- Email: [email protected]
Natural History of Geographic Atrophy Progression Secondary to Age-Related Macular Degeneration (Geographic Atrophy Progression Study)
Natural History of Geographic Atrophy Progression Secondary to Age-Related Macular Degeneration (Geographic Atrophy Progression Study) Steffen Schmitz-Valckenberg, MD,1,7 José-Alain Sahel, MD,2 Ronald Danis, MD,3 Monika Fleckenstein, MD,1 Glenn J. Jaffe, MD,4 Sebastian Wolf, MD,5 Christian Pruente, MD,6 Frank G. Holz, MD1,7 Purpose: The Geographic Atrophy Progression (GAP) study was designed to assess the rate of geographic atrophy (GA) progression and to identify prognostic factors by measuring the enlargement of the atrophic lesions using fundus autofluorescence (FAF) and color fundus photography (CFP). Design: Prospective, multicenter, noninterventional natural history study. Participants: A total of 603 participants were enrolled in the study; 413 of those had gradable lesion data from FAF or CFP, and 321 had gradable lesion data from both FAF and CFP. Methods: Atrophic lesion areas were measured by FAF and CFP to assess lesion progression over time. Lesion size assessments and best-corrected visual acuity (BCVA) were conducted at screening/baseline (day 0) and at 3 follow-up visits: month 6, month 12, and month 18 (or early exit). Main Outcome Measures: The GA lesion progression rate in disease subgroups and mean change from baseline visual acuity. Results: Mean (standard error) lesion size changes from baseline, determined by FAF and CFP, respectively, were 0.88 (0.1) and 0.78 (0.1) mm2 at 6 months, 1.85 (0.1) and 1.57 (0.1) mm2 at 12 months, and 3.14 (0.4) and 3.17 (0.5) mm2 at 18 months. The mean change in lesion size from baseline to month 12 was significantly greater in participants who had eyes with multifocal atrophic spots compared with those with unifocal spots (P < 0.001) and those with extrafoveal lesions compared with those with foveal lesions (P ¼ 0.001). The mean (standard deviation) decrease in visual acuity was 6.2 15.6 letters for patients with image data available. Atrophic lesions with a diffuse (mean 0.95 mm2) or banded (mean 1.01 mm2) FAF pattern grew more rapidly by month 6 compared with those with the “none” (mean, 0.13 mm2) and focal (mean, 0.36 mm2) FAF patterns. Conclusions: Although differences were observed in mean lesion size measurements using FAF imaging compared with CFP, the measurements were highly correlated with one another. Significant differences were found in lesion progression rates in participants stratified by hyperfluorescence pattern subtype. This large GA natural history study provides a strong foundation for future clinical trials. Ophthalmology 2015;-:1e8 ª 2015 by the American Academy of Ophthalmology.
Age-related macular degeneration (AMD) is a multifactorial disease caused by both genetic and environmental factors. Geographic atrophy (GA) is a progressive form of dry AMD that is characterized by irreversible loss of macular retinal tissue, retinal pigment epithelium (RPE), and choriocapillaris. Geographic atrophy is a significant cause of central vision loss, which is irreversible and usually bilateral.1,2 Geographic atrophy is responsible for severe vision loss in approximately 20% of all patients with AMD, and more than 8 million people are affected worldwide.3 Oxidative stress, dysregulation of the complement system, and inflammation are thought to play pathophysiologic roles in the development and progression of AMD, although the relative contribution of each of these pathways and molecular mechanisms is not well established.4e7 Clinical
2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
presentation of GA includes variable lesion topography that typically enlarges over time.6,8 The lesions usually begin to appear in the extrafoveal area with expansion into the foveal center later in the disease course.9 These lesions lead to progressive degenerative changes in the corresponding RPE cell monolayer, inner choroid, and photoreceptors.4,8 There remains a limited understanding of the underlying mechanisms and natural history of GA lesion progression. In clinical studies, mean lesion progression rates vary widely among individuals, ranging from 1.2 to 2.8 mm2 per year.4 The size of the lesion, as well as the topography and number of lesions, may affect progression rates.4 Other risk factors identified include greater distance from the fovea, presence of epiretinal membrane, GA in the fellow eye, and treatment with anti-vascular endothelial growth
http://dx.doi.org/10.1016/j.ophtha.2015.09.036 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 factor medications.10 However, specific reasons for the differences in progression rates are not well understood. Inconsistencies in study design, including imaging technology used, clinical protocols, and follow-up times, may account for differences in lesion progression rates among studies.2,8 In previous investigations, color fundus photography (CFP) has been used to measure lesion progression.11 More recently, fundus autofluorescence (FAF) was used in a multicenter natural history study to follow GA progression (Fundus Autofluorescence Imaging in Age-Related Macular Degeneration [FAM] study).8,12 Fundus autofluorescence imaging of atrophic lesions in GA is primarily based on autofluorescence properties of RPE cells, with a marked reduction of the autofluorescence signal indicative of RPE loss due to the concomitant disappearance of intracellular dominant fluorophores.13 The majority of eyes with GA also show abnormal FAF hyperfluorescence patterns that have recently been classified as banded, patchy, focal, and diffuse.14 The Natural History of Geographic Atrophy Progression (GAP) study was designed to assess disease progression in participants with GA secondary to AMD, by serial measurement of lesion size using CFP and FAF. Lesion progression was also assessed in participants segregated into disease subtypes that included baseline lesion size, location, and distribution, as well as FAF hyperfluorescence pattern. Previous reports from the GAP study have focused on characteristics of reticular drusen and lesion topography.15e17 This report compares lesion progression between the 2 different imaging modalities and the relationship between changes in progression rate and changes in bestcorrected visual acuity (BCVA). The large data set obtained in this study provides additional insights into natural history of GA and can help determine anatomic and functional outcome measures relevant for future clinical trials.
