the discrepancy or had less than 90% confidence in any of the assessments (per reading center guidelines), the case was reviewed by the reading center principal investigator (S.R.S.) for a final determination.

       Macular Atrophy Assessment

      Macular atrophy was assessed from Cirrus HD-OCT volume scans (macular cube 512 × 128) by simultaneous assessment of en face OCT fundus images of the choroid and individual B-scans of the volume SD-OCT cube. Atrophy was identified by a combination of review of en face OCT images for hyperreflective patches suggestive of increased signal transmission, as well as scrutiny of individual SD-OCT B-scans. In addition, graders scrolled through all horizontal B-scans in the cube to assess for CAM criteria. In accordance with the CAM group,
      • Sadda S.R.
      • Guymer R.
      • Holz F.G.
      • et al.
      Consensus definition for atrophy associated with age-related macular degeneration on OCT: Classification of Atrophy Report 3.
      3 criteria had to be fulfilled for diagnosis of MA or complete RPE and outer retinal atrophy (cRORA) on OCT: (1) increased signal transmission through the choroid (choroidal hypertransmission), (2) attenuation of the RPE band, and (3) collapse or thinning of the outer retinal layers.
      • Sadda S.R.
      • Morgenthien E.
      • Gune S.
      A longitudinal SD-OCT analysis of macular atrophy in the HARBOR trial.
      The CAM criteria specified a minimum atrophy size of 250 μm in diameter; however, because the HARBOR OCT grading protocol was finalized before publication of the CAM criteria, a smaller minimum atrophy size of 125 μm or more was used in this analysis. Macular atrophy was considered Definite if 3 criteria were present and considered Questionable if 2 criteria were present. A region meeting only a single criterion (e.g., only hypertransmission) was not considered atrophy, demonstrating why the OCT en face image alone was insufficient for this assessment. It should be noted that Questionable MA as defined in this post hoc analysis is distinct from incomplete RPE and outer retinal atrophy (iRORA) as defined by CAM. The presence of SHRM or RPE elevation did not alter the definition of atrophy.
      Macular atrophy was assessed using data from all available study visits of the analysis population.
      • Abdelfattah N.S.
      • Zhang H.
      • Boyer D.S.
      • et al.
      Progression of macular atrophy in patients with neovascular age-related macular degeneration undergoing antivascular endothelial growth factor therapy.
      ,
      • Schmitz-Valckenberg S.
      • Sahel J.A.
      • Danis R.
      • et al.
      Natural history of geographic atrophy progression secondary to age-related macular degeneration (Geographic Atrophy Progression Study).
      Because the rate of progression could be assessed only in eyes with MA, images from all month 24 visits were assessed first to identify study eyes with MA present at the end of the study. This approach was based on the assumption that as soon as features of atrophy were present, they were not expected to change status to not present at any subsequent visits. In eyes for which the month 24 visit was missing or not gradable, the previous visit (month 23) was examined for presence of atrophy. For study eyes with MA present at month 24, SD-OCT scans for all prior visits were examined to identify the first visit in which 2 or 3 OCT criteria for atrophy were met.

       Definition of Ocular Risk Factors

      Clinical definitions of the significant risk factors are provided in Table 1 and clinical images are provided in Figure 1.
      Table 1Definition of Significant Risk Factors
      Risk FactorDefinition
      Intraretinal cystsFluid collections in the inner retina with intact underlying preserved RPE.
      Subretinal fluidExudation occurring between the outer border of photoreceptors and the inner border of the RPE.
      Neovascularization lesion Types 1, 2, and 3Standard OCT classifications were used to define neovascular lesions (Type 1, fibrovascular PED; Type 2, SHRM with or without PED; Type 3, RAP).
      Drusen volumeDrusen are yellow lipid deposits under the retina. Drusen volume was calculated within 3- and 5-mm circles (square millimeters) centered on the fovea and automatically generated by Cirrus OCT.
      Reticular pseudodrusenThe infrared reflectance image corresponding to the hyperreflective material above the RPE on OCT (unlike drusen, which lie below the RPE).
      Subfoveal choroidal thicknessIn the foveal center, the choroidal thickness is the distance between the outer border of the RPE and the choroid–scleral junction.
      Central foveal thicknessThe distance between the ILM and the outer border of the photoreceptors (in the absence of SHRM or SRF, this should be coincident with the inner border of the RPE). Neither SHRM nor SRF were included in the calculation of central foveal thickness.
      Nascent atrophy
      The definition of nascent atrophy used in the present analysis is distinct from the definition of nascent atrophy from Wu et al.28
      At baseline, 2 variables were used to determine nascent atrophy: RPE loss or thinning and outer retinal layer loss. Nascent atrophy was defined as loss of outer retinal layers with no loss or thinning of RPE. Choroidal hypertransmission was not assessed for the determination of nascent atrophy.
      Atrophy (fellow eye)Atrophy in the fellow eye was diagnosed using the same 3 criteria on OCT as atrophy in the study eye: choroidal hypertransmission, attenuation of the RPE band, and collapse or thinning of the outer retinal layers.
      ILM = inner limiting membrane; PED = pigment epithelial detachment; RAP = retinal angiomatous proliferation; RPE = retinal pigment epithelium; SHRM = subretinal hyperreflective material; SRF = subretinal fluid.
      The definition of nascent atrophy used in the present analysis is distinct from the definition of nascent atrophy from Wu et al.
      • Wu Z.
      • Luu C.D.
      • Ayton L.N.
      • et al.
      Optical coherence tomography–defined changes preceding the development of drusen-associated atrophy in age-related macular degeneration.
      Figure thumbnail gr1
      Figure 1Clinical images showing significant risk factors. NV = neovascularization.

