The Impact of Anti-VEGF Treatments on Naïve AMD Patients

How to choose an agent that is right for your patient.

DR. KAISER: A dependable treatment algorithm for patients with newly diagnosed exudative age-related macular degeneration (AMD) — one that can be modified based on the unique circumstances of each patient — is an indispensable tool for any retina specialist. This algorithm is a useful foundation that can be revised periodically, as new information from clinical trials becomes available and is incorporated over time.

IMAGING FOR DIAGNOSIS

DR. KAISER: Dr. Stewart, could you begin by telling us how you approach a patient we all see on a daily basis — a new patient with wet AMD? When a patient comes to your office and reveals that a first eye has developed symptoms, how do you begin management of the patient?

DR. STEWART: First, the patient is seen by our technicians, who begin the workup. It’s my preference that every patient undergoes OCT, simply because I want to have that data available. In some cases, the patient may not present with a true retina issue. Either way, with an OCT I’m confident that I have the data I need. By the time the patient is dilated, I’ve got the results of the OCT, and can review them while I’m performing the fundus examination. If I find features suggestive of exudative AMD, and the OCT is consistent with that, I know to begin the conversation regarding my suspicions with the patient. At that point, I’ll suggest that we proceed with a fluorescein angiogram (FA).

I still perform an FA on most every patient when I suspect exudative AMD. Only on rare occasions, based on patient preferences or certain clinical findings, will I initiate treatment without an FA. We think it’s important to ­perform FA to obtain a baseline. However, I ­generally don’t perform indocyanine green ­angiography (ICGA) as a matter of routine; I reserve it for when the fluorescein suggests ­polypoidal choroidal vasculopathy (PCV).

DR. KAISER: Dr. Brown, do you look for the same features on fluorescein and OCT before considering an ICGA?

DR. BROWN: I’ll perform an ICGA on two types of patients. The first group of patients would be those in whom I suspect PCV, perhaps because of an obvious polyp on the OCT or one that was discovered during the clinical exam, especially among the very large Asian population in Houston. The second group of patients would be those with a very thick choroid, with predominantly only serous retinal detachment and very little in the way of intraretinal cysts. In this latter group, I may suspect central serous retinopathy (CSR) masquerading syndrome.

DR. KAISER: Could you describe what a polyp on an OCT looks like?

DR. BROWN: It looks like a pigment epithelial detachment (PED), but one that is very bulbous. It’s similar in appearance to one found in the GI tract, which is why I think it’s called a polyp. It looks like an extrusion from the choroid directly into the retinal space, but not violating the retinal pigment epithelium (RPE) or Bruch’s membrane, as with a retinal angiomatous proliferation (RAP) lesion.

DR. KAISER: Dr. Singh, I know how much you value information obtained from imaging. When you request an OCT, do you give your photographers or technicians any instructions about what types of scan patterns to use? Dr. Brown mentioned a focus on the choroid. Do you ask for enhanced-depth imaging (EDI)?

DR. SINGH: My routine is a little different because I work in satellite offices where I don’t have access to fluorescein. I often have to make the presumptive diagnosis of exudative AMD based on an OCT and observe the response to treatment. If there’s no response, I’ll try an angiogram and ICGA, and a fundus autofluorescence. I tend to prefer OCT for initial imaging.

When I do instruct the photographers, typically it’s with regard to horizontal and vertical line scans. We’ll perform the high definition scan and a macula cube. We perform an EDI scan on most patients in the beginning. I think the EDI in its current form is useful; there is a great deal of value in the thickness measurements we obtain. I think the normative database is going to be helpful in future versions of OCT, which should be available soon.

DR. KAISER: Dr. Singh, you do a good deal of work with EMR, and many EMRs will send up maybe one- or two-line scans at best. Do you think that’s adequate to make a diagnosis, or should we really be examining the volume or the cube scan at the different levels?

DR. SINGH: You’ve brought up an important issue. Some recent studies have shown the cube scan is probably the most sensitive agent for finding fluid on the OCT. So, the first few times I see the patient, I do walk through the cube. It’s not too difficult to perform on a Cirrus or Heidelberg, especially with reading software installed. I walk through the cube scan and make sure I agree with the fluid and the changes there.