Methods The GAP study (ClinicalTrials.gov identifier: NCT00599846) was a prospective, multicenter, noninterventional, observational study. It was originally designed to identify risk factors and to quantify atrophic lesion progression in participants with GA secondary to AMD. With the initiation of the Geographic Atrophy Treatment Evaluation (GATE) study, the GAP study was terminated, and participants were allowed to exit early from the GAP study and enroll in the interventional GATE study if inclusion criteria were met. Informed consent was obtained for all participants, and records were maintained in a Health Insurance Portability and Accountability Actecompliant manner. Institutional Review Board/Independent Ethics Committee approval was obtained, and the research was performed in compliance with the ethical principles of the Declaration of Helsinki and Good Clinical Practice. The study included participants who were 55 years of age or older and diagnosed with GA secondary to AMD in at least 1 eye, with no evidence of choroidal neovascularization (CNV) in either eye. To be eligible for enrollment, the study eye needed a welldemarcated area of GA with the following lesion subtype characteristics: For unifocal lesions, the lesion had to be 1.25 mm2 (0.5 disc areas [DA]) and 17.5 mm2 (7 DA). For multifocal lesions, 1 lesion had to be 1.25 mm2 (0.5 DA) and all lesions combined (the total lesion size) could not exceed 17.5 mm2
2
(7 DA). Participants had to have BCVA of 35 letters in the study eye (i.e., 20/200 Snellen equivalent) and drusen 63 mm and/or GA in the fellow eye. Exclusion criteria also included ocular diseases that would confound assessments of the retina (e.g., diabetic retinopathy, uveitis); cataract or ocular surgery within 90 days of baseline visit; or any systemic disease with limited survival prognosis (e.g., cancer). To ensure that the study population was representative of all eligible participants, no participant was excluded because of gender, race, occupation, or socioeconomic status. The number and timing of study visits were preset by the study protocol as follows: visit 1, baseline (day 0); visit 2, day 180; visit 3, day 360; visit 4, day 540/exit. There were no interim analyses planned or conducted during the course of the study. The BCVA was assessed on Early Treatment Diabetic Retinopathy Study charts at baseline and exit visits. Visual acuity data were expressed as the number of letters read on the Early Treatment Diabetic Retinopathy Study chart. The CFP and FAF images were collected at every visit, baseline, and every 6 months for up to 18 months. The CFP was performed with standard fundus cameras that had a minimum resolution of 2000 2000 pixels. Confocal scanning laser ophthalmoscopy FAF was performed with HRAc, HRA2, or Spectralis (Heidelberg Engineering, Heidelberg, Germany) using 488 nm blue light excitation. To minimize variability, study-site technicians and photographers were certified to perform the imaging procedures before any study eye image evaluation. The CFP and FAF images were transmitted to a central reading center, the Duke Reading Center, through a secure, web-based portal. Images were then assigned to trained Duke Reading Center or GRADE Reading Center readers who independently assessed the CFP and FAF images. The lesion progression rate was defined as the change in lesion size from baseline to months 6, 12, and 18. For FAF imaging, comparative grading using 2 computer screens was applied. That is, all confocal scanning laser ophthalmoscopy image data (including blue reflectance and infrared) were available for the analysis of each single visit (whereas CFP and fluorescein angiograms were not available). For each visit, the status of the fovea with regard to any atrophy involvement within a circle of 300 mm in diameter centered on the fovea was classified as “foveal” or “extrafoveal” GA. For each visit, the total size of atrophy was measured by a semiautomatic procedure, which has been described in detail.18,19 Briefly, the reader manually set a seeding point inside the atrophic region to start an automatic region-growing algorithm that detected well-demarcated areas of severely decreased FAF signal. The reader then manually adjusted the threshold of the algorithm. Holes within the detected GA area could be identified, and further GA areas in the same image could be integrated. Furthermore, a second “blood vessel detection” algorithm was used to exclude interfering blood vessels that had intensities similar to those of atrophic areas. In addition, a shadow correction tool was used when there was uneven illumination, and constraints were placed to improve lesion boundary discrimination of atrophic patches. The minimum size of atrophic areas was predefined as 0.05 mm2. When there was confluent peripapillary atrophy in addition to central atrophy, a line constraint tool was used to draw a vertical line at the most narrow part (the “bridge”) of the confluent atrophy. Any atrophy nasal to this line was disregarded for atrophy quantification. For scaling, the individual scaling factor that is registered by the Heidelberg Eye Explorer during image acquisition was used. All follow-up images were aligned to the baseline image, and the scaling factor of the baseline image was used to correct for variable focusing at different study visits.19 This study did not investigate a drug, product, or medical device as defined by the Food Drug and Cosmetic Act. Accordingly, it was not necessary to collect adverse event information. The majority of participants (86%, 317/368) did not complete the study, because the study was terminated when sufficient data were
Schmitz-Valckenberg et al
Geographic Atrophy Natural History Study
Table 1. Demographic and Ocular Baseline Characteristics of the Study Population Characteristics
Participants (No.)
Age (yrs) Visual acuity* (ETDRS letters)
413 405
Mean (SD) 76.9 (7.7) 61.4 (13.9)
Gender
Participants (No.)
Percentage (%)
Male Female
168 245
40.7 59.3
Participants (No.)
Percentage (%)
411 1 1
99.5 0.2 0.2
Participants (No.)
Percentage (%)
114 46 58 162 33
27.6 11.1 14.0 39.2 8.0
Race White Othery Multiracialz Iris color Brown Hazel Green Blue Grey
ETDRS ¼ Early Treatment Diabetic Retinopathy Study; SD ¼ standard deviation. *Eight with no BCVA measurement. y North African. z Black or African American and American Indian.
Table 2. Baseline Lesion Characteristics by Eye Involvement, Localization, and Focality Disease/Lesion Subtypes Total Unilateral Bilateral Foveal Extrafoveal Not gradable Unifocal Multifocal
Participants (No.)
%
413 24 389 63 157 193 106 307
100.0 5.8 94.2 15.3 38.0 46.7 25.7 74.3
imaging. Secondary and additional end points included assessment of GA using CFP, lesion size in different disease subtypes, and mean change from baseline BCVA. SAS software (SAS Inc, Cary, NC) was used to analyze data. Participant demographics and baseline characteristics, as well as lesion progression, were summarized with descriptive statistics (N, mean, standard deviation, median, minimum, and maximum). The GA lesion progression rates, as measured by FAF and CFP, were compared to determine whether there was a correlation between the 2 measurement methods. Exploratory analyses included descriptive assessments of the relationship between the lesion progression rate and the disease subgroups, as determined by FAF.
Results obtained to meet the planned objectives. Active participants were subsequently screened for enrollment into an Alcon-sponsored study (GATE), which tested an investigational drug to treat GA secondary to AMD (GATE Study; ClinicalTrials.gov identifier: NCT00890097).
Statistical Analysis The primary study end point was the mean enlargement rate of the atrophic lesion from baseline to month 12 as assessed by FAF
Figure 1. Comparison of mean lesion size for participants with fundus autofluorescence (FAF) and color fundus photography (CFP) grading data. Mean lesion size (mm2) as assessed by FAF (solid circles) and CFP (open circles). Participant data shown represent the group of participants who had both FAF and CFP images available (N ¼ 321, baseline; N ¼ 237, month 6; N ¼ 115, month 12; N ¼ 30, month 18). P values based on paired t test, FAF versus CFP. Error bars represent standard error (SE).
A total of 603 participants were enrolled; 413 had gradable lesion data from FAF or CFP, of whom 380 were evaluable for the per protocol analysis set, and 321 had gradable lesion data from both FAF and CFP. Forty participants (10.5%) in the per protocol set completed the study. There were 237, 115, and 30 participants at months 6, 12, and 18, respectively, with both FAF and CFP grading data. The majority of participants were white (99.5%) and female (59.3%), and the mean age (standard deviation) was 76.9 (7.7) years. A total of 39.2% had blue eyes. The mean baseline BCVA was 61.4 13.9 letters (Table 1). Of eyes with gradable imaging available, 94.2% had bilateral atrophic lesions. On FAF imaging, 38.0% of the lesions were extrafoveal and 74.3% were multifocal. The majority (55.9%) of the lesions were composed of 2 to 7 distinct atrophic areas, and 18.4% had 8 or more distinct atrophic areas (Table 2). Measured lesion sizes were generally >1.25 mm2 and 12.5 mm2, regardless of the imaging system used. Mean lesion sizes obtained by FAF imaging were significantly smaller than those obtained by CFP at baseline (P < 0.001) and at months 6 (P < 0.001), 12 (P < 0.001), and 18 (P ¼ 0.007) (Fig 1). The baseline mean lesion size measured by FAF, 7.0 0.3 mm2, was significantly less than the mean lesion size measured by CFP, 8.4 0.3 mm2 (P < 0.001). Despite this discrepancy, lesion size measurements assessed using FAF and CFP were highly correlated with one another at baseline (r ¼ 0.90), month 6 (r ¼ 0.89), and month 12 (r ¼ 0.94). The lesion size change from baseline as determined by both imaging modalities was similar: 0.88 0.1 and 0.78 0.1 mm2 at 6 months (P ¼ 0.164), 1.85 0.1 and 1.57 0.1 mm2 at 12 months (P ¼ 0.023), and 3.14 0.4 and 3.17 0.5 mm2 at 18 months (P ¼ 0.944) for FAF and CFP, respectively (Table 3). Baseline lesion size was associated with lesion progression rate; larger lesions had greater area changes over time (Fig 2). An example of lesion area progression for multiple FAF patterns is shown in Figure 3.
