Combination Therapy for Difficult Cases

Reducing the treatment burden is a key potential advantage.

Combination Therapy for Difficult Cases

Reducing the treatment burden is a key potential advantage.

Dr. Hariprasad: The newest treatment options for macular edema secondary to retinal vein occlusion (RVO) have raised our expectations for how effective our interventions can be. At the same time, given the nature of the disease process, successful treatment is rarely “one and done.” Furthermore, we all have patients who don’t respond as robustly as we would like to initial treatment. How do you define a non-responder or perhaps a suboptimal responder?

Dr. Singer: When we see patients once a month, and they return after an anti-VEGF injection with macular edema still present, we may assume they didn’t have a response. I believe, especially with a disease like RVO that produces high levels of VEGF, that there is always some response, but it may be an incomplete response. I have been seeing many patients every 2 weeks instead of every month to evaluate this further. If central macular thickness is more than 350 microns (on Cirrus SD-OCT, Carl Zeiss Meditec) 2 weeks after an injection, I definitely consider that an inadequate response. However, in my experience, most patients have less fluid at 2 weeks after injection compared with before injection, but not all patients are dry (Cirrus SD-OCT less than 300 microns).

Dr. Dugel: I consider a non-responder to be someone who either gets worse or doesn’t get any better with treatment. In my experience, it is unusual for a patient to get worse, but not unusual for a patient to have little or no improvement. Anti-VEGF monotherapy typically stabilizes the situation quickly. I consider a suboptimal response to be one in which macular edema stabilizes but persists.

Dr. Kitchens: We really do not have good definitions of non-responders and suboptimal responders. Are they patients who get 50% better with injections but never achieve total drying of the retina? Are they patients whose edema recurs in 2 weeks rather than 6? In the BRAVO trial of ranibizumab (Lucentis, Genentech) for macular edema secondary to branch retinal vein occlusion (BRVO), 25% of patients required laser therapy. In that context, they would be the non-responders or under-responders to anti-VEGF. For me in daily practice, an under-responder is someone whose macula is not 90% dry with SD-OCT compared to presentation despite at least three injections of an anti-VEGF agent.

Dr. Hariprasad: In BRAVO, and the CRUISE trial of ranibizumab for central retinal vein occlusion (CRVO), significant fluid reduction and more than 80% of the vision gains in a year were achieved by the third or fourth injection. Based on that, residual fluid and minimal vision gain indicate a suboptimal response after four initial monthly injections. That said, some eyes will not improve to 20/40 regardless of what we do. Therefore, it makes more sense to focus on the macular anatomy on SD-OCT. If I don’t see significant drying by the third or fourth anti-VEGF injection, I consider a different approach.

Managing Persistent Macular Edema

Dr. Hariprasad: Once you identify a patient as a suboptimal responder to first-line anti-VEGF treatment, what are your next steps?

Role of Laser in RVO Begins to Evolve

Seenu M. Hariprasad, MD: Many of the ways we currently use laser treatment in retinal vein occlusion (RVO) are largely based on earlier trials, such as the Branch Vein Occlusion Study and the Central Vein Occlusion Study. For example, macular grid laser is a viable option for improving visual acuity compromised by persistent macular edema in branch but not central RVO. From the more recent SCORE trial, we know that grid laser is as effective as intravitreal triamcinolone for improving decreased vision due to macular edema in BRVO, but with a more favorable complications profile.

Regarding panretinal photocoagulation (PRP) for RVO, an interesting new theory has emerged. Now that we have the ability to visualize the retinal periphery with ultra-widefield fluorescein angiography, we are learning that the extent of ischemia in the periphery can be significant even when the macula is not ischemic. The theory, then, is perhaps we can treat the peripheral ischemia with PRP, thereby reducing VEGF production and the inflammation and macular edema that follows.

How do you utilize laser treatment in RVO?

John W. Kitchens, MD: If an RVO patient is developing neovascularization, particularly in the posterior segment, PRP is crucial for preventing neovascular glaucoma. I have tried to use PRP to reduce macular edema in BRVO and CRVO but haven’t achieved the results others have reported. That said, I do use it for BRVO patients who have extensive nonperfusion in an effort to reduce the injection burden, but I don’t count on it to eliminate the need for injections. I also may use it when significant macular edema has persisted after we have already tried anti-VEGF agents and steroids and the next step would be surgery. I tell patients it may reduce edema somewhat, but it’s not a cure.

