A Randomized Clinical Trial of Intravitreal Bevacizumab versus Intravitreal Dexamethasone for Diabetic Macular Edema The BEVORDEX Study Mark C. Gillies, MBBS, PhD,1 Lyndell L. Lim, MBBS,2 Anna Campain, PhD,1 Godfrey J. Quin, MBChB, PhD,1 Wedad Salem, MB BS, MPH,1 Ji Li, MB BS,1 Stephanie Goodwin, BAppSc (Orthoptics) (Hons),1 Christine Aroney, MBBS,1 Ian L. McAllister, MBBS,3 Samantha Fraser-Bell, MBBS, PhD1 Objective: To report the 12-month results of the first head-to-head comparison of a dexamethasone implant (Ozurdex; Allergan, Inc., Irvine, CA) versus bevacizumab (Avastin; Genentech, South San Francisco, CA) for center-involving diabetic macular edema (DME). Design: Phase 2, prospective, multicenter, randomized, single-masked clinical trial (clinicaltrials.gov identifier NCT01298076). Participants: We enrolled 88 eyes of 61 patients with center-involving DME. Methods: Forty-two eyes were randomized to receive bevacizumab every 4 weeks and 46 eyes were randomized to receive a dexamethasone implant every 16 weeks, both pro re nata. Results were analyzed using linear regression with generalized estimation equation methods to account for between-eye correlation. Main Outcome Measures: The primary outcome was the proportion of eyes that improved vision by 10 logarithm of minimum angle of resolution letters. Secondary outcomes included mean change in best-corrected visual acuity (BCVA), change in central macular thickness (CMT), injection frequency, and adverse events. Patient-reported outcomes were measured using the Impact of Vision Impairment (IVI) questionnaire. Results: Improvement in BCVA of 10 or more letters was found in 17 of 42 eyes (40%) treated with bevacizumab compared with 19 of 46 dexamethasone implantetreated eyes (41%; P ¼ 0.83). None of the 42 bevacizumab eyes lost 10 letters or more, whereas 5 of 46 (11%) dexamethasone implant eyes did, mostly because of cataract. Mean CMT decreased by 122 mm for bevacizumab eyes and by 187 mm for dexamethasone implant eyes (P ¼ 0.015). Bevacizumab-treated eyes received a mean of 8.6 injections compared with 2.7 injections for dexamethasone implant eyes. Significant improvement in IVI scores occurred for both treatment groups. Conclusions: Dexamethasone implant achieved similar rates of visual acuity improvement compared with bevacizumab for DME, with superior anatomic outcomes and fewer injections. Both treatments were associated with improvement in visual quality-of-life scores. However, more dexamethasone implantetreated eyes lost vision, mainly because of cataract. Ophthalmology 2014;-:1e9 ª 2014 by the American Academy of Ophthalmology.
Diabetic retinopathy is a common cause of severe vision loss and the leading cause of blindness in individuals between 20 and 65 years of age in developed countries. Diabetic macular edema (DME) is the most common cause of visual loss in diabetic retinopathy,1 affecting approximately 6.8% of people with diabetes.2 Although laser treatment previously was the benchmark treatment for clinically significant DME, the limitations of laser treatment, together with intense clinical research over the last 10 years, have led to laser treatment being surpassed by intravitreal pharmacotherapy as first-line treatment for moderate to severe vision loss caused by DME.3 Vascular endothelial growth factor (VEGF) inhibitors are the most commonly used agents for intravitreal treatment of DME. Ranibizumab (Lucentis; Genentech, Inc., South San 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.
Francisco, CA) consistently has been shown to be effective in treating DME by a series of phase 2 and phase 3 randomized clinical trials.4e8 The cost-effective agent bevacizumab (Avastin; Genentech), which was shown to have similar efficacy to ranibizumab at least for neovascular age-related macular degeneration,9 also was reported to be superior to laser for DME.10 Other anti-VEGF inhibitors that have been reported to be effective for DME include pegaptanib (Macugen; Valeant Inc, Montreal, Canada)11 and, more recently, aflibercept (Eylea; Bayer HealthCare, Whippany, NJ).12 Glucocorticoids were the first class of drug shown by randomized clinical trials to be beneficial for DME after intravitreal injection.13e15 The first agent tested, triamcinolone acetonide (Kenacort; Bristol-Myers Squibb, Anagni, http://dx.doi.org/10.1016/j.ophtha.2014.07.002 ISSN 0161-6420/12
1
Ophthalmology Volume -, Number -, Month 2014 Italy), was not formulated for ocular use and conferred a high risk of cataract and elevated intraocular pressure. Sustained-release steroids devices made specifically for intravitreal injection have been developed for the treatment of posterior segment disease. One such agent is the dexamethasone intravitreal implant (0.7 mg; Ozurdex; Allergan, Inc., Irvine, CA). This implant consists of micronized dexamethasone in a biodegradable copolymer of polylactic-co-glycolic acid. The implant is said to release dexamethasone slowly for up to 6 months,16 although anecdotal evidence together with evidence supplied by the present study suggest that its clinical efficacy is limited to 4 months in most eyes. The dexamethasone implant has been reported to be effective for the treatment of persistent DME by a phase 2 randomized clinical trial17 and for the treatment of DME in vitrectomized eyes.18 Although both glucocorticoids and VEGF inhibitors seem to reduce macular edema and improve vision in the short term, they may have differences that could guide how they are best used. Anti-VEGF drugs do not cause cataract or elevated intraocular pressure to the same extent that steroids do; however, they must be given more frequently (every 4e6 weeks vs. every 4e6 months for the dexamethasone implant). Anti-VEGF treatment may be associated with a small increased risk of stroke or myocardial infarction. It is even suspected by some that chronic VEGF suppression may have a neurotoxic effect on the retina.19 We conducted this study to test the hypothesis that there is a difference in efficacy and treatment frequency between bevacizumab and the dexamethasone implant for DME.
Methods Patient Enrollment This study was conducted in accordance with the Declaration of Helsinki and was approved by the Human Research Ethics Committees of Sydney South West Area Health Service, the University of Sydney, the Royal Victorian Eye and Ear Hospital, and Bellberry Ltd. Safety data were reviewed by an independent safety monitoring committee. Patients were recruited from the retina clinics of Sydney Eye Hospital, Royal Victorian Eye and Ear Hospital in Melbourne, Lions Eye Institute in Perth, and South West Retina in Liverpool, New South Wales, from October 5, 2010, until September 6, 2012. Eyes with DME affecting the central fovea at least 3 months after at least 1 session of laser treatment, or for whom the investigator believed that laser treatment would be unhelpful, with best-corrected visual acuity (BCVA) of 17 to 72 logarithm of the minimum angle of resolution letters (Snellen equivalent, 20/400e20/40) were eligible and identified consecutively as they were seen in the clinics. Exclusion criteria were uncontrolled glaucoma or glaucoma controlled with more than 1 medication, loss of vision because of other causes, intercurrent severe systemic disease, or any condition affecting follow-up or documentation.
Sample Size Calculation There were no good data on which to base sample size estimates when we started this study, so we did not necessarily expect to find a statistically significant difference between the 2 groups. Nevertheless, we regarded as clinically significant an event rate in the order of 50% in one group compared with 20% in the second group. A minimum of 35 eyes per group was required for 80% power of detecting this difference as significant at the 2-sided 5%
2
level. Adjusting for an estimated loss to follow-up of 12% of eyes, we aimed to recruit a minimum of 40 eyes per group.
Treatment Assignment After signing the informed consent form, each patient was randomized to receive study treatment. The sites contacted the coordinating center at the Save Sight Institute, where an unmasked research officer issued the randomization to either bevacizumab or dexamethasone using a master sheet that had been formulated from a list of computer-generated pseudorandom numbers in permuted blocks of variable size. In patients with both eyes eligible for the study, the right eye received the treatment assignment that had been generated and the fellow eye received the other treatment.
