The major angiogenic driver for uncontrolled neovascularization has been introduced as vascular endothelial growth factor (VEGF).1 VEGF-A, VEGF-B, VEGF-C, VEGF-D, and placental growth factor (PlGF) are five members of VEGF family in mammalians.2 Clinical research has demonstrated the significant role of VEGF-A in developing ocular neovascularization, such as diabetic retinopathy and age-related macular degeneration (AMD), with the application of VEGF-A inhibitors.3 Aflibercept (Eylea; Regeneron, Tarrytown, NY), or VEGF Trap-eye, is the late member of the anti-VEGF family.4 Since VEGFR-1 has a higher affinity for VEGF than VEGFR-2, drug developers have employed its binding sequences for VEGF Trap-eye.5 Aflibercept, which is a 115-kDA recombinant protein, was manufactured from portions of the human VEGFR bound to the FC portion of a human Immunoglobulin G1 and approved by the U.S. Food and Drug Administration (FDA) in November 2011.6
Anti-VEGF therapy is the current remedy for AMD, diabetic macular edema (DME), and other uncontrolled ocular neovascularizations.7 Need of monthly visits and intravitreal procedures with undesirable side effects have motivated researchers to find a way to decrease number of injections and monitoring frequency. Therefore, utilizing a new drug with a longer half-life in vitreous that will provide equivalent efficacy in comparison to monthly ranibizumab seems reasonable. The ability of binding tightly to three isoforms of growth factors (VEGF-A, VEGF-B, and PlGF) helps aflibercept to have high binding affinity and a long half-life. On the other hand, aflibercept is a recombinant fusion protein that also targets PlGF. Hence, it may be more efficacious when other anti-VEGFs seems to be tolerated.8
Ziv-aflibercept (Zaltrap; Sanofi-Aventis US, Bridgewater, NJ, and Regeneron, Tarrytown, NY) is an identical fusion protein to aflibercept and also a much cheaper product. Ziv-aflibercept is FDA-approved for the treatment of metastatic colorectal carcinoma. Aflibercept is an iso-osmotic solution (300 mOsm/kg), whereas ziv-aflibercept has an osmolarity of 1,000 mOsm/kg.9 It was shown in rabbits and primates that solutions of less than 500 mOsm caused no damage. Tonicity between 500 and 1,000 mOsm produced inconsistent and often only partial damage. Injections of more than 2,000 mOsm promptly produced severe changes in the retinal pigment epithelium and retinal detachment.10 Although not approved, ziv-aflibercept has been used in management of macular diseases.9
This study was conducted to investigate the retinal safety and vitreous bioavailability of intravitreal injection of ziv-aflibercept in rabbit as experimental model in two doses of complete (1.25 mg/0.05 mL) and half diluted (0.625 mg/0.05 mL).
Materials and Methods
Sixteen male albino white rabbits, weighing 2 kg to 2.5 kg on the day before drug administration, were obtained from Pasteur Institute of Iran. As it was a pilot study, three to five rabbits were proposed for each measurement, and we assigned four rabbits in each group. The research project treated in accommodation with the ARVO assertion for the Use of Animals in Ophthalmic and Vision Research and was approved by the Tehran University of Medical Sciences Research Ethics Committee.
Thirty-two eyes of 16 male albino rabbits were used for the animal study. Before an intravitreal injection, rabbits were systemically anesthetized with a mixture of ketamine 10% (Alfasan, for veterinary use only; Holland) (50 mg/kg) and xylazine hydrochloride 2% (Alfasan, for veterinary use only, Holland) (6 mg/kg), and topically anesthetized with tetracaine (Anestocaine; Sinadarou Laboratories, Tehran, Iran).
Intravitreal ziv-aflibercept injection was performed through a syringe with a 30-gauge needle for each eye under anesthesia and the operating microscope. All right and left eyes received 1,250 μg/0.05 mL and 625 μg/0.05 mL of ziv-aflibercept, respectively, through the same size needle. Drug dilution was done by balanced salt solution. Injection was performed 2 mm from the limbus, superior or superotemporally. Rabbits were allocated to four main groups based on the time of evaluation after intravitreal injection (24, 168, 336, and 720 hours). In each group, electrophysiological recordings were obtained before the animal was euthanized. After euthanization, both eyes were enucleated, and vitreous sampling was obtained after 5-mm sclerotomy and aspiration of around 1-mL vitreous. The vitreous samples were cleared via centrifugation at 12,000 rpm 10 minutes at 4°C and then supernatants were used for the subsequent experiments.