Michael M. Singer, MD: Even though we have more options than ever for treating macular edema associated with RVO, what we don’t have is a way to stop the cycle of ischemia. Therefore, rebound macular edema is the reality in the majority of cases. Our group and others have been working to determine if targeting PRP to only the peripheral ischemic areas of the retina can stop the cycle while preserving more peripheral vision.

We are conducting our own study involving BRVO and CRVO patients who were treated with anti-VEGF injections and/or the dexamethasone intravitreal implant.1We used SD-OCT and wide-field angiography to determine when edema recurred and to quantify ischemic areas. Findings so far have been intriguing. These patients exhibit a great deal of variability in the extent of peripheral nonperfusion at baseline, and the differences seem to correlate with the extent of edema as well as the structural and functional response to treatment. I am becoming more and more convinced that the main driver of edema is peripheral ischemia.

Pravin U. Dugel, MD: In the absence of thick hemorrhage when it’s possible to use laser, it remains an important tool in the armamentarium. As we know, it does lead to permanent structural changes so if we could harness those changes for good rather than bad by using them to turn off the VEGF engine, that would be a great role.

Dr. Hariprasad: The jury is still out on whether decreasing the VEGF load with PRP laser in the periphery can reduce macular edema, but it makes a lot of sense. At the very least, eliminating the peripheral nonperfusion may serve to make our pharmacologic treatments more effective.

Dr. Dugel: When patients have mild or moderate macular edema, I start them on monthly anti-VEGF injections. After about the third injection, if I see a decrease in edema but not improvement in visual acuity, or if fluid persists, I don’t hesitate to use the dexamethasone implant (Ozurdex, Allergan). In some cases, the implant resolves the edema. If not, I continue to supplement with as-needed anti-VEGF injections, an approach that has worked well for me.

When patients present with dense, diffuse macular edema, I am fairly confident they will need more than monotherapy to completely dry the retina, so I initiate combination treatment with an anti-VEGF agent and the dexamethasone implant. With the implant in place, I see patients monthly for at least 3 months to monitor intraocular pressure. If the disease is status quo, i.e., little or no improvement at a visit, I supplement with an anti-VEGF injection. Most of the time, injection frequency drops significantly. Patients tend to reach a period of stability and then I can see them less and less for anti-VEGF supplementation.

In my hands, the durability of the steroid implant has been variable from patient to patient. I don’t have an explanation for that other than differences in patient metabolism and severity of their disease. While the effects may last 5-6 months in some patients, I’m seeing a duration of 3-4 months on average, which is the right amount of time for this particular disease. The pharmacokinetics are such that the amount of drug released declines gradually after an initial bolus release. This drug egress profile is appropriate for RVO. I have many patients who have received up to six steroid implants. I have no qualms about repeat insertions. The needle design has been improved, making it more comfortable for patients and easier for me.

I have also found the dexamethasone implant and an anti-VEGF agent to be a powerful combination for suboptimally responding eyes.

— Seenu M. Hariprasad, MD

Dr. Singer: We are seeing in RVO, as we did with AMD, that some patients who don’t respond well to one anti-VEGF agent may respond to another. So I will switch to another one in some cases before considering steroids. It has been nice to have another FDA-approved option with aflibercept (Eylea, Regeneron). That said, I believe I am quicker than most to introduce the dexamethasone implant. I don’t have too great a concern about managing patients who have a steroid response. In most patients who have an intraocular pressure response, it is almost always controllable with antiglaucomatous drops. If IOP rises to more than 30 mmHg on drops, I have used selective laser trabeculoplasty to control it.

In studies we conducted in our practice to evaluate combination therapy with an anti-VEGF agent and the dexamethasone implant, we were able to keep retinas dry for 3.5 to 4 months and prolong the time between anti-VEGF injections.2 Furthermore, the timing of the rebound edema was very predictable.

The two treatments appear to be synergistic. Moving away from monthly injections may not lead to vision as good as with monthly, but the best possible vision is not every patient’s goal. Some are more concerned with cost of treatment or convenience.

Dr. Kitchens: I, too, tend to work my way through the anti-VEGF agents if macular edema does not improve or recurs quickly with the first choice. When I switch drugs, I see the patient back in a week because I want to know whether I am dealing with a non-responder or an early recurrer. If I see some response, but macular edema is still present, I consider adding steroids or in BRVO perhaps laser.

