Intravitreal Aflibercept for Macular Edema Following Branch Retinal Vein Occlusion The 24-Week Results of the VIBRANT Study Peter A. Campochiaro, MD,1 W. Lloyd Clark, MD,2 David S. Boyer, MD,3 Jeffrey S. Heier, MD,4 David M. Brown, MD,5 Robert Vitti, MD,6 Husain Kazmi, MD,6 Alyson J. Berliner, MD, PhD,6 Kristine Erickson, PhD, OD,6 Karen W. Chu, MS,6 Yuhwen Soo, PhD,6 Yenchieh Cheng, PhD,6 Julia A. Haller, MD7 Purpose: To compare the efficacy and safety of intravitreal aflibercept injection (IAI) with macular grid laser photocoagulation for the treatment of macular edema after branch retinal vein occlusion (BRVO). Design: The VIBRANT study was a double-masked, active-controlled, randomized, phase III trial. Participants: Treatment-naïve eyes with macular edema after BRVO were included in the study if the occlusion occurred within 12 months and best-corrected visual acuity (BCVA) was between 73 and 24 Early Treatment Diabetic Retinopathy Study (ETDRS) letters (20/40e20/320 Snellen equivalent). Methods: Eyes (1 eye per patient) received either IAI 2 mg every 4 weeks (n ¼ 91) from baseline to week 20 or grid laser (n ¼ 92) at baseline with a single grid laser rescue treatment, if needed, from weeks 12 through 20. Main Outcome Measures: The primary outcome measure was the proportion of eyes that gained 15 ETDRS letters from baseline BCVA at week 24. Secondary end points included mean change from baseline BCVA and central retinal thickness (CRT) at week 24. Results: The proportion of eyes that gained 15 ETDRS letters from baseline at week 24 was 52.7% in the IAI group compared with 26.7% in the laser group (P ¼ 0.0003). The mean improvement from baseline BCVA at week 24 was 17.0 ETDRS letters in the IAI group and 6.9 ETDRS letters in the laser group (P < 0.0001). The mean reduction in CRT from baseline at week 24 was 280.5 mm in the IAI group and 128.0 mm in the laser group (P < 0.0001). Traumatic cataract in an IAI patient was the only ocular serious adverse event (SAE) that occurred. There were no cases of intraocular inflammation or endophthalmitis. The incidence of nonocular SAEs was 8.8% in the IAI group and 9.8% in the laser group. One Anti-Platelet Trialists’ Collaborationedefined event of nonfatal stroke (1.1%) and 1 death (1.1%) due to pneumonia occurred during the 24 weeks of the study, both in patients in the laser group. Conclusions: Monthly IAI provided significantly greater visual benefit and reduction in CRT at 24 weeks than grid laser photocoagulation in eyes with macular edema after BRVO. Ophthalmology 2014;-:1e7 ª 2014 by the American Academy of Ophthalmology.
Retinal vein occlusion (RVO) includes central RVO (CRVO), in which the central retinal vein is obstructed with involvement of all 4 retinal quadrants, and branch RVO (BRVO), in which one of the major tributaries of the central retinal vein is obstructed.1 Hemi-RVO (HRVO) refers to obstructions in the first branch of the central retinal vein near the margin of the optic disc and affects an entire hemisphere of the retina.1 A highly prevalent retinal vascular disease, RVO is second only to diabetic retinopathy. Most of this high prevalence is driven by BRVO; the yearly incidence of BRVO in the United States is 150 000 and that of CRVO is 30 000.2 Overall, HRVO comprises approximately 5% of RVO cases.3 In CRVO, hemorrhages and edema develop throughout the retina, whereas in BRVO the pathology is more sectoral, involving the portions of the retina drained by the obstructed 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.
