Combination Therapy for RVO
The complexity of this multifactorial disease state suggests a need for a tailored treatment approach.
For the first time since the 1990s, the treatment options for retinal vein occlusion (RVO) are expanding. It has been more than 1 decade since the data and recommendations from the Central Vein Occlusion Study (CVOS)1-5 were released and more than two decades since the Branch Vein Occlusion Study (BVOS),6,7 the two landmark studies that set the standard of care for central retinal vein occlusion (CRVO) and branch retinal vein occlusion (BRVO).
RVO affects nearly 160,000 eyes each year, according to collected data from the Beaver Dam Eye Study.8 Most of these (80%) are BRVOs, and although some younger patients present with RVO, particularly CRVO, it is most common among patients over age 65.
At the 2009 Retina Congress in New York, more data were released on these diseases than had been for some time, setting the stage for a paradigm shift in the way retina specialists will address this disease.
BRVO. Patients with BRVO commonly complain of a sudden, painless decrease in vision or a visual field defect in one eye.9 In acute BRVO, intraretinal hemorrhages, retinal or macular edema, and cotton wool spots are seen in the portion of the fundus affected by the involved retinal vein.10 In chronic BRVO, hemorrhages may be absent and macular edema may be the only symptom present.
Retinal neovascularization may be seen in eyes with large areas of nonperfusion. This may lead to vitreous hemorrhage and tractional retinal detachments, which may create retinal breaks leading to combined rhegmatogenous and tractional retinal detachments. Neovascular glaucoma and neovascularization at the disc area are rare.10 In patients with reduced vision, fluorescein angiography can help identify vision loss secondary to macular edema or macular ischemia.9
CRVO. Patients with CRVO present in much the same way with sudden, painless loss of vision in one eye. Signs of CRVO include disc edema, with increased dilation and tortuosity of all retinal veins. Widespread deep and superficial hemorrhages, cotton wool spots, retinal edema, and capillary nonperfusion are also usually present.1
CRVO can be ischemic or nonischemic. Ischemic CRVO is a seriously blinding disease, and anterior segment neovascularization leading to neovascular glaucoma is its major complication.11
Nonischemic CRVO is a comparatively benign disease, with permanent central scotoma as the major complications from cystoid macular edema. This type of CRVO less frequently results in the complication of ocular neovascularization. It is estimated that 12.6% to 33% of nonischemic cases may progress to ischemic CRVO within 4 years.11,12
MACULAR EDEMA IN RVO
The exact pathogenesis of RVO is not known, but possible causes include external vascular compression, disease of the vein wall, inflammation, and intravascular thrombus formation.13-15
Once an obstruction has occurred, increased vascular pressure behind the occlusion can cause fluid and small molecules to leak across the vascular wall and into the surrounding retinal tissue, causing macular edema. Macular edema is a common complication of RVO.16
Low-grade, chronic inflammation may also play a role in exacerbating the disease process.15,16
This includes the production of inflammatory mediators (such as prostaglandins and IL-6), increased amounts of vascular permeability factors such as vascular endothelial growth factor (VEGF),17 and may also include the loss of endothelial tight junction proteins.18
The results of the BVOS showed a benefit of laser for vision loss due to macular edema.6 Conversely, the CVOS did not show a significant benefit—any reduction in macular edema from laser did not seem to affect visual acuity.2,3 Interestingly, there was a trend in the CVOS that younger patients with CRVO responded better to laser.2 The BVOS and CVOS both produced the gold standard recommendations for treatment (BVOS) or observation (CVOS), to which there has been little to no challenge—until now.
NEW AND EMERGING DATA IN RVO
New data were released at the 2009 Retina Congress on both CRVO and BRVO regarding visual acuity results with steroids vs laser (SCORE [Standard Care vs Corticosteroid for Retinal Vein Occlusion]), sustaineddelivery dexamethasone vs sham (Ozurdex, Allergan, Inc.), and anti-VEGF agents vs sham for BRVO and CRVO (BRAVO [A Study of the Efficacy and Safety of Ranibizumab Injection in Patients With Macular Edema Secondary to Branch Retinal Vein Occlusion] and CRUISE [Study of the Efficacy and Safety of Ranibizumab Injection in Patients with Macular Edema Secondary to CRVO].
