American Academy of Ophthalmology Web Site: www.aao.org
On The Near Horizon: Better Treatments for DME
Several studies of diabetic macular edema are pushing retina experts to recalibrate their approach to both old and new therapies.
Last summer’s finding by the Diabetic Retinopathy Clinical Research Network (DRCR.net) that laser therapy is more effective—and has fewer side effects—than corticosteroids in the treatment of diabetic macular edema was a resounding affirmation of the gold standard. It also surprised some investigators, who had assumed, perhaps even hoped, that intravitreal triamcinolone (IVT), used off-label since 2001, would beat out the vetted treatment that had been around for more than 20 years.
The trial’s results also affirmed that there are still no pharmacologic treatments for DME proven to be as efficacious as laser, which destroys one part of the retina to save another. But soon that could change. Researchers have been investigating drugs that target the biochemical mechanisms that underlie DME. Specifically, these treatments aim to inhibit the growth of new blood vessels or the activity of certain proteins in the retinal cells.
“VEGF antagonists will be the next line of treatment,” predicted Peter A. Campochiaro MD, who has been involved in a number of DME trials. Dr. Campochiaro, professor of ophthalmology and neuroscience at Johns Hopkins University, attributed the recent flurry of interest in DME to demonstrations that VEGF plays a critical role in the pathogenesis of the disease. “Now that we know what the target is, we will see many, many clinical trials to antagonize VEGF,” he said. He also said researchers will seek to find solutions that avoid frequent intraocular injections.
Old treatments never die. Laser, and perhaps corticosteroids as well, will still hold a place in the armamentarium. “If I was to predict, I would think that laser will certainly be used, particularly for focal macular edema,” Dr. Campochiaro said. Researchers also are looking at ways to continue the use of corticosteroids, most likely in combination with another drug or with laser. But Dr. Campochiaro predicted corticosteroids will be relegated to a second-line treatment because of increased IOP and cataract formation.
Novel delivery approaches, such as topical drugs or sustained-release devices, are also on the horizon, Dr. Campochiaro said. “The future looks pretty good. Over the next five years, you’re going to see several new treatments for DME.” Following is a look at several contemporary trials, beginning with the DRCR.net, for managing DME.
Laser Wins Contest With Corticosteroids
In 2001, doctors began reporting success using intravitreal corticosteroids to treat DME. Improvement, said Michael S. Ip, MD, “at least in the short term, was above and beyond anything we’d seen, including laser.” Dr. Ip is a DRCR.net investigator and associate professor of ophthalmology at the University of Wisconsin in Madison. Widespread use of offlabel corticosteroids followed, despite the known side effects. Not everyone, though, was comfortable using an untested treatment, Dr. Ip recalled. The community was split.
Along came the DRCR.net, a collaborative network supported by the NEI to conduct multicenter clinical research on diabetic retinopathy and associated disorders. The network conducted the first major study comparing the long-term benefits, as well as potential side effects, of laser and IVT.
After two years, laser proved more effective than intravitreal injection of either a 1-mg or 4-mg dose of triamcinolone in reducing vision loss from DME.1 The study found that 28 percent of the corticosteroid group experienced substantial vision loss, compared with 19 percent in the laser group. What’s more, one-third of the laser group experienced substantial improvement in vision, a benefit never before attributed to laser. “The findings clearly indicate that laser is superior to IVT,” said Dr. Ip, who was chairman of the protocol for the DRCR.net study. They also reaffirmed the lesson demonstrated by the Early Treatment Diabetic Retinopathy Study in 1985 that laser is an effective treatment against DME.
Up from behind. It took time for laser to triumph. At four months, mean visual acuity was better in the 4-mg triamcinolone group than in either the laser or 1-mg group. Then at one year, there were no significant differences among groups.
Then at 16 months, and through the primary outcome visit at two years, mean visual acuity was best in the laser group. “The effect of laser is slower than IVT but is steady and more sustained,” Dr. Ip said. Laser also was associated with far fewer adverse effects. Still, the study found that IVT is effective, especially early on. This suggested a possible role for corticosteroids in combination with laser, or even some other drug. “It still does make sense to look at steroids as a potential treatment,” Dr. Ip said. “But not alone.”
Ingrid U. Scott, MD, MPH, agreed, “We can’t conclude from the study that triamcinolone doesn’t work. But as monotherapy it did not do better in the long term than focal laser.” Dr. Scott is a DRCR.net investigator and professor of ophthalmology and public health sciences at Penn State in Hershey. She currently is involved in a follow-up study that is testing intravitreal steroids in conjunction with focal photocoagulation (see “Ranibizumab in DRCR” below).
