Age-related macular degeneration: current treatment and future options

Risk factor reduction

There are a number of risk factors associated with AMD progression, which may be addressed by disease reduction strategies. These risk factors include smoking, sunlight exposure and a low-antioxidant diet. Cigarette smoking is a well-established risk factor for the development of AMD [Thornton et al. 2005] and cessation is advised as there is a dose–response relationship with pack-years of smoking [Chakravarthy et al. 2007; Khan et al. 2006]. Wearing UV light protection is advisable from an early age to limit exposure of the macula to light of blue wavelengths [Ham et al. 1976]. The retina may be vulnerable to oxidative stress if the diet is low in antioxidants, as the ‘free radical’ theory of ageing proposes that oxygen radicals damage cells over time [Finkel and Holbrook, 2000]. Thus, eating a low-fat, balanced diet full of antioxidants, zinc and vitamin D is advised.

Late AMD

There are a number of treatment options for late AMD but the recommended treatment depends on the exact type, location and extent of AMD. In the 1980s and 1990s laser-based therapies were the mainstay of treatment. The Macular Photocoagulation Study Group (MPSG) used focal argon laser photocoagulation to destroy the neovascular complex with heavy confluent burns to try to prevent CNV enlargement or leakage, but recurrences were common [Macular Photocoagulation Study Group, 1991]. Photodynamic therapy (PDT) used with a photosensitive dye (verteporfin) to treat leaking vessels prevented moderate visual loss in predominantly classic lesions that involved the fovea [Bressler et al. 2002]. However, neither of these treatments were particularly successful in improving visual outcome.

Improved treatments have been developed using inhibitors against vascular endothelial growth factor (VEGF) as monotherapy or in combination with other treatments. VEGF increases vascular permeability and affects multiple components in angiogenesis that promote new vessel formation [Ferrara and Davis-Smyth, 1997]. VEGF also has important functions in vascular pathophysiology: in the process of atherothrombosis it induces collateral circulation to protect against ischaemia in areas of vessel narrowing [Simons, 2005], and inside atherosclerotic plaques it promotes microvessel formation [Juan-Babot et al. 2003], which may disrupt vessels and cause intraplaque haemorrhage and disease progression.

Anti-VEGF monotherapies

Several inhibitors of VEGF have been investigated in the treatment of nvAMD and have been unequivocally shown to reduce exudative manifestations with significant beneficial effects on the maintenance of visual function [Rosenfeld et al. 2006; Gragoudas et al. 2004]. These drugs are injected into the vitreous cavity under local anaesthesia, where they reduce exudative manifestations in the macula. The following anti-angiogenic therapies are currently in use:

ranibizumab (Lucentis®, Genentech), a humanized Fab fragment of a monoclonal antibody that binds to and inhibits the action of all isoforms of VEGF-A;

At present, the treatment of choice for any subfoveal CNV is ranibizumab 0.5 mg [Royal College of Ophthalmologists, 2009]. However, this treatment will only improve vision in a third of patients and some 10% will not respond to therapy. The criteria for continuation of treatment at 4-weekly intervals after the three initial loading doses of ranibizumab are:

persistent evidence of lesion activity;

However, despite being administered as an intraocular injection, these drugs may be detected in circulating blood [Gaudreault et al. 2005]. As VEGF has a role in thrombus formation and in the pathophysiology of atherosclerosis, the possibility of adverse cardiovascular effects similar to those reported for systemic anti-VEGF therapy in oncology, such as thrombosis, haemorrhage and hypertension, must be considered [Tunón et al. 2009]. The UK’s Royal College of Ophthalmologists has recommended guidelines on temporarily discontinuing anti-VEGF treatment if a thromboembolic phenomenon, including myocardial infarction or cardiovascular accident, has occurred in the preceding 3 months, or if recurrent thromboembolic phenomena occur which are thought to be related to treatment with ranibizumab [Royal College of Ophthalmologists, 2009].

