To describe the pharmacokinetics, preclinical studies, and clinical trials of the newly approved anti–vascular endothelial growth factor (VEGF) drug aflibercept (Eylea (VEGF Trap-Eye); Regeneron; and Bayer).
Review with editorial commentary.
A review of the medical literature and pertinent Internet postings combined with analysis of key studies with expert opinion regarding the use of aflibercept for the treatment of exudative age-related macular degeneration.
Aflibercept, a fusion protein with binding domains from native VEGF receptors, binds VEGF-A, VEGF-B, and placental growth factors 1 and 2 with high affinity. Preclinical ophthalmologic studies demonstrated that aflibercept suppresses choroidal neovascularization in several animal models. The results of phase 1 and 2 trials showed excellent short-term suppression of choroidal neovascularization in patients with exudative age-related macular degeneration and suggested a longer durability of aflibercept compared with other anti-VEGF drugs. The pivotal phase 3 Vascular Endothelial Growth Factor (VEGF) Trap-Eye: Investigation of Efficacy and Safety in Wet Age-Related Macular Degeneration 1 and 2 trials showed that monthly and bimonthly aflibercept were noninferior to monthly ranibizumab at preventing vision loss (< 15-letter loss) with comparable vision gains and safety. Year 2 treatment involved monthly pro re nata injections with required injections every 3 months and maintained vision gains from the first year, with an average of 4.2 injections of aflibercept and 4.7 injections of ranibizumab.
Aflibercept promises to deliver excellent visual outcomes for exudative age-related macular degeneration patients while undergoing fewer injections compared with ranibizumab. With a wholesale cost of $1850 per dose, the cost per patient with aflibercept treatment promises to be lower than with ranibizumab.
The recent introduction of drugs that inhibit the actions of vascular endothelial growth factor (VEGF) has revolutionized the treatment of exudative age-related macular degeneration (AMD). Current anti-VEGF drugs bind to VEGF in the extracellular space and prevent its activation of transmembrane VEGF receptors. Between 2004 and 2006, 3 anti-VEGF drugs were introduced to ophthalmology either after receiving regulatory approval for the treatment of AMD or being used in an off-label manner. They exhibit important differences in their sites of activity, formulation methods, binding affinities, and biologic activities. Pegaptanib (Macugen; Eyetech, New York, New York, USA) is a ribonucleic acid aptamer that blocks the main pathologic isoform of VEGF (VEGF 165 ) by attaching to its heparin binding domain, whereas ranibizumab (Lucentis; Novartis, Basel, Switzerland; Genentech, South San Francisco, California, USA; and Roche, Basel, Switzerland) and bevacizumab (Avastin; Genentech and Roche) are, respectively, an affinity-matured, humanized, monoclonal antibody fragment and a full-length, humanized, monoclonal antibody to VEGF. Both work by blocking the receptor binding domain of all isoforms of VEGF-A.
Now, more than 5 years later, the next anti-VEGF drug, aflibercept (Eylea; Regeneron, Tarrytown, New York, USA, and Bayer, Berlin, Germany) has been approved by the United States Food and Drug Administration. This soluble decoy receptor is produced by fusing all-human DNA sequences of the second immunoglobulin (Ig) domain of human VEGF receptor (VEGFR) 1 to the third Ig domain of human VEGFR-2, which then is fused to the Fc region of human IgG-1. Aflibercept binds to all VEGF-A and VEGF-B isoforms, as well as the highly related placental growth factor. Aflibercept is produced in a Chinese hamster ovary cell line, and then is specially purified and formulated exclusively for intraocular injection.
