Macula 2011: Highlights From the Conference

The latest tips on treatment and trial results.

Andrew E. Mathis, PhD, Medical Editor

The Macula 2011 conference was held in Philadelphia on January 29, offering retinal physicians presentations on a number of conditions and medical and surgical techniques for treating retinal disease. Here, we present selected highlights from the conference. Next month, we will continue our coverage with recaps of additional presentations from the conference.

ALLEN HO ON COMBINATION THERAPIES

Allen C. Ho, MD, who is professor of ophthalmology at Thomas Jefferson University and an Attending Surgeon at Wills Eye Institute, reviewed combination therapies for AMD in his presentation. Although ranibizumab is a “wondrous drug,” Dr. Ho began, there are still over 60% of patients who do not significantly improve vision, as demonstrated by the monthly dosing results from the ANCHOR and MARINA clinical trials, and the less than monthly dosing SAILOR and PIER trials (Figure 1).

Figure 1. Results from the ANCHOR, MARINA, SAILOR, and PIER studies of ranibizumab.

The goals of combination therapy in AMD should be, Dr. Ho continued, to improve treatment efficacy and the treatment paradigm and to create the best value. The rationale in investigating new treatments, therefore, should be synergistic, potentially combining treatments that are antivascular, antiexudative, anti-inflammatory, and antifibrotic. The current combination strategies in use include verteporfin photodynamic therapy (PDT) (Visudyne, Novartis), intraocular or external focused radiation, other cytokine inhibitors, and inflammatory modulators.

The first strategy, ranibizumab plus PDT, has been investigated in the RADICAL study and the several trials of the SUMMIT clinical program. Reviewing the outcomes data from RADICAL, Dr. Ho noted that ranibizumab monotherapy and three study arms given combination therapy (all three were given a level of reduced-fluence PDT, and two of the three were also given dexamethasone) all improved vision at the two-year mark, in this phase 2 clinical trial (Figure 2). This phase 2 study was not powered to determine if there were significant differences in vision, however. Ranibizumab monotherapy required retreatment on a more frequent basis (ie, a mean of 8.9 injections over two years vs 4.2 injections for triple therapy with half-fluence PDT), and this number of ranibizumab injections was dependent on required injections in that cohort after baseline (Figure 3). “There may be a role for PDT in combination with ranibizumab and dexamethasone for AMD patients who are unable to return for frequent visits,” Dr. Ho said.

Figure 2. Mean visual acuity results from the RADICAL trial.

Figure 3. Retreatment data from RADICAL.

He continued by turning to the second combination strategy, ranibizumab combined with radiation therapy. Here, the pivotal trials have been the CABERNET study of NeoVista's (Fremont, CA) strontium 90 brachytherapy and Oraya's (Newark, CA) office-based I-Ray radiosurgical system. CABERNET trial consists of two combination arms, one with a loading dose of two ranibizumab injections followed by prn injections and the other with the same loading dose followed by injections every three months. Dr. Ho showed slides from four patients in the trial who demonstrated reductions in both choroidal neovascular leakage and subretinal fluid (Figure 4). Complications of radiation retinopathy have been minimal although further follow up is necessary, he noted. Oraya's system exposes the macula to X-rays while the patient is seated in the office (Figure 5). NeoVista's handheld radiation probe delivers focused radiation in the operating room after pars plana vitrectomy. Phase 3 clinical trial results are forthcoming later this year for NeoVista while Oraya is on the threshold of embarking on its phase 3 clinical trial.

Figure 4. Four patients from the CABERNET trial at baseline and one, two, and three years.

Figure 5. Oraya's system allows the patient to be seated while being examined.

The third strategy Dr. Ho discussed was the use of cytokine inhibitors in combination with ranibizumab. The first cytokine-targeted, platelet-derived growth factor (PDGF) is being targeted with the anti-PDGF agent DNA aptamer E10030, developed by Ophthotech (Princeton, which strips pericytes from mature neovascular vessel, leaving them more susceptible to treatment with an anti-VEGF agent. In phase 1 trials of E10030, results were impressive, with some gains of as much as five ETDRS lines of visual acuity (Figure 6). A phase 2 trial is recruiting patients with AMD and classic choroidal neovascularization to test the hypothesis combining ranibizumab with anti-PDGF agent E10030 will create synergistic effects.