       Statistical Methods

      The assessment of images was conducted on evaluable SD-OCT macular cube scans from the study eyes of patients with 24 months of follow-up (N = 941). The statistical analysis population consisted of patients who did not have MA at baseline (N = 761) and excluded 131 patients with Definite MA and 49 patients with Questionable MA at baseline. The end points for analysis were time to Any MA development (defined as either Definite or Questionable MA) and time to Definite MA development. Time to Any MA development (months) was calculated as (date of first visit where Definite or Questionable MA was detected – baseline visit date + 1) / 30.4375. Patients who did not have Definite or Questionable MA detected at the last visit with an evaluable scan were censored as of the date of that visit. Time to Definite MA development was defined similarly.
      Candidate explanatory variables included treatment arm, total number of injections, age, sex, cardiovascular history, baseline BCVA and low-luminance gap (difference between BCVA and low-luminance visual acuity), 25 baseline study eye characteristics determined by SD-OCT (Table S1, available at www.aaojournal.com), and baseline fellow eye atrophy. To include all analysis population patients in multivariable analyses of time to atrophy development, the following missing data handling strategies were used. First, any candidate variable with excessive missing data (>10% of patients) was excluded from multivariable models. Second, logical data corrections were applied as follows: if a characteristic was absent from the general location, it was considered absent from the subfoveal location; if a characteristic was absent from the general location, thickness or volume measurements, or both, were assumed to be 0; or if a characteristic was present subfoveally, it was considered present generally. Third, remaining missing data values (affecting ≤7% of patients for any single variable) were imputed using a single best-guess strategy. If a characteristic was present, but a measurement such as volume or thickness was not reported, the median value from all patients with the characteristic present was used. If a categorical characteristic was missing (such as presence or absence of something), the mode of all patients with a reported value was used. The impact of best-guess imputation was assessed through comparison of multivariable model P values comparing analyses with logical data corrections and logical corrections plus best-guess imputation.
      Multivariable proportional hazards regression was performed for time to atrophy development in months. Backward elimination was performed, keeping treatment arm (required) and statistically significant explanatory variables (P ≤ 0.05). Model robustness was assessed through 5-fold cross-validation. After model selection, hazard ratios (HRs; 95% confidence intervals [CIs]) were estimated. For categorical variables with more than 2 levels, pairwise comparisons were constructed. For treatment arms, each of the 3 other arms was compared with the 0.5-mg monthly arm. For neovascularization types, the following comparisons were made: Type 2 versus Type 1, Types 1 and 2 versus Type 1, Any Type 3 (3, 1 and 3, 2 and 3) versus Type 1, and Not Determined versus Type 1. The Sidak adjustment for multiple comparisons was applied to comparisons among treatments and CNV types. For continuous variables, HR estimates were constructed to provide the increase (or decrease) in hazard corresponding to an X-unit increase in the measurement of the explanatory variable. X was selected based on the unit of measurement and the range of observed data values and had no impact on the significance of the explanatory variable. For example, the HR for CFT (measured from the inner limiting membrane to the outer aspect of the neurosensory retina) was constructed to provide the increase (or decrease) in hazard corresponding to a 200-μm increase in baseline CFT.
      Univariate proportional hazards regression was performed for candidate variables that were excluded from the multivariable analyses because of the extent of missing data, for example, subfoveal choroidal thickness.

      Results

      A summary of patient demographics and selected baseline characteristics overall and by month 24 MA status is presented in Table 2. Baseline risk factors for MA in the study eye were assessed from the population of patients who completed the study with evaluable SD-OCT scans and did not have MA at baseline (N = 761).
      Table 2HARBOR Patient Demographics and Selected Baseline Characteristics, Overall and by Month 24 Macular Atrophy Status, in Completers with Evaluable Scans and No Baseline Atrophy
      Characteristics of the Analysis Population
      Completers with evaluable scans and no baseline atrophy.
      OverallMonth 24 Atrophy Status
      Definite or QuestionableAbsent
      Sample size, no.761246515
      Age (yrs), mean (SD)77.9 (8.26)79.2 (7.41)77.3 (8.58)
      Female450 (59.1)152 (61.8)298 (57.9)
      White736 (96.7)242 (98.4)494 (95.9)
      Smoking, current or former451 (59.3)150 (61.0)301 (58.4)
      Cardiovascular history, yes244 (32.1)78 (31.7)166 (32.2)
      Diabetes mellitus, yes131 (17.2)46 (18.7)85 (16.5)
      Hypertension, yes545 (71.6)182 (74.0)363 (70.5)
      SD = standard deviation.
      Data are number (%).
      Completers with evaluable scans and no baseline atrophy.