For now, however, the laser vs. IVT study should put to rest the use of intravitreal steroids as monotherapy. “This is a definitive study,” Dr. Ip said. “I think therapy with IVT alone for DME will be abandoned as a result of this trial. The findings clearly indicate that laser is superior to IVT.”
Timed-Release: New Life for Corticosteroids?
The DRCR.net trial may have found that laser is more effective than triamcinolone injections for treating DME, but Ken Green, PhD, chief scientific officer at Alimera Sciences, is among those who say there may be a role, still, for corticosteroids. The challenge, he said, will be finding the right corticosteroid with the right delivery system. “Corticosteroids are very effective because they work on lots of different aspects of the condition,” he said, noting that they suppress VEGF production, are anti-inflammatory, inhibit leukostasis and also increase tight junctions between cells, which is helpful in reducing vascular permeability.
The rationale for the use of corticosteroids to treat DME also follows from the observation that the increase in retinal capillary permeability that results in edema may be caused by a breakdown of the blood-retina barrier, mediated in part by VEGF. Corticosteroids have been demonstrated to have antiinflammatory properties and to inhibit the expression of both VEGF and the VEGF gene.1
The downside of corticosteroids is that they have substantial side effects, most notably cataract and glaucoma. But Alimera may have found a way to minimize those side effects with Iluvien, a timed-release insert that slowly releases fluocinolone acetonide into the eye.
Fame: hoping for more than 15 minutes. A three-year, phase 3 clinical trial, FAME (Fluocinolone Acetonide in Diabetic Macular Edema), is under way to test the effect of two different doses delivered by Iluvien in 956 patients at more than 100 clinical sites worldwide. FAME is testing two doses of fluocinolone acetonide (FA)—0.23 µg and 0.45 µg—and is comparing those with a sham injection, with the possibility for laser rescue after six weeks.
Iluvien (formerly known as Medidur) is a tube 3.5 mm long and 0.37 mm in diameter containing a matrix in which the FA is embedded. The device, which is shorter than a grain of rice and thinner than a human hair, is implanted intravitreally with a 25-gauge inserter, small enough to allow for a self-healing wound. Each Iluvien insert is designed to provide a sustained therapeutic effect, up to 24 months for the high dose and 36 months for the low dose. The insert, made of the same nonbioerodible material as the haptics of intraocular lenses, stays in the eye after all the drug is released.
Iluvien builds on the partial success of another FA implant, Retisert, which had good effect in patients with DME but an unacceptable safety record. (Thirty-three percent of patients in a phase 3 clinical trial required an operation to relieve elevated IOP, and 95 percent required cataract surgery.)2 Iluvien’s positioning in the posterior chamber is intended to avoid those side effects. It is designed to take advantage of the fluid dynamics in the vitreous in order to prevent migration of the drug to the anterior chamber, potentially avoiding interaction with the cells of the trabecular meshwork, Dr. Green explained. He added that FA might work better than IVT. Unlike triamcinolone, FA is released in a very controlled manner, he said. “There’s hardly any burst of drug.” And unlike IVT, which has a peak effect at four to six weeks, at which time patients may need another injection, FA releases slowly and steadily over two to three years, depending on the dose, he said.
What’s more, the dosing being tested in the FAME trial (0.23 µg and 0.45 µg) is far smaller than the commonly used 4-mg dose of triamcinolone, and even smaller than the 0.6 µg dose in the Retisert study. “The Retisert studies taught us that, at least with FA, you could administer submicrogram levels on a daily basis and get a profound therapeutic effect. No one could have imagined that,” said Dr. Green.
Ranibizumab in DRCR
Beyond laser and corticosteroids is an approach borrowed from antiangiogenesis research: Could ranibizumab (Lucentis), an effective agent in the treatment of AMD, become part of the DME armamentarium? If so, will the specific anti- VEGF inhibitor be most effective as monotherapy or when used in combination with laser? These are a few of the questions currently being addressed in another DRCR.net trial.
This phase 3 trial follows a phase 2 trial that investigated the short-term effect of intravitreal bevacizumab (Avastin) for DME. The earlier study included 121 eyes of 121 patients randomized to one of five groups: focal laser at baseline; intravitreal injection of 1.25 mg of bevacizumab at baseline and at six weeks; intravitreal injection of 2.5 mg of bevacizumab at baseline and six weeks; intravitreal injection of 1.25 mg of bevacizumab at baseline and sham injection at six weeks; or intravitreal injection of 1.25 mg of bevacizumab at baseline and six weeks, with laser at three weeks.