3
Ophthalmology Volume -, Number -, Month 2015 Table 3. Lesion Size Change from Baseline, by Imaging Modality and Lesion Subtype
Disease/Lesion Subtypes Total, by CFP Total, by FAF P value* Unilateral Bilateral P valuey Foveal Extrafoveal P valuez Unifocal Multifocal P valuex
Lesion Size Change from Baseline, mm2 Mean (SE) Month 6 0.78 (0.1) 0.88 (0.1) 0.164 0.49 (0.2) 0.88 (0.1) 0.084 0.65 (0.1) 0.99 (0.1) 0.011 0.47 (0.1) 1.00 (0.1) <0.001
Month 12 1.57 (0.1) 1.85 (0.1) 0.023 0.56 (0.3) 1.82 (0.1) 0.014 1.28 (0.2) 2.05 (0.2) 0.001 1.05 (0.2) 1.97 (0.1) <0.001
Month 18 3.17 (0.5) 3.14 (0.4) 0.944 NA NA NA NA NA NA
CFP ¼ color fundus photography; FAF ¼ fundus autofluorescence; NA ¼ not available; SE ¼ standard error. *Based on paired t test, FAF versus CFP. y Based on paired t test, unilateral versus bilateral. z Based on paired t test, foveal versus extrafoveal. x Based on paired t test, unifocal versus multifocal.
Lesion progression rates, as determined on FAF images, were then stratified by lesion subtype, foveal versus extrafoveal and unifocal versus multifocal. Atrophy progression was significantly greater among eyes with extrafoveal lesions compared with eyes with foveal lesions at month 6 (extrafoveal 0.99 0.1 vs. foveal 0.65 0.1, P ¼ 0.011) and month 12 (extrafoveal 2.05 0.2 vs. foveal 1.28 0.2, P ¼ 0.001) (Fig 4, Table 3). Atrophy progression was also significantly greater among eyes with
multifocal lesions compared with eyes with unifocal lesions at month 6 (multifocal 1.00 0.1 vs. unifocal 0.47 0.1, P < 0.001) and month 12 (multifocal 1.97 0.1 vs. unifocal 1.05 0.2, P < 0.001) (Fig 4, Table 3). To further differentiate lesion progression rates by lesion subtype, the lesion progression rate in eyes with different hyperfluorescence pattern subtypes at baseline, as determined by FAF imaging, was assessed. The mean (standard error) lesion size change at month 6 and month 12 was 0.13 (0.2) mm2 and 1.00 (0.4) mm2 for the “no abnormal” pattern (n ¼ 21), 0.36 (0.1) mm2 and 0.90 (0.4) mm2 for the focal pattern (n ¼ 26), 1.01 (0.1) mm2 and 1.95 (0.5) mm2 for the banded pattern (n ¼ 30), 0.41 (0.2) mm2 and 1.46 (0.3) mm2 for the patchy pattern (n ¼ 11), and 0.95 (0.2) mm2 and 1.91 (0.1) mm2 for the diffuse pattern (n ¼ 321), respectively. The progression rates at month 6 of the banded (P < 0.001) and the diffuse (P < 0.001) groups were markedly higher compared with the “no abnormal,” focal, and patchy groups (Fig 5). At month 12, the banded and diffuse groups continued to have the largest mean lesion size change among the 5 subgroups, but did not reach statistical significance (P ¼ 0.196 and P ¼ 0.082, respectively). Of the 413 evaluable participants, 343 had visual acuity data at both baseline and exit visit (all time points). For all participants with image data available, the mean decrease in visual acuity was 6.2 15.6 letters from baseline for participants with 12 months of follow-up data (n ¼ 126). Although involvement of the fovea by the atrophic lesion process during progression can affect BCVA, there was no correlation between BCVA and lesion localization (foveal, r ¼ 0.001; extrafoveal, r ¼ 0.05), and lesion size (r ¼ 0.06), or lesion progression rate (r ¼ 0.19) (Fig 6).
Discussion In the present study, we found that lesion size measured by CFP was greater on average than lesion size measured by
Figure 2. Mean change in lesion size at months 6 and 12 by baseline size categories using fundus autofluorescence (FAF) imaging. Baseline lesion categories versus mean change in lesion size (mm2) at months 6 and 12, as assessed by FAF. Graph bar numbers represent N values; error bars represent standard error (SE).
4
Schmitz-Valckenberg et al
Geographic Atrophy Natural History Study
Figure 3. Geographic atrophy (GA) lesion progression for different fundus autofluorescence (FAF) patterns. Representative examples of 4 major FAF patterns: (A) none, (B) focal, (C) banded, and (D) diffuse. For each example, the baseline (first row), month 6 (second row), month 12 (third row), and exit (fourth row) visits are shown, both with the native and the processed images (atrophic lesions outlined by white color). E, The corresponding color fundus camera photograph for each pattern at baseline.
FAF at each time point. However, the lesion progression rate was similar for the 2 imaging modalities, and depended on baseline lesion size, location, and pattern.
Figure 4. Mean lesion size change from baseline by type of atrophic spots and lesion localization assessed using fundus autofluorescence (FAF) imaging. Mean lesion size change based on type of atrophic spots (left, unifocal vs. multifocal) and localization of lesion (right, foveal vs. extrafoveal) as assessed by FAF. P values based on 2-sample t test, unifocal versus multifocal, and foveal versus extrafoveal, at months 6 and 12. Graph bar numbers represent N values; error bars represent standard error (SE).
Variation by imaging modality may be due to differences in the image resolution or data processing.20 The latter possibility is unlikely, because the image reading methods were consistent on each image evaluated. Lesion clarity and contrast on images obtained with FAF were generally better than those seen on CFP. For eyes with multifocal lesions, if one assumes that smaller lesions were detected more readily with FAF than with CFP, then one may have anticipated that total lesion size, which included all lesions, large and small, would have been larger by FAF than by CFP. Conversely, for unifocal lesions, if a higher proportion of small lesions were detected on FAF, then the average lesion size measured on FAF would have been smaller than that measured on CFP. In addition, if depigmented but nonatrophic regions seen on CFP were erroneously identified as atrophic lesions, then lesion size measured on CFP would have been larger than that on FAF. The specific reasons for the size discrepancy on the different modalities remain unknown. We found that despite differences in lesion size, the progression rates measured by CFP and FAF were highly correlated with one another. Fundus autofluorescence may have several advantages over CFP, particularly because the better contrast of FAF images permits more accurate lesion boundary identification, as well as the ability to identify
5
Ophthalmology Volume -, Number -, Month 2015
Figure 5. Mean lesion size from baseline at months 6 and 12 by perifoveal hyperfluorescence patterns assessed using fundus autofluorescence (FAF) imaging. Size reported as mm2. P values based on 2-sample t test, compared with the “no abnormal” group. Diffuse pattern ¼ presence of reticular, branching, fine granular-grainy, dusty, heterogeneous, fine granular with peripheral punctuated spots, or trickling patterns. Graph bar numbers represent N values; error bars represent standard error (SE).
specific lesion autofluorescence that predicts lesion progression rate over time. Nevertheless, we believe that the most accurate assessment is achieved by reference to at least 1 additional image type when grading FAF images (e.g., infrared or CFP), as has been suggested by Sunness et al11 and other investigators.21 The mean lesion change from baseline to month 12, measured on FAF images, was 1.85 mm2. This progression rate is consistent with results from Holz et al in the FAM study.8 In contrast, the mean lesion change from baseline to month 12 was 1.57 mm2 as determined by CFP. The CFP
progression rate was less than the rate reported by Sunness et al,11 possibly because of differences in baseline lesion size and imaging capabilities as described previously. Changes in BCVA were not correlated with baseline lesion area, lesion progression rate, or lesion localization (foveal or extrafoveal). Depending on the clinical study, progression rates vary widely among patients and range from 1.2 to 2.8 mm2 per year.4 The mechanisms responsible for the differences in progression rates among studies are not fully understood and may include different baseline lesion sizes and locations, different trial protocols, imaging systems, and varying genotype or exposure to environmental factors. At month 12, lesion progression rates were significantly greater in extrafoveal and multifocal lesions compared with foveal and unifocal lesions. The findings in this study regarding the distribution of FAF patterns and the lesion progression rates between different patterns were in accordance with those previously reported in the FAM study and represent risk factors predictive of lesion progression.8,9
Study Strengths and Limitations The GAP study was a prospective, multicenter, 18-month, noninterventional, natural history study of participants with GA secondary to AMD. Strengths of this study include standardized follow-up procedures and standardized central reading center image assessment to minimize variation. Participants were instructed to return to the site for repeat autofluorescence and fundus photographic lesion imaging if it was not possible to obtain gradable images taken at the baseline or follow-up visits. The quantification of lesion subtypes by 2 imaging modalities and correlation with
Figure 6. Visual acuity changes from baseline at month 12 based on lesion size, localization, and change. Best-corrected visual acuity (BCVA) change from baseline versus (A) lesion size at baseline, (B) lesion localization (circle, foveal; cross, extrafoveal), or (C) change in lesion area from baseline, in participants with 12 (2) months of follow-up. FAF ¼ fundus autofluorescence.