I think that the registration of the scan is key for following patients over time, especially in a non-fluorescein practice. I tend to use the registered scans every time so the patient can see the change analysis.

DR. KAISER: Dr. Kitchens, what do you look for on OCT? What are some of the findings that may cause you to be concerned, or encourage you to perform a fluorescein?

DR. KITCHENS: We have eight satellite clinics and one home office. So approximately 60% of our patients are seen in satellite offices where, like Dr. Singh, I don’t have access to a fluorescein camera, and must rely on the OCT. When I’m in our home office, I’ll often send the patient back for a fluorescein because, for non-responders, or under-responders, it’s good to be able to reimage them 6 months later if they still have persistent fluid. However, I don’t routinely obtain a fluorescein on straightforward wet AMD patients.

DR. KAISER: What do you look for on the OCT?

DR. KITCHENS: To confirm the presence of wet AMD, I look for subretinal fluid or intraretinal edema, or RPE elevation. I also pay attention to any areas of RPE atrophy, particularly subfoveal RPE atrophy, which is something I consider when talking to patients about potential visual acuity recovery.

DR. KAISER: Dr. Stewart, let’s assume there is choroidal neovascularization (CNV) and not PCV, are there any features you look for on OCT or fluorescein — for example, lesion composition, lesion type, or lesion location — that guide you in a particular direction?

DR. STEWART: I’d say all of the above, and ­particularly lesion location. There are rare circumstances in which I consider laser photo­coagulation: with a small classic lesion well out of the fovea, or a peripapillary lesion. However, if the lesion is within 1 mm of the fovea, or shows predominantly occult features, I would not consider laser.

I agree with Dr. Kitchens that it’s important to get an idea of how much geographic atrophy (GA) there is, and we perform fundus autofluorescence (FAF) routinely with the fluorescein. The information we obtain helps us set expectations for the patient.

DR. MACCUMBER: I want to add that I think the fluorescein is very helpful for guiding the location of the OCT, and determining where to look on the cube scan. It helps determine, for example, whether there’s a lesion outside the center, or whether peripapillary fluid is present.

I also wanted to speak to Dr. Brown’s point about masquerading syndromes. EDI can help you see the choroid, and whether it’s very thick or whether there’s subretinal fluid. In some cases I find that AMD genetic testing can be useful. If the patient has high-risk mutations, I’m inclined to think it’s more likely to be AMD and less likely to be CSR, and vice versa.

GENETIC TESTING

DR. KAISER: Does anyone perform routine genetic testing?

DR. MACCUMBER: I don’t think genetic testing should be used routinely, although I’ve found it useful when the diagnosis of AMD is not clear-cut.

DR. KAISER: If you’re concerned about the possibility of a masquerade, do you perform genetic testing at that point?

DR. MACCUMBER: I’ve done it with a couple of patients. One gentleman had a history of central serous chorioretinopathy 20 years earlier, and he turned out to have a high-risk genetic profile for macular degeneration. I was more aggressive with anti-VEGF treatment, and he did very well.

DR. BROWN: I perform very little genetic test­ing. The problem is that by the nature of the genetic test, if you have mutations or single nucleotide polymorphism in Factor H, for example, you still can have more than an 80% chance of having a normal phenotype at age 65. So, there necessarily will be many people who have high-risk alleles who don’t have AMD, but may have CSR.

DR. KAISER: Does anyone think it’s important to obtain an FAF at baseline in these patients?

DR. BROWN: We obtain FAF, and I think it’s truly only helpful for following how patients do in terms of GA. I don’t think it changes my management unless I can attribute poor vision to the loss of RPE. I’ll still treat the patient aggressively, but paint a darker picture if there is a good deal of GA concomitantly.

DR. KAISER: Yes, the only way we’re going to answer the question regarding GA and anti-VEGF agents in general is to start performing FAF at baseline so we can follow changes over time. I’m glad to hear that’s something you’ve implemented in your practice, Dr. Brown.

