An expert panel led by Retina Today Chief Medical Editor, Dr. Allen C. Ho, discusses strategies for obtaining optimal outcomes in patients with agerelated macular degeneration (AMD). In this monograph, Drs. Ho, Jonathan Prenner, Peter Kaiser, and Richard Kaiser also provide insights into how they individualize therapy for each patient with AMD.

Allen C. Ho, MD, is Director of the Retina Research Unit at Wills Eye Institute and Professor of Ophthalmology at Thomas Jefferson University School of Medicine in Philadelphia, Pennsylvania. He is Chief Medical Editor of Retina Today. Dr. Ho can be reached at acho@att.net.

Jonathan L. Prenner, MD, is Clinical Professor in the Department of Ophthalmology and Pediatrics at the Robert Wood Johnson Medical School in New Brunswick, New Jersey. He is the Co-Chief Medical Editor of New Retina MD and an assistant editor for the journal Retina. Dr. Prenner can be reached at jonathanprenner@gmail.com.

Richard S. Kaiser, MD, is the Co-Director of the Retina Fellowship at Wills Eye Institute and an Associate Professor of Ophthalmology at Thomas Jefferson School of Medicine in Philadelphia, Pennsylvania. He is the Co-Chief Medical Editor of New Retina MD and an assistant editor for the journal Retina. Dr. Kaiser can be reached at kaiserrick@aol.com.

Peter K. Kaiser, MD, is the Chaney Family Endowed Professor of Ophthalmology Research at the Lerner College of Medicine and the Cole Eye Institute, Cleveland, Ohio. He is the Editor-in-Chief of Retinal Physician. Dr. Kaiser can be reached at pkkaiser@gmail.com.

AMD: A GROWING AND COMPLEX PATIENT AND PUBLIC HEALTH PROBLEM

DR. HO: Dr. Prenner, would you please provide some comments on the epidemiology and burden of AMD?

DR. PRENNER: AMD is a surprisingly common problem, which is going to increase in prevalence as the population ages. It is an incredibly impactful disease for patients, often reported as the worst medical problem that a patient might have. One quality-of-life (QOL) study demonstrates that patients with AMD report lower QOL scores than patients with debilitating systemic diseases like chronic obstructive pulmonary disease and acquired immune deficiency syndrome.1 Based on National Institutes of Health data, we project that the advanced form of AMD affects 2.2 million individuals in the United States, and while approximately 85% have dry AMD, wet (neovascular) AMD affects 10-15% of the overall AMD population. This accounts for between 220,000-330,000 individuals in the United States.2

The prevalence of advanced AMD in the United States is high, and our specialty has become, in part, defined by the management of this disorder. This prevalence is going to increase to nearly 3,000,000 people by 2020, which will correspond to 300,000-400,000 patients with wet AMD as the baby boomer population ages.2 Interestingly, a substantial part of that growth is in the 85+ age group. As a result, we are going to be treating a more significantly aged population with this disease as well.

In short, the prevalence of AMD is significant and is going to rise dramatically. The literature suggests that wet AMD affects a comparable number of individuals in Europe, and the demographic trends there are similar.2 Recent survey data suggest that there is a growth in the care of AMD, with two-thirds of physicians reporting an increasing number of injections performed over the last 12 months.3

We're all very busy clinicians, seeing a lot of patients with AMD. I'd like to ask my fellow panelists: Are you're seeing more patients with the disease or treating more often? Is the treatment burden increasing in your practices?

DR. PETER KAISER: I don't know if it's increasing. I think in the past, most general ophthalmologists and even family practitioners looked at AMD as an untreatable disease. So a patient coming in with disease might not get referred to a retina specialist. I think the public awareness of the anti-vascular endothelial growth factor (VEGF) therapies has changed this notion. Now, general ophthalmologists and even primary care doctors are sending us AMD patients earlier and earlier. So, the proportion of patients that we can actually treat as opposed to, say, a dry patient or a disciform-scar patient, is going up.

DR. HO: Retina specialists have been amazingly adept at expanding their capacity to treat these patients, and if you have a disease that is growing in a growing segment of the population, that increases the number of those patients in your practice.

DR. RICHARD KAISER: We've had to make logistical changes to our practices. We've had to adapt our actual office structure to accommodate these patients so that we can (1) move them through in an expedited fashion; (2) present to them the data that we have in a concise package; and (3) still give them as much TLC as we possibly can. But the increased number of patients in our offices does stress the system a little bit. However, I think we've been good at adapting to that.

We all have been practicing long enough to remember when all we had was laser for AMD patients. There is nothing more rewarding than being able to actually help people. We've had to evolve our approach to patients with the advancement of technology, but the reward is well worth it.

SCREENING AND MONITORING AMD IN PATIENTS AND IN THE POPULATION

DR. HO: Please tell us how you screen for and monitor AMD.

DR. PRENNER: As we have discussed, the incidence and prevalence of AMD is very high in the population over age 65, and appropriate screening is critical to help identify those patients at risk for the disease. An annual examination by an experienced ophthalmologist is perhaps the most effective means of screening for AMD. Modifiable risk factors include hypertension, cholesterol control, and control of smoking and obesity, but many of the risk factors for AMD are not modifiable.4

The findings of the screening exam will help determine, in large part, the risk of developing advanced AMD and dictate the need for nutritional supplementation. As far as the pace of follow-up goes, we know that patients with intermediate dry AMD, advanced dry AMD, or wet AMD in 1 eye are at very high risk for development of wet AMD in the other eye. This high-risk population may benefit from examination by a retinal specialist.5,6 Subclinical wet AMD is also not uncommon; we've all seen patients who are asymptomatic with wet AMD, and these patients are best treated before they develop visual dysfunction. These patients can be missed without the high resolution and multifaceted imaging modalities that retina specialists specifically utilize.

At-home screening devices are now coming online, where patients can monitor themselves in a sensitive and automated way for development of advanced AMD. Well-educated patients have started to ask about these at-home screening approaches as they are becoming commercially available.

DR. HO: Those points about at-home screening and monitoring are interesting because our care and our assessments of patients are staccato and intermittent. Does anyone want to comment on their experience with some of the new digital devices that might be helpful in the future for home screening/monitoring?

