FormalPara Features and properties of aflibercept (EYLEA™)

Indication

Wet (neovascular) age-related macular degeneration

Mechanism of action

Neutralizes vascular endothelial growth factor (VEGF) family members VEGF-A, -B and placental growth factor

Dosage and administration

Dose

2 mg per (affected) eye

Route of administration

Intravitreal injection

Frequency of administration

Every month (4 weeks) for the first 3 months followed by every 2 months (8 weeks)

Mean pharmacokinetic parameters for free aflibercept (following repeated intravitreal injections of aflibercept 2 mg/eye)

Mean maximum plasma concentration (Cmax)

0.02 mg/L

Area under the plasma concentration-time curve from time zero to the time of the last measurable concentration

0.15 mg · day/L

Time to Cmax

1–3 days

Most common (incidence ≥5 %) treatment-emergent ocular adverse events

Conjunctival haemorrhage, eye pain, cataract, vitreous detachment, vitreous floaters and increased intraocular pressure

1 Introduction

If left untreated, or inadequately treated, age-related macular degeneration (AMD), is the leading cause of irreversible vision loss among older adults in Western countries [1, 2]. In the US, the number of people with AMD is predicted to rise from 1.75 million in 2000 to 2.95 million in 2020 [1]. The early form of the disease is characterized by the presence of macular drusen with or without changes in the retinal pigment epithelium (RPE), while the late stage encompasses both dry AMD, which is characterized by atrophy of the RPE, and wet (neovascular) AMD, which is characterized by choroidal neovascularization (CNV) and/or RPE detachment [3, 4]. Neovascular AMD only constitutes a minority of AMD cases (10–15 %); however, it accounts for most of the severe vision loss due to the disease (90 % of cases) [1, 5, 6].

The vascular endothelial growth factor (VEGF) family and its receptors are essential regulators of angiogenesis and vascular permeability [7, 8]. The VEGF family member VEGF-A plays a central role in the development of CNV, which is the underlying cause of vision loss in patients with neovascular AMD; this has led to intravitreal anti-VEGF therapies that selectively inhibit VEGF-A being investigated for the treatment of the disease [1, 6]. Pegaptanib, an aptamer that specifically binds to, and blocks, the VEGF-A165 isoform [9], and ranibizumab, a humanized, monoclonal antibody fragment that binds to, and inhibits, multiple VEGF-A isoforms [10, 11], are two such agents that have been approved for the treatment of neovascular AMD in the US; ranibizumab is considered to be the current gold standard for this indication [1]. Bevacizumab, a full-length, humanized, monoclonal antibody, is a third such agent; it is commonly used in the US, despite not being approved by the US FDA for the treatment of neovascular AMD [12]. Intravitreally administered anti-VEGF therapies have improved outcomes compared with previous therapies for neovascular AMD; in the case of ranibizumab and bevacizumab, individualized treatment regimens aimed in part at minimizing the frequency (and associated risks) of injections have emerged as alternatives to traditional monthly dosing [13].

Aflibercept (EYLEA™; also known as ‘VEGF Trap-Eye’) [14, 15] is an intravitreally administered anti-VEGF agent that binds to, and thereby neutralizes, all VEGF-A isoforms. In addition, unlike ranibizumab or bevacizumab, it neutralizes two other VEGF family members (ligands), namely VEGF-B and placental growth factor (PlGF), that have also been implicated in pathological vascular remodelling [16]. Experimental evidence shows that targeting VEGF-B [17] and PlGF [18] inhibits CNV and suggests that PlGF synergizes with VEGF-A in promoting vascular pathology in neovascular AMD [16]. To date, aflibercept has been approved for the treatment of neovascular AMD in the US [19] and Australia [20].

This article reviews the efficacy and tolerability of aflibercept in the treatment of patients with neovascular AMD, and summarizes its pharmacological properties. Medical literature (including published and unpublished data) on the use of aflibercept in neovascular AMD was identified by searching databases for studies published since 1996 (including MEDLINE and EMBASE), bibliographies from published literature, clinical trial registries/databases and websites (including those of regional regulatory agencies and the manufacturer). Additional information (including contributory unpublished data) was also requested from the company developing the drug. Searches were last updated 17 September 2012.