       Confirmation of Previous Risk Factors

      Several baseline risk factors were confirmed from previous literature analyses (Figure 2, Figure 3).
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      • Bailey C.
      • Scott L.J.
      • Rogers C.A.
      • et al.
      Intralesional macular atrophy in anti–vascular endothelial growth factor therapy for age-related macular degeneration in the IVAN trial.
      Presence of intraretinal cysts at baseline was confirmed as a risk factor for Any MA (HR, 1.53; 95% CI, 1.13–2.09; P = 0.007) and Definite MA (HR, 1.54; 95% CI, 1.09–2.19; P = 0.016). Likewise, presence of Type 3 neovascularization at baseline was confirmed as a risk factor for development of Any MA (HR, 2.01; 95% CI, 1.09–3.70; P = 0.019) and Definite MA (HR, 2.15; 95% CI, 1.10–4.21; P = 0.019). Presence of MA in the fellow eye at baseline was also confirmed as a risk factor for Any MA (HR, 1.66; 95% CI, 1.28–2.15; P < 0.001) and Definite MA (HR, 1.92; 95% CI, 1.43–2.57; P < 0.001). Presence of baseline SRF was confirmed to be associated with a lower development rate of Any MA (HR, 0.60; 95% CI, 0.44–0.82; P = 0.001) and Definite MA (HR, 0.59; 95% CI, 0.42–0.83; P = 0.003).
      Figure thumbnail gr2
      Figure 2Graph showing risk factors for development of Any Macular Atrophy (MA). A multivariable proportional hazards regression model was obtained by backward elimination, keeping treatment (required) and all other factors with an overall P value ≤ 0.05. Ninety-five percent confidence intervals are shown. Bolded risk factors indicate new baseline risk factors for MA development identified using spectral-domain OCT. BL = baseline; CFT = central foveal thickness; CNV = choroidal neovascularization; NV = neovascularization; PRN = pro re nata; RPD = reticular pseudodrusen; SRF = subretinal fluid.
      Figure thumbnail gr3
      Figure 3Graph showing risk factors for development of Definite Macular Atrophy (MA). A multivariable proportional hazards regression model was obtained by backward elimination, keeping treatment (required) and all other factors with an overall P value of ≤ 0.05. Ninety-five percent confidence intervals are shown. Bolded risk factors indicate new baseline risk factors for MA development identified using spectral-domain OCT. BL = baseline; CFT = central foveal thickness; CNV = choroidal neovascularization; NV = neovascularization; PRN = pro re nata; SRF = subretinal fluid.

       Identification of New Risk Factors

      Additional risk factors for MA development at 24 months were identified in this analysis of SD-OCT data (Figure 2, Figure 3). Higher central drusen volume (within central 3 mm) at baseline was a risk factor for development of Any MA (HR, 1.23; 95% CI, 1.08–1.39; P = 0.001) and Definite MA (HR, 2.14; 95% CI, 1.29–3.57; P = 0.003). Presence of nascent atrophy at baseline was a risk factor for development of Any MA (HR, 4.22; 95% CI, 2.68–6.64; P < 0.001) and Definite MA (HR, 3.61; 95% CI, 2.21–5.90; P < 0.001). Presence of RPD at baseline was a risk factor for development of Any MA (HR, 1.43; 95% CI, 1.03–1.98; P = 0.032). Reticular pseudodrusen was included as a candidate risk factor for Definite MA but was not retained in the model obtained by backward elimination. In addition, a 200-μm increase in CFT was associated with a higher risk of development of Any MA (HR, 1.37; 95% CI, 1.15–1.63; P < 0.001) and Definite MA (HR, 1.34; 95% CI, 1.10–1.64; P = 0.004). Subretinal fluid was not included in the calculation of CFT.

       Effect of Ranibizumab Treatment on Risk of Macular Atrophy Development

      Ranibizumab treatment regimen and dose level were not risk factors for development of MA at 24 months (Figure 2, Figure 3). Ranibizumab treatment regimen of 0.5 mg PRN versus 0.5 mg monthly was not found to be a risk factor for Any MA (HR, 0.76; 95% CI, 0.49–1.18; P = 0.358) or Definite MA (HR, 0.89; 95% CI, 0.54–1.47; P = 0.928). Ranibizumab treatment regimen of 2.0 mg PRN versus 0.5 mg monthly was not found to be a risk factor for Any MA (HR, 0.86; 95% CI, 0.55–1.33; P = 0.791) or Definite MA (HR, 0.97; 95% CI, 0.59–1.62; P = 0.999). Ranibizumab dose level of 2.0 mg monthly versus 0.5 mg monthly was not found to be a risk factor for Any MA (HR, 0.92; 95% CI, 0.60–1.43; P = 0.960) or Definite MA (HR, 1.06; 95% CI, 0.64–1.74; P = 0.992).

       Risk Factor Identified by Univariate Analysis

      Some variables were excluded from the multivariable analyses because of missing baseline data in more than 25% of patients. For choroidal thickness, SD-OCT scans were obtained in non–enhanced depth imaging mode because enhanced depth imaging mode was not required by the study protocol, thus resulting in approximately 25% of scans not allowing visualization of the choroid at a quality good enough to be considered gradable. Univariate analyses indicated that a 100-μm increase in baseline central subfoveal choroidal thickness was associated with an approximately 25% decreased risk of Any MA (HR, 0.76; 95% CI, 0.61–0.94; P = 0.014; Fig 4) or Definite MA (HR, 0.75; 95% CI, 0.59–0.97; P = 0.025; Fig 4).
      Figure thumbnail gr4
      Figure 4Graph showing that increase in choroidal thickness leads to decreased risk of macular atrophy (MA) development. Determined by univariate proportional hazards regression model. Ninety-five percent confidence intervals are shown. BL = baseline.