COMPARISON STUDIES

DR. KAISER: I’d like to examine several comparison studies and the key lessons we’ve learned from them. Dr. MacCumber, what are some of your thoughts about these studies?

DR. MACCUMBER: Well, first there are the head-to-head comparisons of ranibizumab (Lucentis, Genentech) and bevacizumab (Avastin, Genentech), which were conducted in CATT and then in IVAN.^1,2 With regard to CATT, the key takeaway for me was that visual outcomes for most patients are roughly equivalent with ranibizumab and bevacizumab. IVAN confirmed that the two are roughly equivalent when it comes to efficacy. However, I think CATT showed that there is more fluid in the bevacizumab arm than in the ranibizumab arm after 1 year; and in IVAN, there was some concern that systemic VEGF levels are decreased further by bevacizumab.^1,2

DR. KAISER: The meta-analysis of the CATT and IVAN studies also would suggest that a fixed-dosing scheme works better than an as-needed one.^1,2 Dr. Kitchens, what is your take home from the comparison studies regarding treatment ­regimens?

DR. KITCHENS: Dr. Kaiser, you’re absolutely right about the dosing. We tend to focus on comparing ranibizumab and bevacizumab, but what emerges much more emphatically from the studies is the fact that monthly dosing is better, regardless of what drug you use.^1,2 I still use a treat-and-­extend paradigm, so I must consider whether I may be undertreating my patients.

DR. MACCUMBER: I would like to add that, with a patient’s second eye, or if I have a very functional patient who is driving and working, I tend to treat with more injections based on results from CATT. If it’s a first eye, with relatively poor vision when the patient presents, I’m not quite as aggressive.

DR. KAISER: Why are you less aggressive?

DR. MACCUMBER: Well, for instance, if it’s the first eye and I don’t expect the acuity to be better than 20/100, and the affected eye is still 20/20, and I don’t expect improvement over 20/100 or 20/200, I don’t believe I have to strive to obtain every single letter of improvement. It’s not going to make a difference for that patient. However, if it’s the second eye and we’re trying to preserve reading vision or driving vision, then it’s important to treat monthly, or close to it, to preserve those few letters.

DR. KAISER: With regard to the comparison studies, I agree that essentially the efficacy of ranibizumab and bevacizumab are pretty similar.^1,2 Dr. Kitchens, could you let us know your thoughts about CATT and IVAN in terms of GA?

DR. KITCHENS: There is a difference in the ranibizumab monthly group compared with the other groups — particularly the bevacizumab as-needed group — as far as the formation of GA. It’s about twice as frequent in the ranibizumab monthly group.^1,2 There is some debate about whether this is a matter of a better drying effect. With the drying effect, the GA can be more apparent. The IVAN study actually showed no difference in GA between drugs. There was a difference when it came to dosing regimen, but not between drugs.² So I think the ultimate answer to GA may lie with the HARBOR study, which is specifically examining this issue.³ The reading center may have the information we need in time for AAO or even for ASRS this year.

DR. KAISER: Dr. Singh, what are your thoughts about geographic atrophy? I think Dr. Kitchens made a good point.

DR. SINGH: The big takeaway for me lies in the comparison between continuous treatment and PRN treatment. With continuous treatment, there were higher levels of what we’re calling GA, compared with PRN treatment. I treat PRN, so this made me a little bit more comfortable with the treatments I’m delivering.

I agree that we’re likely to obtain answers from other studies, such as HARBOR. I don’t necessarily believe it’s a drug regimen phenomenon.

I think the last point is that it’s essential for us to ensure that we can accurately identify GA, and be confident that it’s not something else.

DR. KAISER: Yes, I would agree that the jury is still out on GA and whether it’s related to anti-VEGF therapy. Let’s move to the VIEW studies. Dr. Kitchens, could you summarize your thoughts about the VIEW studies, which compared ranibizumab with aflibercept (Eylea, Regeneron)?