6
Schmitz-Valckenberg et al
Geographic Atrophy Natural History Study
changes in BCVA have provided new insights into the natural history of GA and was another study strength. A study limitation was that some of the participants did not have complete follow-up, because they exited to enroll in an interventional trial for GA due to AMD (GATE study). Therefore, the number of participants with gradable image data decreased at each postbaseline study visit, primarily because of study termination. Many participants were excluded because of lack of follow-up FAF images, which caused exclusion because of insufficient points on FAF image analysis. Participants also were excluded if images did not fulfill all predefined lesion criteria for GA according to the reading center assessment. In addition, spectral-domain optical coherence tomography was not available when this study was initiated. In future studies, spectral-domain optical coherence tomography and other technologies could provide further lesion imaging and measurement capabilities.22,23
10. 11.
12. 13.
14.
Acknowledgment. Meridius Health Communications, Inc, provided technical support in preparation of the manuscript. 15.
References 16. 1. Klein R, Meuer SM, Knudtson MD, Klein BE. The epidemiology of progression of pure geographic atrophy: the Beaver Dam Eye Study. Am J Ophthalmol 2008;146:692–9. 2. Sunness JS, Gonzalez-Baron J, Applegate CA, et al. Enlargement of atrophy and visual acuity loss in the geographic atrophy form of age-related macular degeneration. Ophthalmology 1999;106:1768–79. 3. Khan M, Agarwal K, Loutfi M, Kamal A. Present and possible therapies for age-related macular degeneration. ISRN Ophthalmol 2014;2014:608390. 4. Holz FG, Strauss EC, Schmitz-Valckenberg S, van Lookeren Campagne M. Geographic atrophy: clinical features and potential therapeutic approaches. Ophthalmology 2014;121:1079–91. 5. Rudnicka AR, Jarrar Z, Wormald R, et al. Age and gender variations in age-related macular degeneration prevalence in populations of European ancestry: a meta-analysis. Ophthalmology 2012;119:571–80. 6. Clemons TE, Milton RC, Klein R, et al. Risk factors for the incidence of Advanced Age-Related Macular Degeneration in the Age-Related Eye Disease Study (AREDS) AREDS report no. 19. Ophthalmology 2005;112:533–9. 7. Zarbin MA, Rosenfeld PJ. Pathway-based therapies for agerelated macular degeneration: an integrated survey of emerging treatment alternatives. Retina 2010;30:1350–67. 8. Holz FG, 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;143:463–72. 9. Sunness JS, Rubin GS, Applegate CA, et al. Visual function abnormalities and prognosis in eyes with age-related
17.
18.
19. 20.
21.
22.
23.
geographic atrophy of the macula and good visual acuity. Ophthalmology 1997;104:1677–91. Grunwald JE, Pistilli M, Ying GS, et al. Growth of geographic atrophy in the comparison of age-related macular degeneration treatments trials. Ophthalmology 2015;122:809–16. Sunness JS, Margalit E, Srikumaran D, et al. The long-term natural history of geographic atrophy from age-related macular degeneration: enlargement of atrophy and implications for interventional clinical trials. Ophthalmology 2007;114:271–7. Mata NL, Lichter JB, Vogel R, et al. Investigation of oral fenretinide for treatment of geographic atrophy in age-related macular degeneration. Retina 2013;33:498–507. Vaclavik V, Vujosevic S, Dandekar SS, et al. Autofluorescence imaging in age-related macular degeneration complicated by choroidal neovascularization: a prospective study. Ophthalmology 2008;115:342–6. Bindewald A, Schmitz-Valckenberg S, Jorzik JJ, et al. Classification of abnormal fundus autofluorescence patterns in the junctional zone of geographic atrophy in patients with age related macular degeneration. Br J Ophthalmol 2005;89: 874–8. Mauschitz MM, Fonseca S, Chang P, et al. Topography of geographic atrophy in age-related macular degeneration. Invest Ophthalmol Vis Sci 2012;53:4932–9. Schmitz-Valckenberg S, Alten F, Steinberg JS, et al. Reticular drusen associated with geographic atrophy in age-related macular degeneration. Invest Ophthalmol Vis Sci 2011;52: 5009–15. Steinberg JS, Auge J, Jaffe GJ, et al. Longitudinal analysis of reticular drusen associated with geographic atrophy in agerelated macular degeneration. Invest Ophthalmol Vis Sci 2013;54:4054–60. Deckert A, Schmitz-Valckenberg S, Jorzik J, et al. Automated analysis of digital fundus autofluorescence images of geographic atrophy in advanced age-related macular degeneration using confocal scanning laser ophthalmoscopy (cSLO). BMC Ophthalmol 2005;5:8. Fleckenstein M, Schmitz-Valckenberg S, Adrion C, et al. Progression of age-related geographic atrophy: role of the fellow eye. Invest Ophthalmol Vis Sci 2011;52:6552–7. Khanifar AA, Lederer DE, Ghodasra JH, et al. Comparison of color fundus photographs and fundus autofluorescence images in measuring geographic atrophy area. Retina 2012;32: 1884–91. Lindner M, Boker A, Mauschitz MM, et al. Directional kinetics of geographic atrophy progression in age-related macular degeneration with foveal sparing. Ophthalmology 2015;122:1356–65. Wolf-Schnurrbusch UE, Wittwer VV, Ghanem R, et al. Bluelight versus green-light autofluorescence: lesion size of areas of geographic atrophy. Invest Ophthalmol Vis Sci 2011;52: 9497–502. Gocho K, Sarda V, Falah S, et al. Adaptive optics imaging of geographic atrophy. Invest Ophthalmol Vis Sci 2013;54: 3673–80.
Footnotes and Financial Disclosures Originally received: May 14, 2015. Final revision: August 18, 2015. Accepted: September 25, 2015. Available online: ---. 1
2
Université Pierre et Marie Curie and Institut de la Vision, Paris, France.
3
Manuscript no. 2015-780.
Department of Ophthalmology, University of Bonn, Bonn, Germany.
Department of Ophthalmology and Visual Sciences, University of Wisconsin, Madison, Wisconsin.
4
Department of Ophthalmology, Duke University, Durham, North Carolina.
7
Ophthalmology Volume -, Number -, Month 2015 5
Department of Ophthalmology, University Hospital Bern, Inselspital, University Bern, Bern, Switzerland.
Author Contributions:
6
Department of Ophthalmology, Kantonsspital Baselland, Liestal, Switzerland.
Data collection: Schmitz-Valckenberg, Sahel, Danis, Fleckenstein, Jaffe, Wolf, Pruente, Holz
7
Analysis and interpretation: Schmitz-Valckenberg, Sahel, Danis, Fleckenstein, Jaffe, Wolf, Pruente, Holz
GRADE Reading Center, University of Bonn, Bonn, Germany.
Presented at: The Association for Research in Vision and Ophthalmology Annual Meeting, May 3e9, 2009, Fort Lauderdale, Florida; and The Association for Research in Vision and Ophthalmology Annual Meeting, May 2e6, 2010, Fort Lauderdale, Florida. Financial Disclosure(s): The author(s) have made the following disclosure(s): J.-A.S.: Founder and consultant Pixium Vision and GenSight Biologics; Consultant SanofiFovea, Gene Signal, and Vision Medicines, Inc. S.S-V.: Consultant Alcon, Novartis. R.D.: Consultant GlaxoSmithKline. G.J.J.: Consultant Heidelberg Engineering. S.W.: Consultant Alcon, Allergan, Bayer Healthcare, Heidelberg Engineering, Novartis, Roche, and Zeiss. F.G.H.: Consultant Alcon, Allergan, Bayer Healthcare, Heidelberg Engineering, Genentech, Novartis, and Roche.