Measuring trace concentrations of aflibercept in dilute fresh vitreous samples was performed using enzyme-linked immunosorbent assay (ELISA). Optical density was measured using a microplate reader (Autobio PHOmo; Autobio Labtec Instruments, China) at 450 nm with correction at 620 nm. Measurements were repeated for each sample.
Enzyme-Linked Immunosorbent Assay (ELISA) Study
Ziv-aflibercept concentrations in all samples were determined by sandwich ELISA using a bevacizumab (Avastin; Genentech, South San Francisco, CA) ELISA kit (Antibodies Co., LOT AA104-02). The lower limit of quantification of the aflibercept was 30 ng/mL. The kit was used as described by the manufacturer, except that the calibration curve was drawn using different concentrations of aflibercept in order to enable the assay to accurately quantify aflibercept concentrations. A standard curve was prepared with aflibercept ranging from 30 ng/ML to 1,000 ng/mL.
Bevacizumab ELISA belongs to sandwich type ELISA. The kit includes a microliter plate coated with recombinant human VEGF-A in which samples were incubated at room temperature (RT) for the appropriate time. After washing the wells-diluted buffer, a horseradish antibody was added and bound to aflibercept pre-captured by VEGF-A on the well's surface. Then chromogen-substrate was added, and blue color of samples changed to yellow after adding stop solution. The ensuing color was related to the aflibercept concentration in the sample, which was determined by the standard curve.
Measurements and Evaluations
Animals were observed at 24,168,336, and 720 hours after drug administration for signs of possible adverse events, and once daily for qualitative appraisal of food consumption. Body weight was determined before administration. Ophthalmic evaluations of animals were executed once during the screening period, before the dose, and at 24, 168, 336, and 720 hours after drug administration. Aflibercept concentrations in vitreous humor of treated and untreated eyes were determined by sandwich ELISA using Bevacizumab ELISA kit. Samples were screened at proper dilution.
The resultant vitreous ziv-aflibercept concentrations, reported as micrograms per milliliters, were used for analyses. Areas under the curve (AUC) were estimated by a linear-trapezoidal method, and the biological half-life (t 1⁄2) was estimated by linear regression, using at least four concentration time points. The volumes of the vitreous of the rabbits were considered 1.5 mL. Hence, the zero time concentration was calculated based on that the bioavailability of the drug in the injection time considered 100% at the injection time.
Rabbit globes were fixed in 10% formalin and bisected into two calottes axially. After tissue processing and embedding into paraffin blocks, thin tissue sections in three consecutive tissue planes (200 μm apart) were prepared. The hematoxylin and eosin (H&E) stained slides were examined under light microscopy (BX41; Olympus, Tokyo, Japan) by a masked pathologist for the presence of atrophic changes in the retinal layers, retinal necrosis, retinal inflammation, intraretinal hemorrhages, and loss of retinal pigment epithelial cells. Overall, the histopathological findings were categorized into unremarkable retinal histology and remarkable retinal changes based on the presence or absence of the above measures.
Electrophysiological examination was performed at baseline, before intravitreal injection, after general anesthesia while the animal was completely unconscious, and before enucleation. Full-field electroretinography (ERG) responses were recorded according to the International Society for Clinical Electrophysiology of Vision standards using a mini-Ganzfeld bowel and an electroretinography system. The pupils were dilated, and a ground silver wire electrode was attached to the animal's forehead, and a second electrode was placed adjacent to the inferior limbus. After 30 minutes of dark adaptation, scotopic responses were recorded. The rabbits were placed in a lit room for 10 minutes, and photopic responses were recorded after a flash light intensity of 2.0 cd.s/m2. Electroretinographic changes were considered significant if the post-injection differences in a- and b-wave amplitudes were 20% or greater compare to the baseline values.
Statistical testing was performed using SPSS for Windows software (version 16; SPSS, Inc., Chicago, IL). Variables are expressed as mean ± standard deviation. Analysis of variance was used to evaluate the significance of differences among groups. Independent t-test was used to evaluate statistical significance in the difference between means of two groups. P values less than .05 were considered to be statistically significant.