branch vein. This suggests that increased intraluminal pressure behind the obstruction may lead to transudation of blood cells and plasma into the retina. However, recent studies have demonstrated that although increased venous pressure may be the precipitating event for hemorrhages and edema, increased production of vascular endothelial growth factor (VEGF) occurs early in RVO and is a major contributor to their evolution and persistence.4 In addition, the high levels of VEGF contribute to progression of retinal nonperfusion and hence retinal ischemia, which may in turn increase production of VEGF, and may explain why some eyes enter a vicious cycle of worsening disease often referred to as conversion to an ischemic RVO.5 Before the development of VEGF-binding proteins, there was no effective treatment for CRVO; their advent has http://dx.doi.org/10.1016/j.ophtha.2014.08.031 ISSN 0161-6420/12
1
Ophthalmology Volume -, Number -, Month 2014 revolutionized management of the condition. Marked improvements in visual acuity have been demonstrated after intravitreal injection of ranibizumab (Lucentis; Genentech, South San Francisco, CA),4,6,7 bevacizumab (Avastin; Genentech),8,9 and aflibercept.10e12 Intravitreal aflibercept injection (IAI), also known in scientific literature as VEGF Trap-Eye (Regeneron Pharmaceuticals, Inc, Tarrytown, NY), is a 115-kDa soluble receptor fusion protein composed of the second domain of human VEGF receptor 1 and the third domain of VEGF receptor 2 fused to the Fc domain of human immunoglobulin G1.13 The binding affinity of aflibercept for VEGF is substantially greater than that of either bevacizumab or ranibizumab,14 and mathematical modeling predicts that it could have a longer duration of action in the eye.15 Aflibercept also binds placental growth factor, which is a member of the VEGF family of angiogenic factors that can act as a mitogenic, chemotactic, and vascular permeability factor for endothelial cells.16 The COPERNICUS10 and GALILEO11 studies showed that in patients with macular edema after CRVO, 56.1% and 60.2% of eyes given a 2-mg IAI injection every 4 weeks for 24 weeks gained 15 letters in best-corrected visual acuity (BCVA) compared with 12.3% and 22.1% in sham-injected eyes, respectively. Based on these results, IAI was approved by the US Food and Drug Administration for the treatment of patients with macular edema after CRVO. Unlike the situation in macular edema after CRVO, before the development of VEGF binding proteins, there was a treatment for macular edema after BRVO that had been shown to provide benefit: macular grid laser photocoagulation.17 The Ranibizumab for the Treatment of Macular Edema after BRAnch Retinal Vein Occlusion: Evaluation of Efficacy and Safety (BRAVO) study demonstrated that injections of ranibizumab every 4 weeks for 24 weeks are superior to sham injections.18 However, the BRAVO study did not determine whether anti-VEGF injections were superior to laser, because laser was given only as rescue and was available to both groups between weeks 12 and 20. The VIBRANT trial was the first study to compare anti-VEGF treatment with laser. In the first 24 weeks of VIBRANT, eligible eyes with macular edema after BRVO were randomized to receive laser at baseline (with one rescue laser if needed) or 2 mg IAI every 4 weeks. From week 24 onward, eyes in the IAI group received 2 mg IAI every 8 weeks, and if they required rescue treatment, they could receive laser at week 36; eyes in the laser group that required rescue could receive 2 mg IAI every 8 weeks after 3 initial monthly doses. Herein, we report the results at the 24-week primary outcome measure.
Methods The VIBRANT study was a phase III, multicenter, randomized, double-masked, active-controlled, 52-week trial comparing the efficacy and safety of IAI with macular grid laser photocoagulation for treatment of macular edema after BRVO. The study was conducted at 58 sites in North America and Japan (Appendix 1 provides a list of study investigators; available at www.aaojournal.org). Each respective institutional review board or ethics committee approved
2
Figure 1. The VIBRANT study design from baseline to week 24. All eyes were eligible to undergo scatter laser photocoagulation at any time during the study if they developed clinically significant ocular neovascularization. a Eyes in the intravitreal aflibercept injection (IAI) group also received a sham laser treatment at baseline. bMacular laser photocoagulation at baseline and, if eligible, rescue laser at week 12, 16, or 20 (12 weeks apart from the last laser treatment). Eyes in the laser group also received sham injections every 4 weeks from baseline to week 20. BCVA ¼ best-corrected visual acuity; BRVO ¼ branch retinal vein occlusion; CRT ¼ central retinal thickness; ETDRS ¼ Early Treatment Diabetic Retinopathy Study; HRVO ¼ hemi-retinal vein occlusion.
the study protocol. The study was carried out in adherence with guidelines established by the Declaration of Helsinki, the International Conference on Harmonization Guidelines for Good Clinical Practice, and, for patients from the United States, the Health Insurance Portability and Accountability Act of 1996. All patients provided written informed consent to participate in this trial. The VIBRANT study was registered with ClinicalTrials.gov (identifier no. NCT01521559). Data described herein were collected between April 2012 and August 2013.