      Discussion

      Overall, this analysis of monthly SD-OCT data confirmed previously identified baseline risk factors for MA development but also identified several new risk factors. In previous literature reports from the HARBOR, CATT, and IVAN trials, risk factors reported included the presence at baseline of intraretinal cysts in the study eye and atrophy in the fellow eye, whereas presence of baseline SRF was associated with a lower rate of atrophy development (Table 3).
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      • Bailey C.
      • Scott L.J.
      • Rogers C.A.
      • et al.
      Intralesional macular atrophy in anti–vascular endothelial growth factor therapy for age-related macular degeneration in the IVAN trial.
      To further investigate the effect of SRF, we evaluated the relationship between SRF presence at any visit during the monitoring phase in HARBOR (month 3 to study end) and final MA status; the results were consistent with those observed using the baseline data. Of the patients with SRF at any time during monitoring, 34.8% demonstrated MA by month 24, whereas 65.2% of patients without SRF at any time demonstrated MA by month 24 (P < 0.001, chi-square test). The explanation for why SRF is associated with less atrophy is a topic of debate. The SRF may be an indicator of a persistent CNV lesion, and recent reports have suggested that a persistent Type 1 CNV lesion may actually protect against the development of atrophy.
      • Christenbury J.G.
      • Phasukkijwatana N.
      • Gilani F.
      • et al.
      Progression of macular atrophy in eyes with type 1 neovascularization and age-related macular degeneration receiving long-term intravitreal anti-vascular endothelial growth factor therapy: an optical coherence tomographic angiography analysis.
      ,
      • Dhrami-Gavazi E.
      • Balaratnasingam C.
      • Lee W.
      • et al.
      Type 1 neovascularization may confer resistance to geographic atrophy amongst eyes treated for neovascular age-related macular degeneration.
      Thus, a persistent but controlled (i.e., by anti-VEGF therapy) CNV lesion below an intact RPE may be the optimal anatomic outcome for a patient with nAMD. This hypothesis of course needs to be evaluated in future prospective studies. In contrast, Type 3 neovascularization (retinal angiomatous proliferation lesion) was reported to be a risk factor for atrophy development in eyes treated with anti-VEGF.
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Daniel E.
      • Shaffer J.
      • Ying G-s
      • et al.
      Outcomes in eyes with retinal angiomatous proliferation in the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT).
      This also makes sense, because Type 3 lesions are usually preceded by migration of RPE cells from the monolayer into the neurosensory retina,
      • Nagiel A.
      • Sarraf D.
      • Sadda S.R.
      • et al.
      Type 3 neovascularization: evolution, association with pigment epithelial detachment, and treatment response as revealed by spectral domain optical coherence tomography.
      which may leave a gap in the RPE layer. Thus, it is perhaps not surprising that atrophy typically ensues in these eyes. Regardless, the present analysis confirmed these findings and perhaps provides greater confidence in these putative risk factors, given the high sensitivity of SD-OCT for detecting atrophy. Three-dimensional detection methods, compared with the 2-dimensional assessment provided by CFP and FA, provide a more accurate description of the baseline characteristics that are risk factors for development of MA.
      Table 3Literature-Reported Risk Factors for Atrophy Development in Eyes with Neovascular Age-Related Macular Degeneration in HARBOR, Comparison of Age-Related Macular Degeneration Treatments Trials (CATT), and Inhibition of VEGF in Age-related choroidal Neovascularisation (IVAN)
      StudyReferenceImaging MethodNo. of Eyes Included in AnalysisBaseline Factors Associated with Increased Risk of AtrophyBaseline Factors Associated with Decreased Risk of Atrophy
      HARBORSadda et al (2018)
      • Sadda S.R.
      • Tuomi L.L.
      • Ding B.
      • et al.
      Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
      CFP/FA1095
      • Presence of intraretinal cysts (HR, 2.45)
      • Fellow eye atrophy (HR, 2.02)
      • Presence of subretinal fluid (HR, 0.50)
      CATTGrunwald et al (2014)
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      CFP/FA1024
      • BL VA ≤20/200 (aHR, 2.65)
      • Presence of RAP lesions (aHR, 1.69)
      • Atrophy in fellow eye (aHR, 2.07)
      • Intraretinal fluid in foveal center (aHR, 2.10)
      • Monthly vs. PRN dosing (aHR, 1.59)
      • Blocked fluorescence (aHR, 0.49)
      • Subretinal fluid thickness >25 μ (aHR, 0.52)
      • Subretinal tissue complex thickness >275 μ vs. ≤75 μ (aHR, 0.31)
      • Vitreomacular attachment (aHR, 0.55)
      CATTGrunwald et al (2017)
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      CFP/FA1011
      • Older age (≥80 years; aHR, 1.58)
      • Hypercholesterolemia (aHR, 1.47)
      • BL VA ≤20/100 (20/100–20/160 [aHR, 1.76] or 20/200–20/320 [aHR, 1.78])
      • BL CNV area >4 DA (aHR, 1.99)
      • Presence of RAP lesions (aHR, 2.02)
      • Atrophy in fellow eye (aHR, 2.31)
      • Presence of intraretinal fluid (in the foveal center [aHR, 2.27] or not in foveal center [aHR, 1.83])
      • Monthly vs. PRN regimen (aHR, 1.38)
      • Subretinal tissue complex >275 mm (aHR, 0.29)
      • Presence of subretinal fluid (in the foveal center [aHR, 0.37] or not in foveal center [aHR, 0.61])
      IVANChakravarthy et al (2013)
      • Chakravarthy U.
      • Harding S.P.
      • Rogers C.A.
      • et al.
      Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial.
      CFP/OCT (only BL and most recent available follow-up)610
      • Continuous vs. discontinuous treatment (OR, 1.45)
      IVANBailey et al (2019)
      • Bailey C.
      • Scott L.J.
      • Rogers C.A.
      • et al.
      Intralesional macular atrophy in anti–vascular endothelial growth factor therapy for age-related macular degeneration in the IVAN trial.
      CFP/FA/OCT425
      • Extralesional atrophy in fellow eye (OR, 4.96)
      • Intralesional atrophy in fellow eye (OR, 2.34)
      • >50% classic CNV (OR, 0.39)