Data Collection and Masking Measurement of logarithm of the minimum angle of resolution BCVA, the main outcome measure, was performed with Early Treatment Diabetic Retinopathy Study (ETDRS) charts using standardized procedures by certified masked research officers. Intraocular pressures (IOPs) were measured using Goldman applanation tonometry. Cataracts were graded using Age-Related Eye Disease Study photographic standards. Central macular thickness (CMT) was measured from the central 1-mm subfield from spectral-domain ocular coherence tomography images (Cirrus; Zeiss). Participants were seen 1 week after their first treatment, then every 4 weeks. At each of these visits, they underwent dilated fundal examination, measurement of logarithm of the minimum angle of resolution BCVA, IOP, and blood pressure (BP); and grading of anterior and posterior chamber cells and cataract. The primary outcomes were measured at the week 48 visit unless further treatment was indicated (see “Outcomes,” below), in which case the further treatment was provdied and the primary outcome measurements were measured 2 weeks later. Source data were verified by an independent study monitor of all BCVA and ocular coherence tomography data for all patients. Grading of steroid-related adverse events was defined prospectively. Significant elevation of IOP was defined as an increase of more than 5 mmHg compared with the baseline level. Significant progression of cataract was defined as an increase of more than 2 AgeRelated Eye Disease Study grades. The decision to institute glaucoma therapy was made along conventional lines based on the degree of IOP elevation and extent of glaucomatous optic neuropathy, if present. The decision to perform cataract surgery was made in discussion with the patient, taking into account the level of vision in both the affected and fellow eye. All eyes received a repeated injection of their assigned study medication 2 to 4 weeks before surgery. Blood pressure was measured at each study visit. Hypertension was defined as taking an antihypertensive or systolic BP at presentation of 140 mmHg or more and or diastolic BP of 90 mmHg or more. Worsening of systemic hypertension was defined as an increase in BP to more than 150 mmHg systolic or 100 mmHg diastolic and increased by 20 mmHg on at least 2 consecutive visits or the addition of a new antihypertensive medication or an increase in the dose of antihypertensives used.
Treatment Bevacizumab (1.25 mg) or dexamethasone implant (0.7 mg) was injected into the vitreous on the day of the baseline BCVA measurement under sterile conditions in a minor procedures area as an outpatient procedure. Re-treatment was considered at each visit as long as treatments were at least 4 weeks apart for bevacizumab and 16 weeks apart for the dexamethasone implant. All eyes were considered for treatment at the appropriate visit unless the CMT was less than 300 mm or visual acuity was 79 letters or better (20/ 25), in which cases it could be withheld.
Gillies et al
Bevacizumab vs. Dexamethasone for DME
Outcomes
Statistical Analysis
The primary outcome measure was the percentage of eyes in which BCVA improved by 10 or more letters at the 48-week visit, or the 50-week visit if further treatment had been indicated at 48 weeks to ensure that there was an adequate level of drug in the patient’s eye. Secondary outcome measures included change in mean BCVA of the 2 groups, any change in BCVA relative to the baseline level, as well as mean CMT, mean change in CMT, and number of treatments given.
Efficacy data at the 12-month follow-up were analyzed by intention to treat. Data were analyzed using R software version 2.15.0 (R Development core team, R Foundation for Statistical Computing, Vienna, Austria). Because correlation was possible between eyes of the same patient, generalized estimating equation (GEE)24 methods were used where appropriate. The primary outcome (eyes gaining 10 letters or more) was modeled using logistic regression coupled with GEEs; BCVA at baseline visit was used as a covariate. Change in CMT was modeled using generalized linear models with GEEs also using baseline BCVA as a covariate. To compare continuous numerical data, including BCVA, CMT, age, duration of diabetes, and mean glycosylated hemoglobin, t tests were used. Fisher exact tests for trends in proportions were used for gender analysis and change in cataract grade. The last observation carried forward method was used when patients were lost to follow-up. A sensitivity analysis was performed for the primary outcome, improvement of 10 letters, to investigate the potential effect of the missing data, by alternately
Patient-Reported Outcomes Patient-reported outcomes were measured using the Impact of Vision Impairment (IVI) questionnaire, a measure of vision-related quality of life, at baseline and again at 12 months. The IVI has been revised using Rasch analysis20 and has been used and validated for macular diseases, including age-related macular degeneration, macular telangiectasia, and low-vision rehabilitation,21e23 as a measure of patients’ perception of vision-related restriction on their activity and quality of life.
Figure 1. Diagram showing the progression of patients through the study. DEX ¼ dexamethasone; LOCF ¼ last observation carried forward.
3
Ophthalmology Volume -, Number -, Month 2014 assuming that all missing eyes had actually improved by 10 letters or that all missing eyes had not improved by 10 letters.25 Rasch analysis of patient-reported outcome was performed by WINSTEPS software version 3.30.1 (Rasch Measurement Computer Program, Chicago, IL) using a partial credit model.26 The resulting scores are person measures, or logits, which are interval estimates of the ordinal raw scores of questionnaire and calibrated within a range of 5.5 to þ 5.5 (the higher the score, the higher the increase in vision-related quality of life). The logits produced then were analyzed further in SPSS software version 21 (SPSS, Inc., Chicago, IL).
Results We enrolled 88 eyes from 61 patients, of which 42 eyes were randomized to receive bevacizumab and 46 received the dexamethasone implant. Twelve-month data were available for 81 eyes (92%) from 56 patients (92%). Figure 1 shows the flow of eyes through the study. Of the 7 eyes (from 6 patients) that exited the study, 4 had been assigned to bevacizumab and 3 to dexamethasone implant. Four of the 6 patients withdrew from the study because of difficulty meeting the rigor of attending monthly appointments. Two withdrew for perceived lack of efficacy of bevacizumab. Table 1 shows the baseline characteristics for the 2 treatment groups, which were similar. The primary outcome was the proportion of eyes in which BCVA improved by 10 letters or more at the 12-month study visit. Twenty-seven of 38 eyes treated with bevacizumab and 31 of 43 eyes receiving a dexamethasone implant had their primary outcome measured at 48 weeks, whereas the remainder of each group required further treatment, which they received, and the primary outcome was measured at 50 weeks. There was no significant difference between the mean primary outcome visual acuity scores at 48 and 50 weeks for either treatment group (not shown). A gain of 10 letters or more was found in 17 of the 42 eyes (40%) treated with bevacizumab and in 19 of the 46 eyes (41%) treated with dexamethasone implant (P ¼ 0.99) by the 12month visit. There was no correlation in the change in BCVA at 12 months between each eye for patients with both eyes in the study (r ¼ 0.006). Of the 42 eyes treated with bevacizumab, 13 (31%) gained more than 15 letters, compared with 10 (22%) of the dexamethasone implantetreated eyes (Table 2). The mean improvement in BCVA was 8.9 letters (95% confidence interval, 6.27e11.6) for bevacizumab-treated eyes and 5.6 (95%
Table 2. Effect of Treatment on Best-Corrected Visual Acuity from Baseline Change in Best-Corrected Visual Acuity
Bevacizumab
Gain of 15 letters Gain of 10e14 letters Gain of 5e9 letters No change (gain or loss 0.1).
Ocular Adverse Events Most ocular adverse events reported in this study were anticipated, including cataract and increased IOP (Table 3). There were no treatment-related cases of retinal detachment or endophthalmitis in either group. Elevation of Intraocular Pressure. An IOP elevation by at least 5 mmHg from baseline at any follow-up visit during the first
Figure 3. Bar graph showing the distribution of number of treatments given by treatment group for patients completing the 12-month follow-up visit. DEX ¼ dexamethasone.