DR. KITCHENS: Yes, the VIEW 1 and VIEW 2 studies⁴ showed that monthly treatment with aflibercept, the every 8-week (Q8) regimen, which was basically three loading doses followed by 8-week intervals of aflibercept, and the monthly treatment with ranibizumab, were all equivalent, according to top-line data. But to break it down a bit further, these studies actually looked at which patients achieved a dry state more quickly. Patients treated with aflibercept not only achieved total dryness more quickly, but also experienced better resolution of PEDs. Probably the most valuable information came from subdata which showed that after the first three injections, if fluid appeared on any one of the OCTs, putting that patient on monthly aflibercept resulted in almost one line of vision over monthly ranibizumab or aflibercept every other month.⁴

Consequently, when I’ve given patients three monthly injections and have seen persistent fluid 1 month after the last injection, I often tell patients that to obtain a significant improvement in vision, a one-line improvement in vision, it’s necessary to use aflibercept monthly for the next year to achieve and maintain dryness.

DR. KAISER: Dr. Stewart, there has been some debate over the dosing in the VIEW studies. In the first year, because it was a non-inferiority study, ranibizumab had to be given monthly and aflibercept was obviously given both monthly and then Q8. But in the second year of the study, there were modifications that included a modified quarterly PRN dosing scheme.⁴ Do you think there was enough evidence in that second year to determine that there’s a difference in efficacy between these two drugs, from a treatment longevity standpoint?

DR. STEWART: I think after everyone looked at the first year data, there was a good deal of excitement over having a Q8-week treatment, which we’d never had before. Ranibizumab was never tested as a Q8-week treatment. In the EXCITE trial,⁶ they examined patients at 8 weeks after the loading doses and found that patients were beginning to lose vision. So there was reason to believe that there might truly be a difference between drugs; there was a better result with aflibercept Q8 treatment.⁶ Then the second year data from VIEW became available, and there were 4.7 injections on average for ranibizumab. For the patients who received Q4-week aflibercept the first year, there were 4.1 injections the second year.⁴ When the data is broken down to reveal the percentage of patients receiving 11, 10, 9, or fewer injections, at least up until the 3-month mark, patients needed fewer injections of aflibercept compared with the number of injections of ranibizumab that were needed at each time point in the study. Having said that, I think the numbers are pretty close. The difference between the drugs in terms of longevity is real, but small.

SELECTING INITIAL TREATMENT DOSING

DR. KAISER: Dr. MacCumber, how do you select an initial treatment for your patients, and what are your initial thoughts in terms of follow up and injections?

DR. MACCUMBER: I think most patients can begin treatment with bevacizumab; however, with time, I’ve found that a fair percentage of them ultimately end up on a different drug. If they have persistent fluid, I’ll try ranibizumab or aflibercept. I have some patients who have deeper lesions, for instance a PED on initial OCT, and I often start those patients on aflibercept. I feel it’s reasonable to start with bevacizumab, and then make an assessment at 2 to 3 months with regard to vision, hemorrhage and OCT appearance. If we’ve achieved adequate dryness, I begin a treat-and-extend regimen.

DR. KAISER: So at the beginning you treat your patients monthly?

DR. MACCUMBER: Absolutely. I treat monthly for the first two to four injections.

DR. KAISER: And at those monthly visits, do you have them undergo any imaging?

DR. MACCUMBER: I perform OCT at every visit. The only exception would be in a situation in which a patient has a hemorrhage. I usually continue monthly injections until the hemorrhage is cleared. In those cases, I may not perform an OCT, I might just bring the patient back for an injection only.

DR. KAISER: Dr. Stewart, is that similar to your initial treatment for a wet AMD patient?

DR. STEWART: It is. As I mentioned earlier, I’ll be sure to get an idea of how much geographic atrophy there is to help set expectations for the patient. Then I’ll tell the patient that all findings recommend treatment with an anti-VEGF agent, and I’ll explain the options. I’ll usually go with bevacizumab from the beginning. I don’t necessarily ascribe to three monthly injections and then PRN treatment or a treat-and-extend regimen; I start to extend from the first visit. The next visit, if I’m extending, is probably 5 weeks afterward. However, I come close to giving three monthly injections from the beginning. One way to interpret the results of CATT and IVAN is to conclude that more treatment is generally better than less.^1,2 So, I tend to extend probably a little slower than most people; my maximum extension will be approximately 1 week every visit, but often I’ll double up on certain weeks before I extend beyond that. So I extend very slowly, and I start extending from the beginning. By 1 year, I may have extended to 3 months. If I’ve maintained the patient on 3 month intervals for a year, I would discuss stopping therapy and regularly following the patient afterward.