Participants Patients 18 years old with BRVO or HRVO causing edema involving the center of the macula were eligible for enrollment if the occlusion occurred within 12 months, and BCVA was between 73 and 24 Early Treatment Diabetic Retinopathy Study (ETDRS) letters (20/40e20/320 Snellen equivalent). We defined BRVO by the presence of retinal hemorrhages or other biomicroscopic evidence of RVO and a dilated venous system in 1 microaneurysm outside the area of the vein occlusion), ocular inflammation, or uncontrolled glaucoma (intraocular pressure 25 mmHg or previous filtration surgery); periocular corticosteroid use within the last 3 months; or prior treatment with intraocular corticosteroids or antiangiogenic drugs, scatter or panretinal laser, macular grid laser, or sector laser.
Treatments Eyes were randomized 1:1 into the IAI and laser groups according to a predetermined central randomization scheme provided by an interactive voice/web response system to the designated unmasked personnel. Eyes in the IAI group received 2 mg IAI every 4 weeks
Campochiaro et al
Intravitreal Aflibercept for Macular Edema after BRVO
from baseline to week 20 (Fig 1). A sham laser treatment was also performed at baseline. Eyes in the laser group received macular laser photocoagulation at baseline and sham injections every 4 weeks from baseline to week 20 (Fig 1). Eyes in both treatment groups were evaluated for rescue treatment from week 12 onward. Rescue treatment criteria included (a) a >50-mm increase in central retinal thickness (CRT) compared with the lowest previous measurement; (b) presence of new or persistent cystic retinal changes, subretinal fluid, or persistent diffuse edema in the central optical coherence tomography subfield; or (c) loss of 5 letters compared with the best previous measurement because of BRVO in conjunction with any increase in CRT. If 1 rescue treatment criterion was met, eyes in the IAI group received sham laser at either week 12, 16, or 20. Eyes in the laser group eligible for rescue treatment before week 24 received 1 additional laser from week 12 to 20 (Fig 1). The protocol specified how the focal/grid laser should be performed, and this was reviewed at the investigators’ meeting and site initiation visits. All eyes were eligible to receive scatter laser photocoagulation at any time during the study if they developed clinically significant ocular neovascularization.
Outcome Measures The primary efficacy outcome measure was the proportion of eyes that gained 15 ETDRS letters in BCVA from baseline at week 24. The secondary efficacy outcome measures, all at week 24, were a change from baseline in BCVA, CRT, and the National Eye Institute 25-item Visual Function Questionnaire (NEI VFQ-25) total scores. Additional prespecified endpoints at week 24 were the time to first sustained gain of 15 letters, change from baseline in NEI VFQ-25 subscales (near activities, distance activities, and visual dependency), and proportion of patients with a change in retinal perfusion. Safety assessments included ocular and nonocular adverse events (AEs) and serious AEs (SAEs). We evaluated BCVA and CRT every 4 weeks from baseline to week 24. We assessed BCVA using the ETDRS protocol.19 We evaluated CRT with spectral domain optical coherence tomography. Optical coherence tomography images were evaluated by an independent central reading center (Duke Reading Center, Durham, NC). Fundus photography, fluorescein angiography, and assessment of vision-related quality of life were performed at baseline and weeks 12 and 24. Fundus photography and fluorescein angiography were evaluated by an independent central reading center (Digital Angiography Reading Center, New York, NY). Perfused retinas were defined as retinas with 0, 5, 10, and 15 ETDRS letters (post hoc analysis), or had retinal perfusion were analyzed with the Cochran-Mantel-Haenszel test. Missing data were imputed using the last-observation-carried-forward method. The safety analysis set included all randomized patients who received any study treatment.