In the Minimally classic/occult trial of the Anti-VEGF antibody Ranibizumab in the treatment of Neovascular Age-related macular degeneration (MARINA) study, 716 patients with AMD received monthly intravitreal injections of 0.3 or 0.5 mg ranibizumab or a sham injection for 24 months [Rosenfeld et al. 2006]. In the ANti-VEGF antibody for the treatment of predominantly classic CHORoidal neovascularization in age-related macular degeneration (ANCHOR) study, 423 patients with AMD were randomized to receive monthly ranibizumab (0.3 or 0.5 mg) plus sham verteporfin therapy or sham injections plus active verteporfin therapy for 2 years [Brown et al. 2006]. Patients with previous cardiovascular disorders were not excluded from these trials, and no significant difference in the incidence of cardiovascular events was found. However, the sample sizes in these trials are insufficient to permit valid conclusions to be made about the presence or absence of such serious adverse events.

Comparison of anti-VEGF therapies, refinement of treatment strategies and combination therapies

Further refinement of the current treatment regime is needed through research on the optimal treatment frequency and duration and the type of anti-VEGF drug. Studies to compare ranibizumab and bevacizumab in the UK are the randomised controlled trial of alternative treatments to Inhibit VEGF in Age-related choroidal Neovascularisation (IVAN) study and the in USA the Comparison of Age-related Macular Degeneration Treatments Trials: Lucentis-Avastin Trial (CATT) study are ongoing.

There are a number of other agents previously studied, including PDT and steroids, that may be used in combination with any of the above treatments.

Combination therapies

Anti-VEGF monotherapy regimes merely reduce the exudative manifestations and do not cause permanent regression of new vessels in patients who are commenced on these treatments. They require monthly monitoring and treatment for many years. Thus, the burden on scant and stretched clinical resources grows inexorably. With this in mind, combination approaches are being considered using therapies that cause new vessel regression, such as photodynamic therapy and ionizing radiation.

Non-nvAMD

Preventative strategies for non-nvAMD are under investigation that target complement regulation as this is disrupted in AMD. Copaxone® (glatiramer acetate) is an anti-inflammatory agent given by subcutaneous injection that alters immunomodulatory gene expression and is being used in early AMD to reduce the size and extent of drusen [Arnon and Aharoni, 2009]. POT-4 is a derivative of the cyclic peptide Compstatin undergoing phase I testing and is directed against C3, which is a central component of all major complement activation pathways [Rakic, 2003]. ARC1905 is also undergoing study and is an aptamer that inhibits the pro-inflammatory C5 [Usui et al. 2004].

The use of topically administered AL-8309B (Alcon Laboratories) for the treatment of AMD-related geographical atrophy is currently under investigation in the double-masked, randomized phase III GATE trial [Singerman, 2009].

nvAMD

At present, anti-VEGF therapy is the best available treatment for nvAMD but is limited by the need for repeated intravitreal injections and its associated risks and burden on workload. Different methods of VEGF inhibition are being developed that target various stages in the signalling cascade that regulates VEGF production.

Upstream inhibitors

Inhibition of upstream factors in the signalling cascade that leads to angiogenesis includes a protein kinase called mTOR that regulates cell growth and proliferation. The mammalian target of rapamycin (mTOR) activates hypoxia-inducible factor 1 a (HIF-1 a), which causes transcription of multiple genes, including those that produce VEGF [Ma and Blenis, 2009]. Sirolimus (Rapamycin®, Macusight and Santen) is a broad-acting mTOR inhibitor given by the subconjunctival or intravitreal route under investigation in the EMERALD (Phase II Study of an Ocular Sirolimus (Rapamycin) Formulation in Combination With Lucentis® in Patients With Age-Related Macular Degeneration) trial [ClinicalTrials.gov identifier: NCT00766337].

Downstream inhibitors

Targets in the angiogenesis cycle include tyrosine kinase receptors that are expressed in CNVs, such as topical TG100801 (Targegen) [ClinicalTrials.gov identifier: NCT00414999], AL39324 (Alcon) (in phase II testing, WALTZ trial) [ClinicalTrials.gov identifier: NCT00992563], oral vatalanib [ClinicalTrials.gov identifier: NCT00138632] and topical ATG-3 (Athenagen), which is a 7-nicotinic acetylcholine receptor antagonist.

Transmembrane proteins called integrins are essential in angiogenesis during endothelial cell migration and macrophage recruitment [Avraamides et al. 2008]. Phase I trials have been conducted with JM6427, which is a specific integrin antagonist [ClinicalTrials.gov identifier: NCT00536016], and volociximab, which is a monoclonal antibody that block binds integrin [ClinicalTrials.gov identifier: NCT00782093].