Intravitreal triamcinolone acetonide. The SCORE studies were sponsored by the National Eye Institute (NEI) and were designed to evaluate a preservative-free preparation of triamcinolone acetonide (Trivaris, Allergan, Inc.) in 1 mg and 4 mg concentrations in comparison with the standards of care for BRVO (laser in the absence of dense macular hemorrhage)19 and CRVO (observation).20 Both studies enrolled groups of patients similar to those in the BVOS and CVOS studies.
The use of intravitreal triamcinolone acetonide was not shown to provide significant visual acuity benefit over laser in the SCORE-BRVO study. In the first year, steroids appeared to have a better effect on visual acuity, but at year 2, the results evened out and there was no significant benefit. The side-effect profile in the 1 mg group was similar to that of laser, but side effects were significantly higher in the 4 mg group. Thus, the recommendation from the SCORE-BRVO trial was that laser should remain the standard of care in BRVO.
Patients in the SCORE-CRVO trial who were randomized to 1 mg and 4 mg of intravitreal triamcinolone achieved better visual acuity outcomes than those in the laser group. Twenty-seven percent of those in the 1 mg group and 26% of those in the 4 mg group gained three lines in 1 year, compared with 7% in the observation group. Those in the steroid groups lost far less vision than those in the observation groups. The side effects were lower in the 1 mg group, leading to a recommendation that 1 mg nonpreserved triamcinolone acetonide should be considered for patients with CRVO.
Intravitreal dexamethasone implant. The 6-month21 and 12-month22 results for the only US Food and Drug Administration (FDA)-approved drug therapy for BRVO and CRVO, the dexamethasone 0.7-mg intravitreal implant (Ozurdex, Allergan, Inc.), were presented at the 2009 Retina Congress and the 2010 Macula Society meeting, respectively.
Two identical, prospective, multicenter phase 3 clinical trials were conducted to evaluate the safety and efficacy of the dexamethasone implant. Each trial consisted of a 6-month, randomized, sham-controlled, parallel-group, double-masked phase followed by a 6-month open-label extension. In the masked phase of the trial, patients were randomized 1:1 to either sham or the intravitreal dexamethasone implant. In the open-label phase (the second 6 months) all patients who were eligible received the dexamethasone implant.
Both the 6- and 12-month data showed that more patients (30%) injected with the dexamethasone implant (n=427) gained three lines of vision in 1 to 2 months than sham-treated patients (7% to 12%; n=426).9 Improvement with the implant peaked at day 60; 29.3% of patients who received the dexamethasone implant gained three or more lines of vision vs 11.3% of shamtreated patients, a difference that was statistically significant (P<.001). In this trial, patients with CRVO and BRVO were pooled into one group.
Side effects were relatively low with dexamethasone; intraocular pressure (IOP) rise was only 25% higher in the treatment groups with normalization with appropriate topical treatment in essentially all patients, and rates of surgical intervention were very low at 6 months. These data were similar in the 12-month reinjection study. In regard to cataract, the 6-month data showed that one patient required cataract removal at 1 year; at 12 months, four patients required cataract extraction. The side effect profiles of the SCORE and dexamethasone implant studies are seen (compared with the intravitreal fluocinolone implant for uveitis [Retisert, Bausch + Lomb] in Table 1 on page 2.
Intravitreal ranibizumab. The BRAVO23 and CRUISE24 studies evaluated the use of intravitreal ranibizumab (Lucentis, Genentech, Inc.) in two doses (0.3 mg and 0.5 mg) for BRVO and CRVO compared with observation and rescue laser (BRAVO) or observation (CRUISE). All the treatment groups in BRAVO were eligible for rescue laser at 3 months and the observation group was eligible for injections of 0.5 mg ranibizumab if vision was worse than 20/40 and central foveal thickness was greater than or equal to 250 µm. Rescue laser was not available in CRUISE.