Results Patient Disposition, Demographics, and Baseline Characteristics We screened 281 patients; 98 were excluded because they either did not meet the inclusion/exclusion criteria (n ¼ 92) or they withdrew consent (n ¼ 6). Overall, 91 eyes were randomized to IAI and 92 eyes to grid laser (Table 1). All randomized eyes in both treatment groups were included in the full analysis set, except for 2 eyes in the laser group that did not have a post-baseline BCVA assessment. The safety analysis set included all randomized patients. More than 90% of patients completed the study at week 24 (Table 1). Demographics and baseline characteristics of patients were similar in both treatment groups (Table 2). Overall, 1 eye (1.1%) in the IAI group and 3 eyes (3.3%) in the laser group had macular edema after HRVO at baseline. Eyes in the IAI group received a mean of 5.7 injections (range, 2e6 injections) from baseline to week 24. Eyes in the IAI group did not receive active laser rescue treatment. Of a maximum of 2 possible laser treatments (one being rescue), eyes in the laser group received a mean (standard deviation) of 1.7 (0.5) laser treatments from baseline to week 24.
Efficacy The proportions of eyes that gained 15 ETDRS letters from baseline in the IAI and laser groups were 52.7% and 26.7% (P ¼ 0.0003) at week 24, respectively (Fig 2A). The mean time ( standard error) to the first sustained gain of 15 ETDRS letters (gain of 15 ETDRS letters from baseline at 2 consecutive visits)
3
Ophthalmology Volume -, Number -, Month 2014 Table 2. Patient Demographics and Baseline Characteristics
Mean age, years (SD) Women, n (%) Race, n (%) White Black or African American Asian Other* Geographic region, n (%) North America Japan BCVA Mean, letters (SD) >20/200 (35e73 letters), n (%) 20/200 (24e34 letters), n (%) Retinal perfusion status, n (%) Perfusedy Nonperfusedz Cannot grade Missing Mean central retinal thickness, mm (SD) Mean intraocular pressure, mmHg (SD) Time since BRVO diagnosis Mean, days (SD) 0, 5, 10, and 15 ETDRS letters compared with eyes in the IAI group at week 24 (Table 3). The mean change from baseline BCVA in the IAI group compared with the laser group was 17.0 versus 6.9 ETDRS letters (P < 0.0001) at week 24, respectively (Fig 2B). When analyzed by the baseline retinal perfusion status, the mean change from baseline BCVA in the IAI and laser groups was 14.3 versus 5.7 ETDRS letters (P < 0.0001) in the subgroup of eyes considered perfused and 19.1 versus 11.3 ETDRS letters (P ¼ 0.1008) in the subgroup of eyes considered nonperfused, respectively. When analyzed by the baseline BCVA, the mean change from baseline BCVA in the IAI and laser groups was 15.7 versus 6.9 ETDRS letters (P < 0.0001) in eyes with baseline BCVA of >20/200 and 34.5 versus 7.3 ETDRS letters (P ¼ 0.0168) in eyes with baseline BCVA of 20/200, respectively.
4
Figure 2. Visual and anatomic outcomes. The proportion of eyes that gained 15 letters from baseline to week 24 (A) and the mean change from baseline in best-corrected visual acuity (BCVA) (B) and central retinal thickness (C) over 24 weeks are shown. Full analysis set. Missing data were imputed using the last observation carried forward (LOCF) method. a P ¼ 0.0003 and bP < 0.0001 versus laser. ETDRS ¼ Early Treatment Diabetic Retinopathy Study; IAI ¼ intravitreal aflibercept injection.