DR. PETER KAISER: I was recently involved in 2 studies looking at home-based monitoring of AMD: (1) Foresee Home device and (2) an iPhone-based app in which a patient could do a vision test on the smartphone.7,8 Both of these at-home tests measure vision changes and metamorphopsia, and clinical studies are ongoing evaluating whether, once a patient is being treated for wet AMD, can that patient be followed at home so that he/she only needs to come into the office when a reinjection is needed? At this point, both tests are still unreliable in terms of predicting when a patient would require reinjection. They appear to be more reliable in predicting the crossover from dry to wet AMD, but improvements in the algorithm may make the tests even better.

DR. RICHARD KAISER: I believe we need to look at screening from a more global health care perspective. Should we be screening patients before they develop dry disease? Should we be screening patients that have a family history of AMD? What are we going to do if they are identified? Can our health care system afford to provide services for these patients, because you're now talking about millions of patients at risk for AMD? We still do not have any treatments for the dry form other than nutriceuticals, which can only slow down the progression of the disease. Once patients have wet AMD, can and should they continue to monitor themselves at home? Right now, based on our treatment regimens, I think that the money is in educating our patients at highest risk of developing wet AMD, because we have effective interventions.

DR. HO: While patient screening with at-home technology is still evolving, it makes great sense that these technologies will likely mature to provide our increasingly sophisticated senior patients with the ability to monitor disease progression outside of the office setting with improving reliability and ease.

INITIAL EVALUATION OF THE AMD PATIENT

DR. HO: A 75-year-old woman is sent to you with new blurred vision in her right eye. Let's have each faculty member discuss his initial evaluation of this patient.

DR. PRENNER: I'll typically perform a comprehensive ophthalmologic examination and pay careful attention to ophthalmoscopy, often times utilizing a contact lens exam, to get a good first time look at the macular anatomy. Then, pending those results, I'll typically obtain an optical coherence tomography (OCT) and color fundus photograph to document the baseline features. If we assume that the patient in this scenario has only dry AMD, I may initiate the use of AREDS vitamins and exogenous omega 3s, depending on the severity of their disease, and ask the patient to check an Amsler grid daily. I would schedule a follow-up visit in 6-12 months.9

DR. RICHARD KAISER: I agree with the paradigm Dr. Prenner laid out. The exam findings may push you to deviate a bit. If your exam reveals hemorrhage or new clinical changes, I'd consider ordering a fluorescein angiogram (FA), which is still the backbone test for evaluating the pathologic activity of this disease.

DR. PETER KAISER: In addition to an OCT, FA is vital in the diagnosis of this disease. There are masquerade syndromes, so I always obtain a FA at baseline. If you live in a part of the country where you see more pigmented races, the risk of polypoidal choroidal vasculopathy (PCV) is higher and you may also get an indocyanine green (ICG) angiogram at baseline. I practice in a part of the country where most of my patients are Caucasian, and I generally do not obtain an ICG at baseline except in the patients for whom I'm concerned about PCV.

DR. HO: So it appears that our panelists still utilize angiography as an important diagnostic tool in the evaluation of AMD, particularly at the time of initial assessment when neovascular AMD is of concern or when there are clinical changes or unexplained or incongruous clinical signs or symptoms. OCT is an essential diagnostic tool in the initial evaluation and follow-up examination of a patient with AMD, particularly a patient with neovascular AMD.

CURRENT MANAGEMENT OF NEOVASCULAR AMD

The Differences in VEGF Therapies

DR. HO: In the past year, we've had major randomized controlled clinical trials present level 1 evidence for various anti-VEGF agents in the treatment of neovascular AMD. Dr. Peter Kaiser, can you tell us about the differences between these therapies?

DR. PETER KAISER: The nice thing about practicing medicine currently is that we have several commercially available anti-VEGF agents with which to treat our patients. Some of the key differences between these agents are summarized in Table 1. What we see in the chronologic development of these agents is expansion of relevant targets. For example, pegaptanib, the earliest commercially available anti-VEGF agent for use in AMD, only binds to the VEGF165 isoform of VEGF-A, while bevacizumab and ranibizumab bind all isoforms of VEGF-A. These panVEGF agents are associated with higher efficacy than pegaptanib. The latest developed agent, aflibercept, not only binds all isoforms of VEGF-A, it also binds placental growth factor (PLGF) and VEGF-B, which the others don't inhibit. With the more advanced agents, we also see a progression in improved pharmacologic characteristics (increased affinity for relevant targets and/ or penetration into the eye). For example, ranibizumab has a higher binding affinity than bevacizumab, and aflibercept has an even higher affinity, as well as a longer half-life in the eye.11

DR. HO: Is there any relevance about the role of PLGF in neovascular AMD? Or is this unproven?

DR. PETER KAISER: At this point, it's clinically unproven, but there are some good animal data suggesting that PLGF plays a role in neovascularization as well as in CNV. So PLGF plays a role, but it's probably a minor one. We are still learning about the role of PLGF in AMD.15

DR. PRENNER: We have 3 outstanding drugs—certain eyes and certain lesions may respond preferentially to aflibercept, bevacizumab, or ranibizumab. We have all been excited by some of the early responses in aflibercept-treated patients who did not respond to bevacizumab or ranibizumab. I wonder if this preferential response is partially due to PLGF binding. However, these responses may reflect the increased binding affinity of aflibercept to VEGF.

DR. RICHARD KAISER: We have no way of actually measuring a patient's intrinsic PLGF levels to determine which isoform of VEGF is more active in their particular case, etc. Having multiple therapeutic options gives us more opportunity to tailor our therapy for each patient. Ideally we would like to have a biomarker to guide our treatment approach.

DR. HO: How much attention do we pay to this sequence of science—from affinity in a petri dish, to activity in an animal model, to efficacy in our patients with neovascular AMD? Are there meaningful differences among our current treatment options?

DR. PETER KAISER: It's difficult to translate biologic activity from a mathematical model, say, Michael Stewart's model that reported that biologic activity is greatest for aflibercept, then ranibizumab, and finally lowest with bevacizumab.16 However, we do see some clues from the population-based studies. For instance, if you look at the early aflibercept studies in which they followed patients with as-needed treatment, they were able to go a very long time between treatments.17 If you look at the ranibizumab as-needed studies over the long term, the visual acuity does not remain a straight line, like it did in the aflibercept phase 2 studies with as-needed treatment, but instead trails off over time.18

But, what matters is what happens in an individual patient. And the biologic activity in the individual patient is going to be based on a lot of things. For instance, in vitrectomized patients, biologic activity is going to be very different. In my mind, there is a difference in biologic activity in terms of the length of time these products last. Whether it will translate into a much longer time between injections with aflibercept or not remains to be seen. Some patients will respond better to certain agents.

