Article Date: 10/1/2007

Initiating Therapy for Neovascular AMD

Initiating Therapy for Neovascular AMD

Philip J. Rosenfeld, MD, PhD (moderator): As we manage our patients who have neovascular age-related macular degeneration (AMD), one of our goals is to ensure that our treatments leave them with the best possible visual acuity. This discussion centers on the strategies used by vitreoretinal specialists in pursuit of that goal.

We begin with the initial patient encounter. Panel members, after the dilated examination with funduscopy, how do you evaluate patients who present with neovascular AMD?

Rajendra S. Apte, MD, PhD: My test of choice at baseline is fluorescein angiography (FA). For follow-up and evaluating response to treatment, I use optical coherence tomography (OCT) extensively. I do not use autofluorescence or microperimetry often in the management of neovascular AMD.

David M. Brown, MD: I perform FA and OCT. I think a vascular endothelial growth factor (VEGF) inhibitor has a better chance of working if OCT shows a substantial amount of fluid. Also, given the risk of retinal pigment epithelium (RPE) tears with anti-VEGF therapy, I like to evaluate the extent of pigment epithelial detachment (PED).

I have found autofluorescence more helpful in managing dry macular degeneration. I consider microperimetry to be primarily a research tool.

Elias Reichel, MD: I perform FA and OCT at baseline and then monitor the patient's response to treatment primarily with OCT.

THE ROLE OF FLUORESCEIN ANGIOGRAPHY

Dr. Rosenfeld: Why do you perform FAs?
Dr. Reichel: FAs are useful tools for prognosis. For example, in general, certain types of lesions, such as large serous PEDs, can be at high risk for developing RPE tears. Also, a retinal angiomatous proliferation (RAP) lesion may require more treatments compared with a small occult lesion. While these indicators are not exact, they can be useful guides for us and for patients.


Fluorescein angiography also can help to determine why a patient who was expected to do well with treatment suddenly is not doing well.

— Elias Reichel, MD

FA also can help to determine why a patient who was expected to do well with treatment suddenly is not doing well.

Dr. Apte: In subanalyses from the PIER trial of ranibizumab (Lucentis, Genentech), FA was predictive much earlier than OCT as to which patients were going to do better.1 Therefore, perhaps in this era of prn treatment, FA can add to our knowledge in cases where we would like to increase the interval between treatments because the retina appears to be dry on OCT.

Seenu M. Hariprasad, MD: At baseline, FA gives me a good geographic overview of what is going on in the macula. For example, I had a patient who exhibited subretinal and intraretinal fluid on OCT, but FA showed that the lesion was extrafoveal. OCT alone would not have given me that information. The patient responded to 2 treatments — first with ranibizumab and later with MPS laser. The patient is doing well and has been stable for 9 months.

Dr. Rosenfeld: I find FA particularly useful for ruling out conditions that masquerade as neovascular AMD, such as chronic recurrent central serous retinopathy and cuticular drusen with associated leakage. When would you repeat FA?


Perhaps in this era of prn treatment, fluorescein angiography can add to our knowledge in cases where we would like to increase the interval between treatments because the retina appears to be dry on OCT.

— Rajendra S. Apte, MD, PhD

Dr. Hariprasad: I repeat the test in patients who are not responding as expected to anti-VEGF therapy. Something not revealed by OCT, such as an RPE tear, could be revealed.

TREATMENT FOR EXTRAFOVEAL LESIONS

Dr. Rosenfeld: When FA shows an extrafoveal lesion, do you manage that type of lesion differently?

Dr. Brown: I treat truly extrafoveal lesions with thermal laser.

Dr. Apte: I agree, but I would qualify that with size. For a large extrafoveal lesion, I might first shrink it with an anti-VEGF agent and then apply thermal laser.

Dr. Hariprasad: I seriously consider MPS thermal laser for extrafoveal lesions. It is a finite treatment for many patients.

Dr. Rosenfeld: I would use an anti-VEGF agent against an extrafoveal lesion of any size. In my experience, extrafoveal lesions respond with only 1 or 2 treatments; there's less chance of recurrence, and the patient doesn't have a scotoma in their central field. Even with small extrafoveal lesions, recurrence after laser is common and we can end up chasing the lesion as it enlarges, so why not preserve as much retina as possible and start with anti-VEGF therapy.

VISUAL ACUITY AND THE DECISION TO TREAT

Dr. Rosenfeld: Suppose a patient had 20/40 visual acuity for 6 months while OCT clearly showed subretinal fluid and FA showed an occult lesion. Would you begin treatment?

Dr. Reichel: If the patient had been stable for 6 months and has good vision in the fellow eye, I would probably follow him.

