DME: What Trials Tell Us About Treatment

Even though anti-VEGF agents appear to be effective, surgeons would like a long-term solution.

Walter Bethke, Managing Editor
8/9/2012

Diabetic macular edema has responded remarkably well to anti-VEGF treatment in the major prospective trials.
The treatment landscape for diabetic macular edema changed drastically in just a few years with the introduction of anti-vascular endothelial growth factor drugs into the mix. Instead of just getting by using laser to stabilize vision in some patients, with VEGF blockers physicians could now actually improve it. However, keeping track of the recent landmark studies in DME treatment can be challenging. Here, several retinal specialists acquainted with the major trials share insights into the studies’ results and discuss what they mean for treatment in the clinic today and in the future.

The Anti-VEGF Armamentarium

Several studies have taken a look at anti-VEGF agents and where they fit in terms of clinical use. Here’s a look at the major agents and their trials.

First, the VEGF-blocker ranibizumab (Lucentis) has performed well in several studies, particularly RISE and RIDE and RESTORE. RISE and RIDE will be used for Food and Drug Administration approval for a DME indication for Lucentis.

In the RISE study, 377 patients were randomized to monthly injections of 0.3-mg ranibizumab, 0.5-mg ranibizumab or sham. In the nearly identical RIDE study, 382 patients were randomized to the same options. Beginning at three months, macular laser rescue treatment was available to all patients, if necessary. Panretinal laser was available as indicated. After two years, patients in the sham group were eligible to receive monthly injections of 0.5-mg ranibizumab. 

At two years 18.1 percent of sham patients gained at least 15 letters vs. 44.8 percent of 0.3-mg (p<0.0001) and 39.2 percent of 0.5-mg ranibizumab patients (p<0.001). In RIDE, significantly more ranibizumab-treated patients gained 15 letters or more: 12.3 percent of sham patients vs. 33.6 percent of 0.3-mg patients (p<0.0001) and 45.7 percent of 0.5-mg ranibizumab patients (p<0.0001).1

 Surgeons familiar with the studies say they were also pleased with the safety profile of ranibizumab as demonstrated in RISE and RIDE. There were four cases of endophthalmitis in the treatment patients, and the total incidence of nonfatal myocardial infarctions, nonfatal cerebrovascular accidents, and deaths from vascular or unknown causes was 4.9 percent to 5.5 percent of sham patients and 2.4 percent to 8.8 percent of ranibizumab patients at two years.

In RESTORE, researchers compared 0.5-mg ranibizumab monotherapy or combined with laser to laser alone in 345 patients randomized to one of the three treatments. They administered ranibizumab or sham for three months and then p.r.n., and they performed laser or sham at baseline and then p.r.n.

The study doctors found that ranibizumab alone and combined with laser were superior to laser monotherapy in improving the average change in best-corrected letter score from baseline to months one through 12 (+6.1 letters for drug only and +5.9 for drug and laser vs. +0.8 for laser; both p<0.0001). At one year, a significantly greater proportion of ranibizumab patients (22.6 percent) had a BCVA letter score of 15 or better, and 53 percent had a BCVA letter score level greater than 73 (20/40 Snellen equivalent); for the ranibizumab + laser group, it was 22.9 percent and 44.9 percent, respectively. For laser alone, 8.2 percent and 23.6 percent of patients reached these endpoints.2

In RESTORE, the mean central retinal thickness was significantly reduced from baseline with ranibizumab (-118.7 μm) and ranibizumab + laser (-128.3 μm) vs. laser alone (-61.3 μm; both p<0.001). Patients received a mean of seven ranibizumab/sham injections over 12 months. No endoph­thalmitis cases occurred in RESTORE, but increased intraocular pressure was reported for one patient each in the ranibizumab arms. Neither ranibizumab monotherapy nor the drug combined with laser was associated with an increased risk of cardiovascular or cerebrovascular events in the study.

“In RISE/RIDE, anti-VEGF therapy proved to be a marked improvement over laser treatment for center-involved DME,” says Beverly Hills, Calif., surgeon David Boyer, MD, who participated in the study. “Also, this is a vasculopathic group of patients, and everyone in the study was concerned that the patients would have a much higher incidence of Antiplatelet Trialists’ Collaborative events [myocardial infarction, stroke, etc.] but that wasn’t borne out in the study. There was no significant number of APTCs that would cause alarm.”

Along with ranibizumab, another drug being studied for possible approval for the treatment of DME in the United States is Regeneron’s VEGF Trap-Eye (Eylea, aflibercept). 

Aflibercept was studied in the randomized, double-masked, Phase II DAVINCI trial. In DAVINCI, 221 diabetic patients with center-involved DME were randomized to one of five treatments: aflibercept 0.5 mg every four weeks (0.5q4); 2 mg every four weeks (2q4); 2 mg every eight weeks after three initial monthly doses (2q8); or 2 mg dosing as needed after three initial monthly doses (2p.r.n.); or macular laser photocoagulation.