5
Ophthalmology Volume -, Number -, Month 2014
Figure 4. Graph showing the mean central macular thickness by treatment type for each month of follow-up. DEX ¼ dexamethasone.
year of the study was seen in 46% (21/46) of dexamethasone implantetreated eyes and in 19% (8/42) of bevacizumab-treated eyes. Throughout the 12-month study, 12 eyes demonstrated an IOP of more than 25 mmHg at least once during follow-up visits. All 12 observed eyes had been treated with the dexamethasone implant. Eyes with increases in IOP were managed successfully with either observation or topical IOP-lowering medications. One eye from each group required selective laser trabeculoplasty to control IOP; both eyes were from the same patient. Cataract Progression. Twenty-six of the 88 eyes (29.5%) were pseudophakic at baseline, 10 of which were treated with bevacizumab (24% of the bevacizumab-treated eyes), whereas 16 were treated with the dexamethasone implant (35% of the dexamethasone implant eyes). There was no significant effect based on treatment received for the change in BCVA for the pseudophakic eyes (mean change in BCVA for bevacizumab eyes, 7.7 letters; mean change in BCVA for dexamethasone-treated eyes, 10.4 letters; P ¼ 0.47, t test for difference in BCVA change). Four of the
Table 3. Ocular Adverse Events Ocular Adverse Events IOP elevation 5 mmHg IOP elevation 10 mmHg Increase in cataract by >2 grades Cataract surgery Vision decrease of >10 letters Vitreous hemorrhage Floaters Subconjunctival hemorrhage Eye pain Itching Blepharitis Epiphora IOP ¼ intraocular pressure. Data are no. (%).
6
Bevacizumab (n [ 42)
Dexamethasone Implant (n [ 46)
8 (19) 0 (0) 2 (4.8)
21 (45.7) 9 (19.6) 6 (13)
1 0 0 0 3 3 0 1 2
(2.4) (0) (0) (0) (7.1) (7.1) (0) (2.4) (4.8)
3 5 2 5 2 1 3 1 0
(6.5) (10.9) (4.3) (10.9) (4.3) (2.2) (6.5) (2.2) (0)
remaining 62 eyes (6%) had cataract surgery during the first 12 months of the study, 1 treated with bevacizumab and 3 with the dexamethasone implant. Increase in cataract density by 2 grades or more from baseline was reported in 13% (6/46) of eyes in the dexamethasone implant group and in 4.8% (2/42) of eyes in the bevacizumab group. Loss of More Than 10 Letters. None of the bevacizumabtreated eyes lost more than 10 letters; however, 11% (5/46) of the dexamethasone implantetreated eyes did. This vision deterioration was because of an increase in cataract density in 4 cases and syphilitic chorioretinitis in 1 eye that lost 64 letters. This was associated with an attack of generalized secondary syphilis that developed 1 week after the eye had been treated with the dexamethasone implant. Given that syphilitic choroidoretinitis has been described after intraocular steroid treatment in 2 cases,27 the infection was deemed to be related to treatment.
Systemic Adverse Events The dexamethasone implant and bevacizumab injections were well tolerated systemically during the first year of the study. Serious systemic adverse events are summarized in Table 4. The most frequent systemic adverse event was the worsening of hypertension, which occurred in 6.7% (1/15), 10.5% (2/19), and 14.8% (4/ 27) of patients who received bevacizumab only, dexamethasone implant only, and both treatments, respectively.
Discussion This randomized controlled trial provides the first direct comparison of a slow-release steroid formulation made specifically for ophthalmic use with a commonly used antiVEGF agent for the treatment of DME. We found that approximately 40% of both groups achieved the prospectively identified primary outcome of a 10-letter gain over 1 year. The mean gain in visual acuity was more in the group receiving bevacizumab because of progression of cataract and a rare adverse event in the dexamethasone implante treated group, but this was not statistically significant.
Gillies et al
Bevacizumab vs. Dexamethasone for DME
Table 4. Systemic Adverse Events Systemic Adverse Event Worsening hypertension Chest pain Angina Myocardial infarction Congestive cardiac failure Cerebrovascular accident Kidney failure
Dexamethasone Bevacizumab D Bevacizumab Implant Dexamethasone Only (n [ 15) Only (n [ 19) Implant (n [ 27) 1 (6.7)
2 (10.5)
4 (14.8)
1 (6.7) 0 (0) 0 (0)
1 (5.3) 1 (5.3) 0 (0)
1 (3.8) 0 (0) 1 (3.8)
0 (0)
0 (0)
3 (11.5)
0 (0)
0 (0)
2 (7.7)
0 (0)
1 (5.3)
0 (0)
Data are no. (%).
Anatomic outcomes were significantly better in the dexamethasone implantetreated group with far fewer injections. Steroid-related adverse events seemed to be somewhat less frequent in eyes treated with the dexamethasone implant compared with triamcinolone acetonide or the fluocinolone implant (Iluvien; pSivida, Watertown, MA). Significant elevation of IOP (more than 10 mmHg) occurred in approximately 20% of eyes and progression of cataract by 2 grades in 14.5% of eyes. However, it is well established that steroid-related cataract generally appears in the second year after commencing intravitreal steroid therapy, so this rate is likely to increase in the second year of the trial.13 Cataract extraction was performed in 3 dexamethasone implante treated eyes compared with 1 bevacizumab-treated eye. By contrast, cataract surgery was required in 54% of eyes treated for 2 years in 1 randomized clinical trial of triamcinolone acetonide for DME13 and in virtually all phakic eyes receiving the fluocinolone implant.28 This study indicated that steroids may deliver better anatomic results than VEGF inhibitors. The mean reduction of CMT of dexamethasone implantetreated eyes was 187132.2 mm after 12 months compared with 122132.9 mm for the group receiving bevacizumab. This is consistent with the findings of a smaller, shorter study that compared bevacizumab with triamcinolone acetonide.29 The improved anatomic outcomes in the present study did not, however, translate to better visual acuity outcomes. The significance of persistent swelling despite ongoing treatment currently is uncertain and warrants longerterm studies. The much lower treatment intensity required for the dexamethasone implant (mean number of treatments, 2.7) was a distinct advantage over bevacizumab (mean number, 8.6). The dexamethasone implant is not as easy to administer because it requires a 22-gauge needle; however, this was not associated with an increased risk of operative complications. All patients were seen monthly in this study. In routine practice, one would still likely check dexamethasone implantetreated eyes at 4 to 6 weeks, especially to monitor IOP, so the number of visits may not be much less, but the reduced number of injections may be associated with fewer operative risks such as infectious endophthalmitis.
Patients also are likely to prefer fewer injections. When comparing the relative cost efficacy of the 2 classes of drugs, both fewer injections and fewer visits with steroids must be weighed against the substantially greater cost of the dexamethasone implant and the cost and clinical consequences of cataract and its removal, which is much more likely in eyes receiving steroids than in those receiving VEGF inhibitors. Somewhat more people who received both interventions, one in either eye, preferred the dexamethasone implant, although this was not statistically significant. The mistaken belief that the dexamethasone implant provides therapeutic effects for 6 months unfortunately has led to some industry-sponsored studies having difficulty achieving their originally specified primary end points.30,31 The mean CMT in the steroid-treated groups in the present study clearly peaked at 4 and 8 months. Further studies in which the dexamethasone implant is given for only 6-monthly for any condition should be avoided because they bring what could be an effective intervention into disrepute. Bevacizumab was the only VEGF inhibitor to which we had access when we started this study. There are few data on whether it is as effective as other VEGF inhibitors for DME, such as ranibizumab or aflibercept. It generally was believed to be equivalent to ranibizumab for neovascular age-related macular degeneration according to the Comparison of Age-Related Macular Degeneration Treatments Trials study investigators.9 Although local adverse events commonly are cited to relegate steroids to second-line treatment of DME, these need to be balanced against the potential for severe systemic adverse events associated with chronic VEGF inhibition. In the combined data from the RISE (Clinicaltrials.com: NCT00473330) and RIDE (NCT00473382) studies, for example, there was a dose-related increased mortality rate of 4.4% of participants treated with monthly ranibizumab 0.5 mg for 2 years for DME (compared with 1.2% for controls), which rose to 6.4% after 3 years.32 An increased systemic adverse event rate has not been reported with intravitreal corticosteroid therapy. Although there were no deaths in the present study, there were 2 cerebrovascular accidents and 1 myocardial infarction in patients who were receiving both interventions. Limitations of this study include its relatively small size and short duration. Because the proportion of eyes in the 2 groups that achieved the primary outcome, a gain of 10 letters, was essentially the same, larger studies are unlikely to find a significant difference between the 2 interventions at least after 12 months of treatment. Although the present study will continue for a second year, particularly to track the effect of steroid-induced cataract, DME is a chronic disease that conceivably could require life-long treatment in some patients. Randomized clinical trials provide little to no information on these long-term outcomes. There is an urgent need for postmarketing observational studies and data linkage to track long-term outcomes in patients receiving treatment for DME. This study provided direct evidence that the dexamethasone implant can achieve similar rates of vision improvement compared with bevacizumab for DME with superior anatomic outcomes and fewer injections. However, more
7
Ophthalmology Volume -, Number -, Month 2014 patients receiving the dexamethasone implant lost vision, mainly because of cataract. Therefore, the dexamethasone implant may be considered as second-line treatment for DME in phakic eyes and as first-line treatment for pseudophakic eyes or for patients who are averse to frequent intraocular injections or have difficulty attending regularly. Acknowledgment. The authors thank the Safety Monitoring Committee for this study: Dr. Jeremy Smith (Chair), Dr. JieJin Wang, and Paul Power.