Patients enrolled in BRAVO had visual acuity at baseline of between 20/63 and 20/80 and the average retinal thickness showed significant edema at 500 µm. The response to ranibizumab was rapid (7 days). At 6 months, there was a significant difference in threeline gainers in the ranibizumab-treated group compared with the sham group; 55.2% of those in the 0.3-mg group gained three lines and 61.1% of those in the 0.5-mg group gained three lines, compared with 28.8% in the sham group. The foveal thickness measurements were reduced significantly for the ranibizumab- treatment groups at all time points during the 6-month study.
Patients randomized to treatment with either 0.3 mg or 0.5 mg ranibizumab had rapid resolution of macular edema at day 7 (400 µm reduction from baseline mean of 680 µm).
of ranibizumab and 47.7% of patients who received 0.5 mg of ranibizumab gained three lines or more compared with 16.9% of patients in the sham group. The mean foveal thickness (mean baseline measurement of 680 µm) was reduced by almost 400 µm by day 7 in the treatment groups. At the 6-month point, foveal thickness was reduced by at least 430 µm in both treatment groups, while it was reduced by only 168 µm in the sham group.
The side effect profile of ranibizumab in both trials was excellent and consistent with those of the ranibizumab trials for age-related macular degeneration (AMD).
RATIONALE FOR COMBINATION
THERAPY FOR RVO
Macular edema secondary to RVO is often more difficult to treat than exudative AMD. With AMD we are treating the primary action of the disease—neovascularization. When we are treating macular edema secondary to RVO, we are still unable to address the primary mechanism of the disease—the vein occlusion. Although we have many more treatment options available to us in 2010, they only suppress the macular edema, thereby buying time so that the body can recanalize the vessel that is occluded. In RVO, it is clear that the body wants to heal itself, but this process can sometimes take from 9 months to a year—or it may never happen. Thus, the goal in treating RVO is to maximize the treatment with the fewest side effects and discomfort to the patient. Using a combination approach may offer the best treatment while reducing the side effects of any one therapy.
The factors that come into play with macular edema secondary to RVO are multiple. We know that in macular edema a significant amount of VEGF is produced. Thus, VEGF suppression seems a likely path to success; however, there are data showing that the VEGF produced in macular edema secondary to RVO is far more extensive. Based on this knowledge, it is reasonable to assume that more frequent injections could be required.
The results from the BRAVO and CRUISE studies are the most positive that we have seen with any pharmaceutical intervention, and the side effects were minimal. What are the downsides to monthly (or potentially even more frequent) injections of anti-VEGF? The negatives include inconvenience to the patient, more frequent office visits, increased burden on office flow, and high cost. As we know, monthly injections of ranibizumab add up to $24,000 per year. These costs could also potentially rise based on the need for more frequent injections to control the production of VEGF.
We also know from the SCORE-CRVO and the sustained- delivery dexamethasone trials that steroids are effective in mediating the inflammation that causes macular edema secondary to RVO. The downsides to triamcinolone acetonide use are the side effects of cataract and increased IOP. The only steroids that we currently have available for intraocular use are triamcinolone acetonide and dexamethasone. The side effect profile of triamcinolone acetonide has been shown in some case reports to be unfavorable for phakic patients and those patients who are at risk for high IOP. In the SCORE-BRVO study, the 12-month data showed that three patients in the standard-of-care group had cataracts vs none in the 1-mg triamcinolone acetonide group vs four in the 4-mg triamcinolone acetonide group. Between 12 and 24 months, cataracts increased significantly in the 4-mg group. Thirty-five patients in the 4-mg group required cataract surgery vs eight in the 1-mg group and six in the standard-of-care group.19
The 6-month data for the dexamethasone intravitreal implant, however, showed a more favorable profile, with only 4% of patients having cataract progression over the course of 6 months and only one patient in the study requiring cataract removal. At 1 year, only four patients required cataract removal.
