Developments in the treatment of AMD

Age-related macular degeneration (AMD) is a leading cause of blindness and sight loss in the developed world. It usually leads to loss of central vision, resulting in patients able to navigate obstacles but unable to recognise faces, read, or drive. It is often classified as dry (atopic) or wet (neovascular) forms, although these are not always distinct entities and the dry can progress to the more aggressive and damaging wet form.

With wet or neovascular AMD, choroidal vessels invade Bruch’s membrane. Unlike normal retinal vessels, these are fenestrated and result in the leakage of blood and serous fluid, which causes the detachment of both the retinal pigment epithelium and of the overlying retina. This results in the accumulation of intra-retinal fluid and retinal thickening. Haemorrhage under the retina stimulates the proliferation of fibrous tissue, the development of a disciform scar and loss of sight.

Research has focused on possible factors that influence blood vessel growth in the eye. One major area of research has been in growth factors that cause angiogenesis, for example, vascular endothelial growth factor (VEGF). VEGF is not a single entity but is a group incorporating VEGF-A–E and also placental growth factors 1 and 2. VEGF-A has been linked to intraocular neovascularisation in diabetic retinopathy, retinal vein occlusion and neovascular AMD and has provided an attractive target for therapeutic intervention. There are four isoforms of VEGF-A with VEGF-A165 being the major participant in abnormal angiogenesis in the eye.

Pegaptanib (Macugen® Pfizer) was the first anti-VEGF drug marketed for AMD and is an aptamer that binds specifically to the VEGF-A165 isoform. Bevacizumab (Avastin® Roche) and ranibizumab (Lucentis® Novartis) are both humanised monoclonal antibodies that bind to all the isoforms of VEGF-A. This prevents activation of the VEGF receptors and inhibition of angiogenesis.

Currently, ranibizumab is the gold standard for the treatment of neovascular AMD and in the UK it is given monthly for the first three injections and then when necessary, but is recommended that patients are followed up at monthly intervals. This is causing huge problems in capacity in ophthalmic departments and patients are often being left six weeks between appointments. Some ophthalmologists are questioning whether this is causing results in clinical practice to be less effective thanin  the pivotal trials (MARINA, ANCHOR and CATT).(1–3)

Aflibercept (Eylea®; Bayer) or VEG trap is different to the other anti-VEGF drugs because it acts as a decoy receptor for VEGF. VEGF-A binds to two separate receptors – VEGFR-1 and VEGFR-2 –both of which are transmembrane protein tyrosine kinases. Aflibercept is a human IgG1 molecule with its Fc portion fused to two different VEGF receptors. These bind to endogenous VEGF-A and neutralise its effects. Aflibercept also binds to VEGF-B and placental growth factor, but the clinical significance of this is unknown.

Aflibercept is available as a vial containing 4mg aflibercept in 0.1ml.

A dose of 2mg is injected intravitreally monthly for the first three months, then administered every two months up to a year. After a year, the interval for re-injection can be extended based on visual and anatomical outcomes. The schedule for monitoring should be determined by the treating physician and may be more frequent than the schedule for injections.(4) From animal trials there is some evidence that aflibercept has a longer duration of action than ranibizumab.(5)

The VEGF Trap-Eye investigation of efficacy and safety in wet AMD studies (VIEW 1 and VIEW 2) was a prospective double-masked parallel group randomised clinical trial. VIEW 1 recruited patients in Canada and the US (1217 patients) and VIEW 2 (1240 patients) recruited patients in Europe, the Middle East, Asia-Pacific and Latin America.

Patients were divided into four groups. During the first year, group one received aflibercept 0.5mg every four weeks, group two aflibercept 2mg every four weeks, group three aflibercept 2mg every four weeks for the first three injections and then two-monthly thereafter (sham injections given at interim visits to maintain masking), and group four, ranibizumab 0.5mg monthly. During the second year, patients remained on the same drug but the dose changed to an as needed regime. Patients were assessed monthly with treatment given no more frequently than four weeks and no longer than 12 weeks.

The primary endpoint was non-inferiority of aflibercept to ranibizumab in the proportion of patients maintaining vision at 52 weeks defined as losing <15 ETDRS (Early Treatment of Diabetic Retinopathy Study) letters. Results were 94% for ranibizumab and 95–96% for all aflibercept groups. Secondary endpoints included mean change in best corrected visual acuity (BCVA), retinal thickness and persistent fluid as assess by optical coherence tomography (OCT), which also showed non inferiority.(6)

At the end of the second year, preliminary results found that aflibercept 2mg at eight weeks resulted in gain similar to monthly ranibizumab but with fewer injections (11.2 over two years with 4.2 during the second year versus 16.5 over two years with 4.7 in the second year). The proportion of patients who required six or more injections in the second year was lower with aflibercept (15.9%) versus ranibizumab (26.5%).(7)

The results of the VIEW study showed that aflibercept was generally well tolerated with an acceptable safety profile with no obvious difference between aflibercept and ranibizumab. The main side effects were conjunctival haemorrhage, eye pain, reduced visual acuity (0.5% versus 0.5%), retinal haemorrhage (0.3% aflibercept versus 0.5% ranibizumab) and macular degeneration. Most of the serious side effects related to the injection technique and the disease, rather than the drugs administered. Endophthalmitis rates were similar (0.2% afliberecpt versus 0.5% ranibizumab). For non-ocular adverse effects, the incidence of death and withdrawal was similar between groups with arterial thrombotic events being reported in 3.2% of patients receiving ranibizumab and 3.3% of patients receiving aflibercept. The most serious non-ocular side effects included myocardial infarction, pneumonia and atrial fibrillation, which are events expected in an elderly population, and which occurred with similar frequency across all treatment groups.(6,7)

Using the list price, the mean two-year cost of treatment of aflibercept is £15k per eye compared with £14,700 with ranibizumab. However, both companies have confidential discount schemes in place, which changes the overall costs. However, aflibercept may have the advantage of fewer clinic visits and their associated costs. No data on aflibercept use after year two are available.(7)

Aflibercept appears to be as effective and safe as ranibizumab in the treatment of neovacular AMD. In the first year, two-monthly injections appear to be as effective as monthly ranibizumab, which would be good news for overstretched AMD services. However, with the change of protocol in the second year to monthly assessment for all groups, there is a lack of evidence for the efficacy and safety of continuing with two monthly injections for aflibercept in the second year. However, as ranibizumab patients are currently being seen almost two monthly, ophthalmologists and commissioners may feel that, with the evidence of the efficacy of aflibercept administered two-monthly in the first year, it would be clinically more appropriate to give aflibercept rather than ranibizumab.