In the therapeutic setting, treating CNV is not as simple as targeting the angiogenic cascade or any one of its stages. Three antiangiogenic therapies are now widely available: the VEGF aptamer pegaptanib sodium, and the monoclonal antibodies ranibizumab and bevacizumab. Each of these acts by inhibiting the formation of new blood vessels, and has shown some efficacy in reducing CNV.

Rosenfeld et al. (13) showed that ranibizumab could reverse some of the loss in vision caused by CNV, with VA improving in one-quarter to one-third of patients who received monthly intravitreal injections. Gragoudas et al. (14) showed that loss of VA could be slowed with injections of intravitreal pegaptanib every six weeks, and that approximately 20% of patients had a gain of at least 5 ETDRS letters.

Bevacizumab is also widely used to slow or reverse the progress of CNV, although at present, it has not been approved for this purpose by the Food and Drug Administration in the United States, or by any other licensing authorities worldwide. An attractive feature of bevacizumab is that in small studies, it appeared to reverse CNV and led to an improvement in VA with one to three injections, sustained over several months.

But while anti-VEGF monotherapy appears useful for blocking new blood vessel formation, recent research has suggested
that it also stimulates the release of compensatory proangiogenic factors which counteract its anti-VEGF activity, and induce new blood vessel formation. This compensatory reaction was not seen in laboratory animals that received a combination therapy involving three elements: anti-VEGF, an integrin-antagonist, and an antagonist of vascular-endothelial cadherin-mediated adhesion. It was observed in animals treated with anti-VEGF monotherapy (15).

This means that not only will the effectiveness of anti-VEGF therapies be limited when they are used as monotherapy, but that also, whatever role they do play may be temporary. Anti-VEGF will antagonize the CNV, reduce or stop vessel leakage and alleviate the edema, but this effect lasts only as long as the anti-VEGF itself. Once the anti-VEGF effect has worn off, leakage and edema recur. This will happen because the compensatory action of releasing new VEGF will continue to stimulate neovascular growth, even as the anti-VEGF treatment blocks the formation of abnormal vasculature. Patients could be required to undergo anti-VEGF treatments for an unlimited, and unknowable length of time. Furthermore, in view of the upregulation of proangiogenic factors in response to anti-VEGF treatment, this could suggest a rebound effect, should the patient ever discontinue anti-VEGF therapy (16). Therefore, controlling or antagonizing the upregulation of VEGF in any of its isoforms is now thought to be only one of the processes involved in treating CNV.

A second aspect of treating CNV is the eradication of the existing neovascularization. No matter how effective an antiangiogenic drug may be at preventing the sprouting of new blood vessels in the retina, it does nothing to address the issue of new vessel growth that is already established. Laser photocoagulation therapy has been used for many years in treating some forms of cancer, but its efficacy in treating CNV related to AMD has been limited. On one hand, the procedure uses a precise and focused laser beam, but the CNV is usually somewhat diffuse. The risk of destroying surrounding tissue, especially the photoreceptors lying on top of the neovascularization, is considerable. Also, the procedure can only be used for extrafoveal lesions, not subfoveal or juxtafoveal lesions. Finally, laser photocoagulation generally has a high recurrence rate. PDT was approved as a treatment for CNV in 1999 and offers a complementary action to that seen with anti-VEGF agents. While anti-VEGF drugs arrest, or slow the development of new blood vessels, PDT eradicates the CNV that has already taken place.

PDT entails the intravenous infusion of verteporfin (benzoporphyrin derivative), which is a red laser activated monoacid. The photosensitizer, once exposed to the laser light, releases highly reactive short-lived singlet oxygen and other reactive oxygen intermediates which lead to cellular endothelial damage, which then destroys the abnormal blood vessels. There is also further release of vasoactive and thrombogenic signaling molecules, which cause acute thrombosis of the capillaries, followed by decreased vessel leakage. Contrary to laser photocoagulation, only the vessels which have taken up the verteporfin are affected, and therefore, the surrounding tissue is not harmed.

Numerous studies worldwide have now demonstrated the benefit of PDT therapy in patients with classic subfoveal, predominantly classic, and occult CNV (Table 42-1). Most trials with PDT as monotherapy demonstrate a reduction or slowing in vision loss rather than an improvement in VA. Significant improvement of VA in large number of patients has not been seen in any of the major PDT monotherapy trials. The procedure mainly benefits patients with minimally classic CNV or mixed lesions. Smaller lesions respond better to treatment than larger lesions (Table 42-1).

In recent years, certain drawbacks to PDT monotherapy have been identified. First, the therapy requires additional treatments as new CNV arises. Past studies have reported patients receiving an average of 3.1 treatments over 12 months (17) and an average of 5.6 treatments over 24 months (18). Secondly, a small study (12) in elderly patients who had been treated with PDT showed overexpression of VEGF in the endothelial layer of choriocapillaries as well as within larger vessels. Overexpression of VEGF was not evident in eyes that had not been treated with PDT. The same sites showed positive upregulation of vascular endothelial growth factor receptor 3 (VEGFR-3) and pigment epithelium derived growth factor (PEDF). Thus, like anti-VEGF when used as monotherapy, PDT also stimulates VEGF expression, which could, in turn, lead to more neovascular proliferation. Furthermore, PDT is known to cause inflammation which exacerbates the cycle of VEGF upregulation and can trigger neovascularization.

Finally, an in vitro study has demonstrated that PDT resulted in a morphologically and functionally detectable breakdown of the outer blood-retinal barrier function of the RPE, although there was no damage to the RPE cells. Increasing the concentration of verteporfin however, can result in RPE cell damage (19).