DR. RICHARD KAISER: There is always a gap between animal studies, clinical trials, and clinical reality. Clinical trials test a homogeneous population with fresh, active CNV lesions with few complicating clinical features such as significant pigmental epithelial detachments (PEDs), hemorrhage, and early fibrosis. Clinical trials are designed to test the drug on substrate that will best respond to therapy. In clinical practice, we need to treat all lesions and in reality some respond to therapy better than others. Thus, vision outcomes and treatment regimens can vary and may not exactly follow the FDA label in clinical practice.

DR. PRENNER: In a clinical trial, efficacy is judged by the change in ETDRS letters read. For better or for worse, this is not the efficacy metric that we use daily in our clinical practices. In large part, we use surrogate markers rather than ETDRS vision to judge pharmacologic effect in daily patient care. These surrogate markers include anatomic changes on ophthalmoscopy, OCT, and angiography. When one carefully reviews anatomic data from recent prospective AMD trials, aflibercept may demonstrate anatomic advantages in terms of OCT thickness and CNV quiescence on OCT and angiography compared with ranibizumab or bevacizumab.19,20 We may be seeing the scientific advantages of aflibercept at the bench (in terms of durability and efficacy) translate to an improvement of the anatomic biomarkers that we use to judge efficacy as clinicians. These anatomic markers are clinically important, but did not translate into a major difference in ETDRS letters read in a relatively short-term clinical trial.

Reviewing the Evidence from Clinical Trials

The HARBOR Trial-Safety but No Increased Efficacy with Higher-Dose Ranibizumab

DR. HO: The HARBOR trial enrolled approximately 1200 patients and was a comparison of standard-dose ranibizumab (0.5 mg) vs higher-dose ranibizumab (2.0 mg) on a monthly basis as well as those 2 dosages on a PRN basis (after 3 loading doses). All 4 groups showed substantial improvement in mean visual acuity at 1 year (0.5 mg PRN: 8.2 letters, 2.0 mg PRN: 8.6 letters, 0.5 mg monthly: 10.1 letters, 2.0 mg monthly: 9.2 letters), similar to what we saw in the original ranibizumab trials (Figure 1).21 The PRN groups, however, did not meet the strict noninferiority margin (4 letters) of monthly ranibizumab. The other significant information about HARBOR is related to safety. With the ANCHOR and MARINA trials, there was some suggestion of a safety disadvantage with the higher dose of ranibizumab (0.5 mg vs 0.3 mg). This was highlighted in the preliminary analysis of the SAILOR data, which culminated in the Dear Healthcare Provider letter citing numerical difference in cerebrovascular accidents with the higher dose22 in the first 6 months of the study. However, this subsequently did not pan out upon further follow-up. But the HARBOR study showed that there were no significant safety differences between the 2 mg and 0.5 mg dosages of ranibizumab. At least to my thinking, this gives a lot of relief from any safety concerns in a dose response. The disappointment of the HARBOR trial was that we did not raise the bar for efficacy (mean improvement in visual acuity or 3-line gainers) with the higher dose. Therefore, the standard dose of ranibizumab (0.5 mg) remains the benchmark for that drug.21

The VIEW1 and VIEW2 Trials: Aflibercept Efficacy with q8 Week Dosing

DR. PETER KAISER: The VIEW1 and VIEW2 trials enrolled approximately 2400 patients, making VIEW, collectively, the largest AMD trial to date.11,19,20 The VIEW trials were noninferiority studies, which have some specific caveats: (1) The comparator group must be the gold standard therapy (monthly ranibizumab); (2) They must enroll patients similar to those enrolled in the original trials as the gold-standard therapy (ie, patients who were treatment naïve with wet AMD and any type of lesion composition like the MARINA and ANCHOR studies); and (3) They must use a prespecified endpoint for noninferiority. In this case, it was maintenance of vision at 52 weeks (<3 line loss of vision). The noninferiority margin set by the FDA was 10%, but more importantly, if the results were within 5% of ranibizumab, then the drugs could be considered clinically equivalent.

In the VIEW trial, patients were randomized either to ranibizumab 0.5 mg monthly or 1 of 3 aflibercept regimens: 2 mg or 0.5 mg given every month (q4 wk), Figure 1. HARBOR mean change from baseline in BCVA to month 12. All regimens shown are ranibizumab. From Ho AC et al. 2011.21 Figure courtesy and 2 mg given with a loading dose of 3 monthly injections and then every 2 mos (q8 wk) up to 52 weeks. From wks 52 to 96, patients were dosed at least quarterly, with more frequent dosing based on predetermined retreatment criteria.11,19,20

Overall, all the groups prevented moderate vision loss in almost equal amounts. At 52 weeks, all the groups were well within the 5% non-inferiority margin, so all the treatment regimens were clinically equivalent.19 These trends held up at the 96-wk endpoint. For example, the proportion of patients maintaining visual acuity (loss of <15 ETDRS letters) at week 96 was 92% for ranibizumab q4 wk, 92% for aflibercept 2 mg q4 wk, 91% for aflibercept 0.5 mg q4 wk, and 92% for aflibercept loading dose followed by q8-wk dosing. The BCVA gains for these groups were 7.9, 7.6, 6.6, and 7.6 letters, respectively.20 Thus, the important finding of the study was that the aflibercept regimen with a loading dose followed by q8-wk dosing had results equal to ranibizumab monthly. That goes back to this idea of biologic activity. The fact that the every 2-month group was able to equal monthly ranibizumab supports the idea that aflibercept has longer biologic activity. The safety of all aflibercept groups was identical to ranibizumab. There were absolutely no significant or even numerical differences in safety endpoints between the 2 drugs. The results from the second year of the studies, which changed the treatment regimens to a modified as-needed treatment regimen, showed that the vision endpoints and safety were similar up to 2 years.20

THE CATT Trial

DR. RICHARD KAISER: Before we discuss the CATT trial, it is useful to review the body of evidence leading to this study. The early trials with ranibizumab, MARINA, which was sham controlled, and ANCHOR, which had a comparison with PDT, clearly showed that monthly ranibizumab injections were statistically much better than sham or the PDT; with approximately 95% of ranibizumab-treated eyes experiencing visual improvement or stabilization compared to only about 60-65% of sham or PDT-treated eyes. For the first time, we saw a significant group of patients gaining vision, and 41% of patients had more than 3 lines of vision gains between the 2 trials. Overall, the mean improvement in vision was 11 letters.23,24 This was a game-changer. Up until then, we had been trying to prevent significant vision loss, and for the first time, we had clinical trials that showed a significant visual gain. However, the maximum gain in vision has repeatedly been achieved with fixed monthly dosing. But ever since the MARINA and ANCHOR studies, we have been searching for a more efficient and less burdensome means of treating AMD.