The mean reduction from baseline CRT in the IAI and laser groups was 280.5 versus 128.0 mm (P < 0.0001) at week 24, respectively (Fig 2C). At baseline, 60.4% and 68.9% of patients had perfused retinas in the IAI and laser groups, respectively. At week 24, the proportion of patients with perfused retinas in the IAI group increased to 80.2% and the proportion of patients in the laser group
Campochiaro et al
Intravitreal Aflibercept for Macular Edema after BRVO
Table 3. Eyes with Vision Gains and Losses from Baseline at Week 24 Laser (n [ 90) Vision gain (letters), n (%) 30 15* 10 5 0 Vision loss (letters), n (%) >0 5 10 15
IAI (n [ 91)
3 24 35 55 73
(3.3) (26.7) (38.9) (61.1) (81.1)
11 48 67 79 89
(12.1) (52.7) (73.6) (86.8) (97.8)
17 9 6 4
(18.9) (10.0) (6.7) (4.4)
2 1 1 0
(2.2) (1.1) (1.1) (0)
P Value
0.0165 0.0003 25 years. Despite the encouraging preliminary data with ranibizumab injections in macular edema after BRVO,4 in designing the BRAVO study, many investigators expressed concern about withholding the standard of care from the ranibizumab treatment groups for 6 months. Therefore, the BRAVO study compared ranibizumab with sham injections and allowed rescue treatment with grid laser photocoagulation in all patients if prespecified rescue criteria were met.18 At the 6-month primary end point, the BRAVO study showed that ranibizumab injections provided significantly greater visual benefit than sham injections in eyes with macular edema after BRVO, but did not allow comparison of the efficacy of ranibizumab and grid laser therapy. In contrast, the current study provides the first direct comparison of intraocular injection of a VEGF-binding protein with grid laser photocoagulation and demonstrates that IAI provided more edema reduction and greater visual benefit than grid laser photocoagulation over the course of 24 weeks without a relative increase in SAEs. There was no difference in the NEI VFQ-25 composite score for the IAI-treated versus lasertreated eyes at 6 months; this is likely because of combination of factors, including good vision in the fellow eye in almost all patients, only modest vision impairment in some study eyes in each group at baseline, and the clinically meaningful improvement in BCVA that occurred in more than one quarter of study eyes in the grid laser group. The VIBRANT study shows that intravitreal aflibercept is an effective anti-VEGF treatment for macular edema after BRVO for at least 6 months. In addition to the visual acuity benefit, monthly IAI increased the percentage of patients with perfused retinas (defined as those having
Retinal vein occlusion (RVO) includes central RVO (CRVO), in which the central retinal vein is obstructed with involvement of all 4 retinal quadrants, and branch RVO (BRVO), in which one of the major tributaries of the central retinal vein is obstructed.1 Hemi-RVO (HRVO) refers to obstructions in the first branch of the central retinal vein near the margin of the optic disc and affects an entire hemisphere of the retina.1 A highly prevalent retinal vascular disease, RVO is second only to diabetic retinopathy. Most of this high prevalence is driven by BRVO; the yearly incidence of BRVO in the United States is 150 000 and that of CRVO is 30 000.2 Overall, HRVO comprises approximately 5% of RVO cases.3 In CRVO, hemorrhages and edema develop throughout the retina, whereas in BRVO the pathology is more sectoral, involving the portions of the retina drained by the obstructed 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.
branch vein. This suggests that increased intraluminal pressure behind the obstruction may lead to transudation of blood cells and plasma into the retina. However, recent studies have demonstrated that although increased venous pressure may be the precipitating event for hemorrhages and edema, increased production of vascular endothelial growth factor (VEGF) occurs early in RVO and is a major contributor to their evolution and persistence.4 In addition, the high levels of VEGF contribute to progression of retinal nonperfusion and hence retinal ischemia, which may in turn increase production of VEGF, and may explain why some eyes enter a vicious cycle of worsening disease often referred to as conversion to an ischemic RVO.5 Before the development of VEGF-binding proteins, there was no effective treatment for CRVO; their advent has http://dx.doi.org/10.1016/j.ophtha.2014.08.031 ISSN 0161-6420/12
1
