Abstract
Importance
Studies have established the efficacy and safety of aflibercept for the treatment of macular edema due to central retinal vein occlusion. Bevacizumab is used off-label to treat this condition despite the absence of supporting data.
Objective
To investigate whether bevacizumab is noninferior to aflibercept for the treatment of macular edema secondary to central retinal or hemiretinal vein occlusion.
Design, Setting, and Participants
The SCORE2 randomized noninferiority clinical trial was conducted at 66 private practice or academic centers in the United States, and included 362 patients with macular edema due to central retinal or hemiretinal vein occlusion who were randomized 1:1 to receive aflibercept or bevacizumab. The first participant was randomized on September 17, 2014, and the last month 6 visit occurred on May 6, 2016. Analyses included data available as of December 30, 2016.
Interventions
Eyes were randomized to receive intravitreal injection of bevacizumab (1.25 mg; n = 182) or aflibercept (2.0 mg; n = 180) every 4 weeks through month 6.
Main Outcomes and Measures
The primary outcome was mean change in visual acuity (VA) letter score (VALS) from the randomization visit to the 6-month follow-up visit, based on the best-corrected electronic Early Treatment Diabetic Retinopathy Study VALS (scores range from 0-100; higher scores indicate better VA). The noninferiority margin was 5 letters, and statistical testing for noninferiority was based on a 1-sided 97.5% confidence interval.
Results
Among 362 randomized participants (mean [SD] age, 69 [12] years; 157 [43.4%] women; mean [SD] VALS at baseline, 50.3 [15.2] [approximate Snellen VA 20/100]), 348 (96.1%) completed the month 6 follow-up visit. At month 6, the mean VALS was 69.3 (a mean increase from baseline of 18.6) in the bevacizumab group and 69.3 (a mean increase from baseline of 18.9) in the aflibercept group (model-based estimate of between-group difference, −0.14; 97.5% CI, −3.07 to ∞; P = .001 for noninferiority), meeting criteria for noninferiority. Ocular adverse events in the aflibercept group included 4 participants with intraocular pressure (IOP) more than 10 mm Hg greater than baseline; ocular adverse events in the bevacizumab group included 1 participant with endophthalmitis (culture negative), 9 with IOP more than 10 mm Hg greater than baseline, 2 with IOP higher than 35 mm Hg, and 1 with angle-closure glaucoma not attributed to the study drug or procedure.
Conclusions and Relevance
Among patients with macular edema due to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to aflibercept with respect to visual acuity after 6 months of treatment.
Key Points
Question
Is bevacizumab, a commonly used off-label treatment, noninferior to aflibercept, a US Food and Drug Administration–approved treatment, for visual acuity in eyes with macular edema due to central retinal or hemiretinal vein occlusion?
Findings
In this randomized clinical trial that included 362 patients, the mean visual acuity letter score increase from baseline was 18.6 in the bevacizumab group and 18.9 in the aflibercept group (model-based estimate of mean difference, −0.14, meeting criteria for noninferiority).
Meaning
Among patients with macular edema secondary to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to aflibercept for visual acuity after 6 months of treatment.
This randomized clinical trial compares the effects of bevacizumab vs aflibercept on visual acuity among patients with macular edema due to central retinal or hemiretinal vein occlusion.
Introduction
Retinal vein occlusion is the most common retinal vascular disease after diabetic retinopathy, with population-based prevalence estimates in individuals aged 40 years or older of 0.4% in a German cohort, 0.6% in Beaver Dam, Wisconsin, and 1.6% in an Australian cohort. Analysis across 15 population-based studies from the United States, Europe, Asia, and Australia estimated the worldwide prevalence of retinal vein occlusion at 16 million adults. There are 3 general types of retinal vein occlusion based on obstruction site: branch retinal vein occlusion, central retinal vein occlusion, and hemiretinal vein occlusion. Macular edema (central retinal swelling) is the most frequent cause of vision loss in retinal vein occlusion.
Aflibercept is approved by the US Food and Drug Administration for macular edema due to central retinal vein occlusion. Off-label use of bevacizumab for this indication is widespread because its efficacy and safety have been demonstrated for other retinal diseases (age-related macular degeneration [AMD] and diabetic macular edema [DME]) and its cost (approximately $60/dose) is much less than the cost of aflibercept ($1850/dose). This investigation, the Study of Comparative Treatments for Retinal Vein Occlusion 2 (SCORE2) trial, investigated whether bevacizumab was noninferior to aflibercept for treatment of macular edema due to central retinal or hemiretinal vein occlusion.
Methods
Study Conduct and Oversight
The SCORE2 trial was a multicenter, randomized clinical trial and adhered to the tenets of the Declaration of Helsinki. The protocol and informed consent were approved by either a site-specific institutional review board or centralized institutional review board and by a National Eye Institute–appointed data and safety monitoring committee (DSMC). Study participants provided written informed consent.
Study Population
The full trial protocol is shown in the Supplement. Participants had a best-corrected electronic Early Treatment Diabetic Retinopathy Study (E-ETDRS) visual acuity (VA) letter score (VALS) between 19 and 73 (possible range is 0-100 letters; higher scores indicate better VA), center-involved macular edema due to central retinal or hemiretinal vein occlusion on clinical examination, and central retinal thickness on spectral-domain optical coherence tomography (SD-OCT), defined as central subfield thickness of 300 µm or greater if measured with a Carl Zeiss Meditec Cirrus OCT machine or 320 µm or greater if measured with a Heidelberg Spectralis OCT machine (Figure 1). Patients with hemiretinal vein occlusion have been reported to be similar to patients with branch and central retinal vein occlusion with respect to demographic characteristics, with fundus findings of hemiretinal vein occlusion intermediate between those of branch and central retinal vein occlusion. For the results of this trial to be as generalizable as possible, eligibility criteria permitted inclusion of patients with hemiretinal vein occlusion to a maximum of 25% of the total study sample.
Figure 1. Flowchart of Participant Progress in the SCORE2 Trial.
aPatients may be ineligible at screening for multiple reasons; thus, the number of reasons for ineligibility may not sum to the number of ineligible patients.
bEligibility criteria included central retinal thickness on spectral-domain optical coherence tomography (SD-OCT), which was defined as central subfield thickness of 300 µm or greater if measured with a Carl Zeiss Meditec Cirrus OCT machine or 320 µm or greater if measured with a Heidelberg Spectralis OCT machine.
cFor example, chronic alcoholism or drug abuse, personality disorder, or use of major tranquilizers, indicating difficulty in long-term follow-up and likelihood of survival of less than 12 months.
A history of intravitreal corticosteroid use was allowed if it was more than 4 months before randomization, and prior intravitreal anti–vascular endothelial growth factor (VEGF) use was allowed if it was more than 2 months before randomization. Each participant’s eligibility was determined by the site’s principal investigator or coinvestigator. Race and ethnicity were collected per requirements of the National Institutes of Health, they were determined by the participant, and data collection instruments had fixed categories; for race, there was an option to specify free text if the fixed categories were not relevant.
Interventions
Study eyes were randomized 1:1 to intravitreal bevacizumab (1.25 mg) every 4 weeks for 6 months vs intravitreal aflibercept (2.0 mg) every 4 weeks for 6 months. Bevacizumab was purchased with trial funding and repackaged into single-use vials by the University of Pennsylvania Investigational Drug Service. Aflibercept was provided by Regeneron in single-use vials.