8
Conception and design: Schmitz-Valckenberg, Pruente, Holz
Obtained funding: Not applicable Overall responsibility: Schmitz-Valckenberg, Sahel, Danis, Fleckenstein, Jaffe, Wolf, Pruente, Holz Abbreviations and Acronyms: AMD ¼ age-related macular degeneration; BCVA ¼ best-corrected visual acuity; CFP ¼ color fundus photography; DA ¼ disc areas; FAF ¼ fundus autofluorescence; FAM ¼ Fundus Autofluorescence Imaging in AgeRelated Macular Degeneration; GA ¼ geographic atrophy; GAP ¼ Geographic Atrophy Progression; GATE ¼ Geographic Atrophy Treatment Evaluation; RPE ¼ retinal pigment epithelium; SE ¼ standard error. Correspondence: Frank G. Holz, MD, Department of Ophthalmology, University of Bonn, Ernst-Abbe-Str. 2, D-53127 Bonn, Germany. E-mail: Frank.Holz@ukb. uni-bonn.de.
Age-related macular degeneration (AMD) is a multifactorial disease caused by both genetic and environmental factors. Geographic atrophy (GA) is a progressive form of dry AMD that is characterized by irreversible loss of macular retinal tissue, retinal pigment epithelium (RPE), and choriocapillaris. Geographic atrophy is a significant cause of central vision loss, which is irreversible and usually bilateral.1,2 Geographic atrophy is responsible for severe vision loss in approximately 20% of all patients with AMD, and more than 8 million people are affected worldwide.3 Oxidative stress, dysregulation of the complement system, and inflammation are thought to play pathophysiologic roles in the development and progression of AMD, although the relative contribution of each of these pathways and molecular mechanisms is not well established.4e7 Clinical
2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
presentation of GA includes variable lesion topography that typically enlarges over time.6,8 The lesions usually begin to appear in the extrafoveal area with expansion into the foveal center later in the disease course.9 These lesions lead to progressive degenerative changes in the corresponding RPE cell monolayer, inner choroid, and photoreceptors.4,8 There remains a limited understanding of the underlying mechanisms and natural history of GA lesion progression. In clinical studies, mean lesion progression rates vary widely among individuals, ranging from 1.2 to 2.8 mm2 per year.4 The size of the lesion, as well as the topography and number of lesions, may affect progression rates.4 Other risk factors identified include greater distance from the fovea, presence of epiretinal membrane, GA in the fellow eye, and treatment with anti-vascular endothelial growth
http://dx.doi.org/10.1016/j.ophtha.2015.09.036 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 factor medications.10 However, specific reasons for the differences in progression rates are not well understood. Inconsistencies in study design, including imaging technology used, clinical protocols, and follow-up times, may account for differences in lesion progression rates among studies.2,8 In previous investigations, color fundus photography (CFP) has been used to measure lesion progression.11 More recently, fundus autofluorescence (FAF) was used in a multicenter natural history study to follow GA progression (Fundus Autofluorescence Imaging in Age-Related Macular Degeneration [FAM] study).8,12 Fundus autofluorescence imaging of atrophic lesions in GA is primarily based on autofluorescence properties of RPE cells, with a marked reduction of the autofluorescence signal indicative of RPE loss due to the concomitant disappearance of intracellular dominant fluorophores.13 The majority of eyes with GA also show abnormal FAF hyperfluorescence patterns that have recently been classified as banded, patchy, focal, and diffuse.14 The Natural History of Geographic Atrophy Progression (GAP) study was designed to assess disease progression in participants with GA secondary to AMD, by serial measurement of lesion size using CFP and FAF. Lesion progression was also assessed in participants segregated into disease subtypes that included baseline lesion size, location, and distribution, as well as FAF hyperfluorescence pattern. Previous reports from the GAP study have focused on characteristics of reticular drusen and lesion topography.15e17 This report compares lesion progression between the 2 different imaging modalities and the relationship between changes in progression rate and changes in bestcorrected visual acuity (BCVA). The large data set obtained in this study provides additional insights into natural history of GA and can help determine anatomic and functional outcome measures relevant for future clinical trials.
Methods The GAP study (ClinicalTrials.gov identifier: NCT00599846) was a prospective, multicenter, noninterventional, observational study. It was originally designed to identify risk factors and to quantify atrophic lesion progression in participants with GA secondary to AMD. With the initiation of the Geographic Atrophy Treatment Evaluation (GATE) study, the GAP study was terminated, and participants were allowed to exit early from the GAP study and enroll in the interventional GATE study if inclusion criteria were met. Informed consent was obtained for all participants, and records were maintained in a Health Insurance Portability and Accountability Actecompliant manner. Institutional Review Board/Independent Ethics Committee approval was obtained, and the research was performed in compliance with the ethical principles of the Declaration of Helsinki and Good Clinical Practice. The study included participants who were 55 years of age or older and diagnosed with GA secondary to AMD in at least 1 eye, with no evidence of choroidal neovascularization (CNV) in either eye. To be eligible for enrollment, the study eye needed a welldemarcated area of GA with the following lesion subtype characteristics: For unifocal lesions, the lesion had to be 1.25 mm2 (0.5 disc areas [DA]) and 17.5 mm2 (7 DA). For multifocal lesions, 1 lesion had to be 1.25 mm2 (0.5 DA) and all lesions combined (the total lesion size) could not exceed 17.5 mm2
2
(7 DA). Participants had to have BCVA of 35 letters in the study eye (i.e., 20/200 Snellen equivalent) and drusen 63 mm and/or GA in the fellow eye. Exclusion criteria also included ocular diseases that would confound assessments of the retina (e.g., diabetic retinopathy, uveitis); cataract or ocular surgery within 90 days of baseline visit; or any systemic disease with limited survival prognosis (e.g., cancer). To ensure that the study population was representative of all eligible participants, no participant was excluded because of gender, race, occupation, or socioeconomic status. The number and timing of study visits were preset by the study protocol as follows: visit 1, baseline (day 0); visit 2, day 180; visit 3, day 360; visit 4, day 540/exit. There were no interim analyses planned or conducted during the course of the study. The BCVA was assessed on Early Treatment Diabetic Retinopathy Study charts at baseline and exit visits. Visual acuity data were expressed as the number of letters read on the Early Treatment Diabetic Retinopathy Study chart. The CFP and FAF images were collected at every visit, baseline, and every 6 months for up to 18 months. The CFP was performed with standard fundus cameras that had a minimum resolution of 2000 2000 pixels. Confocal scanning laser ophthalmoscopy FAF was performed with HRAc, HRA2, or Spectralis (Heidelberg Engineering, Heidelberg, Germany) using 488 nm blue light excitation. To minimize variability, study-site technicians and photographers were certified to perform the imaging procedures before any study eye image evaluation. The CFP and FAF images were transmitted to a central reading center, the Duke Reading Center, through a secure, web-based portal. Images were then assigned to trained Duke Reading Center or GRADE Reading Center readers who independently assessed the CFP and FAF images. The lesion progression rate was defined as the change in lesion size from baseline to months 6, 12, and 18. For FAF imaging, comparative grading using 2 computer screens was applied. That is, all confocal scanning laser ophthalmoscopy image data (including blue reflectance and infrared) were available for the analysis of each single visit (whereas CFP and fluorescein angiograms were not available). For each visit, the status of the fovea with regard to any atrophy involvement within a circle of 300 mm in diameter centered on the fovea was classified as “foveal” or “extrafoveal” GA. For each visit, the total size of atrophy was measured by a semiautomatic procedure, which has been described in detail.18,19 Briefly, the reader manually set a seeding point inside the atrophic region to start an automatic region-growing algorithm that detected well-demarcated areas of severely decreased FAF signal. The reader then manually adjusted the threshold of the algorithm. Holes within the detected GA area could be identified, and further GA areas in the same image could be integrated. Furthermore, a second “blood vessel detection” algorithm was used to exclude interfering blood vessels that had intensities similar to those of atrophic areas. In addition, a shadow correction tool was used when there was uneven illumination, and constraints were placed to improve lesion boundary discrimination of atrophic patches. The minimum size of atrophic areas was predefined as 0.05 mm2. When there was confluent peripapillary atrophy in addition to central atrophy, a line constraint tool was used to draw a vertical line at the most narrow part (the “bridge”) of the confluent atrophy. Any atrophy nasal to this line was disregarded for atrophy quantification. For scaling, the individual scaling factor that is registered by the Heidelberg Eye Explorer during image acquisition was used. All follow-up images were aligned to the baseline image, and the scaling factor of the baseline image was used to correct for variable focusing at different study visits.19 This study did not investigate a drug, product, or medical device as defined by the Food Drug and Cosmetic Act. Accordingly, it was not necessary to collect adverse event information. The majority of participants (86%, 317/368) did not complete the study, because the study was terminated when sufficient data were
Schmitz-Valckenberg et al
Geographic Atrophy Natural History Study
Table 1. Demographic and Ocular Baseline Characteristics of the Study Population Characteristics
Participants (No.)