If vision were fairly poor in the fellow eye, I would most likely treat.

Dr. Rosenfeld: If vision in the fellow eye was good, what would be the downside to treating?

Dr. Reichel: I would not really know what I was treating or how often I would have to treat. I would wonder whether I was treating subretinal fluid or an ongoing disease process.

Dr. Apte: Given the results we can expect with anti-VEGF therapy and the possibility that the eye may get worse with time, I tend to treat this type of patient. The subretinal fluid is not normal and is in the context of macular degeneration. Perhaps it is not aggressive, but a disease process exists and has the potential of getting worse.

Dr. Brown: I rely on the patient's symptoms. As Sir William Osler said, it is difficult to make an asymptomatic patient feel better. Many chronic occult lesions require a great deal of anti-VEGF therapy to remain dry. If you take an asymptomatic patient and make him dependent on coming to your office every month, he will not be happy.

Dr. Hariprasad: This would be a good time to discuss alternate dosing regimens with the patient. Perhaps he does not need monthly ranibizumab injections. I would consider PIER quarterly dosing in this patient because this dosing regimen has demonstrated vision stabilization in 90% of patients. I consider using this dosing regimen in some patients with 20/40 or better vision. Additionally, I believe that the status of the fellow eye does matter.

Dr. Rosenfeld: The consensus, then, is that we must consider the whole picture in deciding to treat: whether the patient has symptoms, fellow eye status and the patient's ability to follow up and self-monitor for changes in vision.

CHOICE OF FIRST-LINE THERAPY

Dr. Rosenfeld: How do you decide what therapy to use in a treatment-naïve patient?

Dr. Brown: Treatment-naïve patients in our practice primarily get ranibizumab. It is the standard of care; it is well-covered by insurance in my area; and I do not have to talk with patients about off-label use issues. Typically, we reserve other agents or combination therapy for patients with extenuating circumstances, such as visiting from another country, or for those who do not demonstrate a satisfactory response to monotherapy with ranibizumab.


At baseline, fluorescein angiography gives me a good geographic overview of what is going on in the macula.

— Seenu M. Hariprasad, MD

Dr. Rosenfeld: What about patients who cannot follow up on a regular basis?

Dr. Brown: I am likely to talk with them about more finite treatments, such as photodynamic therapy (PDT) with verteporfin (Visudyne, Novartis Ophthalmics) or thermal laser.

Dr. Rosenfeld: What is your treatment of choice for someone who cannot come back for 6 months?

Dr. Brown: This is common in our practice becausewe have patients who come from Central America. I administer triple therapy: a pan-VEGF agent, half or 40% fluence PDT, and a steroid — either dexamethasone or triamcinolone. We do this in 1 day or over the course of 2 days if they are spending the night in town.

Dr. Reichel: I primarily use ranibizumab.

Dr. Apte: Ranibizumab is my agent of choice, although I have used off-label bevacizumab (Avastin, Genentech) when I did not feel ranibizumab was effective after 3 or 4 injections or vice versa.

Dr. Hariprasad: In our practice, we decided to offer ranibizumab to all of our neovascular AMD patients. We feel it is ethical to recommend an FDA-approved treatment first to all patients regardless of insurance status. Undoubtedly, we do think about cost. These are expensive treatments, and we have used the Genentech patient assistance program with great success for numerous patients who do not have insurance or a secondary carrier.

Dr. Rosenfeld: In our practice right now, we use ranibizumab and bevacizumab interchangeably. We just finished analyzing our outcomes retrospectively and found no statistical difference between the medications. We won't know how these drugs compare until a prospective comparative study is performed, but we know enough to have an intelligent discussion with our patients and lay out what we know and don't know about the safety, efficacy and economic impact of these treatments. Therefore, we feel very comfortable moving between the drugs, depending on the patient's financial situation, their understanding of the drugs, their prior experience with either drug and their desire to take cost of therapy into consideration.

Dr. Hariprasad: Before the FDA approved ranibizumab, we frequently used bevacizumab with excellent efficacy.

SIDE EFFECTS, INFORMED CONSENT

Dr. Rosenfeld: What do you discuss with patients regarding the potential side effects of anti-VEGF therapy?

Dr. Reichel: I explain to patients why I use the 0.3-mg dose of ranibizumab rather than the 0.5-mg dose. I tell them that according to some very preliminary data (from the Phase 3b SAILOR study) a risk of stroke may be associated with the higher dose. (See "SAILOR Study.") Therefore, I have chosen to use the lower dose.