Another US Food and Drug Administration–approved drug for macular edema due to central retinal vein occlusion, ranibizumab, was not investigated in this trial. Although similar efficacy and safety of ranibizumab compared with bevacizumab have been demonstrated in AMD and DME, the cost of ranibizumab ($1950/dose) is much higher than the cost of bevacizumab, and its target specificity is similar to that of bevacizumab (both drugs inhibit VEGF-A). In contrast, aflibercept has a broader mechanism of action (inhibits VEGF-A, VEGF-B, and placental growth factor) and a higher in vitro binding affinity than bevacizumab and ranibizumab and was associated with visual outcomes superior to those associated with bevacizumab and ranibizumab at 12 months in a clinical trial for DME.
Randomization to intravitreal bevacizumab or intravitreal aflibercept was performed centrally through a web-based data entry system maintained at the data coordinating center (The Emmes Corporation) using a permuted block design with random block sizes of 4 and 8, stratified into 1 of 3 baseline screening VA groups in the same manner as in the SCORE Study: good (VALS of 73-59 [approximate Snellen VA 20/40 to 20/63]), moderate (VALS of 58-49 [approximate Snellen VA 20/80 to 20/100]), and poor (VALS of 48-19 [approximate Snellen VA 20/125 to 20/400]).
Study visits were scheduled every 4 weeks for 6 months. Following primary outcome assessment at month 6, participants were followed through month 12 according to a protocol-defined treatment strategy based on response at month 6. Eyes with a protocol-defined good response at month 6 were randomized to continue treatment with the originally assigned drug on either a monthly or treat-and-extend schedule. Eyes with a protocol-defined poor response at month 6 were switched to an alternative treatment; eyes in the bevacizumab group with a poor response at month 6 were to receive aflibercept treatment, and eyes in the aflibercept group with a poor response at month 6 were to receive intravitreal dexamethasone implant treatment. This article focuses on the first 6 months of follow-up. At each study follow-up visit, both eyes had VA assessed by the E-ETDRS method, intraocular pressure (IOP) measurement, and slitlamp and dilated ophthalmoscopic examinations. Images from SD-OCT were obtained monthly. Fundus photographs and (at selected sites) ultra–wide-field fluorescein angiograms were obtained at baseline and month 6. Study participants were masked to treatment assignment through month 6. At month 6, VA examiners and SD-OCT technicians were masked to treatment assignments.
Outcomes
The prespecified primary outcome was change in best-corrected E-ETDRS VALS from baseline to month 6, with the noninferiority margin set at 5 letters. The noninferiority margin was selected in part because it was the same margin used in prior trials evaluating anti-VEGF agents for retinal disease. Further, it has been reported that a true difference exceeding 5 letters represents a clinically important difference. Secondary outcomes include mean change from baseline in VALS at other time points, percentage of participants with a VALS decrease from baseline of at least 15, percentage of participants with a VALS gain from baseline of at least 15, and mean change from baseline in central subfield thickness. Post hoc outcomes include the proportion of eyes with a VALS of 70 or better and the proportion of eyes with resolution of macular edema (defined as central subfield thickness <300 µm, no subretinal or intraretinal fluid, and no cystoid spaces within the ETDRS grid based on reading center evaluation). The secondary VA outcomes were selected because a VALS gain or decrease of at least 15 and meeting a VALS of 70 represent clinically meaningful outcomes to the patient. The SD-OCT secondary outcomes of mean change before and after treatment as well as the proportion of eyes with resolution of macular edema were included because they are important measures used by ophthalmologists in assessing treatment response and need for additional treatment in patients with macular edema.
The statistical analysis plan (Supplement) describes other secondary efficacy outcomes not included in this article. These other secondary outcomes include VA categorized into improved, stable, and worse, overall and within baseline VA strata; SD-OCT outcomes of center point thickness, macular volume, measurement of photoreceptor length, outer segment length, integrity of the photoreceptor inner segment–outer segment junction, and percentage change from baseline in calculated retinal thickness at the center of the macula; fundus photography outcome of area of retinal hemorrhage; ultra–wide-field fluorescein angiography outcome of area of peripheral retinal nonperfusion and leakage; and vision-related quality of life as measured by the National Eye Institute 25-Item Visual Function Questionnaire. These outcomes, including data from the analyses at 12 months, will be included in subsequent reports.
Medical Dictionary for Regulatory Activities coding of adverse events was used to select events specified by the Antiplatelet Trialists’ Collaboration.
Statistical Analysis
The target sample size of 360 study eyes, calculated to attain at least 80% power, was based on testing noninferiority of bevacizumab to aflibercept using a VALS margin of 5 and a 1-tailed type I error of .025, assuming an SD of 16 letters for change from baseline in VALS (based on a prior study) and 10% attrition by month 6. Interim efficacy testing was carried out using the Lan-DeMets interim monitoring boundary with a 1-tailed level .025 O’Brien-Fleming–type spending function, adapted for noninferiority testing. The DSMC reviewed outcome data by treatment group, but no formal statistical tests were examined by the DSMC until the sample size reestimation was performed. The DSMC recommended no modification to this trial’s sample size or early stopping due to efficacy based on these analyses.
The noninferiority analysis for the primary outcome was performed based on the treatment group to which study participants were randomized, consistent with intention-to-treat principles. The noninferiority test modeled baseline and 6-month VALS as a 2-step time series in which each 6-month outcome is correlated with its corresponding baseline measure. Sensitivity analysis used multiple imputation modified to explore missing-at-random and missing-not-at-random hypotheses to assess the effect of missing data on the primary outcome finding. The multiple imputation technique was used based on a model that imputed missing month 6 VALS, based on treatment assignment and nonmissing VALS from baseline through month 5, and central subfield thickness values from baseline through month 6. To assess the potential influence of conflict of interest on the primary outcome findings, a secondary analysis of the primary outcome was performed that included a covariate to identify patients from sites with investigators who reported a financial conflict of interest.
Confidence intervals were based on t distributions for means and mean changes and Wald approximation for percentages. P values for analysis of baseline VA stratum (good, moderate, poor), anti-VEGF treatment for macular edema prior to baseline (yes, no), and disease status (central retinal or hemiretinal vein occlusion) subgroups were based on normal-based and logistic regression mixed models in which month 6 data were regressed on treatment assignment, subgroup indicator, and treatment-by-subgroup interaction. Prior anti-VEGF treatment and disease status variables were not listed in the statistical analysis plan, and analyses of these variables need to be interpreted as post hoc. In the monthly analyses, participant-level temporal autocorrelation is modeled as an autoregressive time series, which included a time-by-treatment interaction. P values tested the unweighted average across 6 months of the estimated treatment effect in each month. Controlling for multiple testing of secondary findings was accomplished by adjusting P values using the Hochberg sequentially rejective method. Adjusted 2-tailed P < .05 was considered statistically significant in secondary analyses. Confidence intervals are uncorrected for multiple testing and intended only to describe the uncertainty in the estimates.
SAS statistical software versions 9.3 and 9.4 (SAS Institute Inc) were used to conduct statistical analyses. Analyses included data available as of December 30, 2016.