      aHR = adjusted hazard ratio; BL = baseline; CFP = color fundus photography; CNV = choroidal neovascularization; DA = disc area; FA = fluorescein angiography; HR = hazard ratio; OR = odds ratio; PRN = pro re nata; RAP = retinal angiomatous proliferation; VA = visual acuity.
      In addition, this analysis identified new baseline risk factors for MA development, as determined by time to development of MA over 24 months, which had not been identified previously using CFP and FA methods. Presence of nascent atrophy, higher central drusen volume (within central 3 mm), lower choroidal thickness, presence of RPD (manifesting as subretinal drusenoid deposits on OCT), and higher CFT (excluding SRF) were all identified on OCT as risk factors for MA development. Most of these characteristics are also risk factors for the development of geographic atrophy (GA) in the absence of CNV.
      • Wu Z.
      • Luu C.D.
      • Ayton L.N.
      • et al.
      Optical coherence tomography–defined changes preceding the development of drusen-associated atrophy in age-related macular degeneration.
      • Lindner M.
      • Bezatis A.
      • Czauderna J.
      • et al.
      Choroidal thickness in geographic atrophy secondary to age-related macular degeneration.
      • Adhi M.
      • Lau M.
      • Liang M.C.
      • et al.
      Analysis of the thickness and vascular layers of the choroid in eyes with geographic atrophy using spectral-domain optical coherence tomography.
      • Finger R.P.
      • Wu Z.
      • Luu C.D.
      • et al.
      Reticular pseudodrusen: a risk factor for geographic atrophy in fellow eyes of individuals with unilateral choroidal neovascularization.
      • Finger R.P.
      • Chong E.
      • McGuinness M.B.
      • et al.
      Reticular pseudodrusen and their association with age-related macular degeneration: the Melbourne Collaborative Cohort Study.
      Nascent atrophy as defined by Wu et al
      • Wu Z.
      • Luu C.D.
      • Ayton L.N.
      • et al.
      Optical coherence tomography–defined changes preceding the development of drusen-associated atrophy in age-related macular degeneration.
      is a drusen-associated atrophy preceding the development of cRORA (a subtype of atrophy under which GA is classified)
      • Sadda S.R.
      • Guymer R.
      • Holz F.G.
      • et al.
      Consensus definition for atrophy associated with age-related macular degeneration on OCT: Classification of Atrophy Report 3.
      and on SD-OCT can be observed as subsidence of the outer plexiform layer and inner nuclear layer and development of a hyporeflective wedge-shaped band within the limits of the outer plexiform layer.
      • Wu Z.
      • Luu C.D.
      • Ayton L.N.
      • et al.
      Optical coherence tomography–defined changes preceding the development of drusen-associated atrophy in age-related macular degeneration.
      In our analysis, nascent atrophy was defined as loss of outer retinal layer without the loss of RPE regardless of drusen presence. Although the 2 definitions are slightly different, they both point to the same conclusion of the loss of outer retinal layers being a risk factor for atrophy development. Lower choroidal thickness has been shown to correlate with the presence of GA,
      • Lindner M.
      • Bezatis A.
      • Czauderna J.
      • et al.
      Choroidal thickness in geographic atrophy secondary to age-related macular degeneration.
      ,
      • Adhi M.
      • Lau M.
      • Liang M.C.
      • et al.
      Analysis of the thickness and vascular layers of the choroid in eyes with geographic atrophy using spectral-domain optical coherence tomography.
      and eyes with the fastest progressing fundus autofluorescence subtype of diffuse trickling
      • Holz F.G.
      • Bindewald-Wittich A.
      • Fleckenstein M.
      • et al.
      Progression of geographic atrophy and impact of fundus autofluorescence patterns in age-related macular degeneration.
      have been shown to have a thinner choroid.
      • Lindner M.
      • Bezatis A.
      • Czauderna J.
      • et al.
      Choroidal thickness in geographic atrophy secondary to age-related macular degeneration.
      The presence of RPD is a risk factor for the progression of intermediate AMD to GA,
      • Finger R.P.
      • Wu Z.
      • Luu C.D.
      • et al.
      Reticular pseudodrusen: a risk factor for geographic atrophy in fellow eyes of individuals with unilateral choroidal neovascularization.
      ,
      • Finger R.P.
      • Chong E.
      • McGuinness M.B.
      • et al.
      Reticular pseudodrusen and their association with age-related macular degeneration: the Melbourne Collaborative Cohort Study.
      and conflicting reports suggest that RPD may or may not be associated with a thinner choroid.
      • Garg A.
      • Oll M.
      • Yzer S.
      • et al.
      Reticular pseudodrusen in early age-related macular degeneration are associated with choroidal thinning.
      ,
      • Ho C.Y.
      • Lek J.J.
      • Aung K.Z.
      • et al.
      Relationship between reticular pseudodrusen and choroidal thickness in intermediate age-related macular degeneration.
      In the presence of Type 3 neovascularization, both RPD and a thinner choroid are risk factors for atrophy development.
      • Cho H.J.
      • Yoo S.G.
      • Kim H.S.
      • et al.
      Risk factors for geographic atrophy after intravitreal ranibizumab injections for retinal angiomatous proliferation.
      The current analysis of OCT data is more robust than assessments of CFP or FA data, providing a more comprehensive depiction of MA risk factors. Identification of these OCT-based risk factors is valuable because OCT is nearly universally obtained in patients with nAMD undergoing anti-VEGF therapy and may be useful in patient prognostication.
      As noted previously (vide supra), lower choroidal thickness was identified in the univariate analysis as a risk factor for MA development. These findings would seem to support the growing body of literature suggesting that choroidal insufficiency, in particular the choriocapillaris, may contribute to the development and progression of atrophy.
      • Alagorie A.R.
      • Verma A.
      • Nassisi M.
      • et al.
      Quantitative assessment of choriocapillaris flow deficits in eyes with advanced age-related macular degeneration versus healthy eyes.