2 Pharmacodynamic Profile

Aflibercept is a fully human, soluble, recombinant decoy VEGF receptor (or ‘VEGF-Trap’). It consists of the second extracellular binding domain of VEGF receptor-1 (VEGFR-1) and the third extracellular binding domain of VEGF receptor-2 (VEGFR-2) fused to the Fc portion of immunoglobulin G1 [21, 22].

In vitro, aflibercept binds with high affinity to VEGF-A [dissociation constant (Kd) of 0.36 and 0.49 pmol/L for the human VEGF-A121 and A165 isoforms, respectively] [16, 23]. Because aflibercept binds VEGF-A with higher affinity than both VEGFR-1 (Kd of 9.33 pmol/L for human VEGF-A165) [16] and -2 (Kd of 88.8 pmol/L for human VEGF-A165) [16], it effectively prevents the binding of VEGF-A to, and the activation of, its cognate receptors [24]. Aflibercept forms a stable, inert 1:1 complex with the VEGF-A ligand [25].

The binding affinity of aflibercept for human VEGF-A165 was substantially (≈100-fold) higher than that of ranibizumab (Kd of 46 pmol/L) and bevacizumab (Kd of 58 pmol/L), as assessed under identical experimental conditions. This was primarily attributed to the association rate constant (Ka) for aflibercept binding to human VEGF-A165, which was 256 and 77 times faster than the corresponding Ka values for ranibizumab and bevacizumab, respectively [16].

Aflibercept also binds with high affinity to VEGF-B (Kd of 1.92 pmol/L for the truncated amino acid 10–108 form of human VEGF-B) and PlGF (Kd of 392.0 and 38.9 pmol/L for the human PlGF-1 and -2 isoforms, respectively) [16, 23]. In contrast, neither ranibizumab nor bevacizumab bind to human PlGF-2 [16]. In a similar fashion to that described for VEGF, aflibercept forms a stable, inert 1:1 complex with the PlGF ligand [25], thereby inhibiting the activity of the latter at VEGFR-1 receptors [26].

In vitro, aflibercept was more potent than ranibizumab and bevacizumab in blocking VEGF-A induced activation of VEGFR-1 and -2, and also in inhibiting VEGF-A-induced human umbilical vein endothelial cell (HUVEC) migration. Additionally, aflibercept blocked both PlGF-mediated activation of VEGFR-1 and PlGF-induced HUVEC migration, whereas ranibizumab and bevacizumab showed no such activity [16].

Intravitreally administered aflibercept inhibited CNV in animal models of laser-induced CNV [27, 28]. In particular, aflibercept both prevented the development of new lesions and caused regression of established lesions in non-human primate models of laser-induced CNV [27].

Intravitreal injections of aflibercept (0.05, 0.15, 0.5, 1, 2 and 4 mg/eye) produced functional and anatomical improvements in a phase I single-ascending-dose study in 21 patients with neovascular AMD [CLinical Evaluation of Anti-angiogenesis in the Retina Intravitreal Trial (CLEAR-IT 1)] [29]. At 6 weeks post-injection, the increase from baseline in mean best-corrected visual acuity (BCVA) was 4.4 Early Treatment Diabetic Retinopathy Study (ETDRS) letters for all aflibercept dosage groups combined; the decrease from baseline in mean foveal thickness (as determined by optical coherence tomography) was 104.5 μm.

The improvements in functional and anatomical outcomes were largely dose dependent [29]. At week 6, the increases from baseline in mean BCVA were 0.6, 1.2 and 13.5 ETDRS letters in patients receiving a low (0.05, 0.15 or 0.5 mg/eye; n = 9), intermediate (1.0 mg/eye; n = 6) or high (2 or 4 mg/eye; n = 6) dose of aflibercept, respectively; the corresponding decreases from baseline in mean foveal thickness were 57.4, 157.4 and 88.1 μm, respectively.

Of note, three patients who received a high dose of aflibercept experienced significant vision gain (i.e. a gain of ≥15 ETDRS letters); three patients who received a low, intermediate or high dose of aflibercept showed elimination of (CNV-related) fluorescein leakage and a reduction in CNV area [29].

Functional and anatomical outcomes in patients with neovascular AMD who received aflibercept in larger phase III clinical trials are discussed in detail in Sect. 4.