Results
Demographic Characteristics and Follow-up
Three hundred sixty-two participants were enrolled at 66 private practice or academic centers within the United States between September 17, 2014, and November 18, 2015, and randomly assigned to receive aflibercept (n = 180) or bevacizumab (n = 182). Participants’ mean (SD) age was 69 (12) years, 157 (43.4%) were women, 276 (76.2%) were white, and 54 (14.9%) were black. The mean (SD) VALS at baseline was 50.3 (15.2) (approximate Snellen VA 20/100), and participants had macular edema for a mean of 7 months (range, 0-104 months; median, 1 month [interquartile range, 0-6 months]) before randomization. The mean (SD) central subfield thickness was 665.5 (224.4) µm. Among the 362 participants, 121 (33.4%) had received prior anti-VEGF treatment, 28 (7.7%) had prior intravitreal steroid treatment, and 57 (15.7%) were diagnosed with hemiretinal vein occlusion (Table 1). The month 6 visit was completed by 348 participants (96.1%), including 175 of the 180 participants (97.2%) in the aflibercept group and 173 of the 182 participants (95.1%) in the bevacizumab group (Figure 1).
Table 1. Baseline Characteristics of SCORE2 Participants.
| Characteristic | Aflibercept (n = 180) |
Bevacizumab (n = 182) |
Total (N = 362) |
|---|---|---|---|
| Age | |||
| Mean (SD), y | 69 (11) | 69 (13) | 69 (12) |
| No. (%) | |||
| <50 y | 7 (3.9) | 15 (8.2) | 22 (6.1) |
| 50 to <60 y | 28 (15.6) | 26 (14.3) | 54 (14.9) |
| 60 to <70 y | 59 (32.8) | 48 (26.4) | 107 (29.6) |
| 70 to <80 y | 58 (32.2) | 52 (28.6) | 110 (30.4) |
| ≥80 y | 28 (15.6) | 41 (22.5) | 69 (19.1) |
| Women, No. (%) | 82 (45.6) | 75 (41.2) | 157 (43.4) |
| Race, No. (%) | |||
| White | 131 (72.8) | 145 (79.7) | 276 (76.2) |
| Black | 28 (15.6) | 26 (14.3) | 54 (14.9) |
| Other | 21 (11.7) | 11 (6.0) | 32 (8.8) |
| Hispanic or Latino, No. (%) | 16 (8.9) | 22 (12.1) | 38 (10.5) |
| VALS of study eye, mean (SD) | 50.3 (15.2) | 50.4 (15.3) | 50.3 (15.2) |
| VA strata | |||
| Good, VALS 59-73, Snellen equivalent 20/40 to 20/63 | |||
| No. (%) | 65 (36.1) | 67 (36.8) | 132 (36.5) |
| VALS, mean (SD) | 65.7 (4) | 65.9 (4) | 65.8 (4) |
| Moderate, VALS 49-58, Snellen equivalent 20/80 to 20/100 | |||
| No. (%) | 42 (23.3) | 42 (23.1) | 84 (23.2) |
| VALS, mean (SD) | 52.7 (3) | 53.3 (3) | 53.0 (3) |
| Poor, VALS 19-48, Snellen equivalent 20/125 to 20/400 | |||
| No. (%) | 73 (40.6) | 73 (40.1) | 146 (40.3) |
| VALS, mean (SD) | 35.1 (10) | 34.6 (9) | 34.8 (9) |
| Time between diagnosis of macular edema and randomization | |||
| Mean (SD), mo | 8 (17) | 5 (10) | 7 (14) |
| No. (%) | |||
| <3 mo | 114 (63.3) | 128 (70.3) | 242 (66.9) |
| 3-6 mo | 18 (10.0) | 12 (6.6) | 30 (8.3) |
| 7-12 mo | 17 (9.4) | 18 (9.9) | 35 (9.7) |
| >12 mo | 31 (17.2) | 24 (13.2) | 55 (15.2) |
| Median (IQR), mo | 1 (0-7) | 0 (0-5) | 1 (0-6) |
| SD-OCT central subfield thickness, mean (SD), µma | 652.4 (214.6) | 678.2 (233.3) | 665.5 (224.4) |
| Prior anti-VEGF treatment, No. (%) | 65 (36.1) | 56 (30.8) | 121 (33.4) |
| Prior intravitreal steroid use, No. (%) | 16 (8.9) | 12 (6.6) | 28 (7.7) |
| Hemiretinal vein occlusion, No. (%) | 26 (14.4) | 31 (17.0) | 57 (15.7) |
| Lens status, No. (%) | |||
| Cataract extraction | 43 (23.9) | 55 (30.2) | 98 (27.1) |
| Natural lens | |||
| History of cataract | 108 (60.0) | 95 (52.2) | 203 (56.1) |
| No history of cataract | 29 (16.1) | 32 (17.6) | 61 (16.9) |
| Diabetes mellitus, No. (%) | |||
| Type 1 | 1 (0.6) | 0 | 1 (0.3) |
| Type 2 | 54 (30.0) | 59 (32.4) | 113 (31.2) |
| Hypertensive, No. (%) | 140 (77.8) | 138 (75.8) | 278 (76.8) |
| Coronary artery disease, No. (%) | 26 (14.4) | 30 (16.5) | 56 (15.5) |
| NEI VFQ-25 overall score, mean (SD)b | 77 (15) | 77 (17) | 77 (16) |
Abbreviations: IQR, interquartile range; NEI VFQ-25, National Eye Institute 25-Item Visual Function Questionnaire; SD-OCT, spectral-domain optical coherence tomography; VA, visual acuity; VALS, VA letter score; VEGF, vascular endothelial growth factor.
Spectral-domain OCT displays cross-sectional images of the retina and uses interferometry to provide cross-sectional measurement of the retinal thickness. Spectral-domain OCT uses a Fourier transformation process, allowing for faster scanning time, more scans per unit of area, and fewer artifacts compared with earlier OCT technology.
The NEI VFQ-25 overall score ranges from 0 to 100, with higher scores representing better functioning.
Anti-VEGF Treatments
The mean (SD) number of anti-VEGF injections between randomization and month 5, the last visit with an injection prior to measurement of the primary outcome, was 5.83 (0.64) in the aflibercept group and 5.78 (0.74) in the bevacizumab group (P = .47), with 164 participants (91.1%) in the aflibercept group and 160 participants (87.9%) in the bevacizumab group receiving all 6 expected monthly injections between randomization and month 5 (P = .32).
VA Outcomes
The mean VALS improved from 50.3 at baseline to 64.1 at month 1 and to 69.3 at month 6 in the aflibercept group, and improved from 50.4 at baseline to 62.3 at month 1 and to 69.3 at month 6 in the bevacizumab group (Figure 2). At month 6, bevacizumab was noninferior to aflibercept based on a VALS margin of 5 (bevacizumab minus aflibercept mean difference, −0.14; 1-tailed 97.5% CI, −3.07 to ∞; P for noninferiority = .001; within the noninferiority margin of −5). Missing month 6 outcomes (5 of 180 participants [2.8%] in the aflibercept group; 9 of 182 participants [4.9%] in the bevacizumab group) were ignored in the primary analysis, consistent under a maximum likelihood analysis with the missing-at-random assumption. In the missing-at-random analysis, the bevacizumab minus aflibercept mean difference was −0.49 (1-tailed 97.5% CI, −3.39 to ∞; P for noninferiority = .001). A sensitivity analysis used multiple imputation modified to explore not-missing-at-random hypotheses and showed that the primary analysis was robust to moderate departures from missing at random. In a secondary analysis, after adjusting for all factors in Table 1, the noninferiority result remained (bevacizumab minus aflibercept mean difference, −0.72; 1-tailed 97.5% CI, −3.87 to ∞; P for noninferiority = .01). In another secondary analysis in which sites were grouped according to whether an investigator affirmed a financial conflict of interest or not and in which the primary outcome analysis was repeated including in the model an indicator variable for conflict of interest and the conflict-by-treatment interaction, there was neither a significant effect of conflict (P = .73) nor a conflict × treatment interaction (P = .99).