Several studies of extended dosing are described in Table 2.25-29 The results to date have been disappointing and/or not easily reproduced. Note that there are 2 options for PRN treatment. Treat-and-observe requires regular follow-up of stable patients, with treatment thereafter only in the presence of disease activity. Alternatively, in a treat-and-extend dosing strategy, treatment is administered regardless of detectable activity but the intervals between treatments are extended as long as disease remains quiescent.

In the CATT trial, patients were randomized to ranibizumab and bevacizumab, both monthly and PRN. This trial employed a treat-and-observe regimen, in which the investigator had carte blanche to retreat a patient in the PRN arms based on strict OCT or visual criteria as well as the clinical exam. At 1 year, the CATT results showed that PRN therapy is not as effective as strict monthly on-label therapy. There was actually a decrease of 2.4 letters between PRN and monthly regimens.28

DR. HO: This shows that persistent monthly anti-VEGF inhibition is superior to the tested PRN intermittent anti-VEGF suppression treatment regimens in CATT.

DR. RICHARD KAISER: This brings up a point about the use of OCT in a treat-and-observe manner. We have elevated the OCT to be the gold standard for detecting active disease. But in actuality, the OCT is highly effective at detecting early anatomic change, but the OCT cannot detect biologic activity. Perhaps we put too much emphasis on the OCT to determine if and when a patient should be treated.

DR. PRENNER: We believe that with a “treat-andextend” treatment strategy, we can employ the art of retinal care, taking advantage of our clinical intuition, multifaceted imaging modalities, and interpretation of patients' symptoms, to both maximize patient outcomes and decrease treatment burden. Unfortunately, treatand- extend is not a treatment regimen that's easy to test in a clinical trial.

DR. RICHARD KAISER: To add onto that, if we had a different imaging modality, maybe one that was more effective at OCT, or we had an assay to detect biologic activity, say VEGF or PLGF levels, then our PRN dosing regimens would be much different and we would be treating more frequently. Perhaps we would have better outcomes in the PRN dosing regimens.

DR. HO: With the exception of the q8-week aflibercept regimen (after 3 monthly induction injections) as studied in the VIEW trial, monthly therapy affords the greatest improvement in mean visual acuity in the HARBOR, CATT, and VIEW trials.

I think then we have to translate the data to reality, and it is generally not practical or possible to treat patients every month. It may be possible but the treatment burden often outweighs the benefit. I think the take-home message from this discussion in general is more anti-VEGF therapy is better than less anti- VEGF therapy. If you are going to adopt a treatment regimen that deviates from monthly ranibizumab or bevacizumab, you really have to be cognizant of the PRN data. The aflibercept 2 mg q8-week results are the only less than monthly regimen that is equivalent over 2 years to monthly ranibizumab. Certainly, not all patients require aggressive anti-VEGF inhibition but many do to achieve their best visual potential.

DR. PRENNER: I think we all have patients who clearly don't need to be treated on a monthly basis and who do very well, and we wouldn't want to be heavy-handed in our approach—and take on the liabilities of safety, cost and treatment burden for those particular patients. That group is not a small subset. There are many patients who do quite well with less frequent dosing. We wouldn't want to miss those patients by treating everybody on a monthly basis.

Two-year Update to CATT

DR. HO: If we look at the 2-year CATT, over time, what you saw was that monthly ranibizumab and bevacizumab were very similar. But there are certain potential advantages to receiving monthly ranibizumab. For example, a higher percentage of patients had no OCT fluid at the end of 1 year with monthly ranibizumab vs bevacizumab. If you compare the molecules over 2 years – within the same treatment regimen – they perform similarly with respect to mean visual acuity. However, the PRN-treatment arms failed to meet the gold standard of ranibizumab monthly (ranibizumab monthly 8.5 letters gained, bevacizumab monthly 8.0 letters, ranibizumab PRN 6.8 letters, bevacizumab PRN 5.9 letters, Figure 2).29 We saw this right before the end of year 1 and again at year 2 – the slope of the PRN curves for both ranibizumab and bevacizumab began to fall off and diverge when compared with monthly curves. Although ocular safety comparisons between ranibizumab and bevacizumab are equivalent, bevacizumab use was associated with a higher systemic serious adverse event rate that was observed at 1 and 2 years.

The current discussion of equivalence in my view might apply to an unrealistic monthly comparison of bevacizumab vs ranibizumab. But if you're going to use bevacizumab PRN, at least by the CATT study methodology, you're going to have patients lose vision compared to a monthly treatment.

I think the safety from an ocular standpoint was likely similar between the 2 products. However, in terms of systemic safety, the CATT trial showed a difficult-toexplain disadvantage to bevacizumab compared to ranibizumab (Table 2).29 The disconnect there was that it occurred more frequently in the PRN group, and therefore, people began to think that this undermined the idea that bevacizumab was less safe systemically because it occurred more frequently in the as-needed treatment group.

However, not everything is dose response-related in terms of safety effects. In addition, the other systemic safety events were not typically thought related to an anti-VEGF mechanism that we currently hold (eg, higher rates of hospitalizations for pneumonia). My thinking after year 1 was that this was simply a numerical difference that would likely equalize over time. However, at the end of 2 years, as recently presented, those systemic safety differences persisted.