      • Moreira-Neto C.A.
      • Moult E.M.
      • Fujimoto J.G.
      • et al.
      Choriocapillaris loss in advanced age-related macular degeneration.
      • Nassisi M.
      • Shi Y.
      • Fan W.
      • et al.
      Choriocapillaris impairment around the atrophic lesions in patients with geographic atrophy: a swept-source optical coherence tomography angiography study.
      • Thulliez M.
      • Zhang Q.
      • Shi Y.
      • et al.
      Correlations between choriocapillaris flow deficits around geographic atrophy and enlargement rates based on swept-source OCT imaging.
      • Nassisi M.
      • Baghdasaryan E.
      • Borrelli E.
      • et al.
      Choriocapillaris flow impairment surrounding geographic atrophy correlates with disease progression.
      • Arya M.
      • Sabrosa A.S.
      • Duker J.S.
      • et al.
      Choriocapillaris changes in dry age-related macular degeneration and geographic atrophy: a review.
      Unfortunately, OCT angiography data were not available from the HARBOR trial to assess the choriocapillaris specifically. Nevertheless, data from the present analysis may suggest a need to monitor patients with thin choroids and potentially use this parameter as a prognostic tool for educating patients on the potential for atrophy development.
      Previous literature reports state that monthly, or continuous, treatment of nAMD with anti-VEGF agents is a potential risk factor for the development of atrophy (Table 3).
      • Chakravarthy U.
      • Harding S.P.
      • Rogers C.A.
      • et al.
      Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial.
      ,
      • Grunwald J.E.
      • Daniel E.
      • Huang J.
      • et al.
      Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
      ,
      • Grunwald J.E.
      • Pistilli M.
      • Daniel E.
      • et al.
      Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
      One analysis of CATT reported that ranibizumab was a greater risk for atrophy enlargement than bevacizumab.
      • Grunwald J.E.
      • Pistilli M.
      • Ying G-s
      • et al.
      Growth of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT).
      However, these analyses were performed using quarterly, semiannual, or annual CFP and FA assessments as opposed to the monthly SD-OCT assessments in this study. In the present analysis, neither ranibizumab treatment regimen (monthly vs. PRN) nor ranibizumab dose level (2.0 mg vs. 0.5 mg) were found to be risk factors for development of MA. These results correlate with previous findings in a mouse model that VEGF blockage is not a risk factor for photoreceptor survival, and hence atrophy development.
      • Miki A.
      • Miki K.
      • Ueno S.
      • et al.
      Prolonged blockade of VEGF receptors does not damage retinal photoreceptors or ganglion cells.
      This more sensitive monthly OCT analysis provides results that lead to higher confidence in the findings.
      A limitation of this analysis is that the sample size and power calculation in the HARBOR study were not based on detecting any differences explored in these retrospective analyses, which may lead to lack of power in certain comparisons. Other limitations include the post hoc nature of this analysis, which prevents determination of causal relationships because of the lack of randomization and control of type I error, and the absence of a natural history arm in the HARBOR study to allow for comparison against treatment arms. The recent risk factor analysis performed for the Age-Related Eye Disease Study 2 (AREDS2) provides a cohort not treated with anti-VEGF agents for comparison with HARBOR,
      • Domalpally A.
      • Danis R.P.
      • Trane R.
      • et al.
      Atrophy in neovascular age-related macular degeneration: Age-Related Eye Disease Study 2 report number 15.
      although comparisons between studies may be fraught with confounders of selection bias. The analysis of AREDS2 participants with intermediate AMD enrolled in the fundus autofluorescence ancillary study (eyes not treated with anti-VEGF) concluded that atrophy is common in nAMD and often associated with pre-existing GA.
      • Domalpally A.
      • Danis R.P.
      • Trane R.
      • et al.
      Atrophy in neovascular age-related macular degeneration: Age-Related Eye Disease Study 2 report number 15.
      Because the HARBOR study lasted 24 months, this analysis cannot address the long-term durability of these findings beyond 2 years. Another limitation of this analysis is that the assessment of atrophy criteria was limited in areas with clinical features, causing a shadowing effect on OCT (i.e., subretinal fibrosis or SHRM with fibrous tissue). In addition, no refractive error or axial length information of the eyes was found to adjust or correct choroidal thickness values further.
      In conclusion, baseline risk factors for development of MA were confirmed from prior analyses; presence of intraretinal cysts, presence of Type 3 neovascularization, and presence of MA in the fellow eye were associated with an increased risk of MA, whereas presence of SRF was associated with a lower rate of atrophy development. Several new baseline risk factors were identified using OCT; higher central drusen volume (within central 3 mm), thinner choroid and thicker retina at the fovea, presence of nascent atrophy, and presence of RPD were all associated with an increased risk of MA. Neither monthly ranibizumab treatment nor quadrupling the dose of ranibizumab increased the risk of MA development at 24 months. It is important to note that the risk factors for MA development seem to overlap with risk factors for GA development. Coupled with the lack of connection between dosing amount or frequency and greater rates of MA development, these results suggest the atrophy observed is more likely related to natural progression of the underlying AMD disease rather than any direct effect of ranibizumab treatment. The difference in risk factors from previous reports is presumably reflected by the difference in imaging methods using 2-dimensional CFP and FA versus 3-dimensional SD-OCT analyses. This analysis was performed on outputs obtained from monthly assessments on SD-OCT, the most sensitive method for atrophy assessment, which may give us greater confidence in the observations.