Ophthalmology Volume -, Number -, Month 2014 revolutionized management of the condition. Marked improvements in visual acuity have been demonstrated after intravitreal injection of ranibizumab (Lucentis; Genentech, South San Francisco, CA),4,6,7 bevacizumab (Avastin; Genentech),8,9 and aflibercept.10e12 Intravitreal aflibercept injection (IAI), also known in scientific literature as VEGF Trap-Eye (Regeneron Pharmaceuticals, Inc, Tarrytown, NY), is a 115-kDa soluble receptor fusion protein composed of the second domain of human VEGF receptor 1 and the third domain of VEGF receptor 2 fused to the Fc domain of human immunoglobulin G1.13 The binding affinity of aflibercept for VEGF is substantially greater than that of either bevacizumab or ranibizumab,14 and mathematical modeling predicts that it could have a longer duration of action in the eye.15 Aflibercept also binds placental growth factor, which is a member of the VEGF family of angiogenic factors that can act as a mitogenic, chemotactic, and vascular permeability factor for endothelial cells.16 The COPERNICUS10 and GALILEO11 studies showed that in patients with macular edema after CRVO, 56.1% and 60.2% of eyes given a 2-mg IAI injection every 4 weeks for 24 weeks gained 15 letters in best-corrected visual acuity (BCVA) compared with 12.3% and 22.1% in sham-injected eyes, respectively. Based on these results, IAI was approved by the US Food and Drug Administration for the treatment of patients with macular edema after CRVO. Unlike the situation in macular edema after CRVO, before the development of VEGF binding proteins, there was a treatment for macular edema after BRVO that had been shown to provide benefit: macular grid laser photocoagulation.17 The Ranibizumab for the Treatment of Macular Edema after BRAnch Retinal Vein Occlusion: Evaluation of Efficacy and Safety (BRAVO) study demonstrated that injections of ranibizumab every 4 weeks for 24 weeks are superior to sham injections.18 However, the BRAVO study did not determine whether anti-VEGF injections were superior to laser, because laser was given only as rescue and was available to both groups between weeks 12 and 20. The VIBRANT trial was the first study to compare anti-VEGF treatment with laser. In the first 24 weeks of VIBRANT, eligible eyes with macular edema after BRVO were randomized to receive laser at baseline (with one rescue laser if needed) or 2 mg IAI every 4 weeks. From week 24 onward, eyes in the IAI group received 2 mg IAI every 8 weeks, and if they required rescue treatment, they could receive laser at week 36; eyes in the laser group that required rescue could receive 2 mg IAI every 8 weeks after 3 initial monthly doses. Herein, we report the results at the 24-week primary outcome measure.
Methods The VIBRANT study was a phase III, multicenter, randomized, double-masked, active-controlled, 52-week trial comparing the efficacy and safety of IAI with macular grid laser photocoagulation for treatment of macular edema after BRVO. The study was conducted at 58 sites in North America and Japan (Appendix 1 provides a list of study investigators; available at www.aaojournal.org). Each respective institutional review board or ethics committee approved
2
Figure 1. The VIBRANT study design from baseline to week 24. All eyes were eligible to undergo scatter laser photocoagulation at any time during the study if they developed clinically significant ocular neovascularization. a Eyes in the intravitreal aflibercept injection (IAI) group also received a sham laser treatment at baseline. bMacular laser photocoagulation at baseline and, if eligible, rescue laser at week 12, 16, or 20 (12 weeks apart from the last laser treatment). Eyes in the laser group also received sham injections every 4 weeks from baseline to week 20. BCVA ¼ best-corrected visual acuity; BRVO ¼ branch retinal vein occlusion; CRT ¼ central retinal thickness; ETDRS ¼ Early Treatment Diabetic Retinopathy Study; HRVO ¼ hemi-retinal vein occlusion.
the study protocol. The study was carried out in adherence with guidelines established by the Declaration of Helsinki, the International Conference on Harmonization Guidelines for Good Clinical Practice, and, for patients from the United States, the Health Insurance Portability and Accountability Act of 1996. All patients provided written informed consent to participate in this trial. The VIBRANT study was registered with ClinicalTrials.gov (identifier no. NCT01521559). Data described herein were collected between April 2012 and August 2013.
Participants Patients 18 years old with BRVO or HRVO causing edema involving the center of the macula were eligible for enrollment if the occlusion occurred within 12 months, and BCVA was between 73 and 24 Early Treatment Diabetic Retinopathy Study (ETDRS) letters (20/40e20/320 Snellen equivalent). We defined BRVO by the presence of retinal hemorrhages or other biomicroscopic evidence of RVO and a dilated venous system in 1 microaneurysm outside the area of the vein occlusion), ocular inflammation, or uncontrolled glaucoma (intraocular pressure 25 mmHg or previous filtration surgery); periocular corticosteroid use within the last 3 months; or prior treatment with intraocular corticosteroids or antiangiogenic drugs, scatter or panretinal laser, macular grid laser, or sector laser.