Now, none of the trials we do in ophthalmology are powered to detect significant difference between low adverse event rates. But that numerical persistence of a safety disadvantage with bevacizumab concerns me because we don't have a good explanation for it. I expected it to go away, but it did not.29 I feel obliged in my practice to discuss that with my patients. That poses some dilemma for me when I am proposing off-label bevacizumab. Comments?

DR. PRENNER: CATT demonstrated a statistically significant safety issue that needs to be addressed and looked at going forward.

One should keep in mind that the bevacizumab used in clinical practice is not the same bevacizumab that we had in the clinical trial. As a CATT investigator, I felt very confident giving patients a product that was aliquoted in a safe and uniform manner. The CATT bevacizumab was aliquoted into 2-mL borosilicate glass vials and was delivered to the investigators in a monitored and uniform manner. That is a different process and product than what I am able to obtain from my local compounding pharmacy. There are a number of studies that show that the bioavailability of compounded bevacizumab differs depending on how it is aliquoted, so I wonder about drug efficacy. I also worry when I hear the seemingly annual background noise concerning an outbreak of bevacizumabassociated endophthalmitis that causes significant visual loss.30

IMAGING IN NEOVASCULAR AMD FOLLOW-UP

DR. HO: How frequently do you use FA for follow-up in a patient with neovascular AMD?

DR. RICHARD KAISER: It's easy to fall into the trap of continuing to monitor patients with exams and OCTs. I think it's imperative to periodically perform an angiogram, especially if you're going to alter your treatment regimen and move off label in terms of monthly treatment with ranibizumab or bevacizumab. Many retina specialists don't stick exactly to the label, so I think it's very important to do angiograms periodically, especially as you start to extend the time between your visits. I have seen many cases where the OCT is dry, but the lesion is growing at a rapid rate, and the angiogram is going to reveal that.

DR. PRENNER: We have evidence-based data demonstrating the superiority of frequent, persistent anti- VEGF therapy compared to PRN dosing strategies utilized in recent clinical trials. We balance that data against the burden of treatment involved with monthly injections and the fact that we think that with a very careful treat-andextend paradigm, which has not been formally tested, we can get very close to the outcomes achieved with persistent anti-VEGF inhibition. Treat-and-extend management is a very nuanced process. It has to be managed very carefully by a retina specialist, and it has to incorporate all the tools at our disposal.

I think we tread on somewhat thin ice when we start treating patients in a manner other than with monthly injections. And so, I try to be extremely careful with my treat-and-extend management, and make sure that before I change a dosing interval, that I reimage with OCT and fluorescein angiography, to confirm that there is not a growing CNV component that I can't identify by OCT.

DR. PETER KAISER: I would also add that for patients who don't respond to therapy, then I'm more likely to get ICG in addition to fluorescein to rule out masquerade syndromes and guide future management.

NEXT STEPS IN AMD –WHAT'S IN THE FUTURE?

DR. HO: We've laid out some of the evidence. We do have a lot of evidence for choices now for our patients. And yet, even with those choices and a game-changing kind of treatment, we, for example, still have many patients that don't improve visual acuity. We have maximized the limits on the metric of maintaining or preventing vision loss, but in terms of 3-line gainers, we're still at an opportunity for 60-70% of patients to improve vision with therapies.

Combination therapy has been held to have great hope, but today, my assessment is it's fallen flat in terms of raising the bar and improving visual acuity. That may be changing, however, as we evaluate the information from Ophthotech on their positive phase 2 study combining anti-platelet-derived growth factor (anti-PDGF) and ranibizumab.

DR. RICHARD KAISER: Clearly VEGF is not the only pathway that is driving this disease. As a clinician, I look forward to therapies that block other pathways that may be as important or even more important in treating this disease.

DR. HO: We look forward to the hope that rests on the shoulders of combining anti-VEGF therapy with perhaps noninvasive radiation therapy and other strategies such as anti-PDGF factor treatment.31 We are watching some potentially significant trials of new combination treatment options that have yet to be presented in a peer-reviewed setting. Once we learn more about these combinations, we will have a better sense of the incremental benefit as well as the safety and convenience implications of these more complex approaches.

INDIVIDUALIZING THERAPY OF NEOVASCULAR AMD PATIENTS

DR. HO: When treating individual patients with wet AMD, the art is in interpreting and processing the information from our trials and then tailoring it for that patient. We have choices in 2012 going forward that may benefit 1 patient more than another. The discussion below and the video case discussions eyetube.net/portals/New-Evidence-in-the-Treatmentof- Neovascular-AMD illustrate some of the factors we consider in individualizing therapy.

Initial Therapy for Wet AMD (Case presented by Dr. Allen Ho)

DR. HO: We have a 77-year-old woman, with newonset vision loss, fluid and hemorrhage in the macula, and visual acuity at 20/100 attributable to neovascular AMD. How would you proceed?

DR. PRENNER: I would treat her on the same day that she presents. I would use an approved therapy, either aflibercept or ranibizumab, rather than bevacizumab because of safety concerns with the compounded drug that I can obtain in my clinical practice. Until recently, I most frequently used ranibizumab because of its availability and our extensive experience with the drug. Lately, I've begun using aflibercept as well, more frequently as the drug becomes more readily available to our patients and reimbursement issues are addressed.

DR. RICHARD KAISER: We are restricted by reimbursement issues. Insurance coverage is an issue, so we do not always get to use our first-choice drugs. At this point, we are waiting for a J-code for aflibercept, so that affects our choices. As time goes on and reimbursement issues are resolved, I believe we will be able to offer patients the drugs we want.

DR. PETER KAISER: I agree with everything that has been said. I also prefer aflibercept or ranibizumab as first line in a patient with new onset, treatment-naïve, wet AMD because of their established safety and efficacy, and the fact that fractionation is not required. However, it's important to note that bevacizumab is the most commonly used drug for wet AMD in the United States and all the recent comparison studies show minimal efficacy differences.32 When talking with patients, I present clinical trial data for all 3 drugs, I highlight the differences, and then I let them choose. In general, my patients choose ranibizumab or aflibercept. I use the “mom” test—in other words, what would I use to treat my mother? That would be aflibercept or ranibizumab.

DR. HO: We make treatment decisions based on safety and efficacy. Are there any safety differences that we should be aware of among these agents?

DR. PRENNER: The FDA-approved therapies appear to have similar efficacy and safety. There are some safety concerns with bevacizumab that came out of the 1-year and 2-year CATT data.28,29 While that is an evolving story, it is a concern that I discuss with my patients now that the data are available.