When developing aflibercept, investigators noted that most of the vasoproliferative and hyperpermeability effects of VEGF are mediated by activation of VEGFR-2, yet VEGFR-1 actually binds to VEGF-A with a much higher affinity (K d = 10 to 20 pM vs K d = 100 to 300 pM; binding affinity is inversely proportional to K d ). Therefore, investigators initially created a parent VEGF Trap with 3 binding domains from the high-affinity VEGFR-1 fused to the Fc region of IgG1. Although this molecule bound VEGF tightly, it exhibited unfavorable pharmacokinetic characteristics as a result of rapid sequestration within the intercellular matrix. By substituting the more basic binding sequences from VEGFR-2, Regeneron developed a high-affinity (K d = 0.5 pM) soluble receptor molecule that exhibited minimal binding to the extracellular matrix. This gives aflibercept a significantly higher binding affinity to VEGF 165 than either bevacizumab (K d = 58 pM) or ranibizumab (K d = 46 pM). The intravitreal half-life of aflibercept is 4.7 days in rabbits (Regeneron, data on file), comparable with that of bevacizumab (4.32 days), but longer than that of ranibizumab (2.88 days). Although pharmacokinetic studies have not been performed in human eyes, the molecular size of aflibercept (115 kDa) would suggest that its intraocular half-life would more closely resemble bevacizumab (149 kDa; human half-life, 8.24 days) than the smaller ranibizumab (48 kDa).
The predicted biological activity of a therapeutic macromolecule depends to a large degree on both its intraocular half-life and its VEGF binding affinity. The binding affinity of aflibercept to VEGF-A is substantially greater than any of the other 3 anti-VEGF drugs, but recent biological assays with aflibercept, ranibizumab, and bevacizumab report that the relative inhibition of both endothelial cell proliferation and migration ranges from 1× to 100×. The reason for this wide variation in inhibitory effects is unknown, but the use of different reagent concentrations has been suggested. Based on these data and estimated intraocular drug half-lives, mathematical modeling predicts that a single intravitreal injection of aflibercept 2 mg would last between 48 and 83 days (compared with 30 days for ranibizumab).
Preclinical and Oncology Trials
The development of our currently used anti-VEGF drugs was based on a different long-term therapeutic strategy. Whereas bevacizumab was developed with a long systemic residence time exclusively for the systemic treatment of advanced cancers, ranibizumab was designed to have a short systemic clearance half-life by removing the Fc fragment from the parent IgG molecule, and its affinity for VEGF was enhanced by changing 5 of its amino acids, thereby optimizing it for the intraocular treatment of exudative ocular diseases. Additionally, creating a molecule smaller than the retinal exclusion limit (76 kDa) was believed to be necessary to penetrate the inner retina. Aflibercept, however, was developed to treat both advanced solid tumors and ophthalmic vascular conditions.
Aflibercept has decreased tumor growth successfully in several orthotopic mouse models: ovarian carcinoma, hepatoblastoma, cholangiocarcinoma, pancreatic ductal carcinoma, Wilms tumor, renal cell carcinoma, and glioblastoma. Suppression of experimental neuroblastoma growth by anti-VEGF agents was achieved with the following relative efficacies: aflibercept > monoclonal antibody > aptamer to VEGF 165 .
Phase 1 and 2 oncologic studies evaluated escalating doses of aflibercept in patients with chemotherapy-resistant renal cell carcinoma, Hodgkin lymphoma, glioblastoma, and anaplastic glioma. In a phase 3 trial of patients with advanced adenocarcinoma of the colon, intravenously administered aflibercept (Zaltrap; Regeneron and Sanofi Aventis, Bridgewater, New Jersey, USA) extended progression-free survival from 4.7 to 6.9 months ( P = .00007) and overall survival from 12.1 to 13.5 months ( P = .0032).
In ocular preclinical studies, aflibercept demonstrated in vivo activity against several murine models of choroidal neovascularization (CNV). It prevented the development of CNV after intense laser photocoagulation to the retinal pigment epithelium, inhibited the development of CNV in VEGF-secreting transgenic mice, and prevented the development of CNV in mice receiving exogenous VEGF. This favorable response in part was the result of decreases in both intercellular adhesion molecule-1 and endothelial nitric oxide synthetase synthesis within the CNV. After matrigel-induced CNV in rats, aflibercept injections at 2 and 6 days prevented the development of CNV, whereas injections at 10 days decreased collagen synthesis and leukocyte infiltration. Prolonged, high-dose aflibercept therapy causes loss of both endothelial cells and pericytes, thereby reducing vessels to basement membrane ghosts. These successful preclinical studies justified the development of aflibercept trials for the treatment of exudative AMD in humans.