One of the things that has recently influenced my drug choice is the presence of an RPE detachment. If there’s an RPE detachment with a newly diagnosed choroidal neovascular membrane (CNVM), I’ll begin with aflibercept rather than the other agents, because I think aflibercept is more likely to flatten the RPE detachment.

DR. KAISER: Dr. MacCumber, does it matter to you if the lesion is above the RPE, or below the RPE, or do you use that information at all in terms of your discussion with the patient?

DR. MACCUMBER: I agree with Dr. Stewart. If the fluid is deeper, if there’s a PED, for instance, then those can be harder to treat. In that case, the patient may not respond as well to bevacizumab or ranibizumab, which I often start with, and I may go to aflibercept sooner.

DR. KAISER: Dr. Brown, how do you start treating your patients?

DR. BROWN: We typically start with a brand of drug that’s approved by the FDA; if insurance permits, then we’ll use ranibizumab or aflibercept. We normally give three loading doses. It’s simpler and easier. I tell patients, “You’re going to receive three doses and then we’re going to reassess.” I don’t want to get their expectations too high, so that they assume that there will only be three doses. I think a majority of patients will need ongoing injections, but there’s a maybe 10% to 20% chance that after three shots, the lesions will quiesce.

After that, we reassess and discuss the options of continuing monthly, PRN, or with a treat-and-extend regimen. The only time I think it’s ethically okay to proceed with PRN is if the patient is totally dry after three injections; then I’ll consider seeing whether this will be a patient who re-leaks. After that, I think it’s inappropriate to wait for a patient to have fluid to give an injection, when we know the patient is going to need treatment indefinitely.

From that point forward, it’s a treat-and-extend approach. The durability of each drug varies with each individual. There’s no evidence that ranibizumab lasts longer than 10 weeks. So, if I extend longer than 10 weeks with ranibizumab, I give the patient an option, I’ll ask, “Do you want to try PRN and risk a recurrence, or do you want to continue this 10-week regimen?” I think if you attempt every 12 weeks with ranibizumab, you may be risking too much. I think that may be true with bevacizumab too, but I’m not sure yet with aflibercept. All of these decisions depend on a discussion with the patient, and considering, for example, whether the patient lives alone, whether only one eye still has vision, and what risk the patient is willing to assume.

DR. KAISER: Do you order any imaging besides OCT at these early visits?

DR. BROWN: At the early visits, I don’t. I think imaging becomes important if you’re stretching the envelope at the end. In other words, if at 10 weeks we decide to proceed with PRN and close monthly monitoring, at that point I often will obtain an FA. Some people, especially younger patients or patients with a better RPE pump, may have a dry OCT because their RPE is doing particularly well, and yet on the angiogram you may see late leakage that’s indicative of an active lesion. It’s the same reason I examine patients at least every 2 to 3 months, because you can see hemorrhaging in the absence of leakage on OCT, which in my mind usually means disease activity.

DR. KAISER: Dr. MacCumber, what evidence do you think supports a regimen that begins with a few monthly injections, compared, for example, with one injection followed by PRN treatment?

DR. MACCUMBER: Several studies have shown that most of a patient’s vision gain is within those initial 3 months. I believe that’s why a consensus developed that treatment should begin with three to four injections monthly and then be extended. We know AMD is a very heterogeneous condition and some patients respond promptly. When I’ve caught CNV very early, I’ve sometimes treated with one injection and then just observed, and in some cases have not had to give another injection until 6 months later or longer. But that’s an atypical scenario; most patients need more treatment than that. So, I give about two to four injections monthly before I extend, unless the patient is completely dry earlier.

DR. KAISER: Dr. Kitchens, how do you start your patients?