This study demonstrated intravitreal ziv-aflibercept was generally well tolerated in rabbits. No clinical signs or changes in food consumption or body weight attributable to the drug were observed. None of the cases showed endophthalmitis, uveitis, retinal detachment, or any other adverse reactions. At both 1 day and 3 days after injection, conjunctival hyperemia was obvious, which was partly due to conjunctival washing by povidione iodine 5% before injection.
Bioavailability of Intravitreal Ziv-Aflibercept
The concentrations of ziv-aflibercept in the vitreous humor after intravitreal administration are shown in Figure 1. Remaining vitreal concentrations of ziv-aflibercept injection for the 625 μg/mL group were 416, 349, 124, 41.2, and 18.1 (±10 μg/mL) and for the 1,250 (μg/mL) group were 833, 737, 284, 87.3, and 38.2 (± 10 μg/mL), at 0, 24, 168, 336, and 720 hours after injection, respectively. Ziv-aflibercept concentration in the vitreous humor after intravitreal injection into the treated 1.25 mg/0.05 mL and 0.625 mg/0.05 mL eyes is shown in Figure 1.
Ziv-aflibercept concentration in the vitreous humor after intravitreal injection into the treated 1.25 mg/0.05mL (blue line) and 0.625 mg/0.05 mL (orange line) eyes.
The vitreous concentration of aflibercept was analyzed by one-compartment model. The area under curve from time 0 to the end point (AUC last) was 147,637 hours × μg/mL for the full-dose group (1,250 μg/0.05 mL) and 68,498 hours × μg/mL for the half-dose group (625 μg/0.05 mL). The assessed vitreous half-life of ziv-aflibercept was 113 hours in both groups.
Light Microscopic Examination
Specimens from ziv-aflibercept group were evaluated in simulation to control eyes. Ziv-aflibercept treated eyes illustrated the same microscopic appearance in most specimens. The RPE cells were tightly adherent to each other with eminent nuclei. The different retinal layers were in attendance with the control specimens. No enucleated eye showed remarkable retinal changes. Light micrograph of 1.25 mg/mL intravitreal ziv-aflibercept-treated rabbit eye in 1, 7, 14, and 28 days is shown in Figure 2.
Light micrograph of 1.25 mg/mL intravitreal ziv-aflibercept-treated rabbit eye in (A) 1 day, (B) 7 days, (C) 14 days, and (D) 28 days. Micrographs show retinal cellular integrity and nuclei were prominent. No necrosis or atrophy were seen.
In ERG evaluations, all ERG records show changes less than 20% of pre-injection records. Only in one case, on the seventh day after IVZ injection, were changes more than 20% from the baseline observed.
Precise intraocular drug half-lives permit physicians to exactly foretell drug clearance periods when they have to change patients' treatment modality. To the best of our knowledge, he pharmacokinetics properties of complete or half dose of intravitreal ziv-aflibercept injection has not been studied yet. The present study investigated the safety of IVZ injection, as well as the corresponding bioavailability properties of the drug in rabbit model. The results showed that the half-life of intravitreally administered ziv-aflibercept was 113 hours (4.71 days) in the vitreous, indicating the slow clearance and long vitreous half-life of drug.11
The vitreous half-life of ziv-aflibercept in the present study (4.71 days) was relatively longer in comparison with that reported in the pharmacokinetics study of intravitreal aflibercept (3.92 days).12 This vitreous half-life is nearly equivalent to pharmacokinetic properties of aflibercept given by the drug manufacturer (4.79 days).13 These diversities may owe to factors including performance, vitreous sampling, and drug assay technique. According to the previously published mathematical model that calculated intravitreal drug half-lives in rabbits, the most important determinant of drug half-life is molecular size and less important factors: lipophilicity, drug solubility, dose, salt form factor, and eye pigmentation 14. Therefore, the longer half-life of ziv-aflibercept than that of ranibizumab (MW-48 kD) may be explained by the heavier molecular weight of aflibercept (115kD).