Figure 6. Phase 1 trial results for E10030 have resulted in gains as high as five ETDRS lines.

The final combination strategy Dr. Ho discussed was the use of inflammatory modulators. Reviewing briefly the role of inflammation in AMD, he discussed the compounds under investigation, including intravitreal steroid implants and complement factor modulators for factors C3, C5, CFH and CFD. In closing, Dr. Ho stated that it is early to identify the winners and losers in the competition to develop effective and affordable combination therapies, and the promise of these combination therapies still remains largely unfulfilled.

BAILEY FREUND ON POSTINJECTION IOP

K. Bailey Freund, MD, of Vitreous-Retina-Macula of New York (VRMNY), provided findings from research on the sustained effects of intravitreal anti-VEGF agents on intraocular pressure (IOP). “We know,” Dr. Freund began, “that there is a transient IOP spike following an intravitreal injection of any type.” Citing data from a 2010 report in Archives of Ophthalmology, he stipulated that, although IOP spikes, on average, to between 34.8 and 38.5 mm Hg, that spike usually subsides within 20 minutes. Prophylactic IOP-lowering medications were not shown to be effective against these spikes. A couple of small studies began appearing in 2008 documenting sustained elevation of IOP after injections of bevacizumab or ranibizumab. Two more recent papers continued to show this trend.

Dr. Freund then turned to two papers of particular importance. The first, authored by partners at VRMNY, detected IOP spikes as high as 58 mm Hg, and five of the eyes in this series of 25 required filtration procedures to control IOP. The second paper, which was presented by Dr. Sophie Bakri at the 2010 annual meeting of the American Academy of Ophthalmology, dealt with ocular hypertension and glaucoma-related adverse events from the MARINA and ANCHOR trials of ranibizumab in wet AMD. In this analysis, sustained IOP elevation was defined as a pre-dose IOP of ≥6 mm Hg from baseline at least two consecutive visits or IOP ≥25 mm Hg. However, the events were not dose-dependent, and patients receiving a higher dose of ranibizumab actually had a lower incidence of side effects (5.4% vs 4.5%). Notably, injections of different doses were administered in the same overall amount of fluid (0.5 mL).

Two additional papers that Dr. Freund co-authored continued to identify a trend of sustained IOP elevation following anti-VEGF injections. These papers are currently under consideration for publication, but Dr. Freund presented some of the data. He stated that the baseline demographic characteristics (Figure 7), indicated that a large majority of these eyes (96.41%) received at least one injection of ranibizumab (vs. 32.91% who had been given at least one dose of bevacizumab). This paper, which specifically addresses the clinical predictors for sustained IOP following anti-VEGF therapy, found that the likelihood of sustained IOP elevation increased with the number of injections administered, with the statistical significance of the findings increasing in strength with the number of injections (Figure 8). In patients receiving ≥31 injections, there was a 14.8% probability of sustained IOP elevation.

Figure 7. Baseline demographic characteristics of patients from two studies under consideration for publication.

Figure 8. As total injections increased, so did IOP and the level of statistical significance.

Turning then to the possible mechanisms of this adverse event, Dr. Freund mentioned that he favors repeated injection-related IOP spikes as the most likely cause of this treatment related complication. Also, he detailed the hypothesized pharmacologic effect of VEGF blockade on the trabecular meshwork, leading to inflammation and, thus, impaired outflow. This impaired outflow, he noted, could also be linked to debris such as protein aggregates and/or silicone oil droplets from the syringes themselves. Unfortunately, the evidence here is scant. A case study from the June 2010 issue of Open Ophthalmology, for instance, followed a case of trabeculitis after intravitreal anti-VEGF agents. Also, two other papers published in 2010 have implied that contamination in repackaged bevacizumab originating from compounding pharmacies could be obstructing in aqueous outflow, resulting in increased IOP. Thus, evaluation of the manner in which bevacivumab syringes are shipped and stored following repackaging in compounding pharmacies is essential.