That earlier trial showed that focal photocoagulation alone, intravitreal bevacizumab alone and intravitreal bevacizumab plus laser all reduced DME in some eyes, for at least a short time, said Dr. Scott, who was protocol chairwoman for the phase 2 study and is an investigator for the current phase 3 trial. But the study didn’t answer whether switching to bevacizumab in patients where laser failed would be helpful, harmful or have no benefit, Dr. Scott said.
In the current phase 3 trial, sponsored by the NEI in collaboration with Allergan and Genentech, patients are randomized into four groups: sham injection plus focal photocoagulation; 0.5-mg injection of intravitreal ranibizumab plus laser; 0.5-mg injection of intravitreal ranibizumab plus deferred laser; and 4-mg triamcinolone plus focal photocoagulation.
Cortico question. The last of these groups is poised to answer a question raised by an earlier trial: whether corticosteroids, which as monotherapy didn’t work as well as laser, are nevertheless effective in combination with laser.
Ranibizumab in READ2
“It’s preliminary, but it’s exciting,” Jennifer I. Lim, MD, said, about the improvement observed in another ranibizumab study—the READ2 trial. “The percent of eyes with three or more lines of improvement is very impressive for these diabetic macular edema patients at six months.” Dr. Lim is professor of ophthalmology and director of the retina service at the University of Illinois at Chicago. She is also an executive committee member for READ2. (The acronym stands for “Ranibizumab for Edema of the mAcula in Diabetes.”)
READ1. The earlier READ1, which also tested the role of ranibizumab, demonstrated that VEGF is an important therapeutic target for DME.3 In READ1, 10 patients with chronic DME received intraocular injections of 0.5 mg ranibizumab at baseline, one, two, four and six months. At seven months, mean foveal thickness was reduced from a baseline of 503 µm to 257 µm. READ1 established the dosing interval for READ2, Dr. Lim said. “The investigators found that if you waited between treatments, there was sometimes a larger response than monthly treatment. You got a more robust effect.”
READ2. In READ2, 126 patients were randomized to 0.5 mg ranibizumab alone, laser alone or a combination of ranibizumab followed seven days later by focal laser.
At six months, READ2, which is funded by the Juvenile Diabetes Research Foundation and Genentech, found that 24 percent of patients in the ranibizumab group and 12 percent in the combination group had experienced a three-line improvement. No improvement was seen with laser.
The ranibizumab group also experienced a reduction in thickness of the central retina, Dr. Lim said. There were no adverse events related to the drug or to laser.
Dr. Lim acknowledged that during the two years that READ2 plans to follow patients, the laser group might catch up, as it did in the DRCR.net triamcinolone vs. laser study. But for now, she said, “Ranibizumab does appear to be effective in reducing thickness and improving vision early, without significant side effects.”
A Trap for VEGF
Regeneron Pharmaceuticals is hoping to translate early success with its anti-VEGF product for treating wet age-related macular degeneration into a similar success with diabetic macular edema. VEGF Trap-Eye, which has undergone a phase 2 trial and currently is being studied in phase 3 wet AMD trials, is a soluble, fully human VEGF receptor fusion protein that binds to all forms of VEGF-A and the related placental growth factor. “The binding affinity to VEGF is very high,” said Avner Ingerman, MD, vice president of development at Regeneron. “That’s one of the reasons it was called VEGF Trap.”
In theory, the intravitreally injected VEGF Trap-Eye will bind to VEGF molecules in the eye, blocking the VEGF from reaching receptors on the cell surface. In a normal eye, VEGF that reaches the receptor will set off a cascade of events leading to angiogenesis.
Trappings of success. The one-year results of a multi-center phase 2 trial for wet AMD were promising. In September 2008, Regeneron and Bayer HealthCare reported that, after 12 months, the VEGF Trap-Eye demonstrated improved visual acuity and reduced anatomical markers of neovascular disease.
The double-masked, prospective, randomized, trial involving 157 patients also showed a reduction in the size of the total active choroidal neovascular membrane. Regeneron is now proceeding with two phase 3 trials comparing Trap-Eye with ranibizumab to treat AMD.
Thanks to this momentum, the company is launching a phase 2 study of VEGF Trap-Eye to treat DME. “The rationale for using anti-VEGF has been established in the wet AMD population. If you can extrapolate that, it leads to the rationale for using anti-VEGFs in DME,” Dr. Ingerman said. “The point is to stop the angiogenesis and to stop the leakage.”
2 “Diabetic Macular Edema Trial with Retisert,” media release, pSivida, Feb. 21, 2006.
3 Nguyen, Q. D., et al. Am J Ophthalmol 2006;142(6):961–969.