      Acknowledgment

      Third-party writing assistance (manuscript draft preparation and revision per author direction) was provided by Charlotte A. Osborne, PhD, CMPP, of Envision Pharma Group and funded by Genentech, Inc., a member of the Roche Group.

      Supplementary Data

      References

        • Ferris 3rd, F.L.
        • Wilkinson C.P.
        • Bird A.
        • et al.
        Clinical classification of age-related macular degeneration.
        Ophthalmology. 2013; 120: 844-851
        • Wong T.Y.
        • Chakravarthy U.
        • Klein R.
        • et al.
        The natural history and prognosis of neovascular age-related macular degeneration: a systematic review of the literature and meta-analysis.
        Ophthalmology. 2008; 115: 116-126
        • Spaide R.F.
        Improving the age-related macular degeneration construct: a new classification system.
        Retina. 2018; 38: 891-899
        • Rosenfeld P.J.
        Optical coherence tomography and the development of antiangiogenic therapies in neovascular age-related macular degeneration.
        Invest Ophthalmol Vis Sci. 2016; 57 (OCT14–OCT26)
        • de Oliveira Dias J.R.
        • Zhang Q.
        • Garcia J.M.B.
        • et al.
        Natural history of subclinical neovascularization in nonexudative age-related macular degeneration using swept-source OCT angiography.
        Ophthalmology. 2018; 125: 255-266
        • Brown D.M.
        • Kaiser P.K.
        • Michels M.
        • et al.
        Ranibizumab versus verteporfin for neovascular age-related macular degeneration.
        N Engl J Med. 2006; 355: 1432-1444
        • Rosenfeld P.J.
        • Brown D.M.
        • Heier J.S.
        • et al.
        Ranibizumab for neovascular age-related macular degeneration.
        N Engl J Med. 2006; 355: 1419-1431
        • Busbee B.G.
        • Ho A.C.
        • Brown D.M.
        • et al.
        Twelve-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration.
        Ophthalmology. 2013; 120: 1046-1056
        • Chakravarthy U.
        • Harding S.P.
        • Rogers C.A.
        • et al.
        Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial.
        Ophthalmology. 2012; 119: 1399-1411
        • Martin D.F.
        • Maguire M.G.
        • Ying G.S.
        • et al.
        Ranibizumab and bevacizumab for neovascular age-related macular degeneration.
        N Engl J Med. 2011; 364: 1897-1908
        • Kodjikian L.
        • Souied E.H.
        • Mimoun G.
        • et al.
        Ranibizumab versus bevacizumab for neovascular age-related macular degeneration: results from the GEFAL noninferiority randomized trial.
        Ophthalmology. 2013; 120: 2300-2309
        • Schmidt-Erfurth U.
        • Kaiser P.K.
        • Korobelnik J.F.
        • et al.
        Intravitreal aflibercept injection for neovascular age-related macular degeneration: ninety-six-week results of the VIEW studies.
        Ophthalmology. 2014; 121: 193-201
        • Chakravarthy U.
        • Harding S.P.
        • Rogers C.A.
        • et al.
        Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial.
        Lancet. 2013; 382: 1258-1267
        • Grunwald J.E.
        • Daniel E.
        • Huang J.
        • et al.
        Risk of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials.
        Ophthalmology. 2014; 121: 150-161
        • Grunwald J.E.
        • Pistilli M.
        • Ying G-s
        • et al.
        Growth of geographic atrophy in the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT).
        Ophthalmology. 2015; 122: 809-816
        • Grunwald J.E.
        • Pistilli M.
        • Daniel E.
        • et al.
        Incidence and growth of geographic atrophy during 5 years of Comparison of Age-Related Macular Degeneration Treatments Trials.
        Ophthalmology. 2017; 124: 97-104
        • Sadda S.R.
        • Tuomi L.L.
        • Ding B.
        • et al.
        Macular atrophy in the HARBOR study for neovascular age-related macular degeneration.
        Ophthalmology. 2018; 125: 878-886
        • Bailey C.
        • Scott L.J.
        • Rogers C.A.
        • et al.
        Intralesional macular atrophy in anti–vascular endothelial growth factor therapy for age-related macular degeneration in the IVAN trial.
        Ophthalmology. 2019; 126: 75-86
        • Sadda S.R.
        • Guymer R.
        • Holz F.G.
        • et al.
        Consensus definition for atrophy associated with age-related macular degeneration on OCT: Classification of Atrophy Report 3.
        Ophthalmology. 2018; 125: 537-548
        • Sadda S.R.
        • Morgenthien E.
        • Gune S.
        A longitudinal SD-OCT analysis of macular atrophy in the HARBOR trial.
        Invest Ophthalmol Vis Sci. 2018; 59 (ARVO E-Abstract 4481)
        • Ho A.C.
        • Busbee B.G.
        • Regillo C.D.
        • et al.
        Twenty-four-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration.
        Ophthalmology. 2014; 121: 2181-2192
        • Abdelfattah N.S.
        • Zhang H.
        • Boyer D.S.
        • et al.
        Progression of macular atrophy in patients with neovascular age-related macular degeneration undergoing antivascular endothelial growth factor therapy.
        Retina. 2016; 36: 1843-1850
        • Schmitz-Valckenberg S.
        • Sahel J.A.
        • Danis R.
        • et al.
        Natural history of geographic atrophy progression secondary to age-related macular degeneration (Geographic Atrophy Progression Study).