As far as prevention and management are concerned, Dr. Freund suggested several measures, including baseline nerve fiber layer examination, less frequent treatment when appropriate, use of lower volumes of anti-VEGF drugs and silicone-free needles and syringes, and early referral to a glaucoma specialist. Also, he warned that laser trabeculoplasty appears minimally effective in treatment-sustained IOP elevations following injections, but anti-VEGF agents themselves may actually increase the success of filtration procedures thanks to their pharmacological effects.

JOHN THOMPSON ON GIANT RETINAL TEARS

John T. Thompson, MD, who is a partner with Retina Specialists in Frederick, MD, gave a presentation on the detection and repair of giant retinal tears. While most giant retinal tears, Dr. Thompson began, quickly lead to retinal detachment (Figure 9), patients occasionally present before detachment has occurred. Tears such as these should be treated with at least two to three rows of laser photocoagulation. Then the patients should be monitored carefully for extension of the tear or RD.

Figure 9. Most giant retinal tears quickly lead to retinal detachment.

Dr. Thompson then laid out the epidemiology of giant retinal tears, citing a 2010 paper by Ang and colleagues from Investigative Ophthalmology and Visual Science that found that the incidence of such tears was 0.19 patients per 100,000, that the mean age at diagnosis was 42 years old, that men comprised 72% of such patients, that the macula was detached in 45% of cases, and that there was proliferative vitreoretinopathy (PVR) of grade C or greater in 11% of cases. Regarding the etiology of these giant tears, over half are idiopathic, with the next largest percentages of cases arising from trauma (16.0%) and hereditary vitreoretinopathy (14.5%). High myopia, prior retinal detachment, and uveitis make up the causes for the rest of the cases.

Dr. Thompson then discussed some prognostic factors to consider when repairing giant retinal tears. Citing a 2002 report from Ophthalmology on giant retinal tear management using perfluorocarbons, citing three papers that identified a number of risk factors for favorable and unfavorable outcomes. For instance, two studies found that the odds of recurrence of retinal detachment and probability of poor visual outcome were tied to the extent of the giant retinal tear and the presence and severity of preoperative PVR. Scleral buckles were tied to improved outcomes and lower recurrence.

Turning to specific surgical techniques for the repair, Dr. Thompson began by noting that 20-gauge vitrectomy is preferable when repairing giant retinal tears. Furthermore, he recommended lensectomy in cases where the tear is more than 180° in extent or is accompanied by PVR. Wide-angle viewing and aggressive anterior vitrectomy (Figure 10), in particular, are helpful adjuncts in this surgery, Dr. Thompson noted. Furthermore, because it is necessary to eliminate surface traction in the subretinal and preretinal space, he stated that a small amount of perfluorocarbon liquid filling the post-equatorial vitreous can help remove peripheral epiretinal membranes, following the removal of the posterior membranes. He also recommended cinching up the scleral buckling after fluid-air exchange to allow easy access to the edge of the retinal tear during vitrectomy. When the retina is mobile, the surgeon should inject perfluorocarbon liquid and then conduct fluid-air exchange to remove the liquid. Dr. Thompson then briefly discussed 360° laser in the periphery to help stabilize the vitreous base, noting that cryopexy should be avoided in such cases due to an increased risk of PVR (Figure 11).

Figure 10. Aggressive anterior vitrectomy is a helpful adjunct in treating giant retinal tears.

Figure 11. Cryopexy should be avoided in repairing giant retinal tears due to the risk of PVR.

A problem that can arise during such procedures is retinal slippage during air–perfluorocarbon exchange. To prevent this from happening, Dr. Thompson suggested ensuring that all subretinal fluid has been pushed out by the perfluorocarbon liquid, aspirating at the edge of the giant retinal tear as soon as air enters the eye. A direct perfluorocarbon/silicone oil exchange may also help avoid retinal slippage, although he cautioned that doing this may make it more difficult to assure that all the perfluorocarbon has been removed. If the retinal slippage is substantial, Dr. Thompson suggested trying an “old technique,” ie, prone fluid-air exchange. Then, Dr. Thompson reviewed some of the largest case series of giant retinal tear repair, noting increasing success rates, particularly over the last decade. Common complications in these series included slippage of the retinal tear flap, elevated intraocular pressure in the early postoperative period, recurrent retinal detachment, and others.