Platelet-derived growth factor (PDGF) acts as a mitogen for pericytes and is critical for blood vessel maintenance. E10030 (Ophthotech) is an anti-PDGF agent that, when combined with anti-VEGF, causes neovascular regression [ClinicalTrials.gov identifier: NCT01089517].

VEGF Trap

VEGF Trap is a VEGF–receptor fusion protein that binds all forms of VEGF-A. The Clinical Evaluation of Anti-Angiogenesis in the Retina Intravitreal Trial (CLEAR-IT) of VEGF Trap-Eye in patients with nvAMD found a mean gain in best corrected visual acuity (BCVA) from baseline of 7.3 letters at 3 months, 8.4 letters at 1 year and 7.1 letters at month 6 of the extension study (n = 117). Over the 15-month course of the “as required” Pro re nata (PRN) dosing phase, from month 3 of the original study to month 6 of the extension phase, patients received a mean of 3.5 injections of VEGF Trap-Eye. The visual outcome was similar to that with ranibizumab but with significantly reduced frequency of intravitreal injection [Nguyen et al. 2009].

The VEGF Trap-Eye: Investigation of Efficacy and Safety in Wet AMD (VIEW) study evaluated three dose levels of VEGF Trap-Eye in the treatment of nvAMD in the United States (VIEW 1) and globally (VIEW 2). A statistically significant mean improvement in vision of 10.9 letters was found at 52 weeks in the group injected monthly with VEGF Trap-Eye (2 mg) compared with eight letters in the ranibizumab (0.5 mg) group injected monthly. VEGF Trap-Eye (2 mg) injected every 2 months maintained vision at week 52, and was similar in efficacy and safety to ranibizumab dosed monthly. PRN dosing at least every 3 months (but not more often than monthly) is currently being evaluated.

Genes associated with the complement cascade

Several genes that code for proteins involved in the complement cascade significantly increase the risk of AMD (such as variation in complement factor (CFH) and complement component 3) or decrease the risk of AMD (variation in complement component 2 and factor B and deletion of CFH-related genes CFHR1 and CFHR3).

A mutation in complement factor H, a key regulator of the complement pathway, is strongly associated with increased risk of AMD as carriage of the Y402H polymorphism increases between 2- and 7-fold the risk of developing AMD. This gene polymorphism accounts for up to 50% of the population-attributable risk of AMD [Edwards et al. 2005; Haines et al. 2005].

Variation in C3 has been shown to increase the risk of developing AMD up to 2.6-fold [Yates et al. 2007]. A protective deletion of CFHR1 and 24CFHR3 occurs close to the CFH locus in 8% of AMD patients [Hughes et al. 2006] and a protective haplotype, reducing the risk of AMD, is associated with variation in complement factors C2 and factor B [Gold et al. 2006].

Chromosome 10q locus

Single-nucleotide polymorphisms (SNPs) in chromosome 10q26 are associated with an increased risk of AMD. Further functional studies are needed to confirm whether SNPs identified in the coding region of the LOC387715/ARMS2 (age-related maculopathy susceptibility 2) gene [Rivera et al. 2005] or located in the promoter region of HTRA1 (high temperature requirement factor A1) [Edwards et al. 2005; Haineset al. 2005] represent the true genetic risk variant at this locus. A major genetic risk factor for AMD occurs at this 10q26 locus, with an up to 10-fold increased risk of developing AMD [DeWan et al. 2006].

Other major genetic loci

Meta-analysis of whole genome linkage studies identified several significant genetic loci for AMD [Fisher et al. 2005] on chromosomes 1 (CFH locus), 10q, and 1q, 2 p, 3 p and 16, where undiscovered genetic variants are yet to be revealed.

Pharmacogenetic relationships

Some preliminary pharmacogenetic relationships have been reported to exist between genetic risk factors associated with AMD and the response to treatment. These pharmacogenetic associations could be used in the future to create a personalized therapeutic plan whereby patients with different genotypes are offered different treatments. For example, patients with the CFH Y402H CC genotype had a significantly worse prognosis with intravitreal bevacizumab injections compared with those with the CFH TC or TT genotypes [Brantley et al. 2007]. There is also evidence that carriage of CFH or HTRA1 genetic variants can increase the risk of developing CNV or alter the phenotype of CNV [Leveziel et al. 2008; Weger et al. 2007].