References 1. Moss SE, Klein R, Klein BE. The 14-year incidence of visual loss in a diabetic population. Ophthalmology 1998;105: 998–1003. 2. Yau JW, Rogers SL, Kawasaki R, et al; Meta-Analysis for Eye Disease (META-EYE) Study Group. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care 2012;35:556–64. 3. Mitchell P, Wong TW; Diabetic Macular Edema Treatment Guideline Working Group. Management paradigms for diabetic macular edema. Am J Ophthalmol 2014;157:505–13. 4. Mitchell P, Bandello F, Schmidt-Erfurth U, et al; RESTORE Study Group. The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema. Ophthalmology 2011;118:615–25. 5. Massin P, Bandello F, Garweg JG, et al. Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study. Diabetes Care 2010;33:2399–405. 6. Do DV, Nyuyen QD, Khwaja AA, et al; READ-2 Study Group. Ranibizumab for Edema of the Macula in Diabetes Study: 3-year outcomes and the need for prolonged frequent treatment. JAMA Ophthalmol 2013;131:139–45. 7. Diabetic Retinopathy Clinical Research Network, Elman MJ, Qin H, Aiello LP, et al. Intravitreal ranibizumab for diabetic macular edema with prompt versus deferred laser treatment: three-year randomized trial results. Ophthalmology 2012;119: 2312–8. 8. Nguyen QD, Brown DM, Marcus DM, et al; RISE and RIDE Research Group. Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE. Ophthalmology 2012;119:789–801. 9. Comparison of Age-Related Macular Degeneration Treatments Trials (CATT) Research Group, Martin DF, Maguire MG, Fine SL, et al. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology 2012;119:1388–98. 10. Rajendram R, Fraser-Bell S, Kaines A, et al. A 2-year prospective randomized controlled trial of intravitreal bevacizumab or laser therapy (BOLT) in the management of diabetic macular edema: 24-months data: report 3. Arch Ophthalmol 2012;130:972–9. 11. Sultan MB, Zhou D, Loftus J, et al; Macugen 1013 Study Group. A phase 2/3, multicenter, randomized, double-masked, 2-year trial of pegaptanib sodium for the treatment of diabetic macular edema. Ophthalmology 2011;118:1107–18. 12. Do DV, Nguyen QD, Boyer D, et al; DA VINCI Study Group. One-year outcomes of the DA VINCI Study of VEGF TrapEye in eyes with diabetic macular edema. Ophthalmology 2012;119:1658–65. 13. Gillies MC, Sutter FK, Simpson JM, et al. Intravitreal triamcinolone for refractory diabetic macular edema:
8
14.
15.
16.
17.
18.
19. 20.
21.
22. 23. 24. 25. 26. 27.
28.
29.
30.
two-year results of a double-masked, placebo-controlled, randomized clinical trial. Ophthalmology 2006;113: 1533–8. Gillies MC, Simpson JM, Gaston C, et al. Five-year results of a randomized trial with open-label extension of triamcinolone acetonide for refractory diabetic macular edema. Ophthalmology 2009;116:2182–7. Gillies MC, McAllister IL, Zhu M, et al. Intravitreal triamcinolone prior to laser treatment of diabetic macular edema: 24-month results of a randomized controlled trial. Ophthalmology 2011;118:866–72. Chang-Lin JE, Attar M, Acheampong AA, et al. Pharmacokinetics and pharmacodynamics of a sustained-release dexamethasone intravitreal implant. Invest Ophthalmol Vis Sci 2011;52:80–6. Haller JA, Kuppermann BD, Blumenkranz MS, et al; Dexamethasone DDS Phase II Study Group. Randomized controlled trial of an intravitreous dexamethasone drug delivery system in patients with diabetic macular edema. Arch Ophthalmol 2010;128:289–96. Boyer DS, Faber D, Gupta S, et al; Ozurdex CHAMPLAIN Study Group. Dexamethasone intravitreal implant for treatment of diabetic macular edema in vitrectomized patients. Retina 2011;31:915–23. van Wijngaarden P, Coster DJ, Williams KA. Inhibitors of ocular neovascularization: promises and potential problems. JAMA 2005;293:1509–13. Lamoureux EL, Pallant JF, Pesudovs K, et al. The Impact of Vision Impairment Questionnaire: an evaluation of its measurement properties using Rasch analysis. Invest Ophthalmol Vis Sci 2006;47:4732–41. Lamoureux EL, Pallant JF, Pesudovs K, et al. The effectiveness of low-vision rehabilitation on participation in daily living and quality of life. Invest Ophthalmol Vis Sci 2007;48: 1476–82. Pesudovs K, Caudle LE, Rees G, Lamoureux EL. Validity of a visual impairment questionnaire in measuring cataract surgery outcomes. J Cataract Refract Surg 2008;34:925–33. Finger RP, Fenwick E, Marella M, et al. The impact of vision impairment on vision-specific quality of life in Germany. Invest Ophthalmol Vis Sci 2011;52:3613–9. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach. Biometrics 1988;44:1049–60. Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomised controlled trials. BMJ 1999;319:670–4. Masters GN. A Rasch model for partial credit scoring. Psychometrika 1982;47:149–74. Song JH, Hong YT, Kwon OW. Acute syphilitic posterior placoid chorioretinitis following intravitreal triamcinolone acetonide injection. Graefes Arch Clin Exp Ophthalmol 2008;246:1775–8. Campochiaro PA, Brown DM, Pearson A, et al; FAME Study Group. Long-term benefit of sustained-delivery fluocinolone acetonide vitreous inserts for diabetic macular edema. Ophthalmology 2011;118:626–35. Kriechbaum K, Prager S, Mylonas G, et al; Diabetic Retinopathy Research Group. Intravitreal bevacizumab (Avastin) versus triamcinolone (Volon A) for treatment of diabetic macular edema: one-year results. Eye (Lond) 2014;28:9–15. Callanan DG, Gupta S, Boyer DS, et al; Ozurdex PLACID Study Group. Dexamethasone intravitreal implant in combination with laser photocoagulation for the treatment of diffuse diabetic macular edema. Ophthalmology 2013;120:1843–51.
Gillies et al
Bevacizumab vs. Dexamethasone for DME
31. Haller JA, Bandello F, Belfort R Jr, et al; Ozurdex GENEVA Study Group. Dexamethasone intravitreal implant in patients with macular edema related to branch or central retinal vein occlusion: twelve-month study results. Ophthalmology 2011;118:2453–60.