The mean improvements in BCVA in the aflibercept groups at six months were 11 (0.5q4), 13.1 (2q4), 9.7 (2q8), and 12 letters (2p.r.n.), vs. -1.3 letters for the laser group (p=0.0001). Proportions of eyes with gains in BCVA of 15 or more ETDRS letters at one year in the aflibercept groups were 40.9 percent (0.5q4), 45.5 percent (2q4), 23.8 (2q8), and 42.2 percent (2p.r.n.) vs. 11.4 percent for laser (p=0.0031, p=0.0007, p=0.1608 and p=0.0016, respectively, vs. laser).3

The mean reductions in central retinal thickness in the aflibercept groups at one year were -165.4 μm (0.5q4), -227.4 μm (2q4), -187.8 μm (2q8), and -180.3 μm (2p.r.n.) vs. -58.4 μm for laser (p<0.0001). The researchers say the most frequent ocular adverse events with aflibercept were conjunctival hemorrhage, eye pain, ocular hyperemia and increased intraocular pressure, and common systemic adverse events included hypertension, nausea and congestive heart failure.3

The next step for aflibercept is VISTA, the three-year registration study for FDA approval of the agent for diabetic macular edema. “The recruitment is complete,” says Winter Haven, Fla., retinal specialist Michael Tolentino. “The study design is well-done, and it will answer such questions as aflibercept vs. laser.”

Rounding out the studies of the major anti-VEGF agents is the prospective, randomized BOLT study, which analyzed the use of bevacizumab in DME in 80 patients.4

In BOLT, at two years the average BCVA was 20/50 in the bevacizumab group vs. 20/80 in the macular laser group (p=0.005). The bevacizumab arm gained a median of nine ETDRS letters vs. 2.5 for ML (p=0.005), with a mean gain of 8.6 letters for bevacizumab vs. a mean loss of 0.5 letters for ML. Forty-nine percent of bevacizumab patients gained 10 or more letters (p=0.001) and 32 percent gained at least 15 letters (p=0.004) vs. 7 percent and 4 percent for MLT. The percentage who lost fewer than 15 letters in the ML arm was 86 percent vs. 100 percent for bevacizumab (p=0.03). The mean reduction in central macular thickness was 146 μm in the bevacizumab arm vs. 118 μm in the ML arm. The median number of treatments over 24 months was 13 for bevacizumab and four for MLT.

“It was reassuring that the treatment effect we saw at the end of 12 months was maintained out to 24 months,” says BOLT investigator 
Michel Michaelides, MD, of Moorfield’s Eye Hospital in London. “We didn’t see any significant adverse events including those related to macular perfusion.” Dr. Michaelides cautions about applying the results to all patients. “We were quite broad, but even we excluded patients with HBA1c of greater than 11 percent,” he says. 

Though the anti-VEGF drugs have shown very good results, some physicians think differences in the agents’ molecular makeup will affect their usability over the long term. “I have patents on several molecules that I designed,” says Dr. Tolentino, who consults for Regeneron and Genentech. “In designing a molecule, you can study it and give it properties that are advantageous for your purposes. Avastin, unfortunately, is a monoclonal antibody, which is immunogenic, and we know that DME in diabetes is an inflammatory condition with antibody-mediated inflammation.” He says it’s possible long-term Avastin use could feed the inflammation. “Lucentis is a fab fragment, meaning it’s just one of the arms of an antibody,” Dr. Tolentino continues. “Though it doesn’t have the true inflammatory potential of a full-length antibody like Avastin, it still does things like form immune complexes, so it is somewhat inflammatory. The thing about aflibercept that’s special is that though it uses a backbone of an antibody, it uses the least inflammatory backbone.”

An eye treated with bevacizumab for diabetic macular edema at baseline (top), six months (middle) and a year shows a persistent reduction in macular thickness.
Dr. Michaelides, however, found bevacizumab to be safe in the BOLT trial. “There was nothing significantly different between the drug and the laser arms,” he says. “We found similar side effects. If you remove adverse events such as subconjunctival hemorrhage and ocular discomfort—which we now know are predictable—there really wasn’t a difference between the groups.” Dr. Michaelides adds that he also finds it encouraging that the number of required injections decreased in the second year of the study.

There has also been a discussion about how the molecules’ particular designs might affect their ability to last longer in the eye. “Because of aflibercept’s molecular makeup, it holds onto VEGF for a long time,” says Dr. Boyer. “It seems to have a higher affinity for VEGF. Because of this, it appears its effect may be longer. There have been studies in AMD that show after three loading doses you can give the drug every two months and still get good visual results. Of course, there will be individual patients who require it monthly anyway, but it may last longer in general, so that’s why we’re going to do studies to figure that out.”