DR. KITCHENS: I begin by telling patients that AMD is a chronic disease like hypertension that most likely will require ongoing treatment. I’ll start by giving the patient three monthly injections of whatever drug is covered by — or preferred by — the patient’s insurance. When the patient returns for the second injection, I have them undergo an OCT. I want to confirm that the treatment is working, and I want the patient and his or her family to get the kind of positive feedback that comes from seeing improvement on OCT.

One month after the third injection, if the patient is totally dry, I’ll offer an option. Assuming the patient isn’t monocular and didn’t present with a good deal of hemorrhage, I’ll suggest either proceeding with PRN treatment, and having the patient come back in a month, or trying treat-and-extend, and having the patient return in 6 weeks. I want to know the patient’s preference. There’s nothing worse than announcing that I have decided to continue to treat, and having the patient say “Oh, I hate those shots.” It’s hard to backtrack and say, “Well, we don’t have to treat you.” So, I like to know where each patient stands. Would the patient rather not come as frequently and receive more shots? Is the patient worried about losing vision? Does the patient hate the injections so much that coming in more often is acceptable? I document the motivating factors for each patient.

If a patient is still leaking, I continue to treat, and if the leakage isn’t reduced by 50% or more after the first three injections, I’ll almost certainly switch the patient right away.

I’m never opposed to a therapeutic trial in a 65-year-old patient, when I can’t quite distinguish between CSR — for example, when there is pure subretinal fluid, or maybe a little bit of thickening in their choroid — and wet AMD. If you err on the side of caution, and treat the patient as though the worst-case scenario — wet AMD — and if it gets better, you’ve confirmed your diagnosis. If, after one injection the fluid has not changed, it’s probably safe to conclude that the patient had a nonvascular event, a non-choroidal neovascular event.

DR. BROWN: I agree. I sometimes have difficulty identifying the masquerades initially. I sometimes don’t pick them up until after quite a few injections.

DR. KITCHENS: I will also add, with regard to Dr. MacCumber’s point about selectively choosing aflibercept for some patients, that if I have a patient who presents with hemorrhage, or is monocular, I’ll tend to begin with aflibercept because I want to offer what is theoretically the strongest option available. (See related Case Study on page 12.)

DR. KAISER: When we examine the time to dry in various studies, with aflibercept, at least 50% of patients are dry in 4 to 8 weeks, and with ranibizumab, it is 8 to 12 weeks. With bevacizumab it’s a good deal longer.⁵ So, if you decide to switch if a patient who isn’t dry after 3 months of bevacizumab, aren’t you sort of stacking the deck against bevacizumab?

DR. KITCHENS: Yes, but we do need objective evidence that a drug is producing a change in a patient. Visual acuity with Snellen charts varies, so you have to trust your OCT. I try to imagine what I would prefer if it were my eye. Would I want to have persistent subretinal fluid for 6 months? Or would I want to try something different if I still had persistent fluid 1 month after my third injection? I think I’d want the retina to be as dry as possible, as quickly as possible.

DR. SINGH: With regard to treatment regimens, I understand the concept of treat-and-extend from a theoretical point of view, and I know ASRS is reporting that many practitioners are using treat-and-extend, but I prefer PRN. I guess my theory is that I want to see a patient every month and determine whether the patient has fluid. I’d feel more comfortable with that approach, rather than waiting 8 weeks or 9 weeks and discovering there is fluid and that it’s necessary to shorten the treatment interval.

So, I’ll continue with PRN until I see good evidence to show me otherwise. I also believe that the definition of treat-and-extend has been problematic.

SAFETY CONSIDERATIONS

DR. KAISER: Dr. Singh, do you discuss the safety of the anti-VEGF drugs with your patients before treatment, or do you focus primarily on efficacy?

DR. SINGH: At least for patients who are insured, I find that there isn’t really much of a discussion. I present the data to patients and often find that the drug they want is the one that’s the FDA-approved and on label.

DR. KAISER: Dr. Brown, you have many patients who seem to have insurance issues. Does safety ever come up in your early discussions?