32. Brown DM, Nguyen QD, Marcus DM, et al; RIDE and RISE Research Group. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology 2013;120: 2013–22.
Footnotes and Financial Disclosures Originally received: April 22, 2014. Final revision: June 26, 2014. Accepted: July 2, 2014. Available online: ---.
Allergan Pharmaceutical. The author(s) have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2014-605.
1
The Save Sight and Eye Health Institute, Sydney Medical School, The University of Sydney, Sydney NSW, Australia. 2 Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Hospital, East Melbourne, Victoria, Australia. 3
Lions Eye Institute, Centre for Ophthalmology and Visual Science, University of Western Australia, Australia. Financial Disclosure(s): Funded by a project grant from the National Health and Medical Research Council, which was supplemented by an unrestricted educational grant from
Abbreviations and Acronyms: BCVA ¼ best-corrected visual acuity; BP ¼ blood pressure; CMT ¼ central macular thickness; DME ¼ diabetic macular edema; GEE ¼ generalized estimating equation; IOP ¼ intraocular pressure; IVI ¼ Impact of Vision Impairment; SD ¼ standard deviation; VEGF ¼ vascular endothelial growth factor. Correspondence: Samantha Fraser-Bell, MBBS, PhD, Save Sight and Eye Health Institute, University of Sydney, 8 Macquarie Street, Sydney NSW 2000, Australia. E-mail: [email protected].
9
Diabetic retinopathy is a common cause of severe vision loss and the leading cause of blindness in individuals between 20 and 65 years of age in developed countries. Diabetic macular edema (DME) is the most common cause of visual loss in diabetic retinopathy,1 affecting approximately 6.8% of people with diabetes.2 Although laser treatment previously was the benchmark treatment for clinically significant DME, the limitations of laser treatment, together with intense clinical research over the last 10 years, have led to laser treatment being surpassed by intravitreal pharmacotherapy as first-line treatment for moderate to severe vision loss caused by DME.3 Vascular endothelial growth factor (VEGF) inhibitors are the most commonly used agents for intravitreal treatment of DME. Ranibizumab (Lucentis; Genentech, Inc., South San 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.
Francisco, CA) consistently has been shown to be effective in treating DME by a series of phase 2 and phase 3 randomized clinical trials.4e8 The cost-effective agent bevacizumab (Avastin; Genentech), which was shown to have similar efficacy to ranibizumab at least for neovascular age-related macular degeneration,9 also was reported to be superior to laser for DME.10 Other anti-VEGF inhibitors that have been reported to be effective for DME include pegaptanib (Macugen; Valeant Inc, Montreal, Canada)11 and, more recently, aflibercept (Eylea; Bayer HealthCare, Whippany, NJ).12 Glucocorticoids were the first class of drug shown by randomized clinical trials to be beneficial for DME after intravitreal injection.13e15 The first agent tested, triamcinolone acetonide (Kenacort; Bristol-Myers Squibb, Anagni, http://dx.doi.org/10.1016/j.ophtha.2014.07.002 ISSN 0161-6420/12
1
Ophthalmology Volume -, Number -, Month 2014 Italy), was not formulated for ocular use and conferred a high risk of cataract and elevated intraocular pressure. Sustained-release steroids devices made specifically for intravitreal injection have been developed for the treatment of posterior segment disease. One such agent is the dexamethasone intravitreal implant (0.7 mg; Ozurdex; Allergan, Inc., Irvine, CA). This implant consists of micronized dexamethasone in a biodegradable copolymer of polylactic-co-glycolic acid. The implant is said to release dexamethasone slowly for up to 6 months,16 although anecdotal evidence together with evidence supplied by the present study suggest that its clinical efficacy is limited to 4 months in most eyes. The dexamethasone implant has been reported to be effective for the treatment of persistent DME by a phase 2 randomized clinical trial17 and for the treatment of DME in vitrectomized eyes.18 Although both glucocorticoids and VEGF inhibitors seem to reduce macular edema and improve vision in the short term, they may have differences that could guide how they are best used. Anti-VEGF drugs do not cause cataract or elevated intraocular pressure to the same extent that steroids do; however, they must be given more frequently (every 4e6 weeks vs. every 4e6 months for the dexamethasone implant). Anti-VEGF treatment may be associated with a small increased risk of stroke or myocardial infarction. It is even suspected by some that chronic VEGF suppression may have a neurotoxic effect on the retina.19 We conducted this study to test the hypothesis that there is a difference in efficacy and treatment frequency between bevacizumab and the dexamethasone implant for DME.
Methods Patient Enrollment This study was conducted in accordance with the Declaration of Helsinki and was approved by the Human Research Ethics Committees of Sydney South West Area Health Service, the University of Sydney, the Royal Victorian Eye and Ear Hospital, and Bellberry Ltd. Safety data were reviewed by an independent safety monitoring committee. Patients were recruited from the retina clinics of Sydney Eye Hospital, Royal Victorian Eye and Ear Hospital in Melbourne, Lions Eye Institute in Perth, and South West Retina in Liverpool, New South Wales, from October 5, 2010, until September 6, 2012. Eyes with DME affecting the central fovea at least 3 months after at least 1 session of laser treatment, or for whom the investigator believed that laser treatment would be unhelpful, with best-corrected visual acuity (BCVA) of 17 to 72 logarithm of the minimum angle of resolution letters (Snellen equivalent, 20/400e20/40) were eligible and identified consecutively as they were seen in the clinics. Exclusion criteria were uncontrolled glaucoma or glaucoma controlled with more than 1 medication, loss of vision because of other causes, intercurrent severe systemic disease, or any condition affecting follow-up or documentation.
Sample Size Calculation There were no good data on which to base sample size estimates when we started this study, so we did not necessarily expect to find a statistically significant difference between the 2 groups. Nevertheless, we regarded as clinically significant an event rate in the order of 50% in one group compared with 20% in the second group. A minimum of 35 eyes per group was required for 80% power of detecting this difference as significant at the 2-sided 5%
2
level. Adjusting for an estimated loss to follow-up of 12% of eyes, we aimed to recruit a minimum of 40 eyes per group.
Treatment Assignment After signing the informed consent form, each patient was randomized to receive study treatment. The sites contacted the coordinating center at the Save Sight Institute, where an unmasked research officer issued the randomization to either bevacizumab or dexamethasone using a master sheet that had been formulated from a list of computer-generated pseudorandom numbers in permuted blocks of variable size. In patients with both eyes eligible for the study, the right eye received the treatment assignment that had been generated and the fellow eye received the other treatment.
Data Collection and Masking Measurement of logarithm of the minimum angle of resolution BCVA, the main outcome measure, was performed with Early Treatment Diabetic Retinopathy Study (ETDRS) charts using standardized procedures by certified masked research officers. Intraocular pressures (IOPs) were measured using Goldman applanation tonometry. Cataracts were graded using Age-Related Eye Disease Study photographic standards. Central macular thickness (CMT) was measured from the central 1-mm subfield from spectral-domain ocular coherence tomography images (Cirrus; Zeiss). Participants were seen 1 week after their first treatment, then every 4 weeks. At each of these visits, they underwent dilated fundal examination, measurement of logarithm of the minimum angle of resolution BCVA, IOP, and blood pressure (BP); and grading of anterior and posterior chamber cells and cataract. The primary outcomes were measured at the week 48 visit unless further treatment was indicated (see “Outcomes,” below), in which case the further treatment was provdied and the primary outcome measurements were measured 2 weeks later. Source data were verified by an independent study monitor of all BCVA and ocular coherence tomography data for all patients. Grading of steroid-related adverse events was defined prospectively. Significant elevation of IOP was defined as an increase of more than 5 mmHg compared with the baseline level. Significant progression of cataract was defined as an increase of more than 2 AgeRelated Eye Disease Study grades. The decision to institute glaucoma therapy was made along conventional lines based on the degree of IOP elevation and extent of glaucomatous optic neuropathy, if present. The decision to perform cataract surgery was made in discussion with the patient, taking into account the level of vision in both the affected and fellow eye. All eyes received a repeated injection of their assigned study medication 2 to 4 weeks before surgery. Blood pressure was measured at each study visit. Hypertension was defined as taking an antihypertensive or systolic BP at presentation of 140 mmHg or more and or diastolic BP of 90 mmHg or more. Worsening of systemic hypertension was defined as an increase in BP to more than 150 mmHg systolic or 100 mmHg diastolic and increased by 20 mmHg on at least 2 consecutive visits or the addition of a new antihypertensive medication or an increase in the dose of antihypertensives used.