3 Pharmacokinetic Profile

The pharmacokinetics of aflibercept have been investigated after intravitreal administration (in phase I–III studies in patients with neovascular AMD) and intravenous administration (in phase I studies in healthy volunteers). The main findings are presented in the US prescribing information [19] and/or the US FDA regulatory review [30].

Following intravitreal injection, aflibercept can be detected in the plasma as free drug (i.e. unbound to circulating endogenous VEGF) or, more predominantly, in the form of a stable, inactive aflibercept:VEGF complex (see Sect. 2) [19, 30].

The mean maximum plasma concentration (Cmax) and area under the plasma concentration-time curve from time zero to the time of the last measurable concentration of free aflibercept were 0.02 (range 0–0.05) μg/mL and 0.15 (range 0–0.59) μg·day/L, respectively, following repeated intravitreal administration of aflibercept 2 mg (per eye) in patients with neovascular AMD; Cmax was attained in 1–3 days. Free aflibercept was not detectable in the plasma 2 weeks after dosing, nor did it accumulate in the plasma after repeated intravitreal administration of aflibercept at intervals of 4 weeks [19, 30].

Following intravitreal administration of aflibercept 2 mg to patients with neovascular AMD, the mean Cmax of free aflibercept was estimated to be more than 100-fold lower than the concentration of aflibercept required for half-maximal systemic VEGF binding [19].

The volume of distribution of free aflibercept was ≈6 L following intravenous administration of aflibercept [19, 30].

The metabolism of aflibercept has not been studied; the drug is expected to be eliminated both as a result of binding to free endogenous VEGF and as a result of proteolysis [19, 30]. Free aflibercept is not expected to be eliminated by the kidney, because of its relatively large molecular size (115 kilodaltons) [30].

The terminal elimination half-life of free aflibercept in the plasma was ≈5–6 days following intravenous administration of aflibercept doses of 2–4 mg/kg [19, 30].

Plasma concentrations of free aflibercept or the aflibercept:VEGF complex were not influenced to a clinically significant extent by age, sex or renal function, according to exploratory subgroup analyses from a phase III clinical study [VEGF-trap-eye Investigation of Efficacy and safety in Wet AMD (VIEW) 2; see Sect. 4 for further details] [19, 30].

The US prescribing information [19] indicates that aflibercept dose adjustment based on renal impairment is not needed; however, it does not include a (similar) recommendation regarding hepatic impairment [19]. As is the case with renal impairment, there is no pharmacokinetic basis for dose adjustment based on hepatic impairment [30].

Aflibercept is not expected to interact with other medications; no in vivo drug–drug interactions studies have been conducted (or are needed) [30].

4 Therapeutic Efficacy

The efficacy of repeated intravitreal administration of aflibercept in the treatment of neovascular AMD has been compared with that of intravitreal ranibizumab in two pivotal phase III noninferiority studies of virtually identical design (VIEW 1 [31] and 2 [32]; see Table 1 for trial eligibility criteria and patient baseline characteristics).

Table 1 Randomized, double-masked, multicentre VIEW 1 [31] and 2 [32] studies: eligibility criteria and baseline patient characteristics
Full size table

In the first year of these studies (weeks 0–52), patients were randomly assigned to the recommended dosing regimen of ranibizumab (0.5 mg every 4 weeks) or to one of three different dosing regimens of aflibercept [0.5 mg every 4 weeks; 2 mg every 4 weeks; or 2 mg every 8 weeks (with the first three doses administered at 4-week intervals)] that had previously been identified in a phase II dose- and interval-ranging study in 157 patients with neovascular AMD (CLEAR-IT 2 [24]). The primary endpoint was the proportion of patients who maintained their vision (i.e. lost <15 ETDRS letters) at week 52; secondary endpoints included the proportion of patients who experienced a clinically meaningful improvement in vision (i.e. gained ≥15 ETDRS letters) at week 52 and the mean change from baseline in BCVA at week 52 [19,33–35].

During the second year (weeks 52–96), patients were treated on a modified quarterly basis, i.e. they were assessed monthly and, if necessary, treated (with the same drug and dose as in the first year) no more frequently than every month and no less frequently than every 3 months [33]. This treatment approach has also been referred to as a ‘capped pro re nata dosing schedule’ [36].

First-year results are available from the US prescribing information [19], conference abstracts (VIEW 1 [34] and 2 [35]) and a press release [33]; second-year results are available from a press release [37].