Ziv-aflibercept with the same mechanism of action of aflibercept, but with different osmolarity, has the potential of intraocular use despite the report of Marmor,14,15 which indicated the retinal toxicity and even retinal detachment. The osmolarity of ziv-aflibercept is considerably more than aflibercept (815 to 820 mOsm vs. 250 to 260 mOsm, respectively).11,15
After injection of 0.05 mL of a solution with osmolarity greater than 500 mOsm, it was showed no difference between groups receiving Ziv-aflibercept or aflibercept, neither histopathologically, nor electrophysiologically.11 Costa de Andrade et al. have reported no significant change in all viability of both human primary RPE and Müller-glia cells, exposed to clinical concentrations of aflibercept (40 mg/mL) and ziv-aflibercept (25 mg/mL) and even progressive concentrations of NaCl up to 10,000 mOsm/kg.16
It should also be mentioned that the vitreous volume of the rabbit is one-half to one-third of the human vitreous and, hence, we may have tested the two- or three-fold the concentration used for human vitreous in this experimental model. It is not so far to conclude that ziv-aflibercept by 2-3-folds concentration used recently is shown to be safe.
In our study light microscopic examination of H & E slides showed similar findings with unremarkable retinal changes in both ziv-aflibercept and control groups (at 1, 7, 14, and 28 days). Tight adherence of RPE cell to each other with eminent nuclei along with regular other retinal layers was evident. Normal limit for electrophysiological recordings was considered less than 20% of the pre-injection records. In our study, all measured photopic and scotopic ERG findings, including a-wave, b-wave amplitudes, and latencies, were within predetermined normal limits. Only in one rabbit, on the seventh day after IVZ injection, did ERG records show changes of more than 20% from pre-injection records. Regarding the report of Dias et al.,11 which indicated the safety of 25 mg/mL (1.25 mg/0.05 mL) ziv-aflibercept, our study examined the safety of two doses (1.25 mg and 0.625 mg) of intravitreal ziv-aflibercept in rabbit eyes, which proved to be safe, both by histopathologic and electrophysiologic measures. Additionally, we evaluated the bioavailability of ziv-aflibercept at 1, 7, 14, and 30 days.
The ELIZA kit we used is often used to monitor VEGF levels in exudative AMD. In previous reports, it has been mentioned that aflibercept binds to a receptor-binding region of VEGF. Capture antibodies are presumed to bind to the receptor-binding region, and anti-VEGF antibody drugs interfere with the capture antibodies, leading to evaluation of the actual concentration of the determined drug. Hence, this kit can be used for different anti-VEGF drugs.17
In conclusion, this study presents, for the first time, precise pharmacokinetic properties of IVZ, by meticulous enzyme-linked immunosorbent assay in a rabbit model. The finding also affirmed IVZ will be safe for intraocular tissue. We anticipate that our consequences could assist IVZ treatment as a cost-effective therapeutic option for the treatment of retinal vascular diseases.
- Otani A, Takagi H, Oh H, et al. Vascular endothelial growth factor family and receptor expression in human choroidal neovascular membranes. Microvasc Res. 2002;64(1):162–169. https://doi.org/10.1006/mvre.2002.2407 PMID: doi:10.1006/mvre.2002.2407 [CrossRef]12074642
- Ferrara N. Vascular endothelial growth factor. Arterioscler Thromb Vasc Biol. 2009;29(6):789–791. https://doi.org/10.1161/ATVBAHA.108.179663 PMID: doi:10.1161/ATVBAHA.108.179663 [CrossRef]19164810
- Tah V, Orlans HO, Hyer J, et al. Anti-VEGF therapy and the retina: an update. J Ophthalmol. 2015;2015:627674. doi doi:10.1155/2015/627674 [CrossRef] PMID: 26417453