DR. RICHARD KAISER: I share those concerns. There are potential ocular issues with bevacizumab related to the formulation (ie, the source of bevacizumab). In the CATT trial, serious adverse events occurred in 32% of patients receiving ranibizumab vs 39% of patients receiving bevacizumab.29 This was statistically significant. We may not have a rational explanation at this point, but we need to investigate it further, and it is something we should discuss with our patients.

DR. PETER KAISER: There are 2 different safety issues associated with use of an off-label molecule like bevacizumab. Rick outlined the safety issues from the molecule itself that arose in CATT and other comparison studies. However, the significant differences in adverse events seen in these studies are not the ones we usually associate with anti-VEGF agents. So we really do not know what to do with these data. It is important to note that the CATT trial was not powered to measure safety events, such as myocardial infarction and stroke. However, larger retrospective case series using the Medicare database have shown a possible increased risk.33 So the jury is still out in terms of the safety of the molecule. The clinical trials for the approved drugs, aflibercept and ranibizumab, suggest that they are safe. To me, the main safety issue regarding bevacizumab is related to risk introduced with compounding of the molecule. In CATT, the molecule was supplied in a sterile single-use vial like we receive ranibizumab and aflibercept that was produced in a sterile fashion by a pharmaceutical company. This is not how we get bevacizumab when we use it. In practice, we have to trust that the bevacizumab that our compounding pharmacy gives us is free of microbiologic contaminants and contains the appropriate amount of the drug. We know that this can be an issue, given the case reports of clusters of bevacizumab-associated endophthalmitis due to compounding pharmacy errors.30

DR. HO. Let's say it is your mother, and there are no payer issues, how would you treat her?

DR. PRENNER: I have extensive experience with ranibizumab, but I am very encouraged by my early experience with aflibercept. Right now, I would be comfortable giving either aflibercept or ranibizumab. That answer may change in a few months, as I gain more experience with aflibercept.

DR. RICHARD KAISER: I agree. I am comfortable using ranibizumab or aflibercept as a primary therapy. We are fortunate to have multiple excellent drugs.

DR. PETER KAISER: I would give my mom the best medication possible. In my opinion, I would choose aflibercept. While I agree that aflibercept and ranibizumab are similar in efficacy, aflibercept has a small safety advantage, in that I can give her fewer injections. That lowers the risk of an adverse event simply because of the reduced numbers of injections.

DR. HO: For my mother, I would be comfortable with aflibercept—I do have a lot of experience with that agent, since we were involved in the trials. I do like the potential advantage of fewer injections. However, we don't know whether ranibizumab would be effective in a similar dosage regimen, since it's never really been studied in a less than monthly fashion and compared with aflibercept. Certainly, I am very comfortable with ranibizumab as well.

When is it too Late to Treat Wet AMD? (Case presentation by Dr. Richard Kaiser)

DR. RICHARD KAISER: Let's discuss the case of an 89-year-old gentlemen, status post stroke 10 years ago, who was left in a somewhat debilitated state. He was followed by another retina specialist. Left eye had a disciform scar. His right eye was treated from 2005 to 2007 with regular ranibizumab injections. The right eye treatment was halted in 2007 when it was deemed not worth continuing (he had counting-finger vision in both eyes). The patient was observed only until November 2011, when his daughter brought him into our clinic for a second opinion. I was unable to get an angiogram because he had a known history of fluorescein dye allergy, but the OCT and the red-free image show scarring in the central macular, atrophy, some fibrovascular change, a lot of edema, and subretinal fluid (Figure 3a). As poor as his vision was, I didn't get a sense that this was an end-stage eye.

We opted to proceed with treatment with ranibizumab. After 1 injection, he did not notice any change in vision, but the OCT shows some improvement (Figure 3b). Fast forward, after 5 injections with ranibizumab, his OCT is greatly improved, although obviously his scarring is still there (Figure 3c). At that point his vision was 20/200, and his quality of life was greatly improved.

This raises an important point—how do we know when it is too late to treat wet AMD? We don't have clinical trials or guidelines to inform us on this issue.

DR. HO: The status in the fellow eye helps me to determine how aggressive I'll be with an eye with a fibrotic lesion. Rick's right—there is no clear guidance. When a patient is down with 2 eyes with wet AMD, I'll talk with the patient, asking them what his/her preferred eye is. I'll suggest being more aggressive with that eye and talk with the patient and the family, since that patient will need to be brought in on a regular basis. It's the art of retina care. I will get aggressive if they have 2 bad eyes, trying to figure out which eye has more potential.

DR. PETER KAISER: I agree with Allen. I would get a lot of information from the prior retina specialist about previous responses to anti-VEGF. If there was at least some response in the past, then it's probably worth trying 1 injection to evaluate the clinical response. One question I have about this case is whether the patient gave up on therapy because of the central atrophy and wasn't getting a visual improvement. When the fluid came back after treatment was stopped, the visual field and quality of life were much worse. Obviously, this patient has had a great outcome with anti-VEGF therapy, but this is not the case in all late AMD cases.

DR. RICHARD KAISER. This patient would never qualify for a clinical trial, and it's tough to measure the benefit of these therapies in a near-end stage case like this one. But for this patient, going from counting fingers vision to 20/200 was life-changing.

Management of Suboptimal Responses and Extending the Dosing Interval (Cases provided by Dr. Jonathan Prenner)

DR. PRENNER: I'd like to discuss some patients with partial or suboptimal responses. The first is a monocular, 75-year-old widow, who is very independent, and has a visual acuity 20/50. We treated her with monthly ranibizumab for more than 2 years, and despite visual and anatomic improvement over the first year or so, her progress stalled. She had some residual intraretinal fluid, and some subretinal tissue consistent with a CNV membrane. Despite monthly treatments, she continued to leak on FA and OCT, and we confirmed the absence of IPCV on ICG angiography. So I felt that she was a partial responder to ranibizumab. She looked forward to ongoing developments with aflibercept and the potential for less frequent injections. When available, we treated her with aflibercept and evaluated her at her 1 week post injection for signs of a biologic response. She looked quite good, with functional improvement; she was a 2-hands up, hugging, kissing kind of patient. Five weeks out—she had significant improvement.