One scenario I try very hard to avoid with RVO patients is them losing faith in what we’re trying to accomplish. If monthly injections are helping but are not keeping them dry for longer than a month, the visits can become too burdensome for them. They may question the importance of coming back so frequently. In my experience, they tend to have higher expectations than patients with other conditions. They may be of working age or may not be as accustomed to or accepting of living with a chronic disease as our AMD or diabetic patients are. When they regain some visual acuity they may think, well, I can see OK and my other eye is fine so I don’t want to keep going back for treatment. The dexamethasone implant can be a good way to gain their confidence because it can allow us to see them less frequently.

Finally, for suboptimally responding eyes, significant epiretinal membrane formation or vitreomacular traction may mean surgery is the necessary path forward. If fluorescein angiography shows peripheral nonperfusion, I add panretinal photocoagulation (PRP) to the vitrectomy and membrane peeling. I also insert the dexamethasone implant and inject sub-Tenon’s triamcinolone. It is an aggressive approach, but I want to do everything I can for the patient at the time of surgery.

Figure 1. In the BRAVO clinical trials of ranibizumab for the treatment of macular edema associated with retinal vein occlusion, 80% of the first-year gains in vision were seen by month 4.

Dr. Hariprasad: I have also found the dexamethasone implant and an anti-VEGF agent to be a powerful combination for suboptimally responding eyes. If the macula isn’t significantly flatter and vision isn’t better by the patient’s third or fourth anti-VEGF injection, I recommend we add the dexamethasone implant. As we discussed previously, in the BRAVO and CRUISE trials of ranibizumab in RVO, more than 80% of the first year reduction in macular thickness and the gain in visual acuity occurred by that point. Therefore, I don’t want to continue giving anti-VEGF injections alone if further vision gains are not going to be accomplished.

Once I proceed with a dexamethasone implant, I continue to see patients monthly. If the retina is dry, I send them home for another month. If fluid is present on SD-OCT, I proceed with an anti-VEGF injection. At 4 months, if the fluid has not resolved, I insert another dexamethasone implant and stay with the same follow-up regimen. It is interesting to note that in the GENEVA studies, approximately 21% of BRVO patients and 17% of CRVO patients required only one implant to keep the retina dry for the entire year. Dry was defined as 20/20 visual acuity and retinal thickness of 250 microns or less on OCT.3 This durability is rarely seen with anti-VEGF therapies and is a feature more commonly seen with the dexamethasone implant.

Building on the Progress That Has Been Made

Dr. Hariprasad: Keeping the macula fluid-free is our main goal, our best strategy for safeguarding vision, and I think we all strive for the additional goals of getting each patient to that point with the least burdensome treatment regimen and cost as possible.

Dr. Singer: Yes, and we shouldn’t feel restricted to one drug or class of drug if we are not getting the desired response. I believe using the entire arsenal available to us, much like oncologists do, provides our best chance for success. Even though our current interventions may be prolonging the course of the disease process, they are preferable to the scarring and atrophy that would occur without them. Until we have a way to fix the underlying problem of the occlusion, we should continue to explore ways to utilize anti-inflammatory and anti-VEGF agents as well as laser treatments to our advantage against macular edema, hopefully breaking the cycle of recurrence.

Figure 2. In the CRUISE clinical trials of ranibizumab for the treatment of macular edema associated with retinal vein occlusion, 80% of the first-year gains in vision were seen by month 4 as well.

Dr. Dugel: There are some studies under way that may shed more light on the challenges of RVO and macular edema, but the number of treatment options to compare and the number of patients that would be required for drawing solid conclusions are prohibitive. I don’t think we will have an obvious patient management blueprint in the foreseeable future. We will have to make our best clinical judgments based on studies that are not comparable. In other words, we will continue to be doctors and researchers, studying the studies in order to understand what we can and cannot extrapolate from the results.


1. Tan CS, Singer M, Bell D, Sadda SR. Predictive value of the area of peripheral retinal non-perfusion on treatment response in branch and central retinal vein occlusion. Paper presented during the annual meeting of the Association for Research in Vision and Ophthalmology, May 7, 2012, Fort Lauderdale, Fla.

2. Singer MA, Bell DJ, Woods P, et al. Effect of combination therapy with bevacizumab and dexamethasone intravitreal implant in patients with retinal vein occlusion. Retina 2012; 32(7):1289-1294.

3. Haller JA, Bandello F, Belfort R Jr, et al., the Ozurdex GENEVA Study Group. Randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Ophthalmology 2010;117(6):1134-1146.