SAILOR Study
In November 2005, Genentech began enrollment in the Phase 3b SAILOR study to make ranibizumab (Lucentis, Genentech) available to eligible patients. SAILOR (Safety Assessment of Intravitreal Lucentis fOR AMD) includes patients with all subtypes of new or recurrent active subfoveal neovascular AMD. It is a 1-year study designed to evaluate the safety of 2 different doses (0.3 mg and 0.5 mg) of ranibizumab administered once a month for 3 months and thereafter as needed, based on retreatment criteria. The study will enroll up to 5000 patients at more than 100 sites in the United States, but only about 2400 patients will be followed for 1 year.

Reference
  1. Genentech letter to healthcare providers providing latest data from the SAILOR trial. Issued Jan. 24, 2007.

Dr. Rosenfeld: It is important to note that a risk of stroke exists for both doses, and outcomes with the 0.3-mg dose have not been as good as the outcomes seen with the 0.5-mg dose.

These interim SAILOR results are far too preliminary to draw any conclusions yet, and the FDA hasn't recommended any change in the label or the clinical practice recommendations. Even if we assume these preliminary results hold up at 1 year, do you think there really is a difference in terms of potential side effects between selective VEGF blockade treatments, ie, pegaptanib sodium (Macugen, [OSI] Eyetech) and nonselective VEGF blockade (ie, ranibizumab and bevacizumab)?


Many chronic occult lesions require a great deal of anti-VEGF therapy to remain dry. If you take an asymptomatic patient and make him dependent on coming to your office every month, he will not be happy.

— David M. Brown, MD

Dr. Reichel: I do not think we have enough data to make that determination. However, in looking just within the ranibizumab data, there is something suggestive of risk with the 0.5-mg dose. Whether it is for thromboembolic or hemorrhagic events, we may be seeing a trend.

Dr. Rosenfeld: What dose of ranibizumab is the rest of the panel using?

Dr. Brown: We are treating with the 0.5-mg dose. The phase 3 ranibizumab data did not reveal a safety signal. I agree that the SAILOR study2 preliminary data is concerning, but we will see what the 1-year data show.

In terms of selective vs. nonselective VEGF blockade, selective agents do not seem to work very well anatomically, so it is a moot point. The only time I emphasize the possible risk for thromboembolic events is when a patient has had a previous stroke. I tell the patient that the potential risk exists, and we will try to minimize it by using the least possible number of injections. Then, it is up to the patient and his family. I have many patients who are willing to take the risk if their vision is declining. Other patients would rather have PDT.

Dr. Apte: I use the 0.5-mg dose of ranibizumab in all patients. I am waiting for the final outcome of the SAILOR study before deciding whether I would want to switch to the 0.3-mg dose. Based on the animal and in vitro data, there may be a theoretical difference in risk between the selective and nonselective anti-VEGF agents. But none of the clinical trials to date has shown a significant difference in systemic safety profile. We will have to wait for the final SAILOR analysis.


The consensus is that we must consider the whole picture in deciding to treat: fellow eye status and the patient's ability to follow up and self-monitor for changes in vision.

— Philip J. Rosenfeld, MD, PhD

Dr. Hariprasad: The MARINA and ANCHOR trials of ranibizumab may have included a different population of patients compared with the VISION trials of pegaptanib in terms of preexisting heart disease. We have to compare apples to apples before concluding that nonselective anti-VEGF agents are less safe than selective agents.


The only time I emphasize the possible risk for thromboembolic events is when a patient has had a previous stroke. I tell the patient that the potential risk exists, and we will try to minimize it by using the least possible number of injections.

— David M. Brown, MD

At this time, I treat with the 0.5-mg dose of ranibizumab. I am reserving my final decision on lowering the dose until I see the 1-year SAILOR study data. If a patient is sick, having a history of stroke in the past 6 months or a recent myocardial infarction, I engage in a more careful discussion than I do with a healthy elderly patient who needs ranibizumab treatment.

Dr. Rosenfeld: Of course, the SAILOR study included more than 2400 patients in the group responsible for these preliminary results, and pegaptanib, to my knowledge, has never been studied in that number of patients, so the pegaptanib studies were never powered to pick up a 1% increase in strokes. Since patients with prior strokes were excluded from the pegaptanib trials, we really don't know if pegaptanib is any safer. It is also important to note that the rate of thromboembolic events in VISION was 6%. RP

Reference
  1. Kaiser PK. Subgroups analyses of one-year results of the PIER study of ranibizumab in neovascular AMD. Presented Nov. 13, 2006 during the annual meeting of the American Academy of Ophthalmology in Las Vega, Nevada.
  2. Genentech letter to healthcare providers providing latest data from the SAILOR trial. Issued Jan. 24, 2007.


Retinal Physician, Issue: October 2007