Finally, the dexamethasone intravitreal implant study, BRAVO, and CRUISE all evaluated monotherapy. If we accept that laser treatment is effective in BRVO, it is likely that when combined with laser, fewer injections of the chosen pharmacotherapy can be given, minimizing the inconvenience to the patient and the side effects of a single treatment, while increasing the duration between patient visits. Furthermore, it is possible that a combination approach may prove to be more effective than monotherapy due to a multi-pronged approach to treating the disease.
COMBINATION STRATEGIES USING
Our options for combination therapy for treating macular edema secondary to RVO using FDA-approved treatments will likely include the following: 1) immediate injection of one or several ranibizumab injections followed by laser when there is clearance of any hemorrhaging; 2) injection of the dexamethasone implant followed by laser; and 3) combination of an injection of anti-VEGF and dexamethasone implant with or without laser.
Pretreatment with an injection prior to laser is important for several reasons. First, the geographic area requiring treatment in the macula often is “smaller” after an injection compared with when one does not pretreat with an injection. Second, in a fresh RVO with macular edema, the retina is thick and boggy, requiring more laser power and decreasing the accuracy of the treatment. With pretreatment, either with anti-VEGF or steroid, we can thin the macula (eg, from 600 µm to 400 µm), and when the laser is applied the precision of burn placement is more accurate with less power required.
This past year in retina has been landmark. We have gone from having no FDA-approved pharmacologics to having one approved (dexamethasone intravitreal implant) and one with excellent data to show that it is effective in our patients with macular edema secondary to RVO (ranibizumab), and FDA approval seems likely. How do we make choices as to what combinations will bring the best benefit to our patients? Currently, we do not have hard data to support any one combination over another. We do know, however, that some choices are obvious; for example, in a patient who has advanced glaucoma or a patient with a crystal clear lens, an anti-VEGF agent would be more favorable than steroid. For pseudophakic patients with no history of glaucoma, the dexamethasone implant may be more favorable because of its long durability.
Further data are needed to develop solid algorithms for treating RVO. Many factors come into play: severity and location of macular edema, natural history, response to therapy, side effects, and cost.
Although there will be continued debate over which combination is best until we have these hard data, there is no doubt that our treatments for macular edema secondary to RVO have expanded and improved and that the benefit to our patients is significant.
Seenu M. Hariprasad, MD, is an Associate Professor and the Director of Clinical Research at the University of Chicago Department of Surgery, Section of Ophthalmology and Visual Science. He serves as Chief of the Vitreoretinal Service and is Director of the Surgical Retina Fellowship Program. Dr. Hariprasad states that he is a consultant for OD-OS, Alcon Laboratories, Inc., Baxter, Optos, Ocular Therapeutix, Pfizer, Inc., Genentech, and Allergan and is on the speakers’ bureau for Genentech, Inc., Allergan, Inc., Pfizer, and Alcon Laboratories, Inc. He can be reached via e-mail at email@example.com.
Veeral S. Sheth, MD, is a Clinical Assistant Professor with North Shore University Health Systems and the University of Chicago- Vitreoretinal Service. Dr. Sheth states that he has no financial interests to disclose. He can be reached via e-mail at firstname.lastname@example.org.
Paulpoj Chiranand, MD, is a Vitreoretinal Surgical Fellow at the University of Chicago. Dr. Chiranand states that he has no financial relationships to disclose. He can be reached via e-mail at email@example.com.
Rachel M. Renshaw, Editor-in-Chief of Retina Today, provided medical writing assistance for this article.
Combination Therapy for BRVO and CRVO
BY SEENU M. HARIPRASAD, MD
A 92-year-old woman presented with a 2-month historyof branch retinal vein occlusion (BRVO) in her right eye andsecondary vision loss. The patient had not been treated previously,and her vision in the right eye at presentation wascounting fingers at 2 feet. On fundus photography and fluoresceinangiography (FA), intraretinal hemorrhages wereseen along with cotton wool spots along the inferotemporalarcade and the inferior macula (Figures 1, 2, and 3). Thesefindings are consistent with BRVO.