Figure 2. Electronic Early Treatment Diabetic Retinopathy Study Visual Acuity Letter Score at Baseline and Monthly Through Month 6.
The center horizontal line of each box indicates the median; circle, mean; and top and bottom borders of each box, 75th and 25th percentiles, respectively. The whiskers extend to the minimum observation above the lower fence and maximum observation below the upper fence, which are defined as 1.5 times the interquartile range below the 25th percentile and above the 75th percentile, respectively. The outliers (squares) mark any observation below the lower fence or above the upper fence.
In the aflibercept group, 114 eyes (65.1%) had a VALS gain of at least 15 at month 6 compared with 106 eyes (61.3%) in the bevacizumab group. The odds of a VALS gain of at least 15 averaged over months 1 to 6 was not significantly different for aflibercept relative to bevacizumab (odds ratio [OR] = 0.85; 95% CI, 0.62-1.17; P = .89). Less than 2% in each group had a VALS decrease of at least 15 at month 6 (3 of 175 eyes [1.7%] in the aflibercept group; 3 of 173 eyes [1.7%] in the bevacizumab group). In post hoc analysis, the number of eyes achieving a VALS of at least 70 (approximate Snellen VA 20/40) at month 6 was 101 (57.7%) in the aflibercept group and 99 (57.2%) in the bevacizumab group, with the odds of a VALS of at least 70 over months 1 to 6 for the aflibercept group not significantly different from those of the bevacizumab group (OR = 0.82; 95% CI, 0.59-1.13; P = .89). The treatment effect comparing bevacizumab with aflibercept was homogeneous in the identified subgroups (Table 2).
Table 2. Electronic ETDRS VALS Secondary Outcomes, Overall and Within Subgroup Through Month 6.
| Participants | Participants Attending Visit, No. | Change From Baseline in VALS | Study Eyes With VALS of 70 (Snellen Equivalent 20/40) or Better, No. (% [95% CI])a,b | |||||
|---|---|---|---|---|---|---|---|---|
| Mean (95% CI)a | Study Eyes With Gain of ≥15 Letters in VALS, No. (% [95% CI])a | |||||||
| Aflibercept | Bevacizumab | Aflibercept | Bevacizumab | Aflibercept | Bevacizumab | Aflibercept | Bevacizumab | |
| All participants by month | ||||||||
| 1 | 179 | 179 | 13.94 (12.26-15.63) | 11.73 (9.77-13.68) | 75 (41.9 [34.7-49.1]) | 65 (36.3 [29.3-43.4]) | 89 (49.7 [42.4-57.0]) | 70 (39.1 [32.0-46.3]) |
| 2 | 177 | 179 | 16.63 (14.74-18.53) | 14.29 (11.93-16.65) | 93 (52.5 [45.2-59.9]) | 86 (48.0 [40.7-55.4]) | 99 (55.9 [48.6-63.2) | 82 (45.8 [38.5-53.1]) |
| 3 | 175 | 175 | 17.54 (15.55-19.53) | 16.29 (13.97-18.61) | 92 (52.6 [45.2-60.0]) | 96 (54.9 [47.5-62.2]) | 92 (52.6 [45.2-60.0]) | 90 (51.4 [44.0-58.8]) |
| 4 | 173 | 172 | 17.87 (15.61-20.12) | 17.48 (15.07-19.88) | 108 (62.4 [55.2-69.6]) | 93 (54.1 [46.6-61.5]) | 101 (58.4 [51.0-65.7]) | 97 (56.4 [49.0-63.8]) |
| 5 | 169 | 171 | 19.15 (17.01-21.30) | 17.49 (15.04-19.94) | 103 (60.9 [53.6-68.3]) | 99 (57.9 [50.5-65.3]) | 105 (62.1 [54.8-69.4]) | 93 (54.4 [46.9-61.9]) |
| 6 | 175 | 173 | 18.89 (16.62-21.15) | 18.62 (16.16-21.09) | 114 (65.1 [58.1-72.2]) | 106 (61.3 [54.0-68.5]) | 101 (57.7 [50.4-65.0]) | 99 (57.2 [49.9-64.6]) |
| Statistical testc | Estimate of bevacizumab minus aflibercept treatment effect averaged over months 1-6, −1.52 (95% CI, −4.21 to 1.17; P = .89)a | OR = 0.85 (95% CI, 0.62-1.17; P = .89)a,d; RD, −3.85% (95% CI, −7.54% to −0.15%)e | OR = 0.82 (95% CI, 0.59-1.13; P = .89)a,d; RD, −5.35% (95% CI, −10.25% to −0.46%)e | |||||
| Month 6 VALS by baseline VA stratum | ||||||||
| Good, VALS 73-59, Snellen equivalent 20/40 to 20/63 | 64 | 65 | 14.73 (12.29-17.18) | 12.35 (9.77-14.93) | 38 (59.4 [47.3-71.4]) | 31 (47.7 [35.6-59.8]) | 56 (87.5 [79.4-95.6]) | 54 (83.1 [74.0-92.2]) |
| Moderate, VALS 58-49, Snellen equivalent 20/80 to 20/100 | 41 | 40 | 21.02 (17.03-25.02) | 15.90 (10.95-20.85) | 27 (65.9 [51.3-80.4]) | 24 (60.0 [44.8-75.2]) | 26 (63.4 [48.7-78.2]) | 21 (52.5 [37.0-68.0]) |
| Poor, VALS 48-19, Snellen equivalent 20/125 to 20/400 | 70 | 68 | 21.43 (16.82-26.04) | 26.22 (21.69-30.75) | 49 (70.0 [59.3-80.7]) | 51 (75.0 [64.7-85.3]) | 19 (27.1 [16.7-37.6]) | 24 (35.3 [23.9-46.7]) |
| P value for baseline VA stratum × treatment group interactionc | .80 | .89 | .89 | |||||
| Month 6 VALS by status of prior treatment with anti-VEGF at baseline | ||||||||
| Prior anti-VEGF treatment | 63 | 53 | 15.17 (11.78-18.57) | 13.98 (9.93-18.03) | 32 (50.8 [38.4-63.1]) | 30 (56.6 [43.3-69.9]) | 32 (50.8 [38.4-63.1]) | 24 (45.3 [31.9-58.7]) |
| No prior anti-VEGF treatment | 112 | 120 | 20.97 (18.03-23.91) | 20.68 (17.64-23.71) | 82 (73.2 [65.0-81.4]) | 76 (63.3 [54.7-72.0]) | 69 (61.6 [52.6-70.6]) | 75 (62.5 [53.8-71.2]) |
| P value for status of prior treatment with anti-VEGF at baseline × treatment group interactionc | .89 | .89 | .89 | |||||
| Month 6 VALS, by disease type | ||||||||
| Central retinal vein occlusion | 149 | 142 | 18.21 (15.71-20.72) | 18.84 (16.01-21.67) | 93 (62.4 [54.6-70.2]) | 85 (59.9 [51.8-67.9]) | 84 (56.4 [48.4-64.3]) | 75 (52.8 [44.6-61.0]) |
| Hemiretinal vein occlusion | 26 | 31 | 22.73 (17.50-27.96) | 17.65 (12.66-22.63) | 21 (80.8 [65.6-95.9]) | 21 (67.7 [51.3-84.2]) | 17 (65.4 [47.1-83.7]) | 24 (77.4 [62.7-92.1]) |
| P value for disease type × treatment group interactionc | .89 | .89 | .89 | |||||
Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; OR, odds ratio; RD, risk difference; VA, visual acuity; VALS, VA letter score; VEGF, vascular endothelial growth factor.