Age (yrs) Visual acuity* (ETDRS letters)
413 405
Mean (SD) 76.9 (7.7) 61.4 (13.9)
Gender
Participants (No.)
Percentage (%)
Male Female
168 245
40.7 59.3
Participants (No.)
Percentage (%)
411 1 1
99.5 0.2 0.2
Participants (No.)
Percentage (%)
114 46 58 162 33
27.6 11.1 14.0 39.2 8.0
Race White Othery Multiracialz Iris color Brown Hazel Green Blue Grey
ETDRS ¼ Early Treatment Diabetic Retinopathy Study; SD ¼ standard deviation. *Eight with no BCVA measurement. y North African. z Black or African American and American Indian.
Table 2. Baseline Lesion Characteristics by Eye Involvement, Localization, and Focality Disease/Lesion Subtypes Total Unilateral Bilateral Foveal Extrafoveal Not gradable Unifocal Multifocal
Participants (No.)
%
413 24 389 63 157 193 106 307
100.0 5.8 94.2 15.3 38.0 46.7 25.7 74.3
imaging. Secondary and additional end points included assessment of GA using CFP, lesion size in different disease subtypes, and mean change from baseline BCVA. SAS software (SAS Inc, Cary, NC) was used to analyze data. Participant demographics and baseline characteristics, as well as lesion progression, were summarized with descriptive statistics (N, mean, standard deviation, median, minimum, and maximum). The GA lesion progression rates, as measured by FAF and CFP, were compared to determine whether there was a correlation between the 2 measurement methods. Exploratory analyses included descriptive assessments of the relationship between the lesion progression rate and the disease subgroups, as determined by FAF.
Results obtained to meet the planned objectives. Active participants were subsequently screened for enrollment into an Alcon-sponsored study (GATE), which tested an investigational drug to treat GA secondary to AMD (GATE Study; ClinicalTrials.gov identifier: NCT00890097).
Statistical Analysis The primary study end point was the mean enlargement rate of the atrophic lesion from baseline to month 12 as assessed by FAF
Figure 1. Comparison of mean lesion size for participants with fundus autofluorescence (FAF) and color fundus photography (CFP) grading data. Mean lesion size (mm2) as assessed by FAF (solid circles) and CFP (open circles). Participant data shown represent the group of participants who had both FAF and CFP images available (N ¼ 321, baseline; N ¼ 237, month 6; N ¼ 115, month 12; N ¼ 30, month 18). P values based on paired t test, FAF versus CFP. Error bars represent standard error (SE).
A total of 603 participants were enrolled; 413 had gradable lesion data from FAF or CFP, of whom 380 were evaluable for the per protocol analysis set, and 321 had gradable lesion data from both FAF and CFP. Forty participants (10.5%) in the per protocol set completed the study. There were 237, 115, and 30 participants at months 6, 12, and 18, respectively, with both FAF and CFP grading data. The majority of participants were white (99.5%) and female (59.3%), and the mean age (standard deviation) was 76.9 (7.7) years. A total of 39.2% had blue eyes. The mean baseline BCVA was 61.4 13.9 letters (Table 1). Of eyes with gradable imaging available, 94.2% had bilateral atrophic lesions. On FAF imaging, 38.0% of the lesions were extrafoveal and 74.3% were multifocal. The majority (55.9%) of the lesions were composed of 2 to 7 distinct atrophic areas, and 18.4% had 8 or more distinct atrophic areas (Table 2). Measured lesion sizes were generally >1.25 mm2 and 12.5 mm2, regardless of the imaging system used. Mean lesion sizes obtained by FAF imaging were significantly smaller than those obtained by CFP at baseline (P < 0.001) and at months 6 (P < 0.001), 12 (P < 0.001), and 18 (P ¼ 0.007) (Fig 1). The baseline mean lesion size measured by FAF, 7.0 0.3 mm2, was significantly less than the mean lesion size measured by CFP, 8.4 0.3 mm2 (P < 0.001). Despite this discrepancy, lesion size measurements assessed using FAF and CFP were highly correlated with one another at baseline (r ¼ 0.90), month 6 (r ¼ 0.89), and month 12 (r ¼ 0.94). The lesion size change from baseline as determined by both imaging modalities was similar: 0.88 0.1 and 0.78 0.1 mm2 at 6 months (P ¼ 0.164), 1.85 0.1 and 1.57 0.1 mm2 at 12 months (P ¼ 0.023), and 3.14 0.4 and 3.17 0.5 mm2 at 18 months (P ¼ 0.944) for FAF and CFP, respectively (Table 3). Baseline lesion size was associated with lesion progression rate; larger lesions had greater area changes over time (Fig 2). An example of lesion area progression for multiple FAF patterns is shown in Figure 3.
3
Ophthalmology Volume -, Number -, Month 2015 Table 3. Lesion Size Change from Baseline, by Imaging Modality and Lesion Subtype
Disease/Lesion Subtypes Total, by CFP Total, by FAF P value* Unilateral Bilateral P valuey Foveal Extrafoveal P valuez Unifocal Multifocal P valuex
Lesion Size Change from Baseline, mm2 Mean (SE) Month 6 0.78 (0.1) 0.88 (0.1) 0.164 0.49 (0.2) 0.88 (0.1) 0.084 0.65 (0.1) 0.99 (0.1) 0.011 0.47 (0.1) 1.00 (0.1) <0.001
Month 12 1.57 (0.1) 1.85 (0.1) 0.023 0.56 (0.3) 1.82 (0.1) 0.014 1.28 (0.2) 2.05 (0.2) 0.001 1.05 (0.2) 1.97 (0.1) <0.001
Month 18 3.17 (0.5) 3.14 (0.4) 0.944 NA NA NA NA NA NA
CFP ¼ color fundus photography; FAF ¼ fundus autofluorescence; NA ¼ not available; SE ¼ standard error. *Based on paired t test, FAF versus CFP. y Based on paired t test, unilateral versus bilateral. z Based on paired t test, foveal versus extrafoveal. x Based on paired t test, unifocal versus multifocal.