Treatment Bevacizumab (1.25 mg) or dexamethasone implant (0.7 mg) was injected into the vitreous on the day of the baseline BCVA measurement under sterile conditions in a minor procedures area as an outpatient procedure. Re-treatment was considered at each visit as long as treatments were at least 4 weeks apart for bevacizumab and 16 weeks apart for the dexamethasone implant. All eyes were considered for treatment at the appropriate visit unless the CMT was less than 300 mm or visual acuity was 79 letters or better (20/ 25), in which cases it could be withheld.
Gillies et al
Bevacizumab vs. Dexamethasone for DME
Outcomes
Statistical Analysis
The primary outcome measure was the percentage of eyes in which BCVA improved by 10 or more letters at the 48-week visit, or the 50-week visit if further treatment had been indicated at 48 weeks to ensure that there was an adequate level of drug in the patient’s eye. Secondary outcome measures included change in mean BCVA of the 2 groups, any change in BCVA relative to the baseline level, as well as mean CMT, mean change in CMT, and number of treatments given.
Efficacy data at the 12-month follow-up were analyzed by intention to treat. Data were analyzed using R software version 2.15.0 (R Development core team, R Foundation for Statistical Computing, Vienna, Austria). Because correlation was possible between eyes of the same patient, generalized estimating equation (GEE)24 methods were used where appropriate. The primary outcome (eyes gaining 10 letters or more) was modeled using logistic regression coupled with GEEs; BCVA at baseline visit was used as a covariate. Change in CMT was modeled using generalized linear models with GEEs also using baseline BCVA as a covariate. To compare continuous numerical data, including BCVA, CMT, age, duration of diabetes, and mean glycosylated hemoglobin, t tests were used. Fisher exact tests for trends in proportions were used for gender analysis and change in cataract grade. The last observation carried forward method was used when patients were lost to follow-up. A sensitivity analysis was performed for the primary outcome, improvement of 10 letters, to investigate the potential effect of the missing data, by alternately
Patient-Reported Outcomes Patient-reported outcomes were measured using the Impact of Vision Impairment (IVI) questionnaire, a measure of vision-related quality of life, at baseline and again at 12 months. The IVI has been revised using Rasch analysis20 and has been used and validated for macular diseases, including age-related macular degeneration, macular telangiectasia, and low-vision rehabilitation,21e23 as a measure of patients’ perception of vision-related restriction on their activity and quality of life.
Figure 1. Diagram showing the progression of patients through the study. DEX ¼ dexamethasone; LOCF ¼ last observation carried forward.
3
Ophthalmology Volume -, Number -, Month 2014 assuming that all missing eyes had actually improved by 10 letters or that all missing eyes had not improved by 10 letters.25 Rasch analysis of patient-reported outcome was performed by WINSTEPS software version 3.30.1 (Rasch Measurement Computer Program, Chicago, IL) using a partial credit model.26 The resulting scores are person measures, or logits, which are interval estimates of the ordinal raw scores of questionnaire and calibrated within a range of 5.5 to þ 5.5 (the higher the score, the higher the increase in vision-related quality of life). The logits produced then were analyzed further in SPSS software version 21 (SPSS, Inc., Chicago, IL).
Results We enrolled 88 eyes from 61 patients, of which 42 eyes were randomized to receive bevacizumab and 46 received the dexamethasone implant. Twelve-month data were available for 81 eyes (92%) from 56 patients (92%). Figure 1 shows the flow of eyes through the study. Of the 7 eyes (from 6 patients) that exited the study, 4 had been assigned to bevacizumab and 3 to dexamethasone implant. Four of the 6 patients withdrew from the study because of difficulty meeting the rigor of attending monthly appointments. Two withdrew for perceived lack of efficacy of bevacizumab. Table 1 shows the baseline characteristics for the 2 treatment groups, which were similar. The primary outcome was the proportion of eyes in which BCVA improved by 10 letters or more at the 12-month study visit. Twenty-seven of 38 eyes treated with bevacizumab and 31 of 43 eyes receiving a dexamethasone implant had their primary outcome measured at 48 weeks, whereas the remainder of each group required further treatment, which they received, and the primary outcome was measured at 50 weeks. There was no significant difference between the mean primary outcome visual acuity scores at 48 and 50 weeks for either treatment group (not shown). A gain of 10 letters or more was found in 17 of the 42 eyes (40%) treated with bevacizumab and in 19 of the 46 eyes (41%) treated with dexamethasone implant (P ¼ 0.99) by the 12month visit. There was no correlation in the change in BCVA at 12 months between each eye for patients with both eyes in the study (r ¼ 0.006). Of the 42 eyes treated with bevacizumab, 13 (31%) gained more than 15 letters, compared with 10 (22%) of the dexamethasone implantetreated eyes (Table 2). The mean improvement in BCVA was 8.9 letters (95% confidence interval, 6.27e11.6) for bevacizumab-treated eyes and 5.6 (95%
Table 2. Effect of Treatment on Best-Corrected Visual Acuity from Baseline Change in Best-Corrected Visual Acuity
Bevacizumab
Gain of 15 letters Gain of 10e14 letters Gain of 5e9 letters No change (gain or loss 0.1).
Ocular Adverse Events Most ocular adverse events reported in this study were anticipated, including cataract and increased IOP (Table 3). There were no treatment-related cases of retinal detachment or endophthalmitis in either group. Elevation of Intraocular Pressure. An IOP elevation by at least 5 mmHg from baseline at any follow-up visit during the first
Figure 3. Bar graph showing the distribution of number of treatments given by treatment group for patients completing the 12-month follow-up visit. DEX ¼ dexamethasone.
5
Ophthalmology Volume -, Number -, Month 2014
Figure 4. Graph showing the mean central macular thickness by treatment type for each month of follow-up. DEX ¼ dexamethasone.
year of the study was seen in 46% (21/46) of dexamethasone implantetreated eyes and in 19% (8/42) of bevacizumab-treated eyes. Throughout the 12-month study, 12 eyes demonstrated an IOP of more than 25 mmHg at least once during follow-up visits. All 12 observed eyes had been treated with the dexamethasone implant. Eyes with increases in IOP were managed successfully with either observation or topical IOP-lowering medications. One eye from each group required selective laser trabeculoplasty to control IOP; both eyes were from the same patient. Cataract Progression. Twenty-six of the 88 eyes (29.5%) were pseudophakic at baseline, 10 of which were treated with bevacizumab (24% of the bevacizumab-treated eyes), whereas 16 were treated with the dexamethasone implant (35% of the dexamethasone implant eyes). There was no significant effect based on treatment received for the change in BCVA for the pseudophakic eyes (mean change in BCVA for bevacizumab eyes, 7.7 letters; mean change in BCVA for dexamethasone-treated eyes, 10.4 letters; P ¼ 0.47, t test for difference in BCVA change). Four of the
Table 3. Ocular Adverse Events Ocular Adverse Events IOP elevation 5 mmHg IOP elevation 10 mmHg Increase in cataract by >2 grades Cataract surgery Vision decrease of >10 letters Vitreous hemorrhage Floaters Subconjunctival hemorrhage Eye pain Itching Blepharitis Epiphora IOP ¼ intraocular pressure. Data are no. (%).