Confidence intervals were not adjusted for multiple testing.
Post hoc analysis.
P values were adjusted for multiple comparisons using the Hochberg sequentially rejective method, considering Table 2 and Table 3 jointly in a single Hochberg procedure.
Aflibercept was the reference group. Odds ratios were generated by a contrast that incorporates data from all 6 of the monthly visits.
Risk differences were calculated as the unweighted average of percentage in bevacizumab cell means minus average of percentage in aflibercept cell means from the data from each of the monthly visits.
SD-OCT Outcomes
Both groups showed statistically significant SD-OCT central subfield thickness decreases from baseline through month 6 (Table 3 and Figure 3). With a mean (SD) baseline central subfield thickness of 652 (215) µm in the aflibercept group, there was a mean decrease of 394 µm (95% CI, −429 to −360 µm) at month 1 and 425 µm (95% CI, −461 to −389 µm) at month 6. For the bevacizumab group, with a mean (SD) baseline central subfield thickness of 678 (233) µm, there was a mean decrease of 333 µm (95% CI, −366 to −299 µm) at month 1 and 387 µm (95% CI, −426 to −348 µm) at month 6. The bevacizumab minus aflibercept estimate of the change from baseline treatment effect averaged over the 6 months was 49.3 µm (95% CI, 1.93 to 96.74 µm) in favor of aflibercept, a difference that is not statistically significant (P = .83).
Table 3. Spectral-Domain Optical Coherence Tomography Central Subfield Thickness Secondary Outcomes, Overall and Within Subgroup Through Month 6.
| All Study Eyes | Participants With Central Subfield Thickness Measurement, No. | Mean (95% CI), μma | Study Eyes With Central Subfield Thickness <300 μm, No Subretinal Fluid, No Intraretinal Fluid, and No Cystoid Spaces, No./Total No. (% [95% CI])a,b | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Central Subfield Thickness | Change in Central Subfield Thickness | ||||||||
| Aflibercept | Bevacizumab | Aflibercept | Bevacizumab | Aflibercept | Bevacizumab | Aflibercept | Bevacizumab | ||
| Baseline | 173 | 178 | 652.4 (620.2 to 684.6) | 678.2 (643.7 to 712.7) | |||||
| Month | |||||||||
| 1 | 167 | 171 | 260.4 (249.9 to 270.9) | 340.8 (319.1 to 362.6) | −394 (−429 to −360) | −333 (−366 to −299) | 63/170 (37.1 [29.8 to 44.3]) | 24/171 (14.0 [8.8 to 19.2]) | |
| 2 | 169 | 176 | 240.3 (231.5 to 249.0) | 311.9 (292.2 to 331.6) | −418 (−453 to −383) | −368 (−405 to −332) | 80/171 (46.8 [39.3 to 54.3]) | 37/178 (20.8 [14.8 to 26.7]) | |
| 3 | 167 | 174 | 238.2 (228.3 to 248.9) | 304.3 (284.1 to 324.6) | −420 (−456 to −384) | −369 (−405 to −332) | 96/170 (56.5 [49.0 to 63.9]) | 46/174 (26.4 [19.9 to 33.0]) | |
| 4 | 164 | 169 | 233.7 (223.6 to 243.7) | 299.3 (279.7 to 318.9) | −421 (−457 to −386) | −378 (−415 to −340) | 96/168 (57.1 [49.7 to 64.6]) | 39/171 (22.8 [16.5 to 29.1]) | |
| 5 | 161 | 166 | 229.7 (219.7 to 239.6) | 308.8 (284.9 to 332.8) | −429 (−466 to −393) | −365 (−406 to −324) | 93/164 (56.7 [49.1 to 64.3]) | 47/170 (27.6 [20.9 to 34.4]) | |
| 6 | 164 | 171 | 231.3 (220.8 to 241.8) | 287.9 (269.4 to 306.4) | −425 (−461 to −389) | −387 (−426 to −348) | 92/169 (54.4 [46.9 to 61.9]) | 49/172 (28.5 [21.7 to 35.2]) | |
| Statistical testc | Estimate of bevacizumab minus aflibercept treatment effect averaged over months 1-6, 68.6 (95% CI, 50.79 to 86.50; P < .001)a | Estimate of bevacizumab minus aflibercept treatment effect averaged over months 1-6, 49.3 (95% CI, 1.93 to 96.74; P = .83)a | OR = 0.28 (95% CI, 0.20 to 0.39; P < .001)a,d; RD = −28.07 (95% CI, −32.22 to −23.91)a,e | ||||||
| Month 6 central subfield thickness by baseline VA stratum | |||||||||
| Good, VALS 73-59, Snellen equivalent 20/40 to 20/63 | |||||||||
| Baseline | 567.4 (527.8 to 607.0) | 553.3 (507.5 to 599.2) | |||||||
| Month 6 | 61 | 65 | 235.4 (225.7 to 245.0) | 283.7 (261.3 to 306.2) | −338 (−380 to −296) | −268 (−323 to −212) | 32/62 (51.6 [39.2 to 64.1]) | 18/65 (27.7 [16.8 to 38.6]) | |
| Moderate, VALS 58-49, Snellen equivalent 20/80 to 20/100 | |||||||||
| Baseline | 645.7 (587.7 to 703.7) | 652.7 (593.2 to 712.2) | |||||||
| Month 6 | 38 | 39 | 229.6 (216.4 to 242.8) | 293.2 (252.4 to 334.0) | −416 (−475 to −358) | −359 (−420 to −299) | 28/38 (73.7 [59.7 to 87.7]) | 14/39 (35.9 [20.8 to 51.0]) | |
| Poor, VALS 48-19, Snellen equivalent 20/125 to 20/400 | |||||||||
| Baseline | 739.2 (676.2 to 802.2) | 801.0 (742.5 to 859.5) | |||||||
| Month 6 | 65 | 67 | 228.5 (204.3 to 252.7) | 288.9 (253.1 to 324.6) | −515 (−586 to −444) | −517 (−583 to −451) | 32/69 (46.4 [34.6 to 58.1]) | 17/68 (25.0 [14.7 to 35.3]) | |
| P value for baseline VA stratum × treatment interactionc | .89 | .89 | .89 | ||||||
| Month 6 central subfield thickness values by status of prior treatment with anti-VEGF at baseline | |||||||||
| Prior anti-VEGF treatment | |||||||||
| Baseline | 608.5 (559.1 to 657.9) | 661.1 (589.4 to 732.8) | |||||||
| Month 6 | 56 | 52 | 235.9 (208.5 to 263.3) | 286.0 (248.9 to 323.0) | −370 (−424 to −316) | −375 (−454 to −297) | 19/61 (31.1 [19.5 to 42.8]) | 8/52 (15.4 [5.6 to 25.2]) | |
