Dry AMD research might hold key to unexplained vision loss

Loss of visual acuity could be related to ongoing disease progression that anti-VEGF agents cannot touch.

ByCarmen A. Puliafito, MD, MBA

Carmen A. Puliafito

It is interesting to see how the insights from our clinical trials can help inform our consideration of things that are unrelated to what the clinical trial set out to study. As an example, I would like to talk about a case from the PrONTO trial.

We all have patients who are treated with Avastin (bevacizumab, Genentech) or Lucentis (ranibizumab, Genentech) who have experienced drying of the retina and restoration of normal anatomy who subsequently go on to lose vision.

Phillip A. Rosenfeld, MD, worked with Genentech and tried to analyze why these patients lose vision. It is crucial to understand which lesion characteristics have a statistically significant association with visual acuity loss. We are beginning to understand that dry AMD plays a more important role than we might have thought in vision loss.

Vision loss and geographic atrophy

The percent of the lesion with classic choroidal neovascularization (CNV) was not relevant. Nor was the CNV leakage area, total area of blood, subretinal fibrotic tissue or serous retinal pigment epithelium detachment.

In looking at what element of the lesion area is relevant, the only factor that was associated in a photographic, angiographic way was geographic atrophy.

We had a patient from the PrONTO trial who was followed for 24 months. The patient entered with an entry criteria of 20/63. At 3 months, the patient was 20/30, so it would seem this person responded to therapy. Yet at 6 months the patient was 20/50, at 12 months, 20/40, and at 24 months, 20/100. The patient had a 10-letter loss, going from 61 to 51 letters on the ETDRS chart.

In this patient, the anatomy is quite disorganized. There you see pigment epithelial detachment (PED), subretinal fluid, intraretinal fluid, all of which are associated with vision loss in macular degeneration.

Of course the wonderful thing about the PrONTO trial is that we looked at patients in ways that we would not normally look at them in clinical practice. We looked at this patient 1, 2, 4 and 7 days after treatment.

There was not a lot of change after day 1. By day 2, the intraretinal fluid was going down a little bit, but there was still not a lot of change.

By day 7, the intraretinal fluid was down, but the PED was still there. We know that PEDs are the hardest things to change anatomically, and after a week, not a lot had happened. After 14 days, the patient’s vision had improved by four letters, but the PED was still there and there was a small amount of fluid.

For the first 18 months, it seemed as though the patient was improving and holding steady. Then, at month 19, vision declined to 20/80. Even though the anatomy looked good, the patient was re-treated.

That is part of our treatment paradigm: If the patient has been a success and the vision goes back down, even if the optical coherence tomography (OCT) looks absolutely fine, we re-treat those patients because there may be fluid there that we cannot detect.

In following this patient, it was eventually noted that there was a difference on OCT – an area of hyper-reflectivity. The patient had developed geographic atrophy, and it was visible on color photographs, too.

Progression of dry AMD

What we know now from MARINA, ANCHOR and PrONTO is that most failures do not result from continued growth and leakage of the CNV. Our problem is that in treating neovascular AMD with anti-VEGF agents, we are dealing with one side of the disease pathogenesis equation and leaving the other side out because loss of visual acuity results from progression of the dry AMD.

You might use the word “underlying” dry AMD, but of course, many patients develop wet AMD without a lot of dry AMD. There is not a logical progression. Some of those patients go on to develop geographic atrophy, and some of the patients develop wet AMD.

Why do some high-risk eyes develop geographic atrophy and other high-risk eyes develop neovascular AMD? We do not know, but what we do know is that ranibizumab or bevacizumab can alter the disease progression.

You can start with wet AMD, which we can treat with anatomic success, and then it can be converted back into high-risk dry AMD, associated with visual loss from the drusen and destruction of the photoreceptors or geographic atrophy.

I do not think this is related to the treatment with the anti-VEGF agents, but rather to ongoing disease progression that anti-leakage agents do not touch. This is obviously a fruitful area for investigation. If you want to know what the follow-up is from all these anti-VEGF studies, it is exactly this: Why do some patients do well and some not?

Future strategies to improve outcomes

There are a couple of ways to approach the study of dry AMD. First of all, we can take patients who are 20/40 and follow them for 10 years.

Or we can take those patients we are treating for wet AMD with ranibizumab or bevacizumab and treat them with something else to see whether or not we can actually halt this disease progression in eyes that are already being treated.

Combination trials are difficult to design, but I suggest that what you will probably see are studies in which you use an anti-VEGF agent in combination with another agent, such as the ciliary neurotrophic factor implant NT-501 (Neurotech Pharmaceuticals), the oral fenretinide compound ST-602 (Sirion Therapeutics) or the topical treatment OT-551 (Othera Pharmaceuticals) for dry AMD.

The take-home message is that, when we are seeing and treating so many patients with wet macular degeneration and they want to know why they are not improving, and we want to know why they are not improving, the answer is that they probably have dry AMD again after successful treatment of wet AMD.

For more information:

Carmen A. Puliafito, MD, MBA, can be reached at Office of the Dean, Keck School of Medicine, University of Southern California, 1975 Zonal Ave., KAM 500, Los Angeles, CA 90033; 323-442-1900; e-mail: cpuliafito@usc.edu. Dr. Puliafito has no direct financial interest in the products mentioned in this article.