I have 2 other cases with a similar response—ie, incomplete or partial response to ranibizumab, who responded well to aflibercept. Thoughts?

DR. HO. Jon discussed a case of partial or incomplete responses to the existing therapy ranibizumab. Before the introduction of aflibercept, for such a patient my strategy would be to go to an every 2-week cycle, alternating between bevacizumab and ranibizumab. We would get a response to that strategy in about 30% of cases. With such a regimen, we get more VEGF inhibition. Currently, for such a patient, I would consider switching to aflibercept, because mechanistically it has both VEGF and potentially relevant PLGF activity, perhaps such a switch can be effective. We've had similar responses as Jon has outlined.

DR. PETER KAISER. Jon presented a type 1 lesion, which is marked by CNV beneath the retinal pigmental epithelium (RPE).34 Usually, such lesions have relatively good visual acuity, a persistent RPE detachment, and require a lot of injections. I would have taken a similar approach as Allen (ie, alternating between bevacizumab and ranibizumab prior to the introduction of aflibercept). These patients generally do well when switched to aflibercept, so I do offer that as an option. However, this wasn't studied in the clinical trials, so we don't have guidance on how or when to do this. I generally bring them back 4 weeks after the change in drugs, then add approximately 2 weeks every time they come back dry (treat-and-extend) and stable after they are reinjected. We don't know the correct way to extend the interval. But I'm cautiously optimistic, because I have seen good responses to aflibercept after suboptimal response to other therapies in some patients.

DR. PRENNER: We've looked carefully at this treatment group—patients with partial or incomplete response to other therapies who then receive aflibercept. About a quarter of eyes have complete resolution of their anatomic abnormalities, a quarter have partially improved anatomy, and 40% remain unchanged. Interestingly about 10% of patients do not do as well anatomically with aflibercept as they did on ranibizumab or bevacizumab. For example, Figure 4 shows results in a monocular woman who had a stable response to ranibizumab (Figure 4a). She had received 32 injections of ranibizumab. But I couldn't extend the interval between doses. When dosed at 5-week intervals, she was reliably dry and 20/20. But if I tried to extend the interval, she leaked. So she was looking forward to the potential benefits of less frequent dosing with aflibercept. So I switched her to aflibercept and gave her 1 dose. When I looked at her at 5 weeks post injection, she had some retinal fluid that she recognized and was 20/25 (Figure 4b). When we switched her back to ranibizumab every 5 weeks, this went away. So I agree with Peter, we are cautiously optimistic about the benefits of aflibercept, but this is a nuanced process.

DR. RICHARD KAISER. We are limited in our ability to understand the biology. We use OCTs, but they show anatomic but not biologic benefits. As a clinician, it's great to have choices, so we can see what works for each patient.

DR. HO: Monthly treatment is beneficial. But it's used in only a minority of patients. Many retinal specialists use as needed, PRN treatment.

DR. PRENNER: Persistent anti-VEGF therapy gives you the best chance of a good visual outcome. It is important to consider the status of the fellow eye. Many 1-eye patients will choose to have monthly treatment in their remaining eye. For many binocular patients, I use a treat–and-extend approach.

DR. RICHARD KAISER: We have level 1 evidence for monthly treatment, and level 1 evidence for treat-andobserve approaches. In clinical practice, we try to extend the interval between treatments-and we don't have evidence for or against that approach.

DR. HO: How do you extend the interval? Visual and anatomic response?

DR. RICHARD KAISER. I monitor them with OCT, evaluate their vision, and I also use an angiogram to confirm the disease is quiescent. Once I am convinced that the disease is stable, then I will extend the intervals between the visits. I will try and keep them quiet with slightly less frequent visits, and then I will try and extend the time intervals. I will treat even when there is no disease. I'll do an angiogram or contact lens exam once they enter this stage to look for subtle forms of activity.

DR. PETER KAISER. Most of us don't want to do what the level 1 evidence tells us to do. We have the HARBOR showing that monthly therapy did meet the non inferiority endpoint compared to as needed treatment. The IVAN 1-year, CATT 1- and 2-year data show that as needed treatment was not as good as monthly. So, following level 1 evidence we should all be using monthly treatment. We don't. While I also use a treat-and-extend approach, we don't have any level 1 to say whether it works. The LUCAS study is studying a treat-and-extend regimen using a level 1 paradigm, but not against monthly therapy. So, I tell my patients that monthly is best. Most do not want monthly treatment, so in those patients I usually move to treat-and-extend. But I do have some patients who stay monthly.

CONCLUDING REMARKS

DR. HO. In 2012, our patients with neovascular AMD have a brighter and more clear future than ever. We have several excellent treatment options from which to choose. We've raised the bar in neovascular AMD—we have “above the line” vision improvements and we hope to have combination approaches that might raise the bar even higher. We still struggle with patient treatment burden, specifically the need to visit a retina specialist frequently to assess the need for injection. We look forward to longer-acting drug delivery platforms to treat neovascular AMD and toward new ways to ward off the development of choroidal neovascularization in the first place.