Based on a plausible biologic rationale for suspected efficacy and an acceptable safety profile in preclinical animal studies, aflibercept first was administered intravenously to patients with neovascular AMD in a placebo-controlled clinical trial. In 2 of the 5 patients receiving 3.0 mg/kg aflibercept, systemic toxicity developed (1 patient had grade 2 proteinuria and 1 patient had grade 4 hypertension). A subsequent phase 1 study of intravitreal aflibercept showed that up to 4-mg dosing decreased macular edema and subretinal fluid for at least 6 weeks and was tolerated well with no ocular inflammation. A phase 2 clinical trial of 159 patients explored 5 different aflibercept dosing regimens with a primary outcome at 12 weeks: 0.5 mg every 4 weeks, 2 mg every 4 weeks, 0.5 mg every 12 weeks, 2 mg every 12 weeks (thus, only 1 treatment before the primary outcome), and 4 mg every 12 weeks. At 12 weeks, quarterly dosing reduced macular thickening and improved best-corrected visual acuity, but monthly dosing was more effective. The same group of patients subsequently was followed up monthly and was retreated pro re nata with their assigned dose based on the following criteria: central retinal or lesion thickness increased 100 μm or more from best previous reading or best-corrected vision dropped 5 or more letters with recurrent fluid on optical coherence tomography if persistent fluid was seen on optical coherence tomography or if there was new CNV, persistent leakage on fluorescein angiography, or a new macular hemorrhage. Pro re nata dosing maintained the efficacy established in each of the 5 groups during the first part of the trial. At 52 weeks of follow-up, patients in the initial monthly dosing regimens tended to have improved visual acuity outcomes compared with those in the quarterly dosing groups, regardless of the aflibercept dose, indicating the importance of loading dose before the less frequent dosing.
Based on the results from these phase 1 and 2 studies, 2 parallel phase 3 pivotal clinical trials were designed and conducted to establish the noninferiority of 3 regimens of aflibercept compared with monthly ranibizumab 0.5 mg. These trials were the Vascular Endothelial Growth Factor (VEGF) Trap-Eye: Investigation of Efficacy and Safety in Wet Age-Related Macular Degeneration Study 1, a study of 1217 patients in the United States and Canada, and the Vascular Endothelial Growth Factor (VEGF) Trap-Eye: Investigation of Efficacy and Safety in Wet Age-Related Macular Degeneration Study 2, a study of 1240 patients in Europe, Asia, Japan, and Latin America. In both trials, patients were randomized 1:1:1:1 to 1 of 4 groups: ranibizumab 0.5 mg given monthly, aflibercept 0.5 mg given monthly, aflibercept 2 mg given monthly, and aflibercept 2 mg given monthly for 3 injections followed by continued injections every 2 months. The end point for both trials was at 52 weeks. The primary outcome measure, termed maintenance of vision, was the percentage of eyes losing fewer than 3 lines of visual acuity on the Early Treatment Diabetic Retinopathy chart. A secondary visual acuity outcome was mean improvement in visual acuity measured in number of additional letters read on the Early Treatment Diabetic Retinopathy chart compared with baseline.
Both studies showed that all 3 regimens of aflibercept were noninferior to the ranibizumab monthly regimen ( Table ). Of greatest interest to clinicians was the aflibercept arm featuring bimonthly injections after 3 monthly loading doses. This regimen offers fewer injections for the patient, less risk of endophthalmitis, and no loss of efficacy at least through 52 weeks.
|Aflibercept 0.5 mg Monthly
|Aflibercept 2 mg Monthly
|Aflibercept 2 mg Bimonthly after 3 Monthly Injections
|Ranibizumab 0.5 mg Monthly
|VIEW 1/maintenance of VA (%)
|VIEW 2/maintenance of VA (%)
|VIEW 1/mean VA improvement (letters)
|VIEW 2/mean VA improvement (letters)