- Stewart MW. Clinical and differential utility of VEGF inhibitors in wet age-related macular degeneration: focus on aflibercept. Clin Ophthalmol. 2012;6:1175–1186. https://doi.org/10.2147/OPTH.S33372 PMID: doi:10.2147/OPTH.S33372 [CrossRef]22973088
- Matthews W, Jordan CT, Gavin M, Jenkins NA, Copeland NG, Lemischka IR. A receptor tyrosine kinase cDNA isolated from a population of enriched primitive hematopoietic cells and exhibiting close genetic linkage to c-kit. Proc Natl Acad Sci USA. 1991;88(20):9026–9030. https://doi.org/10.1073/pnas.88.20.9026 PMID: doi:10.1073/pnas.88.20.9026 [CrossRef]1717995
- Brown DM, Kaiser PK, Michels M, et al. ANCHOR Study Group. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006;355(14):1432–1444. https://doi.org/10.1056/NEJMoa062655 PMID: doi:10.1056/NEJMoa062655 [CrossRef]17021319
- Tolentino M. Systemic and ocular safety of intravitreal anti-VEGF therapies for ocular neovascular disease. Surv Ophthalmol. 2001;56(2):95–113. https://doi.org/10.1016/j.survophthal.2010.08.006 PMID: doi:10.1016/j.survophthal.2010.08.006 [CrossRef]
- Browning DJ, Kaiser PK, Rosenfeld PJ, Stewart MW. Aflibercept for age-related macular degeneration: a game-changer or quiet addition?Am J Ophthalmol. 2012;154(2):222–226. https://doi.org/10.1016/j.ajo.2012.04.020 PMID: doi:10.1016/j.ajo.2012.04.020 [CrossRef]22813448
- Mansour AM, Al-Ghadban SI, Yunis MH, El-Sabban ME. Ziv-aflibercept in macular disease. Br J Ophthalmol. 2015;99(8):1055–1059. https://doi.org/10.1136/bjophthalmol-2014-306319 PMID: doi:10.1136/bjophthalmol-2014-306319 [CrossRef]25677668
- Marmor MF. Retinal detachment from hyperosmotic intravitreal injection. Invest Ophthalmol Vis Sci. 1979;18(12):1237–1244. PMID:116971
- de Oliveira Dias JR, Badaró E, Novais EA, et al. Preclinical investigations of intravitreal ziv-aflibercept. Ophthalmic Surg Lasers Imaging Retina. 2014;45(6):577–584. https://doi.org/10.3928/23258160-20141118-15 PMID: doi:10.3928/23258160-20141118-15 [CrossRef]25423640
- Park SJ, Choi Y, Na YM, et al. Intraocular pharmacokinetics of intravitreal aflibercept (eylea) in a rabbit model. Invest Ophthalmol Vis Sci. 2016;57(6):2612–2617. https://doi.org/10.1167/iovs.16-19204 PMID: doi:10.1167/iovs.16-19204 [CrossRef]27258433
- Stewart MW, Rosenfeld PJ. Predicted biological activity of intravitreal VEGF Trap. Br J Ophthalmol. 2008;92(5):667–668. https://doi.org/10.1136/bjo.2007.134874 PMID: doi:10.1136/bjo.2007.134874 [CrossRef]18356264
- Durairaj C, Shah JC, Senapati S, Kompella UB. Prediction of vitreal half-life based on drug physicochemical properties: quantitative structure-pharmacokinetic relationships (QSPKR). Pharm Res. 2009;26(5):1236–1260. https://doi.org/10.1007/s11095-008-9728-7 PMID: doi:10.1007/s11095-008-9728-7 [CrossRef]
- Braimah IZ, Agarwal K, Mansour A, Chhablani JZiv-aflibercept Study Group. One-year outcome of intravitreal ziv-aflibercept therapy for non-responsive neovascular age-related macular degeneration. Br J Ophthalmol. 2018;102(1):91–96. https://doi.org/10.1136/bjophthalmol-2017-310318 PMID: doi:10.1136/bjophthalmol-2017-310318 [CrossRef]
- Costa de Andrade G, Wertheimer C, Eibl K, et al. Viability of Primary Human Pigment Epithelium Cells and Muller-Glia Cells after Intravitreal Ziv-Aflibercept and Aflibercept. Ophthalmologica. 2016;236(4):223–227. https://doi.org/10.1159/000452677 PMID: doi:10.1159/000452677 [CrossRef]27997919
- Takahashi H, Nomura Y, Nishida J, Fujino Y, Yanagi Y, Kawashima H. Vascular Endothelial Growth Factor (VEGF) Concentration Is Underestimated by Enzyme-Linked Immunosorbent Assay in the Presence of Anti-VEGF Drugs. Invest Ophthalmol Vis Sci. 2016;57(2):462–466. https://doi.org/10.1167/iovs.15-18245 PMID: doi:10.1167/iovs.15-18245 [CrossRef]26868748