| No prior anti-VEGF treatment | |||||||||
| Baseline | 678.5 (634.5 to 722.5) | 680.4 (639.6 to 721.1) | |||||||
| Month 6 | 108 | 119 | 228.9 (221.1 to 236.8) | 288.8 (267.4 to 310.2) | −453 (−500 to −406) | −392 (−437 to −347) | 73/108 (67.6 [58.8 to 76.4]) | 41/120 (34.2 [25.7 to 42.7]) | |
| P value for status of prior treatment with anti-VEGF at baseline × treatment interactionc | .89 | .89 | .89 | ||||||
| Month 6 central subfield thickness values by disease type | |||||||||
| Central retinal vein occlusion | |||||||||
| Baseline | 653.7 (617.9 to 689.5) | 700.8 (661.2 to 740.5) | |||||||
| Month 6 | 139 | 140 | 233.0 (220.9 to 245.1) | 289.1 (268.4 to 309.9) | −424 (−463 to −385) | −413 (−457 to −369) | 79/144 (54.9 [46.7 to 63.0]) | 45/141 (31.9 [24.2 to 39.6]) | |
| Hemiretinal vein occlusion | |||||||||
| Baseline | 653.4 (553.8 to 753.1) | 552.2 (486.7 to 617.7) | |||||||
| Month 6 | 25 | 31 | 222.1 (206.6 to 237.6) | 282.4 (239.6 to 325.2) | −431 (−534 to −329) | −266 (−344 to −189) | 13/25 (52.0 [32.4 to 71.6]) | 4/31 (12.9 [1.1 to 24.7]) | |
| P value for disease type × treatment interaction | .89 | .75 | .89 | ||||||
Abbreviations: OR, odds ratio; RD, risk difference; VA, visual acuity; VALS, VA letter score; VEGF, vascular endothelial growth factor.
Confidence intervals were not adjusted for multiple testing.
Post hoc analysis.
P values were adjusted for multiple comparisons using the Hochberg sequentially rejective method, considering Table 2 and Table 3 jointly in a single Hochberg procedure.
Aflibercept was the reference group. Odds ratios were generated by a contrast that incorporates data from all 6 of the monthly visits.
Risk differences were calculated as the unweighted average of percentage in bevacizumab cell means minus average of percentage in aflibercept cell means from the data from each of the monthly visits.
Figure 3. Spectral-Domain Optical Coherence Tomography Central Subfield Thickness at Baseline and Monthly Through Month 6.
The center horizontal line of each box indicates the median; circle, mean; and top and bottom borders of each box, 75th and 25th percentiles, respectively. The whiskers extend to the minimum observation above the lower fence and maximum observation below the upper fence, which are defined as 1.5 times the interquartile range below the 25th percentile and above the 75th percentile, respectively. The outliers (squares) mark any observation below the lower fence or above the upper fence.
Post hoc analyses showed that at month 6, 92 of 169 eyes (54.4%) in the aflibercept group had resolution of macular edema compared with 49 of 172 eyes (28.5%) in the bevacizumab group. The odds of complete resolution of fluid averaged over months 1 to 6 was significantly lower in the bevacizumab group compared with the aflibercept group (OR = 0.28; 95% CI, 0.20 to 0.39; P < .001). The treatment effect comparing bevacizumab with aflibercept was homogeneous in the identified subgroups (Table 3). An example of both baseline and month 6 color fundus photographs and SD-OCT scans in a participant in this trial is shown in Figure 4.
Figure 4. Color Fundus Photograph and Optical Coherence Tomogram of the Left Eye at Baseline and Month 6 in a Representative SCORE2 Participant With a Central Retinal Vein Occlusion.
At baseline, the color fundus photograph of the central 30° of the retina showed extensive retinal hemorrhages and dilated retinal veins. The optic nerve was not visible because of retinal edema and blood. Retinal edema was present throughout the retina and is best seen on optical coherence tomography (OCT), which provides a cross-sectional image of the retina. The OCT scan shown here was taken through the center of the retina. The participant’s visual acuity was 20/400 at baseline. At month 6, the color fundus photograph of the retina in the same participant showed nearly complete resolution of intraretinal blood after receiving monthly injections of aflibercept for 6 months. The optic nerve was visible and appeared normal. The OCT scan showed that retinal edema resolved and the fovea returned to its normal contour. At month 6, the participant’s visual acuity improved to 20/40.
Adverse Events
Most ocular adverse events of interest were rare (Table 4). One case of endophthalmitis (culture negative) occurred in the bevacizumab group. Two participants receiving bevacizumab had IOP higher than 35 mm Hg at a monthly visit through month 6, and 4 participants (2.2%) receiving aflibercept and 9 (4.9%) receiving bevacizumab had IOP more than 10 mm Hg greater than baseline, with 1 participant receiving bevacizumab experiencing angle-closure glaucoma not attributed to the anti-VEGF agent or procedure. Antiplatelet Trialists’ Collaboration events occurred in 2 participants (1.1%) receiving aflibercept (including a fatal myocardial infarction) and 2 participants (1.1%) receiving bevacizumab. Two deaths (myocardial infarction and metastatic bladder cancer) occurred, 1 in each group, through 6 months.
Table 4. Ocular and Systemic Events Within 6 Months Among Study Eyes.