The excellent efficacy anti-VEGF agents have shown in their trials has led physicians to change how they use laser treatment for DME, and it’s no longer the instant first-line treatment it once was. “If you have a patient with a focal area of leakage well away from the foveal center, laser might be a better option than a drug because you have a high chance of curing that leakage with laser therapy,” says Dr. Michaelides. Andrew Antoszyk, MD, of Charlotte, N.C., agrees. “A lot of people will continue to use laser to try to reduce the treatment burden of intravitreal injections using a Protocol-I kind of approach [a protocol from the Diabetic Retinopathy Clinical Research Network’s study of DME],”5 he says. “This would involve using an intravitreal anti-VEGF agent followed by laser coagulation, then treating with anti-VEGF and reassessing for the need for supplemental laser every three to four months.”

Steroid Inserts

Surgeons who had been hoping for an implantable steroid for select patients with DME had their hopes dashed when the FDA denied approval of Alimera’s Iluvien fluocinolone insert in November of 2011, citing issues with the risks of adverse reactions, including intraocular pressure increases and cataract, that were manifest during the 36-month FAME study. The insert, which is designed to deliver drug for up to three years, is approved in several European countries.

In FAME, at three years 33 percent of patients receiving a 0.2-mg insert and 31.9 percent of those receiving a 0.5-mg insert gained at least 15 letters of vision, vs. 21.4 percent in a sham group (p=0.030). However, patients with DME with a duration of three or more years had an even better result: 34 percent of low-dose patients and 28.8 percent of high-dose patients improved by 15 letters or more, vs. just 13.4 percent of sham. In terms of adverse events, 82 percent of low-dose patients underwent cataract surgery, as did 88 percent of the high-dose group. Dr. Antoszyk, who took part in the study, says 38 percent of the 0.2-mg group and 47 percent of the 0.5-mg group required glaucoma drops. Also, 4.8 percent of the low-dose group and 8.1 percent of the high-dose group required incisional glaucoma surgery.

Many physicians, though, think the benefits of Iluvien outweigh the side effects. “I wish Iluvien were available,” says Szilard Kiss, MD, assistant professor of ophthalmology at Weill Cornell Medical College in Manhattan, “because there are certain patients I’d like to use it on. Remember, in the Iluvien trials, patients with macular edema longer than three years benefited the most. In someone with longstanding edema, there are other things going on of an inflammatory nature that steroid would address that simple anti-VEGF treatment would not. It’s one of those options where you know there’s a side-effect profile that includes glaucoma and cataract, but if the macular edema and vision improve you’re willing to take such a risk—or at least I would be.” When asked about the fate of the implant in the United States, Alimera representatives said they couldn’t discuss any possible plans.

U.S. ophthalmologists aren’t all out of implantable steroid options, however, as Allergan’s Ozurdex dexamethasone implant is making its way through the approval process for a DME indication (it’s already approved for macular edema due to retinal vein occlusion). The Ozurdex has a shorter duration of action compared to the Iluvien, lasting two to three months according to physicians.

Though the latest data from the trial hasn’t been made available in the run-up to its application for FDA approval, a study of Ozurdex in a subset of DME patients showed it might fill a niche that anti-VEGF drugs can’t: vitrectomized eyes. “Because of the vitrectomy, the pharmacokinetics of all drugs is shortened dramatically, so even the anti-VEGF drugs don’t work well in vitrectomized eyes,” says Dr. Boyer. “So Ozurdex and Iluvien would both be welcomed in those cases. In a small, Phase II trial, the Ozurdex implant was shown to be very effective in vitrectomized eyes.” In the study of 55 patients with treatment-resistant DME, at week eight, 30.4 percent of them had gained at least 10 letters in best-corrected visual acuity.6

Though anti-VEGF drugs currently yield the best results with the fewest side effects, Dr. Tolentino says the idea behind the sustained-release steroid is a good one, because it cuts down on the number of injections. “Let’s say we start a diabetic patient on anti-VEGF injections in his 40s, and he lives to be 80,” he muses. “He’s going to be getting these injections for almost 40 years. What’s the long-term consequence of this? You have to somehow reduce the number of injections.”  REVIEW



1. Nguyen QD, Brown DM, Marcus DM, et al. Ranibizumab for DME: Results from 2 phase III randomized trials: RISE and RIDE. Ophthalmology 2012 Apr;119:4:789-801. Epub 2012 Feb 11.
2. Mitchell P, Bandello F, Schmidt-Erfurth U, et al. The RESTORE study: Ranibizumab monotherapy or combined with laser versus laser monotherapy for DME. Ophthalmology 2011 Apr;118:4:615-25.
3. Do DV, Nguyen QD, Boyer D, et al. One-Year Outcomes of the DA VINCI Study of VEGF Trap-Eye in Eyes with Diabetic Macular Edema. Ophthalmology 2012 Apr 24. [Epub ahead of print]
4. Michaelides M, 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-Month Data: Report 3. Arch Ophthalmol 2012 Apr 9. [Epub ahead of print]
5. Elman MJ, Aiello LP, Beck RW, et al. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2010;117:6:1064-1077.
6. Boyer DS, Faber D, Gupta S, et al. Dexamethasone intravitreal implant for treatment of diabetic macular edema in vitrectomized patients. Retina 2011;31:5:915-23.