Optical coherence tomography (OCT) at presentationdemonstrated massive thickening with subfoveal serous retinaldetachment (Figure 4). These findings precluded effectivelaser treatment at this time, so we chose to inject thepatient with the intravitreal dexamethasone 0.7 mg implant(Ozurdex, Allergan, Inc.).
The patient returned for a follow-up visit 3 weeks later.On OCT, the macular edema was seen to have improvedmarkedly, and the serous retinal detachment had resolved(Figure 5). Her visual acuity remained the same as beforeinjection with the dexamethasone implant (counting fingersat 2 feet), and her intraocular pressure (IOP) hadremained stable (preinjection IOP: 14 mm Hg; postinjectionIOP: 12 mm Hg).
At this visit, we applied focal grid laser in her right eye.The decreased edema allowed a more accurate treatment with less power to a smaller geographicalarea. Two months following laser treatment,the foveal architecture was restored (Figure6), and vision improved to 20/400. IOP at thisfollow-up visit was 18 mm Hg. At mostrecent follow-up, her vision had improved to20/100.
A 47-year-old man from Abu Dhabi presentedwith central retinal vein occlusion(CRVO) and macular edema in both eyes.The patient has severe hypertension fromkidney disease. His history revealed vision lossfor at least 7 months, and he had notreceived any prior treatment. At presentation,the patient’s vision in the right eye was20/400 with an IOP of 8 mm Hg and 20/70 in the lefteye with an IOP of 8 mm Hg. The macular edema inboth eyes is seen on OCT in Figure 7. Fundus photographyand FA showed intraretinal hemorrhages consistentwith CRVO and macular edema (Figure 8).
We injected the intravitreal dexamethasoneimplant in the patient’s right eye 2 weeks after presentationand in his left eye 3 days later. Two weeksfollowing the right eye injection, visual acuity in hisright eye improved to 20/50 and the macular edemadecreased on OCT (Figure 9A). His IOP at 2 weekswas 10 mm Hg. At this visit we applied gentle focalgrid laser for the subtle macular edema immediatelyoutside the fovea in the right eye. The visual acuityin the left eye had improved to 20/50 with an IOP of 12 mm Hg and a decrease in macular edema (Figure 9B).On fundus photography and FA, we saw decreased vasculartortuosity and decreased intraretinal hemorrhagingin both eyes (Figure 10). We applied laser to the left eye3 days later for mild residual macular edema.
Approximately 2 weeks after combination therapy withthe intravitreal dexamethasone implant and laser in botheyes, the patient returned for follow-up. The patient’s OCTscans showed significant improvement in macular edema(Figure 11).
At the most recent follow-up, approximately 3 monthslater, the patient’s vision had stabilized to 20/50 in botheyes with complete resolution of macular edema.
Anti-VEGF for Recurring Macular Edema
BY DANTE PIERAMICI, M
A 63-year-old woman presented with an acute loss ofvision over several recent weeks in her left eye from a centralretinal vein occlusion (CRVO). The baseline visual acuitywas 20/200, and central retinal thickness (CRT) on opticalcoherence tomography (OCT) was 685 μm. Clinical examinationshowed no signs of neovascularization of the iris,and fundus examination showed a dense intraretinal hemorrhagein all four quadrants, swelling of the optic nerve,and macular edema. Fluorescein angiography confirmedswelling of the optic nerve, with cystoid macular edema inthe late phases of the angiogram. OCT testing confirmedmarked cystic edema of the retina, as well as areas of subretinalfluid (Figure 1).
The patient appeared to have what is typically called aperfused CRVO. Although there was no evidence of neovascularizationof the iris or neovascularization of the angle,the patient had significantly decreased visual acuity secondaryto severe macular edema.