4.1 Year 1 Results

In both trials, all three aflibercept regimens were found to be noninferior to monthly dosing with ranibizumab in terms of the primary endpoint of the proportion of patients who maintained their vision at week 52 (Table 2) [19, 33–35].

Table 2 Comparative efficacy of intravitreal aflibercept and ranibizumab for the treatment of neovascular age-related macular degeneration. Summary of results at week 52a from two virtually identical, randomized, double-masked, multicentre, phase III trialsb
Full size table

Across both trials, ≈95 % of patients receiving aflibercept 2 mg every month or every second month (after treatment initiation with three monthly doses) maintained their vision at week 52, as did ≈95 % of patients receiving ranibizumab 0.5 mg every month (Table 2) [19].

As regards the secondary endpoints, there were no significant differences between the aflibercept and ranibizumab regimens in each trial, with the exception that the mean change from baseline in BCVA at week 52 was greater in the aflibercept 2 mg every month group compared with the ranibizumab group in VIEW 1 [10.9 vs. 8.1 ETDRS letters (p < 0.01); Table 2] [19, 33].

Across both trials, patients receiving aflibercept 2 mg every month or every second month (after treatment initiation with three monthly doses) had a mean increase from baseline in BCVA at week 52 that ranged from 7.6 to 10.9 letters; approximately one-third of these individuals (29–38 %) experienced a clinically meaningful improvement in vision (Table 2) [19].

4.2 Year 2 Results

In a pooled analysis of the two trials, patients receiving aflibercept 2 mg every second month (after three initial monthly doses) during the first year followed by modified quarterly treatment during the second year had an increase from baseline in mean BCVA of 7.6 letters at week 96, as compared with an increase of 8.4 letters at week 52. They received an average of 11.2 injections over the 96-week period, including 4.2 injections between weeks 52 and 96. Sixteen per cent of patients required frequent (six or more) injections during the second year [37].

In comparison, patients receiving ranibizumab 0.5 mg every month during the first year followed by modified quarterly treatment during the second year had an increase from baseline in mean BCVA of 7.9 letters at week 96, as compared with an increase of 8.7 letters at week 52. They received an average of 16.5 injections over the 96-week period and 4.7 injections between weeks 52 and 96. Approximately one-quarter (26.5 %) of patients required frequent (six or more) injections during the second year [37].

5 Tolerability

The following tolerability profile of intravitreal aflibercept is based on pooled findings from the pivotal phase III VIEW studies discussed in Sect. 4; these have been reported in the US prescribing information [19] and/or the US FDA’s regulatory review [23, 38].

A total of 1,824 patients (911 and 913 in VIEW 1 and 2, respectively) were treated with aflibercept for 52 weeks, including 613 (304 and 309 in VIEW 1 and 2, respectively) who received 2 mg every month and 610 (303 and 307 in VIEW 1 and 2, respectively) who received 2 mg every second month [19, 31, 32].

5.1 Ocular Adverse Events

Of the 1,824 aflibercept recipients, 1,263 (69.2 %) reported at least one treatment-emergent ocular adverse event (in the study eye); 36 (2.0 %) reported at least one treatment-emergent serious ocular adverse event [38]. Among the 595 patients who received ranibizumab in the two VIEW studies, 433 (72.8 %) reported at least one treatment-emergent ocular adverse event; 19 (3.2 %) reported at least one treatment-emergent serious ocular adverse event [38].

The most common (incidence ≥5 %) treatment-emergent ocular adverse events occurring in aflibercept recipients were conjunctival haemorrhage, eye pain, cataract, vitreous detachment, vitreous floaters and increased intraocular pressure (IOP). The incidences of these (and other frequently reported ocular) adverse events in aflibercept recipients did not differ appreciably from those in ranibizumab recipients [19] (Fig. 1).