Lesion progression rates, as determined on FAF images, were then stratified by lesion subtype, foveal versus extrafoveal and unifocal versus multifocal. Atrophy progression was significantly greater among eyes with extrafoveal lesions compared with eyes with foveal lesions at month 6 (extrafoveal 0.99 0.1 vs. foveal 0.65 0.1, P ¼ 0.011) and month 12 (extrafoveal 2.05 0.2 vs. foveal 1.28 0.2, P ¼ 0.001) (Fig 4, Table 3). Atrophy progression was also significantly greater among eyes with
multifocal lesions compared with eyes with unifocal lesions at month 6 (multifocal 1.00 0.1 vs. unifocal 0.47 0.1, P < 0.001) and month 12 (multifocal 1.97 0.1 vs. unifocal 1.05 0.2, P < 0.001) (Fig 4, Table 3). To further differentiate lesion progression rates by lesion subtype, the lesion progression rate in eyes with different hyperfluorescence pattern subtypes at baseline, as determined by FAF imaging, was assessed. The mean (standard error) lesion size change at month 6 and month 12 was 0.13 (0.2) mm2 and 1.00 (0.4) mm2 for the “no abnormal” pattern (n ¼ 21), 0.36 (0.1) mm2 and 0.90 (0.4) mm2 for the focal pattern (n ¼ 26), 1.01 (0.1) mm2 and 1.95 (0.5) mm2 for the banded pattern (n ¼ 30), 0.41 (0.2) mm2 and 1.46 (0.3) mm2 for the patchy pattern (n ¼ 11), and 0.95 (0.2) mm2 and 1.91 (0.1) mm2 for the diffuse pattern (n ¼ 321), respectively. The progression rates at month 6 of the banded (P < 0.001) and the diffuse (P < 0.001) groups were markedly higher compared with the “no abnormal,” focal, and patchy groups (Fig 5). At month 12, the banded and diffuse groups continued to have the largest mean lesion size change among the 5 subgroups, but did not reach statistical significance (P ¼ 0.196 and P ¼ 0.082, respectively). Of the 413 evaluable participants, 343 had visual acuity data at both baseline and exit visit (all time points). For all participants with image data available, the mean decrease in visual acuity was 6.2 15.6 letters from baseline for participants with 12 months of follow-up data (n ¼ 126). Although involvement of the fovea by the atrophic lesion process during progression can affect BCVA, there was no correlation between BCVA and lesion localization (foveal, r ¼ 0.001; extrafoveal, r ¼ 0.05), and lesion size (r ¼ 0.06), or lesion progression rate (r ¼ 0.19) (Fig 6).
Discussion In the present study, we found that lesion size measured by CFP was greater on average than lesion size measured by
Figure 2. Mean change in lesion size at months 6 and 12 by baseline size categories using fundus autofluorescence (FAF) imaging. Baseline lesion categories versus mean change in lesion size (mm2) at months 6 and 12, as assessed by FAF. Graph bar numbers represent N values; error bars represent standard error (SE).
4
Schmitz-Valckenberg et al
Geographic Atrophy Natural History Study
Figure 3. Geographic atrophy (GA) lesion progression for different fundus autofluorescence (FAF) patterns. Representative examples of 4 major FAF patterns: (A) none, (B) focal, (C) banded, and (D) diffuse. For each example, the baseline (first row), month 6 (second row), month 12 (third row), and exit (fourth row) visits are shown, both with the native and the processed images (atrophic lesions outlined by white color). E, The corresponding color fundus camera photograph for each pattern at baseline.
FAF at each time point. However, the lesion progression rate was similar for the 2 imaging modalities, and depended on baseline lesion size, location, and pattern.
Figure 4. Mean lesion size change from baseline by type of atrophic spots and lesion localization assessed using fundus autofluorescence (FAF) imaging. Mean lesion size change based on type of atrophic spots (left, unifocal vs. multifocal) and localization of lesion (right, foveal vs. extrafoveal) as assessed by FAF. P values based on 2-sample t test, unifocal versus multifocal, and foveal versus extrafoveal, at months 6 and 12. Graph bar numbers represent N values; error bars represent standard error (SE).
Variation by imaging modality may be due to differences in the image resolution or data processing.20 The latter possibility is unlikely, because the image reading methods were consistent on each image evaluated. Lesion clarity and contrast on images obtained with FAF were generally better than those seen on CFP. For eyes with multifocal lesions, if one assumes that smaller lesions were detected more readily with FAF than with CFP, then one may have anticipated that total lesion size, which included all lesions, large and small, would have been larger by FAF than by CFP. Conversely, for unifocal lesions, if a higher proportion of small lesions were detected on FAF, then the average lesion size measured on FAF would have been smaller than that measured on CFP. In addition, if depigmented but nonatrophic regions seen on CFP were erroneously identified as atrophic lesions, then lesion size measured on CFP would have been larger than that on FAF. The specific reasons for the size discrepancy on the different modalities remain unknown. We found that despite differences in lesion size, the progression rates measured by CFP and FAF were highly correlated with one another. Fundus autofluorescence may have several advantages over CFP, particularly because the better contrast of FAF images permits more accurate lesion boundary identification, as well as the ability to identify
5
Ophthalmology Volume -, Number -, Month 2015
Figure 5. Mean lesion size from baseline at months 6 and 12 by perifoveal hyperfluorescence patterns assessed using fundus autofluorescence (FAF) imaging. Size reported as mm2. P values based on 2-sample t test, compared with the “no abnormal” group. Diffuse pattern ¼ presence of reticular, branching, fine granular-grainy, dusty, heterogeneous, fine granular with peripheral punctuated spots, or trickling patterns. Graph bar numbers represent N values; error bars represent standard error (SE).
specific lesion autofluorescence that predicts lesion progression rate over time. Nevertheless, we believe that the most accurate assessment is achieved by reference to at least 1 additional image type when grading FAF images (e.g., infrared or CFP), as has been suggested by Sunness et al11 and other investigators.21 The mean lesion change from baseline to month 12, measured on FAF images, was 1.85 mm2. This progression rate is consistent with results from Holz et al in the FAM study.8 In contrast, the mean lesion change from baseline to month 12 was 1.57 mm2 as determined by CFP. The CFP
progression rate was less than the rate reported by Sunness et al,11 possibly because of differences in baseline lesion size and imaging capabilities as described previously. Changes in BCVA were not correlated with baseline lesion area, lesion progression rate, or lesion localization (foveal or extrafoveal). Depending on the clinical study, progression rates vary widely among patients and range from 1.2 to 2.8 mm2 per year.4 The mechanisms responsible for the differences in progression rates among studies are not fully understood and may include different baseline lesion sizes and locations, different trial protocols, imaging systems, and varying genotype or exposure to environmental factors. At month 12, lesion progression rates were significantly greater in extrafoveal and multifocal lesions compared with foveal and unifocal lesions. The findings in this study regarding the distribution of FAF patterns and the lesion progression rates between different patterns were in accordance with those previously reported in the FAM study and represent risk factors predictive of lesion progression.8,9
Study Strengths and Limitations The GAP study was a prospective, multicenter, 18-month, noninterventional, natural history study of participants with GA secondary to AMD. Strengths of this study include standardized follow-up procedures and standardized central reading center image assessment to minimize variation. Participants were instructed to return to the site for repeat autofluorescence and fundus photographic lesion imaging if it was not possible to obtain gradable images taken at the baseline or follow-up visits. The quantification of lesion subtypes by 2 imaging modalities and correlation with
Figure 6. Visual acuity changes from baseline at month 12 based on lesion size, localization, and change. Best-corrected visual acuity (BCVA) change from baseline versus (A) lesion size at baseline, (B) lesion localization (circle, foveal; cross, extrafoveal), or (C) change in lesion area from baseline, in participants with 12 (2) months of follow-up. FAF ¼ fundus autofluorescence.
6
Schmitz-Valckenberg et al
Geographic Atrophy Natural History Study
changes in BCVA have provided new insights into the natural history of GA and was another study strength. A study limitation was that some of the participants did not have complete follow-up, because they exited to enroll in an interventional trial for GA due to AMD (GATE study). Therefore, the number of participants with gradable image data decreased at each postbaseline study visit, primarily because of study termination. Many participants were excluded because of lack of follow-up FAF images, which caused exclusion because of insufficient points on FAF image analysis. Participants also were excluded if images did not fulfill all predefined lesion criteria for GA according to the reading center assessment. In addition, spectral-domain optical coherence tomography was not available when this study was initiated. In future studies, spectral-domain optical coherence tomography and other technologies could provide further lesion imaging and measurement capabilities.22,23
10. 11.