The OCT at the top of Figure 2 shows a large amount ofintraretinal and subretinal fluid with swelling at the opticnerve edge. We injected intravitreal ranibizumab (Lucentis,Genentech, Inc.) and within 1 week (Figure 2, second row) amarked decrease in the intraretinal edema is evident; however,there is still persistent subretinal fluid. At 1 month followingtreatment (Figure 2, third row), subretinal fluid is stillpresent, but the edema and retina continue to thin, andthere is associated improvement in visual acuity. It is interestingto note that the subretinal fluid is slightly more resilient to the anti-VEGF agent. Additionally, macularedema can resolve significantly despite the patient not subjectivelynoticing the improvement for some time.
For this patient, we injected again at 1 month. We sawadditional reduction in edema at 2 months (Figure 2, fourthrow) and injected again. By 3 months, the OCT appearedfairly normal, and the visual acuity improved from 20/200 atbaseline to 20/40 (Figure 2, fifth row).
At month 3, not only did we see improvement on OCT,but we also saw marked improvement in the fundus photographsand fluorescein angiogram (Figure 3). There is clearlyrapid reduction in the intraretinal hemorrhaging and opticnerve swelling. The fluorescein angiogram confirms decreasedswelling of the optic nerve and the macula.
Anti-VEGF agents seem to have an effect on antipermeabilityof fluid, but they also seem to reduce leakage ofintraretinal hemorrhage and leakage of the optic nerve. Earlycollateralization that is apparent on the optic nerve head isseen in this patient.
Because the patient was doing well at 3 months, observationcould be an option at this point. After discussion withthe patient, however, we chose to inject again at month 4(Figure 4, top row). Interestingly, at 5 months there was anincrease in subretinal fluid (Figure 4, second row). Afterinjecting again at 5 months, the edema was fairly resolvedby month 6 (Figure 4, third row), after which we chose toobserve. At month 7 the edema recurred along with slightreduction in visual acuity (Figure 4, fourth row). For thispatient we decided to treat again, and by month 8 (Figure 4,fifth row), the edema and visual acuity began to improve.The fundus and fluorescein images from month 6 showmarked improvements and collateralization at the opticnerve head that correlate with the OCT findings (Figure 5).
Although we saw some recurrence of the edema atmonth 9 (Figure 6, top row), we decided to continue theperiod of observation. By month 12, the patient continuedto do well (Figure 6, third row) and over the next year,although there was some fluctuation in OCT findings, thevisual acuity and fundus findings continued to improve outto 2 years (Figures 7-9). After more than 1 year without another injection, the patient’s visual acuity is nearly 20/20,and the OCT is normalized.
Although many of our patients do not have the recurringedema that was seen with this patient, approximately onethirdof CRVO cases will require extended treatment.
What should the clinician do for cases that require 3 to4 years of treatment? If a patient is responding to treatmentand tolerates the injections well, I consider this a reasonableapproach.
The other option for patients who require a long courseof treatment may be a combination approach. It may helppatients to add a steroid injection or a sustained-deliverydevice such as the intravitreal dexamethasone implant(Ozurdex, Allergan, Inc.) to extend the treatment period.
Another option in combination therapy is to use laseralong with anti-VEGF injections. It is believed that recurrenceof edema is caused by increased or persistent VEGFexpression, so patients with recalcitrant edema might benefitfrom the addition of panretinal laser photocoagulation to the areas of ischemia. Although there are no data toshow that this approach is successful for these patients,from our knowledge of the effect of laser, this seems a reasonableapproach. The use of laser may reduce peripheral ornight vision, so it is important to discuss these side effectswith the patient.
It is important to note that we must be aware of the possibilityof iris or retinal neovascularization, particularly duringperiods when we withhold intraocular VEGF therapy.
Dante J. Pieramici, MD, practices atCalifornia Retina Consultants in SouthernCalifornia. He is the Director of the CaliforniaRetina Research Foundation and a ClinicalAssistant Professor of Ophthalmology at theDoheny Eye Institute. He states that he is a consultant forGenentech, Inc., and that Genentech, Inc., sponsored theBRAVO and CRUISE trials and an Investigator SponsoredTrial for which he was the principal investigator.Dr. Pieramici may be reached at +1 805 963 1648; fax:+1 805 965 5214; or e-mail at firstname.lastname@example.org.