| Event | Study Eyes, No. (%) | ||
|---|---|---|---|
| Aflibercept (n = 180) |
Bevacizumab (n = 182) |
Total (N = 362) |
|
| Ocular events | |||
| Elevated IOP | |||
| >35 mm Hg | 0 | 2 (1.1) | 2 (0.6) |
| >10 mm Hg greater than baseline | 4 (2.2) | 9 (4.9) | 13 (3.6) |
| Ocular adverse events | |||
| Culture-negative endophthalmitis | 0 | 1 (0.5) | 1 (0.3) |
| Angle-closure glaucoma | 0 | 1 (0.5) | 1 (0.3) |
| Neovascular glaucoma | 1 (0.6) | 0 | 1 (0.3) |
| Ischemic central retinal vein occlusion | 1 (0.6) | 0 | 1 (0.3) |
| Central retinal artery nonperfusion | 0 | 1 (0.5) | 1 (0.3) |
| Retinal arterial occlusion | 0 | 2 (1.1) | 2 (0.6) |
| Retinal detachment | 0 | 1 (0.5) | 1 (0.3) |
| Vitreous hemorrhage | 0 | 1 (0.5) | 1 (0.3) |
| Any ocular adverse events | 2 (1.1) | 6 (3.3) | 8 (2.2) |
| Ocular surgery in study eye | |||
| YAG capsulotomy | 1 (0.6) | 0 | 1 (0.3) |
| Glaucoma surgery with laser | 0 | 1 (0.5) | 1 (0.3) |
| Cataract extraction | 1 (0.6) | 1 (0.5) | 2 (0.6) |
| Pars plana vitrectomy | 0 | 1 (0.5) | 1 (0.3) |
| Peripheral iridotomy | 0 | 1 (0.5) | 1 (0.3) |
| Any ocular surgery | 2 (1.1) | 3 (1.6) | 5 (1.4) |
| Systemic events | |||
| APTC events | |||
| Nonfatal myocardial infarction | |||
| Acute myocardial infarction | 0 | 1 (0.5) | 1 (0.3) |
| Myocardial infarction | 1 (0.6) | 1 (0.5) | 2 (0.6) |
| Nonfatal stroke | |||
| Cerebrovascular accident | 1 (0.6) | 0 | 1 (0.3) |
| Vascular death | |||
| Myocardial infarction | 1 (0.6) | 0 | 1 (0.3) |
| Any APTC event | 2 (1.1) | 2 (1.1) | 4 (1.1) |
| Adverse and serious adverse events, ocular and systemic, not limited to study eye | |||
| Death from any cause | 1 (0.6) | 1 (0.5) | 2 (0.6) |
| Participants with any adverse event | 82 (45.6) | 98 (53.8) | 180 (49.7) |
| Total No. of adverse events | 184 | 263 | 447 |
| Participants with any serious adverse eventa | 14 (7.8) | 14 (7.7) | 28 (7.8) |
| Total No. of serious adverse eventsa | 18 | 25 | 43 |
Abbreviations: APTC, Antiplatelet Trialists’ Collaboration; IOP, intraocular pressure.
An adverse event or suspected adverse reaction is considered serious if it results in any of the following outcomes: death, a life-threatening adverse event, inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant incapacity or substantial disruption of the ability to conduct normal life functions, a congenital anomaly or birth defect, or an important medical event that may require medical or surgical intervention to prevent one of the noted outcomes.
Discussion
In this study of patients with macular edema due to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to intravitreal aflibercept for VA after 6 months of treatment, based on a VALS noninferiority margin of 5. In secondary analyses, there were no significant differences between bevacizumab and aflibercept in terms of VA throughout 6 months of follow-up. The magnitude of improvement in mean VALS from baseline to month 6 in the aflibercept and bevacizumab groups (18.9 and 18.6 letters, respectively) was comparable with that reported with monthly ranibizumab injections in the CRUISE Study (14.9 letters) and with monthly aflibercept injections in the COPERNICUS (17.3 letters) and GALILEO (18.0 letters) trials. The proportion of eyes with a VALS gain of at least 15 from baseline to month 6 in the aflibercept and bevacizumab groups (65.1% and 61.3%, respectively) is 13% to 17% higher than that reported in CRUISE (47.7%), and more similar to those with gains reported in both the COPERNICUS (56.1%) and GALILEO (60.2%) trials. These trials support the favorable effect of anti-VEGF therapy on vision loss due to macular edema from central retinal vein occlusion. While the COPERNICUS and GALILEO trials included only treatment-naive patients with central retinal vein occlusion, this trial included both treatment-naive patients as well as patients who had received prior anti-VEGF treatment. In this trial, there was no significant difference in treatment effect between treatment-naive and previously treated participants.
In Protocol T conducted by the Diabetic Retinopathy Clinical Research Network (DRCR.net) to evaluate the relative efficacy and safety of aflibercept, bevacizumab, and ranibizumab for the treatment of DME, there was a greater treatment effect on VALS at 1 year associated with aflibercept compared with bevacizumab in the poor VA stratum (20/50 to 20/320) and no difference in treatment effect in the good VA stratum (20/32 to 20/40) (P < .0001); the treatment advantage of aflibercept over bevacizumab among eyes with worse baseline VA persisted at 2 years. The current trial found no such interaction of treatment effect with baseline VA. This difference in findings between this trial and Protocol T may be due to variation in the a priori definition of baseline VA strata (there were only 38 eyes in this trial with VA of 20/40 or better). Additionally, response to treatment may vary based on the retinal disease studied.
Both aflibercept and bevacizumab were associated with significant central subfield thickness reduction, consistent with previous clinical trials of anti-VEGF therapy for macular edema due to central retinal vein occlusion. In this trial, DRCR.net Protocol T, and the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT) (which evaluated the relative efficacy and safety of ranibizumab and bevacizumab for treatment of neovascular AMD), bevacizumab was associated with a significantly lower proportion of eyes that achieved resolution of macular edema than the comparative on-label anti-VEGF agents (ranibizumab in CATT; aflibercept and ranibizumab in Protocol T). In this trial and CATT, the lower proportion of eyes with resolution of fluid among eyes treated with bevacizumab did not translate into poorer VA outcomes at the primary outcome visits of these 2 clinical trials. This was also true at 12 months (primary outcome) in the participants with better baseline VA in Protocol T. Continued follow-up of participants in this trial will allow evaluation of the cumulative effect of the presence of fluid on VA in all treatment groups and on the number of injections administered in participants assigned to the treatment groups not defined by a fixed-dosing schedule.
Baseline characteristics were balanced between the 2 treatment groups except for the shorter time between diagnosis of macular edema and randomization in the bevacizumab group. The extent to which baseline characteristics like duration between diagnosis of macular edema and randomization may confound the primary results was examined through a secondary noninferiority analysis that adjusted for all factors in Table 1. The adjusted secondary analysis supported the noninferiority conclusion reached in the primary analysis.
Rates of adverse events in the current study were similar for both drugs and are consistent with those reported in other phase 3 trials evaluating anti-VEGF therapy for retinal vein occlusion, neovascular AMD, and DME. Although this trial was not powered to identify differences between treatment groups in rare adverse events, new safety concerns associated with anti-VEGF therapy were not identified. The bevacizumab used in this trial was commercially acquired and repackaged into single-use vials by the University of Pennsylvania Investigational Drug Service, which performed a similar service in CATT and DRCR.net Protocol T. The adverse event profile and efficacy associated with commercially available single-use syringes of repackaged bevacizumab may differ from the adverse event profile and efficacy associated with the repackaged bevacizumab study drug evaluated in this trial. However, in a retrospective cohort study using medical claims data from ambulatory care centers across the United States from 2005 through 2012, the rate of postinjection endophthalmitis was not significantly different among the 296 565 injections of compounded bevacizumab packaged into single-use syringes compared with the 87 245 injections of ranibizumab packaged in single-use glass vials.
Cost is one of many factors that may contribute to drug selection when treating patients. The cost of a single dose of aflibercept is substantially greater than a single dose of off-label repackaged bevacizumab, and because most patients treated with anti-VEGF therapy for macular edema associated with central retinal or hemiretinal vein occlusion are treated with multiple doses of anti-VEGF drug, the cost differential between the 2 drugs has important economic implications.
Limitations
Limitations of the current trial include lack of a ranibizumab treatment group, lack of as-needed treatment groups that were evaluated from baseline, and relatively short follow-up. Caution should be exercised when extrapolating results from clinical trials to clinical practice. For instance, findings of the current trial must be interpreted in light of the high rate of treatment adherence, with approximately 90% of study participants receiving all 6 of the monthly anti-VEGF injections. Cross-trial comparisons include such caveats as differences among trials in the timing of outcome assessments; for instance, the primary outcome of the current trial is 6 months, compared with 1 year in CATT and DRCR.net Protocol T.