6
Bevacizumab (n [ 42)
Dexamethasone Implant (n [ 46)
8 (19) 0 (0) 2 (4.8)
21 (45.7) 9 (19.6) 6 (13)
1 0 0 0 3 3 0 1 2
(2.4) (0) (0) (0) (7.1) (7.1) (0) (2.4) (4.8)
3 5 2 5 2 1 3 1 0
(6.5) (10.9) (4.3) (10.9) (4.3) (2.2) (6.5) (2.2) (0)
remaining 62 eyes (6%) had cataract surgery during the first 12 months of the study, 1 treated with bevacizumab and 3 with the dexamethasone implant. Increase in cataract density by 2 grades or more from baseline was reported in 13% (6/46) of eyes in the dexamethasone implant group and in 4.8% (2/42) of eyes in the bevacizumab group. Loss of More Than 10 Letters. None of the bevacizumabtreated eyes lost more than 10 letters; however, 11% (5/46) of the dexamethasone implantetreated eyes did. This vision deterioration was because of an increase in cataract density in 4 cases and syphilitic chorioretinitis in 1 eye that lost 64 letters. This was associated with an attack of generalized secondary syphilis that developed 1 week after the eye had been treated with the dexamethasone implant. Given that syphilitic choroidoretinitis has been described after intraocular steroid treatment in 2 cases,27 the infection was deemed to be related to treatment.
Systemic Adverse Events The dexamethasone implant and bevacizumab injections were well tolerated systemically during the first year of the study. Serious systemic adverse events are summarized in Table 4. The most frequent systemic adverse event was the worsening of hypertension, which occurred in 6.7% (1/15), 10.5% (2/19), and 14.8% (4/ 27) of patients who received bevacizumab only, dexamethasone implant only, and both treatments, respectively.
Discussion This randomized controlled trial provides the first direct comparison of a slow-release steroid formulation made specifically for ophthalmic use with a commonly used antiVEGF agent for the treatment of DME. We found that approximately 40% of both groups achieved the prospectively identified primary outcome of a 10-letter gain over 1 year. The mean gain in visual acuity was more in the group receiving bevacizumab because of progression of cataract and a rare adverse event in the dexamethasone implante treated group, but this was not statistically significant.
Gillies et al
Bevacizumab vs. Dexamethasone for DME
Table 4. Systemic Adverse Events Systemic Adverse Event Worsening hypertension Chest pain Angina Myocardial infarction Congestive cardiac failure Cerebrovascular accident Kidney failure
Dexamethasone Bevacizumab D Bevacizumab Implant Dexamethasone Only (n [ 15) Only (n [ 19) Implant (n [ 27) 1 (6.7)
2 (10.5)
4 (14.8)
1 (6.7) 0 (0) 0 (0)
1 (5.3) 1 (5.3) 0 (0)
1 (3.8) 0 (0) 1 (3.8)
0 (0)
0 (0)
3 (11.5)
0 (0)
0 (0)
2 (7.7)
0 (0)
1 (5.3)
0 (0)
Data are no. (%).
Anatomic outcomes were significantly better in the dexamethasone implantetreated group with far fewer injections. Steroid-related adverse events seemed to be somewhat less frequent in eyes treated with the dexamethasone implant compared with triamcinolone acetonide or the fluocinolone implant (Iluvien; pSivida, Watertown, MA). Significant elevation of IOP (more than 10 mmHg) occurred in approximately 20% of eyes and progression of cataract by 2 grades in 14.5% of eyes. However, it is well established that steroid-related cataract generally appears in the second year after commencing intravitreal steroid therapy, so this rate is likely to increase in the second year of the trial.13 Cataract extraction was performed in 3 dexamethasone implante treated eyes compared with 1 bevacizumab-treated eye. By contrast, cataract surgery was required in 54% of eyes treated for 2 years in 1 randomized clinical trial of triamcinolone acetonide for DME13 and in virtually all phakic eyes receiving the fluocinolone implant.28 This study indicated that steroids may deliver better anatomic results than VEGF inhibitors. The mean reduction of CMT of dexamethasone implantetreated eyes was 187132.2 mm after 12 months compared with 122132.9 mm for the group receiving bevacizumab. This is consistent with the findings of a smaller, shorter study that compared bevacizumab with triamcinolone acetonide.29 The improved anatomic outcomes in the present study did not, however, translate to better visual acuity outcomes. The significance of persistent swelling despite ongoing treatment currently is uncertain and warrants longerterm studies. The much lower treatment intensity required for the dexamethasone implant (mean number of treatments, 2.7) was a distinct advantage over bevacizumab (mean number, 8.6). The dexamethasone implant is not as easy to administer because it requires a 22-gauge needle; however, this was not associated with an increased risk of operative complications. All patients were seen monthly in this study. In routine practice, one would still likely check dexamethasone implantetreated eyes at 4 to 6 weeks, especially to monitor IOP, so the number of visits may not be much less, but the reduced number of injections may be associated with fewer operative risks such as infectious endophthalmitis.
Patients also are likely to prefer fewer injections. When comparing the relative cost efficacy of the 2 classes of drugs, both fewer injections and fewer visits with steroids must be weighed against the substantially greater cost of the dexamethasone implant and the cost and clinical consequences of cataract and its removal, which is much more likely in eyes receiving steroids than in those receiving VEGF inhibitors. Somewhat more people who received both interventions, one in either eye, preferred the dexamethasone implant, although this was not statistically significant. The mistaken belief that the dexamethasone implant provides therapeutic effects for 6 months unfortunately has led to some industry-sponsored studies having difficulty achieving their originally specified primary end points.30,31 The mean CMT in the steroid-treated groups in the present study clearly peaked at 4 and 8 months. Further studies in which the dexamethasone implant is given for only 6-monthly for any condition should be avoided because they bring what could be an effective intervention into disrepute. Bevacizumab was the only VEGF inhibitor to which we had access when we started this study. There are few data on whether it is as effective as other VEGF inhibitors for DME, such as ranibizumab or aflibercept. It generally was believed to be equivalent to ranibizumab for neovascular age-related macular degeneration according to the Comparison of Age-Related Macular Degeneration Treatments Trials study investigators.9 Although local adverse events commonly are cited to relegate steroids to second-line treatment of DME, these need to be balanced against the potential for severe systemic adverse events associated with chronic VEGF inhibition. In the combined data from the RISE (Clinicaltrials.com: NCT00473330) and RIDE (NCT00473382) studies, for example, there was a dose-related increased mortality rate of 4.4% of participants treated with monthly ranibizumab 0.5 mg for 2 years for DME (compared with 1.2% for controls), which rose to 6.4% after 3 years.32 An increased systemic adverse event rate has not been reported with intravitreal corticosteroid therapy. Although there were no deaths in the present study, there were 2 cerebrovascular accidents and 1 myocardial infarction in patients who were receiving both interventions. Limitations of this study include its relatively small size and short duration. Because the proportion of eyes in the 2 groups that achieved the primary outcome, a gain of 10 letters, was essentially the same, larger studies are unlikely to find a significant difference between the 2 interventions at least after 12 months of treatment. Although the present study will continue for a second year, particularly to track the effect of steroid-induced cataract, DME is a chronic disease that conceivably could require life-long treatment in some patients. Randomized clinical trials provide little to no information on these long-term outcomes. There is an urgent need for postmarketing observational studies and data linkage to track long-term outcomes in patients receiving treatment for DME. This study provided direct evidence that the dexamethasone implant can achieve similar rates of vision improvement compared with bevacizumab for DME with superior anatomic outcomes and fewer injections. However, more
7
Ophthalmology Volume -, Number -, Month 2014 patients receiving the dexamethasone implant lost vision, mainly because of cataract. Therefore, the dexamethasone implant may be considered as second-line treatment for DME in phakic eyes and as first-line treatment for pseudophakic eyes or for patients who are averse to frequent intraocular injections or have difficulty attending regularly. Acknowledgment. The authors thank the Safety Monitoring Committee for this study: Dr. Jeremy Smith (Chair), Dr. JieJin Wang, and Paul Power.