Treatments Eyes were randomized 1:1 into the IAI and laser groups according to a predetermined central randomization scheme provided by an interactive voice/web response system to the designated unmasked personnel. Eyes in the IAI group received 2 mg IAI every 4 weeks
Campochiaro et al
Intravitreal Aflibercept for Macular Edema after BRVO
from baseline to week 20 (Fig 1). A sham laser treatment was also performed at baseline. Eyes in the laser group received macular laser photocoagulation at baseline and sham injections every 4 weeks from baseline to week 20 (Fig 1). Eyes in both treatment groups were evaluated for rescue treatment from week 12 onward. Rescue treatment criteria included (a) a >50-mm increase in central retinal thickness (CRT) compared with the lowest previous measurement; (b) presence of new or persistent cystic retinal changes, subretinal fluid, or persistent diffuse edema in the central optical coherence tomography subfield; or (c) loss of 5 letters compared with the best previous measurement because of BRVO in conjunction with any increase in CRT. If 1 rescue treatment criterion was met, eyes in the IAI group received sham laser at either week 12, 16, or 20. Eyes in the laser group eligible for rescue treatment before week 24 received 1 additional laser from week 12 to 20 (Fig 1). The protocol specified how the focal/grid laser should be performed, and this was reviewed at the investigators’ meeting and site initiation visits. All eyes were eligible to receive scatter laser photocoagulation at any time during the study if they developed clinically significant ocular neovascularization.
Outcome Measures The primary efficacy outcome measure was the proportion of eyes that gained 15 ETDRS letters in BCVA from baseline at week 24. The secondary efficacy outcome measures, all at week 24, were a change from baseline in BCVA, CRT, and the National Eye Institute 25-item Visual Function Questionnaire (NEI VFQ-25) total scores. Additional prespecified endpoints at week 24 were the time to first sustained gain of 15 letters, change from baseline in NEI VFQ-25 subscales (near activities, distance activities, and visual dependency), and proportion of patients with a change in retinal perfusion. Safety assessments included ocular and nonocular adverse events (AEs) and serious AEs (SAEs). We evaluated BCVA and CRT every 4 weeks from baseline to week 24. We assessed BCVA using the ETDRS protocol.19 We evaluated CRT with spectral domain optical coherence tomography. Optical coherence tomography images were evaluated by an independent central reading center (Duke Reading Center, Durham, NC). Fundus photography, fluorescein angiography, and assessment of vision-related quality of life were performed at baseline and weeks 12 and 24. Fundus photography and fluorescein angiography were evaluated by an independent central reading center (Digital Angiography Reading Center, New York, NY). Perfused retinas were defined as retinas with 0, 5, 10, and 15 ETDRS letters (post hoc analysis), or had retinal perfusion were analyzed with the Cochran-Mantel-Haenszel test. Missing data were imputed using the last-observation-carried-forward method. The safety analysis set included all randomized patients who received any study treatment.
Results Patient Disposition, Demographics, and Baseline Characteristics We screened 281 patients; 98 were excluded because they either did not meet the inclusion/exclusion criteria (n ¼ 92) or they withdrew consent (n ¼ 6). Overall, 91 eyes were randomized to IAI and 92 eyes to grid laser (Table 1). All randomized eyes in both treatment groups were included in the full analysis set, except for 2 eyes in the laser group that did not have a post-baseline BCVA assessment. The safety analysis set included all randomized patients. More than 90% of patients completed the study at week 24 (Table 1). Demographics and baseline characteristics of patients were similar in both treatment groups (Table 2). Overall, 1 eye (1.1%) in the IAI group and 3 eyes (3.3%) in the laser group had macular edema after HRVO at baseline. Eyes in the IAI group received a mean of 5.7 injections (range, 2e6 injections) from baseline to week 24. Eyes in the IAI group did not receive active laser rescue treatment. Of a maximum of 2 possible laser treatments (one being rescue), eyes in the laser group received a mean (standard deviation) of 1.7 (0.5) laser treatments from baseline to week 24.
