PHARMACOKINETIC STUDY OF
`INTRAVITREAL AFLIBERCEPT IN
`HUMANS WITH NEOVASCULAR
`AGE-RELATED MACULAR
`DEGENERATION
`
`DIANA V. DO, MD,* WILLIAM RHOADES, MD,† QUAN DONG NGUYEN, MD, MSC*
`
`Purpose: To investigate the half-life of aflibercept in aqueous humor after a single
`intravitreal injection in patients with neovascular age-related macular degeneration.
`Methods: Prospective, noncomparative, interventional case series of five eyes with
`neovascular age-related macular degeneration naive to anti–vascular endothelial growth
`factor therapy were enrolled and treated with intravitreal aflibercept. At baseline, best-
`corrected visual acuity, optical coherence tomography imaging, and aqueous humor (treat-
`ment eye) and blood/plasma samples were taken. Patients underwent best-corrected visual
`acuity, optical coherence tomography imaging, and sampling of aqueous humor from the
`eye and blood/plasma at six additional post-treatment time points of 4 hours and Days 1, 3,
`7, 14, and 28. Concentrations of aflibercept were quantified using an enzyme-linked immu-
`nosorbent assay.
`Results: Median peak concentration (Cmax) of free aflibercept in the aqueous was 122
`mg/L. The median half-life of free aflibercept was 11 days in the eye. In plasma, the
`concentrations of free aflibercept were low and transient, reaching undetectable levels
`during the first week after injection, and undetectable in all patients at time points beyond 7
`days.
`Conclusion: The pharmacokinetic profile in the aqueous humor described here together
`with the previously reported affinity of aflibercept for vascular endothelial growth factor is
`consistent with and adds to our understanding for the duration of its clinical efficacy.
`RETINA 00:1–5, 2019
`
`Intravitreal antiangiogenic therapy is currently the
`
`primary treatment for neovascular age-related macu-
`lar degeneration (AMD). Antiangiogenic agents used
`to treat neovascular AMD target vascular endothelial
`growth factor (VEGF), a positive regulator of angio-
`
`From the *Byers Eye Institute, Stanford University School of
`Medicine, Palo Alto, California; and †Associated Retinal Consul-
`tants, Grand Rapids, Michigan.
`Supported by Regeneron Pharmaceuticals through an investiga-
`tor initiated study designed and conducted by D. V. Do.
`D. V. Do reports consulting fees from Clearside Biomedical Inc,
`Genentech, Kodiak Sciences, Regeneron, and Santen; and research
`funding from Genentech, Regeneron, and Santen. Q. D. Nguyen
`reports advisory board fees from Regeneron.
`Data from this study were presented in part at the Retina Sub-
`specialty Day held in conjunction with the American Academy of
`Ophthalmology Annual Meeting, New Orleans, LA, 2017.
`Reprint
`requests: Diana V. Do, MD, Byers Eye Institute,
`Stanford University School of Medicine, 2370 Watson Court, Suite
`228, Palo Alto, CA 94303; e-mail: dianado@stanford.edu
`
`genesis that has been implicated in the pathogenesis of
`neovascular AMD.1,2 Among the VEGF isoforms,
`VEGF-A is known to be most strongly linked to
`angiogenesis.3 Therapeutic VEGF-A inhibitors such
`as intravitreal aflibercept and intravitreal ranibizumab
`have been approved by the Food and Drug Adminis-
`tration (FDA) for use in neovascular AMD patients. In
`addition, bevacizumab, which is indicated for systemic
`therapy of colorectal cancer,
`is frequently com-
`pounded and prescribed off-label to treat choroidal
`neovascularization.4,5
`Information on the pharmacokinetics of intravitreal
`VEGF inhibitors is clinically useful and can be
`supportive in the understanding of dosing intervals
`in patients. Published studies have evaluated the
`pharmacokinetics of both ranibizumab and bevacizu-
`mab in humans.6,7 In these studies, eyes with both
`
`1
`
`Copyright © by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
`
`Mylan Exhibit 1143
`Mylan v. Regeneron, IPR2021-00881
`Page 1
`
`

`

`2
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2019  VOLUME 00  NUMBER 00
`
`macular edema and visually significant cataract were
`treated with either bevacizumab or ranibizumab. Then
`aqueous humor samples were obtained from each sub-
`ject at a single time point during cataract surgery.
`Based on the varying distribution of time intervals
`between injection and aqueous harvest in the individ-
`ual patients, the investigators, based on the totality of
`the data with only a single determination per patient,
`estimated a half-life of 7.19 days for ranibizumab and
`9.82 days for bevacizumab.