Conclusions
Among patients with macular edema secondary to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to aflibercept with respect to VA after 6 months of treatment.
Trial Protocol
References
- 1.Ponto KA, Elbaz H, Peto T, et al. Prevalence and risk factors of retinal vein occlusion: the Gutenberg Health Study. J Thromb Haemost. 2015;13(7):1254-1263. [DOI] [PubMed] [Google Scholar]
- 2.Klein R, Klein BE, Moss SE, Meuer SM. The epidemiology of retinal vein occlusion: the Beaver Dam Eye Study. Trans Am Ophthalmol Soc. 2000;98:133-141. [PMC free article] [PubMed] [Google Scholar]
- 3.Mitchell P, Smith W, Chang A. Prevalence and associations of retinal vein occlusion in Australia: the Blue Mountains Eye Study. Arch Ophthalmol. 1996;114(10):1243-1247. [DOI] [PubMed] [Google Scholar]
- 4.Rogers S, McIntosh RL, Cheung N, et al. ; International Eye Disease Consortium . The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology. 2010;117(2):313-319, e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Central Vein Occlusion Study Group Natural history and clinical management of central retinal vein occlusion. Arch Ophthalmol. 1997;115(4):486-491. [DOI] [PubMed] [Google Scholar]
- 6.Boyer D, Heier J, Brown DM, et al. Vascular endothelial growth factor Trap-Eye for macular edema secondary to central retinal vein occlusion: six-month results of the phase 3 COPERNICUS study. Ophthalmology. 2012;119(5):1024-1032. [DOI] [PubMed] [Google Scholar]
- 7.Holz FG, Roider J, Ogura Y, et al. VEGF Trap-Eye for macular oedema secondary to central retinal vein occlusion: 6-month results of the phase III GALILEO study. Br J Ophthalmol. 2013;97(3):278-284. [DOI] [PubMed] [Google Scholar]
- 8.Prager F, Michels S, Kriechbaum K, et al. Intravitreal bevacizumab (Avastin) for macular oedema secondary to retinal vein occlusion: 12-month results of a prospective clinical trial. Br J Ophthalmol. 2009;93(4):452-456. [DOI] [PubMed] [Google Scholar]
- 9.Wu L, Arevalo JF, Berrocal MH, et al. Comparison of two doses of intravitreal bevacizumab as primary treatment for macular edema secondary to central retinal vein occlusion: results of the pan American collaborative retina study group at 24 months. Retina. 2010;30(7):1002-1011. [DOI] [PubMed] [Google Scholar]
- 10.Martin DF, Maguire MG, Ying GS, Grunwald JE, Fine SL, Jaffe GJ; CATT Research Group . Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;364(20):1897-1908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wells JA, Glassman AR, Ayala AR, et al. ; Diabetic Retinopathy Clinical Research Network . Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. N Engl J Med. 2015;372(13):1193-1203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Office of Inspector General, Department of Health and Human Services Medicare Payments for Drugs Used to Treat Wet Age-Related Macular Degeneration. Washington, DC: Dept of Health & Human Services; 2012. Publication OEI-03-10-00360. [Google Scholar]
- 13.American Academy of Pediatrics Red Book Online [database online]. Elk Grove Village, IL: American Academy of Pediatrics; 2016.
- 14.World Medical Association World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. [DOI] [PubMed] [Google Scholar]
- 15.Scott IU, Vanveldhuisen PC, Oden NL, et al. ; SCORE Study Investigator Group . Baseline characteristics and response to treatment of participants with hemiretinal compared with branch retinal or central retinal vein occlusion in the standard care vs corticosteroid for retinal vein occlusion (SCORE) study: SCORE study report 14. Arch Ophthalmol. 2012;130(12):1517-1524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Brown DM, Campochiaro PA, Singh RP, et al. ; CRUISE Investigators . Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology. 2010;117(6):1124-1133, e1. [DOI] [PubMed] [Google Scholar]
- 17.Stewart MW. Aflibercept (VEGF Trap-eye): the newest anti-VEGF drug. Br J Ophthalmol. 2012;96(9):1157-1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ip MS, Oden NL, Scott IU, et al. ; SCORE Study Investigator Group . SCORE Study report 3: study design and baseline characteristics. Ophthalmology. 2009;116(9):1770-1777, e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Beck RW, Moke PS, Turpin AH, et al. A computerized method of visual acuity testing: adaptation of the early treatment of diabetic retinopathy study testing protocol. Am J Ophthalmol. 2003;135(2):194-205. [DOI] [PubMed] [Google Scholar]
- 20.Gross JG, Glassman AR, Jampol LM, et al. ; Writing Committee for the Diabetic Retinopathy Clinical Research Network . Panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy: a randomized clinical trial. JAMA. 2015;314(20):2137-2146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Beck RW, Maguire MG, Bressler NM, Glassman AR, Lindblad AS, Ferris FL. Visual acuity as an outcome measure in clinical trials of retinal diseases. Ophthalmology. 2007;114(10):1804-1809. [DOI] [PubMed] [Google Scholar]
- 22.Antiplatelet Trialists’ Collaboration Collaborative overview of randomised trials of antiplatelet therapy—I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ. 1994;308(6921):81-106. [PMC free article] [PubMed] [Google Scholar]
- 23.Lan KKG, DeMets DL. Discrete sequential boundaries for clinical trials. Biometrika. 1983;70(3):659-663. doi: 10.1093/biomet/70.3.659 [DOI] [Google Scholar]
- 24.Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika. 1988;75(4):800-802. doi: 10.1093/biomet/75.4.800 [DOI] [Google Scholar]
- 25.Wells JA, Glassman AR, Ayala AR, et al. ; Diabetic Retinopathy Clinical Research Network . Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema: two-year results from a comparative effectiveness randomized clinical trial. Ophthalmology. 2016;123(6):1351-1359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kitchens JW, Do DV, Boyer DS, et al. Comprehensive review of ocular and systemic safety events with intravitreal aflibercept injection in randomized controlled trials. Ophthalmology. 2016;123(7):1511-1520. [DOI] [PubMed] [Google Scholar]
- 27.Heier JS, Clark WL, Boyer DS, et al. Intravitreal aflibercept injection for macular edema due to central retinal vein occlusion: two-year results from the COPERNICUS study. Ophthalmology. 2014;121(7):1414-1420, e1. [DOI] [PubMed] [Google Scholar]
- 28.Campochiaro PA, Brown DM, Awh CC, et al. Sustained benefits from ranibizumab for macular edema following central retinal vein occlusion: twelve-month outcomes of a phase III study. Ophthalmology. 2011;118(10):2041-2049. [DOI] [PubMed] [Google Scholar]
- 29.Brown DM, Campochiaro PA, Bhisitkul RB, et al. Sustained benefits from ranibizumab for macular edema following branch retinal vein occlusion: 12-month outcomes of a phase III study. Ophthalmology. 2011;118(8):1594-1602. [DOI] [PubMed] [Google Scholar]
- 30.VanderBeek BL, Bonaffini SG, Ma L. Association of compounded bevacizumab with postinjection endophthalmitis. JAMA Ophthalmol. 2015;133(10):1159-1164. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol