New hypotheses emerge on causes of wet AMD

By Daniel M. Keller

WASHINGTON, DC — Vision scientists have long known of the existence of the wet form of age related macular degeneration (AMD) — the abnormal formation of blood vessels under the retina — but until now its causes have remained a mystery.

Scientists presented new hypotheses on the disease and potential treatments at a Research to Prevent Blindness Foundation seminar.
While choroidal neovascularisation (CNV) under the retina is a central feature of wet AMD, little is known about its origin. The thinking has been that new vessels arise from neighbouring quiescent choriocapillaris when normal blood vessel cells just under the retina begin to grow. But investigator Karl Csaky MD, PhD outlined an alternative hypothesis.

“The choroid is an active tissue where there’s intrachoroidal cellular recruitment and new vessel formation. So abnormal new blood vessels arise from deeper within the choroidal tissue, giving rise to both intrachoroidal and subretinal new vessels. This suggests that now the choroid is playing a crucial role in the formation of this disease,” Dr Csaky said.

Animal experiments indicate that the abnormal new vessels may arise from circulating bone marrow-derived precursor cells. Transplantation of genetically ‘marked’ bone marrow into animals shows that marrow-derived cells are recruited to the site of experimentally induced choroidal neovascularisation.

In the presence of specific growth factors, the cells can grow and form part of the neovascular complex. Clinical samples of CNV have shown over-expression of vascular endothelial growth factor (VEGF), he reported.
Dr Csaky transplanted bone marrow cells containing a marker gene that turns blue or green under the control of a specific endothelial cell promoter into marrow-depleted animals.

Agent	Company	Status
VEGF F(ab) VEGF aptamer PKC antagonist PKC antagonist Anecortave acetate fluocinolone acetonide 2-methoxyestradiol Somatostatin analogue MMP antagonist Integrin antagonist Integrin antagonist	Genentech Eyetech Lilly Novartis Alcon Bausch & Lomb Allergan/Entremed/Oculex Novartis Agouron Ixis, Medimmune Merck, KgaA	Phase II Phase I/II/III Phase II (diabetic retinopathy) Phase II/III Phase II/III Phase III Phase II Phase II Trial halted, side effects Phase I Phase I

TABLE: Anti-angiogenic pipeline at a glance

He showed ‘marked’ blood vessel cells in the choroid of animals with experimentally induced CNV, suggesting differentiation of marrow cells into blood vessel cells.
Clinically, many patients with wet AMD have an intrachoroidal component to their CNV complex. Because of fluorescein’s spectral properties, the usual fluorescein angiography shows only the part of the CNV complex that is above the plane of the choroid and choriocapillaris.

Dr Csaky showed that high speed indocyanine green (HS-ICG) angiography highlights the intrachoroidal component as well, including the choriocapillaris, since this fluorescent dye is activated with infrared light which penetrates the choroid.
Using pinpoint laser photocoagulation, Dr Csaky could temporarily close the intrachoroidal vessels in about 75% of cases, markedly reducing subretinal fluid and fluorescein leakage.

The small laser spots are aimed at the long intrachoroidal vessels which themselves are distant from the centre of vision but supply subretinal vessels under the macula.
Even though repeat treatments are needed, so far there do not appear to be any side effects from these treatments.

Unfortunately, current treatments with thermal laser or photodynamic photocoagulation are usually aimed only at that part of the CNV which can be visualised by fluorescein angiography — a dye which clearly underestimates the size and extent of the CNV lesion.

And re-treatment with both these treatments is often necessary within weeks or months as abnormal blood vessels reappear, he noted.
HS-ICG is very different from standard static ICG angiography. In static ICG, images of the choroid are taken every 15 to 30 seconds for 10 minutes in an attempt to show the CNV.

In high speed ICG angiography, however, a scanning laser illuminates the choroid and pictures are captured at a rate of 12 frames/second for 10 to 20 seconds.
Dr Csaky said that HS-ICG is the only one that can show the filling of normal and abnormal choroidal vessels and even show the direction of filling.

Repeat examinations can monitor therapy. By superimposing the HS-ICG images on the computer with those from fluorescein angiography, Dr Csaky demonstrated that in patients with various forms of AMD, a majority of the new intrachoroidal vessels connect to or ‘feed’ the subretinal neovascular complex as seen on fluorescein angiography.
He noted that choroidal involvement in AMD is not yet well accepted and is still at the proof-of-concept stage.

“Until we can start to target our treatments at the choroid, there’s really not going to be a significant breakthrough.
“But treating a small component in the choroid may still yield benefit, almost completely turning things around from what’s right now being done,” he said.
US ophthalmologist Dwight Cavanagh MD, PhD said Dr Csaky’s animal model is “extremely convincing”, indicating that CNV could proceed in the absence of a break in Bruch’s membrane to cause haemorrhage under the retina.

He indicated that many of the findings in the animal model would be considered “heresy” by many AMD experts but with “enormous implications in potentially redefining all of our concepts about whether dry macular degeneration always precedes wet and whether a break in Bruch’s membrane is always required”.
Dr Cavanagh added: “And the answer to that question on the basis of his data would appear to be ‘No,’ and that is a major new finding.”

Anti-angiogenic agents in the AMD treatment pipeline
A VAST array of molecules can participate in angiogenesis, but ultimately new blood vessel formation involves proliferation, migration and differentiation from existing vessels or endothelium.

Martin Friedlander MD, PhD discussed several pathways involved and potential approaches to inhibiting neo-angiogenesis.
One approach is to take advantage of adhesion receptors which direct the interaction between vascular endothelial cells and their immediate environment.

Lack of signal transduction between these components can lead to apoptosis in the forming vessel. Antagonists targeted here may prevent abnormal vessel formation.
Fragments of angiogenic molecules may down-regulate neovascularisation. Examples are endostatin (a collagen XVIII fragment), angiostatin (from plasminogen) and fragments of matrix metalloproteinases (MMPs) and tRNA synthetases. Small molecule ‘mimetics’ of such anti-angiogenic molecules may mimic their actions.

Small molecule antagonists of MMPs, zinc-requiring matrix-degrading enzymes, which allow cell migration are already in clinical trials. A recent trial of an agent developed by Agouron Pharmaceuticals was shut down because of side effects, according to Dr Friedlander.
Gene therapy approaches are also on the horizon. Genes coding for anti-angiogenic molecules may be introduced into ocular cells, which would then synthesise the compounds locally. Adenoviruses have been ‘re-targeted’ to selectively infect specific cells in the back of the eye and may be good vectors for these genes.

Adult bone marrow-derived stem cells can hone in on developing vessels and deliver anti-angiogenic proteins. Disease targets might include diabetic retinopathy, AMD and inherited retinal degeneration.

Conversely, mouse experiments have shown the ability of endothelial precursor-enriched haematopoietic stem cells to rescue normal vasculature and reconstitute cells in the photoreceptor cell layer. The idea is to stabilise or rescue a degenerating retina, he explained.

Dr Friedlander said major pharmaceutical companies are beginning clinical trials with compounds that may affect angiogenesis, including inhibitors of protein kinase C-beta and VEGF and a potent angiostatic steroid (anecortave acetate from Alcon Laboratories) lacking glucocorticoid activity.
Since the body has redundant pathways for most functions, it may be able to get around any one inhibitor. It may be necessary to have ‘Cocktails’ of inhibitors to the major angiogenic stimulators.

“Giving angiogenesis regulators systemically can be a risky proposition. So I think if you want to treat AMD in elderly individuals with possible cerebral or myocardial ischemia or diabetics who have widespread ischemia, you have to think about local delivery,” Dr Friedlander said.

Top