`Aflibercept
`is a fusion protein, which acts as
`a soluble decoy receptor that binds VEGF-A, VEGF-
`B, and placental growth factor (another member of the
`VEGF family).8 These two angiogenic factors have
`been found to stimulate endothelial cell proliferation
`and migration, as well as vascular permeability.9 Phase
`3 clinical trials in neovascular AMD have demon-
`strated that every 8-week intravitreal dosing of 2-mg
`aflibercept (Eylea) injection is noninferior to both
`monthly dosing of 2-mg aflibercept
`(Eylea) and
`monthly dosing of 0.5-mg ranibizumab.10
`Little information is known about the intraocular
`pharmacokinetics of aflibercept in human eyes. We
`conducted a pharmacokinetic study of aflibercept
`injection in nonvitrectomized eyes with wet AMD to
`characterize the intraocular and systemic concentra-
`tions of free and bound aflibercept in humans with
`neovascular AMD. We sampled aqueous at multiple
`times after injection in the treatment eye in each of the
`five patients, thus permitting characterization of the
`pharmacokinetics including half-life of free aflibercept
`in aqueous humor. This research will fill in these gaps
`in the knowledge base for
`intravitreal aflibercept
`injection.
`
`Methods
`
`This clinical research was approved by the Institu-
`tional Review Board and conducted at the University
`of Nebraska Medical Center. Eligible participants had
`a diagnosis of new-onset neovascular AMD without
`previous treatment in the study eye. In addition, no
`active neovascular AMD or anti-VEGF treatment
`could be present
`in the fellow eye. Subjects were
`followed for a 28-day period and had 6 study visits
`(baseline, Days 1, 3, 7, 14, and 28). All study eyes
`underwent the following procedures at baseline and at
`Day 28 measurement of best-corrected visual acuity
`(BCVA), dilated fundoscopic examination,
`fundus
`photograph, fluorescein angiography, and optical
`coherence tomography (OCT). In addition, anterior
`chamber paracentesis and a blood draw were per-
`formed at enrollment and before Eylea was given to
`
`free and bound
`determine the concentrations of
`concentrations of aflibercept at baseline in aqueous
`humor and plasma. After baseline ocular and plasma
`samples were obtained, intravitreal aflibercept injec-
`tion (2 mg) into the study eye was administered and
`considered to be Time 0. Four hours after intravitreal
`aflibercept
`injection, anterior chamber paracentesis
`and blood draw were repeated. These two procedures,
`as well as BCVA and OCT, were also performed on
`each study visit.
`Aflibercept concentrations in aqueous humor and
`blood plasma samples were quantified in triplicate by
`enzyme-linked immunosorbent assay (ELISA).
`
`Results
`
`Five subjects provided informed consent and were
`enrolled, leading to five study eyes with new-onset wet
`AMD. The median age was 78 years. All patients were
`treatment-naive and had no previous vitrectomy
`surgery. The median baseline BCVA was 68 letters
`(Snellen equivalent of 20/50). In addition, the mean
`OCT central subfield retinal thickness was 304 mm
`(using the Heidelberg Spectralis machine). Four of
`the five subjects completed all study visits. One sub-
`ject did not have aqueous fluid collection at Day 28,
`and that missing data point was not included in the
`Day 28 analyses.
`At baseline (before intravitreal aflibercept was
`administered), there were no detectable concentrations
`of free or bound aflibercept in the aqueous humor or
`plasma. Four hours after the aflibercept injection, the
`median peak concentration (Cmax) of free aflibercept in
`the aqueous was 64.4 mg/L. At Days 1, 3, 7, 14, and
`28, the median concentrations of free aflibercept in the
`aqueous were 51.9, 35.3, 36.9, 17.1, and 9.4 mg/L
`(Table 1). The half-life of free aflibercept in the aque-
`ous humor, calculated from the median values at each
`time point, was approximately 11 days. Adjusted
`bound aflibercept in the aqueous was undetectable at
`all time points.
`At the 4-hour mark, the median plasma concentra-
`tion of free and adjusted bound aflibercept were 0 mg/
`L. At Days 1, 3, 7, 14, and 28, the median plasma
`concentrations of free aflibercept were 0.03, 0.01, and
`undetectable from Day 7 onward. Similarly,
`the
`median concentration of adjusted bound aflibercept
`in the plasma was 0, 0.08, 0.11, 0.17, and 0.18 mg/L at
`those respective time points (Table 2).