DR. BROWN: I may differ a little in that I’m truly concerned about how bevacizumab is produced. Compounding pharmacies produce products that log units more unsafe than drugs that are monitored by the FDA in sterile, aseptic conditions. Approximately one-third of our patients are in HMOs or Medicare Advantage plans that will not reimburse for anything but bevacizumab. In this scenario, I tell the patient there are three drugs for this condition, two that are specifically intended to go in the eye and that are monitored by the FDA, and one that is unregulated and not meant to go in the eye. I’ll explain that the last one is cheapest, and that the insurance company will threaten not to reimburse me unless I choose it. I explain, however, that there are increased risks associated with the cheapest option. However, I don’t think it’s worth paying cash to mitigate those risks, because they are relatively small.

DR. KAISER: I would agree. I think it’s important to outline the differences between the drugs, and allow patients to make an informed decision. If your patient has had a previous history of stroke or transient ischaemic attack, what is your discussion with the patient?

DR. STEWART: In those cases I tend to get a little more granular with the discussion, especially if the event was within the past 6 months. I’ll say that we know these drugs have the potential for causing thromboembolic events, but the incidence is not clear. Based on various studies, we have inconclusive evidence regarding relative risk among the drugs.

If it’s a recent thromboembolic event and I’m initiating therapy, I’ll probably go with ranibizumab because of that short, systemic half-life. I feel like I’m thinking of something that may tell the patient, “We’re doing whatever we theoretically can to lower your incidence of a subsequent stroke.” And that’s despite the fact that in SAILOR patients with previous strokes did have a higher incidence with ranibizumab. I still go by that low systemic half-life, and I’m hoping that that means something in patients with the highest risk.⁷

DR. MACCUMBER: Moving to another safety issue, I think it’s important to note that aflibercept is a drug that is more sensitive to temperature changes. Improper refrigeration can lead to inflammation.

DR. KAISER: Has anybody else experienced this?

DR. KITCHENS: Yes, and I agree with Dr. MacCumber that the problem does seem to be temperature dependent. We keep aflibercept refrigerated until just before use.

DR. KAISER: That is interesting. I hadn’t heard that to be a source of inflammatory events. At the Macula Society meeting, I presented a summary analysis of safety based on all of the aflibercept studies. One of the things we specifically evaluated was intraocular inflammation, which didn’t seem to be greater in the aflibercept groups compared with control groups, except in the diabetes phase 2 studies. In these studies there is, of course, much greater control of how the drug is stored.

I’ve heard that some cases of inflammation have been related to the use of the special 32-gauge needle. Dr. Brown, do you think using a 32-gauge needle has any bearing on inflammation?

DR. BROWN: I designed that needle and, it’s important to note, I donate all the royalties. To answer your question, I don’t think so. All of my partners use it. We use as much aflibercept as anybody in the country. We aren’t seeing any increased inflammation.

DR. MACCUMBER: I don’t think the cases mentioned earlier were infectious cases. The patients in those situations cleared spontaneously with no injection of antibiotics.

DR. KAISER: Let’s talk about the injection itself. After you give the injection, do you administer antibiotics?

DR. SINGH: No.

DR. KAISER: At the Cleveland Clinic, we stopped giving post-injection antibiotics almost 2 years ago, for multiple reasons, including drug-resistant organisms being bred on the surface of the conjunctiva, and the risk that would pose in the future. In general, I think the standard of care has certainly shifted to not using postoperative antibiotics with these injections.

DR. SINGH: Can I ask whether anyone uses post-injection antibiotics immediately after the injection? Just in the office, but not giving them to the patient to take home?

DR. MACCUMBER: I do. What I’m doing is essentially flushing out the Betadine. I probably could use BSS, but I happen to use an antibiotic. However, I don’t give the patient a bottle to use after that.

DR. BROWN: We do the same.

DR. SINGH: I don’t see the point in administering an antibiotic after an injection. I don’t think any of the antibiotic drugs are penetrating into the vitreous cavity. Whether you’re giving gentamicin or another drug, I think the penetration is quite low. So I’ve stopped administering antibiotics, even after the injection. If anything, I believe it makes more sense to give antibiotics before the injection, if you want to decrease ocular flora, but we’ve actually moved away from that practice, as well. We aren’t really sure if antibiotics help in any discernible way. ■

This discussion continues with a second supplement to be published with the July/August issue of Retinal Physician.

REFERENCES