Efficacy The proportions of eyes that gained 15 ETDRS letters from baseline in the IAI and laser groups were 52.7% and 26.7% (P ¼ 0.0003) at week 24, respectively (Fig 2A). The mean time ( standard error) to the first sustained gain of 15 ETDRS letters (gain of 15 ETDRS letters from baseline at 2 consecutive visits)
3
Ophthalmology Volume -, Number -, Month 2014 Table 2. Patient Demographics and Baseline Characteristics
Mean age, years (SD) Women, n (%) Race, n (%) White Black or African American Asian Other* Geographic region, n (%) North America Japan BCVA Mean, letters (SD) >20/200 (35e73 letters), n (%) 20/200 (24e34 letters), n (%) Retinal perfusion status, n (%) Perfusedy Nonperfusedz Cannot grade Missing Mean central retinal thickness, mm (SD) Mean intraocular pressure, mmHg (SD) Time since BRVO diagnosis Mean, days (SD) 0, 5, 10, and 15 ETDRS letters compared with eyes in the IAI group at week 24 (Table 3). The mean change from baseline BCVA in the IAI group compared with the laser group was 17.0 versus 6.9 ETDRS letters (P < 0.0001) at week 24, respectively (Fig 2B). When analyzed by the baseline retinal perfusion status, the mean change from baseline BCVA in the IAI and laser groups was 14.3 versus 5.7 ETDRS letters (P < 0.0001) in the subgroup of eyes considered perfused and 19.1 versus 11.3 ETDRS letters (P ¼ 0.1008) in the subgroup of eyes considered nonperfused, respectively. When analyzed by the baseline BCVA, the mean change from baseline BCVA in the IAI and laser groups was 15.7 versus 6.9 ETDRS letters (P < 0.0001) in eyes with baseline BCVA of >20/200 and 34.5 versus 7.3 ETDRS letters (P ¼ 0.0168) in eyes with baseline BCVA of 20/200, respectively.
4
Figure 2. Visual and anatomic outcomes. The proportion of eyes that gained 15 letters from baseline to week 24 (A) and the mean change from baseline in best-corrected visual acuity (BCVA) (B) and central retinal thickness (C) over 24 weeks are shown. Full analysis set. Missing data were imputed using the last observation carried forward (LOCF) method. a P ¼ 0.0003 and bP < 0.0001 versus laser. ETDRS ¼ Early Treatment Diabetic Retinopathy Study; IAI ¼ intravitreal aflibercept injection.
The mean reduction from baseline CRT in the IAI and laser groups was 280.5 versus 128.0 mm (P < 0.0001) at week 24, respectively (Fig 2C). At baseline, 60.4% and 68.9% of patients had perfused retinas in the IAI and laser groups, respectively. At week 24, the proportion of patients with perfused retinas in the IAI group increased to 80.2% and the proportion of patients in the laser group
Campochiaro et al
Intravitreal Aflibercept for Macular Edema after BRVO
Table 3. Eyes with Vision Gains and Losses from Baseline at Week 24 Laser (n [ 90) Vision gain (letters), n (%) 30 15* 10 5 0 Vision loss (letters), n (%) >0 5 10 15
IAI (n [ 91)
3 24 35 55 73
(3.3) (26.7) (38.9) (61.1) (81.1)
11 48 67 79 89
(12.1) (52.7) (73.6) (86.8) (97.8)
17 9 6 4
(18.9) (10.0) (6.7) (4.4)
2 1 1 0
(2.2) (1.1) (1.1) (0)
P Value
0.0165 0.0003 25 years. Despite the encouraging preliminary data with ranibizumab injections in macular edema after BRVO,4 in designing the BRAVO study, many investigators expressed concern about withholding the standard of care from the ranibizumab treatment groups for 6 months. Therefore, the BRAVO study compared ranibizumab with sham injections and allowed rescue treatment with grid laser photocoagulation in all patients if prespecified rescue criteria were met.18 At the 6-month primary end point, the BRAVO study showed that ranibizumab injections provided significantly greater visual benefit than sham injections in eyes with macular edema after BRVO, but did not allow comparison of the efficacy of ranibizumab and grid laser therapy. In contrast, the current study provides the first direct comparison of intraocular injection of a VEGF-binding protein with grid laser photocoagulation and demonstrates that IAI provided more edema reduction and greater visual benefit than grid laser photocoagulation over the course of 24 weeks without a relative increase in SAEs. There was no difference in the NEI VFQ-25 composite score for the IAI-treated versus lasertreated eyes at 6 months; this is likely because of combination of factors, including good vision in the fellow eye in almost all patients, only modest vision impairment in some study eyes in each group at baseline, and the clinically meaningful improvement in BCVA that occurred in more than one quarter of study eyes in the grid laser group. The VIBRANT study shows that intravitreal aflibercept is an effective anti-VEGF treatment for macular edema after BRVO for at least 6 months. In addition to the visual acuity benefit, monthly IAI increased the percentage of patients with perfused retinas (defined as those having