References 1. Moss SE, Klein R, Klein BE. The 14-year incidence of visual loss in a diabetic population. Ophthalmology 1998;105: 998–1003. 2. Yau JW, Rogers SL, Kawasaki R, et al; Meta-Analysis for Eye Disease (META-EYE) Study Group. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care 2012;35:556–64. 3. Mitchell P, Wong TW; Diabetic Macular Edema Treatment Guideline Working Group. Management paradigms for diabetic macular edema. Am J Ophthalmol 2014;157:505–13. 4. Mitchell P, Bandello F, Schmidt-Erfurth U, et al; RESTORE Study Group. The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema. Ophthalmology 2011;118:615–25. 5. Massin P, Bandello F, Garweg JG, et al. Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study. Diabetes Care 2010;33:2399–405. 6. Do DV, Nyuyen QD, Khwaja AA, et al; READ-2 Study Group. Ranibizumab for Edema of the Macula in Diabetes Study: 3-year outcomes and the need for prolonged frequent treatment. JAMA Ophthalmol 2013;131:139–45. 7. Diabetic Retinopathy Clinical Research Network, Elman MJ, Qin H, Aiello LP, et al. Intravitreal ranibizumab for diabetic macular edema with prompt versus deferred laser treatment: three-year randomized trial results. Ophthalmology 2012;119: 2312–8. 8. Nguyen QD, Brown DM, Marcus DM, et al; RISE and RIDE Research Group. Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE. Ophthalmology 2012;119:789–801. 9. Comparison of Age-Related Macular Degeneration Treatments Trials (CATT) Research Group, Martin DF, Maguire MG, Fine SL, et al. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology 2012;119:1388–98. 10. Rajendram R, Fraser-Bell S, Kaines A, et al. A 2-year prospective randomized controlled trial of intravitreal bevacizumab or laser therapy (BOLT) in the management of diabetic macular edema: 24-months data: report 3. Arch Ophthalmol 2012;130:972–9. 11. Sultan MB, Zhou D, Loftus J, et al; Macugen 1013 Study Group. A phase 2/3, multicenter, randomized, double-masked, 2-year trial of pegaptanib sodium for the treatment of diabetic macular edema. Ophthalmology 2011;118:1107–18. 12. Do DV, Nguyen QD, Boyer D, et al; DA VINCI Study Group. One-year outcomes of the DA VINCI Study of VEGF TrapEye in eyes with diabetic macular edema. Ophthalmology 2012;119:1658–65. 13. Gillies MC, Sutter FK, Simpson JM, et al. Intravitreal triamcinolone for refractory diabetic macular edema:
8
14.
15.
16.
17.
18.
19. 20.
21.
22. 23. 24. 25. 26. 27.
28.
29.
30.
two-year results of a double-masked, placebo-controlled, randomized clinical trial. Ophthalmology 2006;113: 1533–8. Gillies MC, Simpson JM, Gaston C, et al. Five-year results of a randomized trial with open-label extension of triamcinolone acetonide for refractory diabetic macular edema. Ophthalmology 2009;116:2182–7. Gillies MC, McAllister IL, Zhu M, et al. Intravitreal triamcinolone prior to laser treatment of diabetic macular edema: 24-month results of a randomized controlled trial. Ophthalmology 2011;118:866–72. Chang-Lin JE, Attar M, Acheampong AA, et al. Pharmacokinetics and pharmacodynamics of a sustained-release dexamethasone intravitreal implant. Invest Ophthalmol Vis Sci 2011;52:80–6. Haller JA, Kuppermann BD, Blumenkranz MS, et al; Dexamethasone DDS Phase II Study Group. Randomized controlled trial of an intravitreous dexamethasone drug delivery system in patients with diabetic macular edema. Arch Ophthalmol 2010;128:289–96. Boyer DS, Faber D, Gupta S, et al; Ozurdex CHAMPLAIN Study Group. Dexamethasone intravitreal implant for treatment of diabetic macular edema in vitrectomized patients. Retina 2011;31:915–23. van Wijngaarden P, Coster DJ, Williams KA. Inhibitors of ocular neovascularization: promises and potential problems. JAMA 2005;293:1509–13. Lamoureux EL, Pallant JF, Pesudovs K, et al. The Impact of Vision Impairment Questionnaire: an evaluation of its measurement properties using Rasch analysis. Invest Ophthalmol Vis Sci 2006;47:4732–41. Lamoureux EL, Pallant JF, Pesudovs K, et al. The effectiveness of low-vision rehabilitation on participation in daily living and quality of life. Invest Ophthalmol Vis Sci 2007;48: 1476–82. Pesudovs K, Caudle LE, Rees G, Lamoureux EL. Validity of a visual impairment questionnaire in measuring cataract surgery outcomes. J Cataract Refract Surg 2008;34:925–33. Finger RP, Fenwick E, Marella M, et al. The impact of vision impairment on vision-specific quality of life in Germany. Invest Ophthalmol Vis Sci 2011;52:3613–9. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach. Biometrics 1988;44:1049–60. Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomised controlled trials. BMJ 1999;319:670–4. Masters GN. A Rasch model for partial credit scoring. Psychometrika 1982;47:149–74. Song JH, Hong YT, Kwon OW. Acute syphilitic posterior placoid chorioretinitis following intravitreal triamcinolone acetonide injection. Graefes Arch Clin Exp Ophthalmol 2008;246:1775–8. Campochiaro PA, Brown DM, Pearson A, et al; FAME Study Group. Long-term benefit of sustained-delivery fluocinolone acetonide vitreous inserts for diabetic macular edema. Ophthalmology 2011;118:626–35. Kriechbaum K, Prager S, Mylonas G, et al; Diabetic Retinopathy Research Group. Intravitreal bevacizumab (Avastin) versus triamcinolone (Volon A) for treatment of diabetic macular edema: one-year results. Eye (Lond) 2014;28:9–15. Callanan DG, Gupta S, Boyer DS, et al; Ozurdex PLACID Study Group. Dexamethasone intravitreal implant in combination with laser photocoagulation for the treatment of diffuse diabetic macular edema. Ophthalmology 2013;120:1843–51.
Gillies et al
Bevacizumab vs. Dexamethasone for DME
31. Haller JA, Bandello F, Belfort R Jr, et al; Ozurdex GENEVA Study Group. Dexamethasone intravitreal implant in patients with macular edema related to branch or central retinal vein occlusion: twelve-month study results. Ophthalmology 2011;118:2453–60.
32. Brown DM, Nguyen QD, Marcus DM, et al; RIDE and RISE Research Group. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology 2013;120: 2013–22.
Footnotes and Financial Disclosures Originally received: April 22, 2014. Final revision: June 26, 2014. Accepted: July 2, 2014. Available online: ---.
Allergan Pharmaceutical. The author(s) have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2014-605.
1
The Save Sight and Eye Health Institute, Sydney Medical School, The University of Sydney, Sydney NSW, Australia. 2 Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Hospital, East Melbourne, Victoria, Australia. 3
Lions Eye Institute, Centre for Ophthalmology and Visual Science, University of Western Australia, Australia. Financial Disclosure(s): Funded by a project grant from the National Health and Medical Research Council, which was supplemented by an unrestricted educational grant from
Abbreviations and Acronyms: BCVA ¼ best-corrected visual acuity; BP ¼ blood pressure; CMT ¼ central macular thickness; DME ¼ diabetic macular edema; GEE ¼ generalized estimating equation; IOP ¼ intraocular pressure; IVI ¼ Impact of Vision Impairment; SD ¼ standard deviation; VEGF ¼ vascular endothelial growth factor. Correspondence: Samantha Fraser-Bell, MBBS, PhD, Save Sight and Eye Health Institute, University of Sydney, 8 Macquarie Street, Sydney NSW 2000, Australia. E-mail: [email protected].
9