Fig. 1
figure 1

Comparative ocular tolerability of intravitreally administered aflibercept and ranibizumab in patients with neovascular age-related macular degeneration. The figure shows the pooled incidences of commonly occurring (incidence ≥1 %) treatment-emergent ocular adverse events in patients randomized to receive one of three aflibercept dosing regimens [0.5 mg every 4 weeks; 2 mg every 4 weeks; and 2 mg every 8 weeks (with the first three doses administered at 4-week intervals)] or the recommended regimen of ranibizumab (0.5 mg every 4 weeks) in two pivotal phase III studies (VIEW 1 and 2; see Table 1 for study design details) [19, 38]. IAI intravitreal aflibercept injection, IOP intraocular pressure, RBZ ranibizumab, RPE retinal pigment epithelium

Full size image

Three (0.2 %) of the 1,824 aflibercept recipients developed endophthalmitis; all three patients received the 2 mg every month regimen in VIEW 1. In comparison, three (0.5 %) of the 595 ranibizumab recipients developed endophthalmitis; all three patients were participants in the VIEW 1 trial [38].

Fifty-one (2.8 %) of the 1,824 patients treated with aflibercept, as compared with 22 (3.7 %) of the 595 patients treated with ranibizumab, had an absolute IOP value of ≥35 mm Hg during the study. The corresponding proportions of patients who had a ≥10 mm Hg increase in IOP from baseline during the study were 2.0 % and 3.2 %, respectively [38]. Increased IOP, which was noted within 1 h of intravitreal injection (of aflibercept), resolved in most patients before the next pre-dose measurement [19].

5.2 Non-Ocular Adverse Events

Of the 1,824 aflibercept recipients, 1,324 (72.6 %) reported at least one treatment-emergent non-ocular adverse event; 252 (13.8 %) reported at least one treatment-emergent serious non-ocular adverse event [38]. Among the 595 ranibizumab recipients, 415 (69.7 %) reported at least one treatment-emergent non-ocular adverse event; 83 (13.9 %) reported at least one treatment-emergent serious non-ocular adverse event [38].

Intravitreal administration of anti-VEGF agents to patients with neovascular AMD has been associated with the occurrence of arterial thromboembolic events (ATEs) [defined as non-fatal stroke, non-fatal myocardial infarction and vascular death (including deaths of unknown cause)] [19]. Across both trials, the incidence of ATEs was 1.8 % in patients treated with aflibercept (all dosage groups combined), as compared with 1.7 % in patients treated with ranibizumab [19, 38].

Nasal septal deviation, rhinorrhoea, epistaxis, nasal polyps, turbinate hypertrophy, nasal dryness and cysts were occasionally observed in a subset of VIEW 2 participants (n = 160; 40 per treatment group) who underwent a special ear, nose and throat examination. However, there were no reports of nasomucosal erosions (which have been seen in animal studies of aflibercept) [19, 23, 38].

6 Dosage and Administration

In the US, the recommended dosing schedule (per affected eye) for intravitreally injected aflibercept is 2 mg every month for the first 3 months followed by 2 mg every second month. Although aflibercept (2 mg per affected eye) may be administered every month from the outset of treatment, this regimen did not demonstrate greater efficacy compared with administration every second month after the three initial monthly doses [19] (see Sect. 4).

Aflibercept is contraindicated in patients with ocular or periocular infections and in patients with active intraocular inflammation [19].

Local prescribing information should be consulted for full details of dosage and administration instructions, as well as contraindications, warnings and precautions relating to the use of intravitreal aflibercept.

7 Current Status

Intravitreally administered aflibercept has been approved for the treatment of neovascular AMD in the US [19] and Australia [20].

This approval was based on the results of two well designed phase III studies (VIEW 1 and 2), which compared aflibercept with ranibizumab, the current gold standard for the treatment of neovascular AMD in the US. In both trials, which were of virtually identical design, the recommended regimen of aflibercept [2 mg every second month (after three initial monthly doses)] was shown to be noninferior to the recommended regimen of ranibizumab (0.5 mg every month) in terms of the primary endpoint of maintaining vision after 1 year of treatment; similar results were seen when monthly dosing with aflibercept (0.5 or 2 mg) was compared with ranibizumab.

Furthermore, over a 96-week period, patients receiving aflibercept 2 mg every second month (after three initial monthly doses) during the first year followed by modified quarterly treatment during the second year had a similar visual acuity gain to those receiving ranibizumab every month during first year followed by modified quarterly treatment during the second year, but on average required five fewer injections.

Aflibercept was generally well tolerated in the VIEW studies; the ocular and non-ocular adverse event profile of the drug was similar to that of ranibizumab.

In the EU, the EMA’s CHMP has adopted a positive opinion for the use of intravitreal aflibercept in the treatment of neovascular AMD [39].