`The median OCT retinal thickness decreased to 252
`mm one day after aflibercept injection. Subsequently,
`the median OCT thickness was 235, 233, 218, and
`218 mm at Days 3, 7, 14, and 28. Two of the subjects
`
`Copyright © by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
`
`Mylan Exhibit 1143
`Mylan v. Regeneron, IPR2021-00881
`Page 2
`
`

`

`PHARMACOKINETICS OF AFLIBERCEPT  DO ET AL
`
`3
`
`Table 1. Concentrations of Free Aflibercept in Aqueous
`Humor by Time After a Single Intravitreal Injection of
`Aflibercept
`
`TE002
`0
`24.8
`38
`
`Time
`TE003 TE004 TE005 Median
`(Days) TE001
`0
`0
`0
`0
`0
`0
`222
`18.3
`129
`64.4
`0.166
`104
`65.9
`24.5
`106
`51.95
`1
`122
`15.4
`35.3
`67.6
`35.3
`3
`88.5
`10.8
`31.9
`42.9
`36.95
`44.6
`7
`42
`4.46
`20.3
`29.1
`17.15
`28.6
`14
`12.9
`28
`0.793 8.770001 NA
`9.92
`13.1
`9.4
`
`Baseline treated as zero.
`All concentrations reported as mg/L.
`NA, not available.
`
`had a more rapid decline in aflibercept levels in the
`aqueous, which was associated with a smaller reduc-
`tion in their central subfield thickness measurement
`(Figure 1). The median BCVA was 74 letters, 72
`letters, 71 letters, 73 letters, and 73 letters (Snellen
`equivalent of 20/32, 20/40, 20/40, 20/40, and 20/40,
`respectively) at the time points described above.
`
`Discussion
`
`This study provides novel information on the half-
`life of free aflibercept in the aqueous humor from
`patients with neovascular AMD. To the best of our
`knowledge, this research is the first study to inves-
`tigate the pharmacokinetics of aflibercept in human
`eyes with serial
`intensive sampling. Our
`study
`estimated pharmacokinetic parameters for free afli-
`bercept in the aqueous humor including the median
`half-life, which was estimated to be approximately 11
`days, which is longer than the reported half-life of
`ranibizumab (7 days) and bevacizumab (9 days). In
`addition, the concentration of free aflibercept in the
`aqueous is 1,000-fold greater
`than observed in
`plasma.
`The range of half-lives among the five subjects is
`interesting. Perhaps,
`the variability results from
`differential vitreal liquefaction in different eyes or
`the presence of additional exit paths in some eyes.
`We may speculate that variability in duration of
`treatment among patients may be related to persis-
`tence of high levels of aflibercept in some eyes and
`not others. However, these are speculations, and the
`mechanism and significance of this variability in half-
`lives are currently unknown and remain the subject
`for future investigation. It seems unlikely that the
`difference in half-lives among patients is due to
`difference in disease burden,
`that
`is, differential
`target-mediated clearance in the eye because some
`
`Table 2. Concentrations of Free Aflibercept in Plasma by
`Time After a Single Intravitreal Injection of Aflibercept
`
`Time
`(Days) TE001 TE002 TE003 TE004 TE005 Median
`0
`0
`0
`0
`0
`0
`0
`0.166
`0
`0
`0.0218
`0
`0
`0
`1
`0.0342
`0
`0.064
`0
`0.0456 0.0342
`3
`0.0582
`0
`0.0195
`0
`0.0239 0.0195
`7
`0.0283
`0
`0
`0
`0
`0
`14
`0
`0
`0
`0
`0
`0
`28
`0
`0
`NA
`0
`0
`0
`
`All concentrations reported as mg/L.
`Baseline is treated as 0.
`NA, not available.
`
`patients are producing so much more VEGF. If this
`were the case, we would see differential accumulation
`of VEGF-bound aflibercept in the aqueous correlated
`with half-life. In fact, levels of bound aflibercept in
`the aqueous are below the level of quantitation in all
`patients at all time points. It is also of interest to note
`that no free aflibercept at all was detected in the
`
`Fig. 1. The figure illustrates the concentration of free aflibercept (mg/L)
`in the aqueous humor over time for each of the five subjects after
`a single intravitreal injection of 2-mg aflibercept on Day 0. The lower
`figure illustrates the percent change in OCT central subfield thickness
`from baseline for each of the subjects. Two of the subjects had a more
`rapid decline in aflibercept levels in the aqueous, which was associated
`with a smaller reduction in their central subfield thickness measure-
`ments.
`
`Copyright © by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
`
`Mylan Exhibit 1143
`Mylan v. Regeneron, IPR2021-00881
`Page 3
`
`

`

`4
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2019  VOLUME 00  NUMBER 00
`
`plasma of patients that never achieved aqueous Cmax
`values of 100 mg/L. This may indicate that
`the
`primary exit route for aflibercept is through aqueous
`outflow, and detectable plasma levels are related to
`peak aqueous levels.