12. 13.
14.
Acknowledgment. Meridius Health Communications, Inc, provided technical support in preparation of the manuscript. 15.
References 16. 1. Klein R, Meuer SM, Knudtson MD, Klein BE. The epidemiology of progression of pure geographic atrophy: the Beaver Dam Eye Study. Am J Ophthalmol 2008;146:692–9. 2. Sunness JS, Gonzalez-Baron J, Applegate CA, et al. Enlargement of atrophy and visual acuity loss in the geographic atrophy form of age-related macular degeneration. Ophthalmology 1999;106:1768–79. 3. Khan M, Agarwal K, Loutfi M, Kamal A. Present and possible therapies for age-related macular degeneration. ISRN Ophthalmol 2014;2014:608390. 4. Holz FG, Strauss EC, Schmitz-Valckenberg S, van Lookeren Campagne M. Geographic atrophy: clinical features and potential therapeutic approaches. Ophthalmology 2014;121:1079–91. 5. Rudnicka AR, Jarrar Z, Wormald R, et al. Age and gender variations in age-related macular degeneration prevalence in populations of European ancestry: a meta-analysis. Ophthalmology 2012;119:571–80. 6. Clemons TE, Milton RC, Klein R, et al. Risk factors for the incidence of Advanced Age-Related Macular Degeneration in the Age-Related Eye Disease Study (AREDS) AREDS report no. 19. Ophthalmology 2005;112:533–9. 7. Zarbin MA, Rosenfeld PJ. Pathway-based therapies for agerelated macular degeneration: an integrated survey of emerging treatment alternatives. Retina 2010;30:1350–67. 8. Holz FG, 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;143:463–72. 9. Sunness JS, Rubin GS, Applegate CA, et al. Visual function abnormalities and prognosis in eyes with age-related
17.
18.
19. 20.
21.
22.
23.
geographic atrophy of the macula and good visual acuity. Ophthalmology 1997;104:1677–91. Grunwald JE, Pistilli M, Ying GS, et al. Growth of geographic atrophy in the comparison of age-related macular degeneration treatments trials. Ophthalmology 2015;122:809–16. Sunness JS, Margalit E, Srikumaran D, et al. The long-term natural history of geographic atrophy from age-related macular degeneration: enlargement of atrophy and implications for interventional clinical trials. Ophthalmology 2007;114:271–7. Mata NL, Lichter JB, Vogel R, et al. Investigation of oral fenretinide for treatment of geographic atrophy in age-related macular degeneration. Retina 2013;33:498–507. Vaclavik V, Vujosevic S, Dandekar SS, et al. Autofluorescence imaging in age-related macular degeneration complicated by choroidal neovascularization: a prospective study. Ophthalmology 2008;115:342–6. Bindewald A, Schmitz-Valckenberg S, Jorzik JJ, et al. Classification of abnormal fundus autofluorescence patterns in the junctional zone of geographic atrophy in patients with age related macular degeneration. Br J Ophthalmol 2005;89: 874–8. Mauschitz MM, Fonseca S, Chang P, et al. Topography of geographic atrophy in age-related macular degeneration. Invest Ophthalmol Vis Sci 2012;53:4932–9. Schmitz-Valckenberg S, Alten F, Steinberg JS, et al. Reticular drusen associated with geographic atrophy in age-related macular degeneration. Invest Ophthalmol Vis Sci 2011;52: 5009–15. Steinberg JS, Auge J, Jaffe GJ, et al. Longitudinal analysis of reticular drusen associated with geographic atrophy in agerelated macular degeneration. Invest Ophthalmol Vis Sci 2013;54:4054–60. Deckert A, Schmitz-Valckenberg S, Jorzik J, et al. Automated analysis of digital fundus autofluorescence images of geographic atrophy in advanced age-related macular degeneration using confocal scanning laser ophthalmoscopy (cSLO). BMC Ophthalmol 2005;5:8. Fleckenstein M, Schmitz-Valckenberg S, Adrion C, et al. Progression of age-related geographic atrophy: role of the fellow eye. Invest Ophthalmol Vis Sci 2011;52:6552–7. Khanifar AA, Lederer DE, Ghodasra JH, et al. Comparison of color fundus photographs and fundus autofluorescence images in measuring geographic atrophy area. Retina 2012;32: 1884–91. Lindner M, Boker A, Mauschitz MM, et al. Directional kinetics of geographic atrophy progression in age-related macular degeneration with foveal sparing. Ophthalmology 2015;122:1356–65. Wolf-Schnurrbusch UE, Wittwer VV, Ghanem R, et al. Bluelight versus green-light autofluorescence: lesion size of areas of geographic atrophy. Invest Ophthalmol Vis Sci 2011;52: 9497–502. Gocho K, Sarda V, Falah S, et al. Adaptive optics imaging of geographic atrophy. Invest Ophthalmol Vis Sci 2013;54: 3673–80.
Footnotes and Financial Disclosures Originally received: May 14, 2015. Final revision: August 18, 2015. Accepted: September 25, 2015. Available online: ---. 1
2
Université Pierre et Marie Curie and Institut de la Vision, Paris, France.
3
Manuscript no. 2015-780.
Department of Ophthalmology, University of Bonn, Bonn, Germany.
Department of Ophthalmology and Visual Sciences, University of Wisconsin, Madison, Wisconsin.
4
Department of Ophthalmology, Duke University, Durham, North Carolina.
7
Ophthalmology Volume -, Number -, Month 2015 5
Department of Ophthalmology, University Hospital Bern, Inselspital, University Bern, Bern, Switzerland.
Author Contributions:
6
Department of Ophthalmology, Kantonsspital Baselland, Liestal, Switzerland.
Data collection: Schmitz-Valckenberg, Sahel, Danis, Fleckenstein, Jaffe, Wolf, Pruente, Holz
7
Analysis and interpretation: Schmitz-Valckenberg, Sahel, Danis, Fleckenstein, Jaffe, Wolf, Pruente, Holz
GRADE Reading Center, University of Bonn, Bonn, Germany.
Presented at: The Association for Research in Vision and Ophthalmology Annual Meeting, May 3e9, 2009, Fort Lauderdale, Florida; and The Association for Research in Vision and Ophthalmology Annual Meeting, May 2e6, 2010, Fort Lauderdale, Florida. Financial Disclosure(s): The author(s) have made the following disclosure(s): J.-A.S.: Founder and consultant Pixium Vision and GenSight Biologics; Consultant SanofiFovea, Gene Signal, and Vision Medicines, Inc. S.S-V.: Consultant Alcon, Novartis. R.D.: Consultant GlaxoSmithKline. G.J.J.: Consultant Heidelberg Engineering. S.W.: Consultant Alcon, Allergan, Bayer Healthcare, Heidelberg Engineering, Novartis, Roche, and Zeiss. F.G.H.: Consultant Alcon, Allergan, Bayer Healthcare, Heidelberg Engineering, Genentech, Novartis, and Roche.
8
Conception and design: Schmitz-Valckenberg, Pruente, Holz
Obtained funding: Not applicable Overall responsibility: Schmitz-Valckenberg, Sahel, Danis, Fleckenstein, Jaffe, Wolf, Pruente, Holz Abbreviations and Acronyms: AMD ¼ age-related macular degeneration; BCVA ¼ best-corrected visual acuity; CFP ¼ color fundus photography; DA ¼ disc areas; FAF ¼ fundus autofluorescence; FAM ¼ Fundus Autofluorescence Imaging in AgeRelated Macular Degeneration; GA ¼ geographic atrophy; GAP ¼ Geographic Atrophy Progression; GATE ¼ Geographic Atrophy Treatment Evaluation; RPE ¼ retinal pigment epithelium; SE ¼ standard error. Correspondence: Frank G. Holz, MD, Department of Ophthalmology, University of Bonn, Ernst-Abbe-Str. 2, D-53127 Bonn, Germany. E-mail: Frank.Holz@ukb. uni-bonn.de.