Dr. Thompson then briefly covered some controversies in the care of giant retinal tears, including the debates over gas vs. oil tamponade (most surgeons prefer oil in tears greater than 180° and/or comorbid with PVR) and whether or not to use a scleral buckle (there is research that supports both points of view but he favors use of a scleral buckle). In closing, Dr. Thompson reminded the audience that visual acuity should be ≤20/200 in up to two-thirds of operated eyes using these surgical techniques (Figure 12).

Figure 12. After repair, eyes with giant retinal tears should recover VA of 20/200 or better.

PETER CAMPOCHIARO ON LUCENTIS FOR RVO

A presentation on the long-term outcomes of ranibizumab in the treatment of RVO was given by Peter A. Campochiaro, MD, of the Wilmer Eye Institute at Johns Hopkins University. In two large phase 3 studies, BRAVO for patients with BRVO and CRUISE for patients with CRVO, patients were given six monthly injections of 0.3 mg or 0.5 mg of ranibizumab or sham. At the month 6 primary endpoint in BRAVO, patients in the two ranibizumab groups showed a mean improvement in ETDRS letter score of 18.3 (0.5 mg) and 16.6 (0.3 mg), compared to 7.3 in the sham group.

During the next six months, patients in all three groups were evaluated each month and could receive an injection of ranibizumab if they met retreatment criteria. This “treatment as needed” protocol provided good maintenance of visual gains in the ranibizumab groups; the change from baseline in ETDRS letter score at month 12 was 18.3 (0.5 mg) and 16.4 (0.3 mg). It also resulted in substantial improvement in the sham group, for which the change from baseline letter score at month 12 was 12.1.

At the month 6 primary endpoint in CRUISE, patients in the two ranibizumab groups showed a mean improvement in ETDRS letter score of 14.9 (0.5 mg) and 12.7 (0.3 mg) compared to 0.8 in the sham group. During the next six months, patients in all 3 groups received “treatment as needed,” and it provided good maintenance of visual gains in the ranibizumab groups; the change from baseline in ETDRS letter score at month 12 was 13.9 (0.5 mg) and 13.9 (0.3 mg). It also resulted in substantial improvement in the sham group for which the change from baseline letter score at month 12 was 7.3.

The only data available for time points longer than a year are from relatively small pilot trials. Patients with BRVO given three monthly injections of ranibizumab had a mean improvement of 16.1 letters. After that, they were seen every two months and were given an injection of ranibizumab if foveal thickness was ≥250µm on a Stratus III OCT. At month 24, the mean improvement in BCVA was 17.8 letters, 60% had a Snellen equivalent of 20/40 or better, and 65% had foveal thickness ≤250µm. This indicates that visits every two months with injections of ranibizumab as needed is sufficient to control edema and maintain vision in patients with BRVO.

Patients with CRVO given three monthly injections of ranibizumab had a mean improvement of 12.0 letters. After that, they were seen every two months and were given an injection of ranibizumab if foveal thickness was ≥250µm on a Stratus III OCT. At month 24, the mean improvement in BCVA was 8.5 letters, 30% had a Snellen equivalent of 20/40 or better, and 30% had foveal thickness ≤250µm. This indicates that visits every two months with injections of ranibizumab as needed is not sufficient to control edema and maintain vision in patients with CRVO.

Two baseline characteristics that predicted a poor visual outcome in patients with BRVO or CRVO were the presence of macular edema for greater than one year and extensive closure of perifoveal capillaries. Thus, the long term outcomes for treatment of CRVO or BRVO with ranibizumab are excellent, but many patients with CRVO require more frequent follow up and a greater number of injections to control edema and maintain visual benefits.

Dr. Campochiaro closed with treatment recommendations, suggesting monthly injections of ranibizumab for six months in both CRVO and BRVO and then monthly follow with injections only if there is recurrent edema. If frequent injections are needed during the second six months in patients with BRVO, grid laser therapy could be considered. If frequent injections are needed during the second six months in patients with CRVO, treatment with Ozurdex could be considered.

CONCLUSION

Next year's conference dates have not yet been announced, but it is expected that the conference will return to New York. Keep an eye on our Calendar section for an announcement once details become available. RP