        Ophthalmology. 2016; 123: 361-368
        • Christenbury J.G.
        • Phasukkijwatana N.
        • Gilani F.
        • et al.
        Progression of macular atrophy in eyes with type 1 neovascularization and age-related macular degeneration receiving long-term intravitreal anti-vascular endothelial growth factor therapy: an optical coherence tomographic angiography analysis.
        Retina. 2018; 38: 1276-1288
        • Dhrami-Gavazi E.
        • Balaratnasingam C.
        • Lee W.
        • et al.
        Type 1 neovascularization may confer resistance to geographic atrophy amongst eyes treated for neovascular age-related macular degeneration.
        Int J Retina Vitreous. 2015; 1: 15
        • Daniel E.
        • Shaffer J.
        • Ying G-s
        • et al.
        Outcomes in eyes with retinal angiomatous proliferation in the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT).
        Ophthalmology. 2016; 123: 609-616
        • Nagiel A.
        • Sarraf D.
        • Sadda S.R.
        • et al.
        Type 3 neovascularization: evolution, association with pigment epithelial detachment, and treatment response as revealed by spectral domain optical coherence tomography.
        Retina. 2015; 35: 638-647
        • Wu Z.
        • Luu C.D.
        • Ayton L.N.
        • et al.
        Optical coherence tomography–defined changes preceding the development of drusen-associated atrophy in age-related macular degeneration.
        Ophthalmology. 2014; 121: 2415-2422
        • Lindner M.
        • Bezatis A.
        • Czauderna J.
        • et al.
        Choroidal thickness in geographic atrophy secondary to age-related macular degeneration.
        Invest Ophthalmol Vis Sci. 2015; 56: 875-882
        • Adhi M.
        • Lau M.
        • Liang M.C.
        • et al.
        Analysis of the thickness and vascular layers of the choroid in eyes with geographic atrophy using spectral-domain optical coherence tomography.
        Retina. 2014; 34: 306-312
        • Finger R.P.
        • Wu Z.
        • Luu C.D.
        • et al.
        Reticular pseudodrusen: a risk factor for geographic atrophy in fellow eyes of individuals with unilateral choroidal neovascularization.
        Ophthalmology. 2014; 121: 1252-1256
        • Finger R.P.
        • Chong E.
        • McGuinness M.B.
        • et al.
        Reticular pseudodrusen and their association with age-related macular degeneration: the Melbourne Collaborative Cohort Study.
        Ophthalmology. 2016; 123: 599-608
        • Holz F.G.
        • Bindewald-Wittich A.
        • Fleckenstein M.
        • et al.
        Progression of geographic atrophy and impact of fundus autofluorescence patterns in age-related macular degeneration.
        Am J Ophthalmol. 2007; 143: 463-472.e2
        • Garg A.
        • Oll M.
        • Yzer S.
        • et al.
        Reticular pseudodrusen in early age-related macular degeneration are associated with choroidal thinning.
        Invest Ophthalmol Vis Sci. 2013; 54: 7075-7081
        • Ho C.Y.
        • Lek J.J.
        • Aung K.Z.
        • et al.
        Relationship between reticular pseudodrusen and choroidal thickness in intermediate age-related macular degeneration.
        Clin Exp Ophthalmol. 2018; 46: 485-494
        • Cho H.J.
        • Yoo S.G.
        • Kim H.S.
        • et al.
        Risk factors for geographic atrophy after intravitreal ranibizumab injections for retinal angiomatous proliferation.
        Am J Ophthalmol. 2015; 159: 285-292.e1
        • Alagorie A.R.
        • Verma A.
        • Nassisi M.
        • et al.
        Quantitative assessment of choriocapillaris flow deficits in eyes with advanced age-related macular degeneration versus healthy eyes.
        Am J Ophthalmol. 2019; 205: 132-139
        • Moreira-Neto C.A.
        • Moult E.M.
        • Fujimoto J.G.
        • et al.
        Choriocapillaris loss in advanced age-related macular degeneration.
        J Ophthalmol. 2018; 2018: 8125267
        • Nassisi M.
        • Shi Y.
        • Fan W.
        • et al.
        Choriocapillaris impairment around the atrophic lesions in patients with geographic atrophy: a swept-source optical coherence tomography angiography study.
        Br J Ophthalmol. 2019; 103: 911-917
        • Thulliez M.
        • Zhang Q.
        • Shi Y.
        • et al.
        Correlations between choriocapillaris flow deficits around geographic atrophy and enlargement rates based on swept-source OCT imaging.
        Ophthalmol Retina. 2019; 3: 478-488
        • Nassisi M.
        • Baghdasaryan E.
        • Borrelli E.
        • et al.
        Choriocapillaris flow impairment surrounding geographic atrophy correlates with disease progression.
        PLoS One. 2019; 14e0212563
        • Arya M.
        • Sabrosa A.S.
        • Duker J.S.
        • et al.
        Choriocapillaris changes in dry age-related macular degeneration and geographic atrophy: a review.
        Eye Vis (Lond). 2018; 5: 22
        • Miki A.
        • Miki K.
        • Ueno S.
        • et al.
        Prolonged blockade of VEGF receptors does not damage retinal photoreceptors or ganglion cells.
        J Cell Physiol. 2010; 224: 262-272
        • Domalpally A.
        • Danis R.P.
        • Trane R.
        • et al.
        Atrophy in neovascular age-related macular degeneration: Age-Related Eye Disease Study 2 report number 15.
        Ophthalmol Retina. 2018; 2: 1021-1027