  1. Williams RA, Brody BL, Thomas RG, Kaplan RM, Brown SI. The psychosocial impact of macular degeneration. Arch Ophthalmol. 1998;116:514-520.
  2. Friedman DS, O'Colmain BJ, Muñoz B, et al.; Eye Diseases Prevalence Research Group. Prevalence of age-related macular degeneration in the United States. Arch Ophthalmol. 2004;122:564-572.
  3. American Society of Retina Specialists Annual Preferences and Trends Survey. Available at http://www.asrs.org.
  4. Chakravarthy U, Wong TY, Fletcher A, Piault E, Evans C, Zlateva G, Buggage R, Pleil A, Mitchell P. Clinical risk factors for age-related macular degeneration: a systematic review and meta-analysis. BMC Ophthalmol. 2010;10:31.
  5. Age-Related Eye Disease Study Research Group. A randomized, placebo-controlled clinical trial of high-dose supplementation with vitamins C and E, beta carotene and zinc for age-related macular degeneration and vision loss: AREDS Report No. 8. Arch Ophthalmol. 2001;119:1417-1436.
  6. Klein R, Klein BE, Jensen SC, Meuer SM. The five-year incidence and progression of age-related maculopathy: the Beaver Dam Eye Study. Ophthalmology. 1997;104:7-21.
  7. National Eye Institute. Home Vision Monitoring in AREDS2 for Progression to Neovascular Age-Related Macular Degeneration. http://clinicaltrials.gov/ct2/show/NCT01314430?term=home+screening+and+AMD&rank=1. Accessed May 15, 2012.
  8. MyVisionTrack. http://myvisiontrack.com/myvisiontrack/. Accessed May 15, 2012.
  9. Gess AJ, Fung AE, Rodriquez JG. Imaging in neovascular age-related macular degeneration. Semin Ophthalmol. 2011;26:225-233.
  10. Ciardella AP, Donsoff IM, Yannuzzi LA. Polypoidal choroidal vasculopathy. Ophthalmol Clin North Am. 2002;15:537-554.
  11. Ohr M, Kaiser PK. Intravitreal aflibercept injection for neovascular (wet) age-related macular degeneration. Expert Opin Pharmacother. 2012;13:585-591.
  12. Gragoudas ES, Adamis AP, Cunningham ET Jr, Feinsod M, Guyer DR; VEGF Inhibition Study in Ocular Neovascularization Clinical Trial Group. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med. 2004;351:2805-2816.
  13. Magdelaine-Beuzelin C, Paintaud G, Watier H. Therapeutic antibodies in ophthalmology: Old is new again. MAbs. 2010;2:176–180.
  14. Mordenti J, Cuthbertson RA, Ferrara N, et al. Comparisons of the intraocular tissue distribution, pharmacokinetics, and safety of 125I-labeled full-length and Fab antibodies in rhesus monkeys following intravitreal administration. Toxicol Pathol. 1999;27:536-544.
  15. Rakic JM, Lambert V, Devy L, et al. Placental growth factor, a member of the VEGF family, contributes to the development of choroidal neovascularization. Invest Ophthalmol Vis Sci. 2003;44:3186-3193.
  16. Stewart MW, Rosenfeld PJ. Predicted biological activity of intravitreal VEGF Trap. Br J Ophthalmol. 2008;92:667- 668.
  17. Heier JS, Boyer D, Nguyen QD, et al. The 1-year results of CLEAR-IT 2, a phase 2 study of vascular endothelial growth factor trap-eye dosed as-needed after 12-week fixed dosing. Ophthalmology. 2011;118:1098-1106.
  18. Martin DF, Maguire MG, Fine SL, et al.; Comparison of Age-related Macular Degeneration Treatments Trials (CATT) Research Group Writing Committee:. Ranibizumab and bevacizumab for treatment of neovascular agerelated macular degeneration: two-year results. Ophthalmology. 2012 May 1. [Epub ahead of print] PubMed PMID: 22555112.
  19. Heier JS. Vascular endothelial growth factor (VEGF) trap-eye, 1-year results. Paper presented at: The American Society of Retinal Specialists; August 20-24, 2011; Boston, MA.
  20. Heier JS, VIEW1 and VIEW2 Investigators. 96 weeks results from the VIEW 1 and VIEW 2 studies: intravitreal aflibercept injection versus ranibizumab for neovascular AMD shows sustained improvements in visual acuity. Presented at the Annual Meeting of the Association for Research in Vision and Ophthalmology; May 10, 2012; Fort Lauderdale, FL.
  21. Ho AC, Busbee B, Kaiser PK, Brown DM, Heier J and the HARBOR Study Group. HARBOR study: 1-year results of efficacy and safety of 2.0 mg versus 0.5 mg ranibizumab in patients with subfoveal choroidal neovascularization secondary to age-related macular degeneration. Presented at the American Academy of Ophthalmology 2011 Retina Subspecialty Day, October 21-22, 2011, Orlando, FL.
  22. Genentech. Dear Healthcare Provider Letter. Lucentis Sailor Letter. Released January 24, 2007.
  23. Brown DM, Kaiser PK, Michels M, et al; ANCHOR Study Group. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006;355:1432-1444.
  24. Rosenfeld PJ, Brown DM, Heier JS, et al; MARINA Study Group. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006;355:1419-1431.
  25. Regillo CD, Brown DM, Abraham P, Yue H, Ianchulev T, Schneider S, Shams N. Randomized, double-masked, sham-controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER Study year 1. Am J Ophthalmol. 2008;145:239-248.
  26. Schmidt-Erfurth U, Eldem B, Guymer R, et al; EXCITE Study Group. Efficacy and safety of monthly versus quarterly ranibizumab treatment in neovascular age-related macular degeneration: the EXCITE study. Ophthalmology. 2011;118:831-839.
  27. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An optical coherence tomography-guided, variable dosing regimen with intravitreal ranibizumab (Lucentis) for neovascular age-related macular degeneration. Am J Ophthalmology. 2007;143:566-583.
  28. CATT Research Group, Martin DF, Maguire MG, Ying GS, Grunwald JE, Fine SL, Jaffe GJ. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;364:1897-1908.
  29. Martin DF, Maguire MG, Fine SL, et al.; Comparison of Age-related Macular Degeneration Treatments Trials (CATT) Research Group Writing Committee. Ranibizumab and bevacizumab for treatment of neovascular agerelated macular degeneration: two-year results. Ophthalmology. 2012 May 1. [Epub ahead of print] PubMed PMID: 22555112.
  30. Food and Drug Administration.http://www.fda.gov/Drugs/DrugSafety/ucm270296.htm. Accessed June 8, 2012.
  31. Lally DR, Gerstenblith AT, Regillo CD. Preferred therapies for neovascular age-related macular degeneration. Curr Opin Ophthalmol. 2012;23:182-188.
  32. Brechner RJ, Rosenfeld PJ, Babish JD, et al. Pharmacotherapy for neovascular age-related macular degeneration: an analysis of the 100% 2008 Medicare fee for-service Part B claims file. Am J Ophthalmol. 2011;151:887-895.
  33. Gower EW, Cassard S, Chu L, et al. Adverse event rates following intravitreal injection of Avastin or Lucentis for treating age-related macular degeneration. Paper presented at: Annual Meeting of the Association for Research in Vision and Ophthalmology; Fort Lauderdale, FL; May 3, 2011.
  34. Gass JD. Stereoscopic Atlas of Macular Diseases. 4th ed. CV Mosby: St Louis, 1997; 26–30.