`Concentrations of free aflibercept in the plasma
`were low and transient and were undetectable 1 week
`after administration. The extremely low and transient
`concentrations of
`free aflibercept
`in the plasma
`suggest that aflibercept is unlikely to bind to circu-
`lating VEGF to appreciable extent in the plasma. To
`date, clinical trials of intravitreal VEGF inhibitors
`have not been sufficiently powered to detect small
`differences in rare adverse events, such as arterial-
`thrombotic events. Although Avery et al suggested
`that monthly aflibercept or
`ranibizumab possibly
`increases the risk of death and cerebral vascular
`accidents in DME patients, other systematic reviews
`and clinical trials have not shown any increased risk
`of serious systemic adverse events associated with
`ranibizumab.11–13
`aflibercept,
`bevacizumab,
`or
`Recently, Jampol et al13 in the Diabetic Retinopathy
`Clinical Research (DRCR) Network evaluated sys-
`temic VEGF levels in subjects treated with intravi-
`treal VEGF blockers (aflibercept, ranibizumab, and
`bevacizumab) for diabetic macular edema and found
`no data to suggest that patients with lower VEGF
`blood levels were at an increased risk of systemic
`complications.
`Changes in BCVA and OCT retinal thickness were
`variable among the five study eyes. The baseline
`median BCVA was of 68 letters (20/40), and retinal
`thickness (349 mm) indicates that this new-onset cho-
`roidal neovascularization was detected early in these
`eyes. Some of the study eyes had more rapid reduc-
`tions in retinal thickness, and there was a trend to
`suggest that eyes which had higher concentrations of
`free aflibercept in the aqueous had more robust reduc-
`tions in OCT compared with eyes in which aflibercept
`exited the eye more rapidly. We hypothesize that a con-
`stant, high concentration of the drug in the eye likely
`provides more VEGF inhibition over a longer period,
`and leads to less fluid on the OCT.8 However, normal
`variability of drug levels in humans can also contribute
`to different anatomical responses on OCT, and larger
`sample sizes are necessary to truly confirm the corre-
`lation between aflibercept
`levels and OCT retinal
`thickness changes.
`This study has several strengths. Serial samples of
`intraocular fluid and blood plasma were obtained
`from the same five subjects.
`In addition, high
`frequency of
`sampling (7 specimens) during
`a 28-day period provide for greater data points on
`the pharmacokinetics of aflibercept. This design
`
`permitted the calculation of the half-life of the drug
`in the eye.
`We recognize there are some limitations with this
`research. The sample size of five subjects is small. In
`addition, longer follow-up of subjects with additional
`data beyond 28 days would provide more long-term
`data on how free and bound aflibercept concentration
`changes after multiple intravitreal aflibercept injec-
`tions. Systemic VEGF levels were not measured in
`this study, and we recognize that
`information on
`systemic VEGF levels could be helpful in identifying
`if aflibercept has any effect on systemic VEGF
`blockade.
`This research provides useful information on the
`pharmacokinetics of aflibercept
`in patients with
`neovascular AMD. This new information compli-
`ments the existing medical literature on intravitreal
`aflibercept and gives ophthalmologists new insights
`into the ocular half-life on this drug. Additional
`pharmacokinetic studies in subjects with different
`retinal diseases, such as diabetic macular edema and
`diabetic retinopathy, or with previous vitrectomy
`may be helpful
`in determining how aflibercept
`concentrations vary in patients with these ocular
`disorders.
`Key words: age-related macular degeneration, phar-
`macokinetics, half-life, aflibercept, intravitreal VEGF
`inhibitors, VEGF.
`
`References
`
`1. Miller JW, Adamis AP, Aiello LP. Vascular endothelial
`growth factor in ocular neovascularization and prolifera-
`tive diabetic retinopathy. Diabetes Metab Rev 1997;13:
`37–50.
`2. Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial
`growth factor in ocular fluid of patients with diabetic retinop-
`athy and other retinal disorders. N Engl J Med 1994;331:1480–
`1487.
`3. Bhisitkul RB. Vascular endothelial growth factor biology: clin-
`ical implications for ocular treatments. Br J Ophthalmol 2006;
`90:1542–1547.
`4. Epstein P. Trials that matter: two faces of progress in the
`treatment of age-related macular degeneration. Ann Intern
`Med 2007;146:532–534.
`5. Brechner RJ, Rosenfeld PJ, Babish JD, Caplan S. Pharmaco-
`therapy for neovascular age-related macular degeneration: an
`analysis of the 100% 2008 medicare fee-for-service part B
`claims file. Am J Ophthalmol 2011;151:887–895.
`6. Krohne TU, Eter N, Holz FG, Meyer CH. Intraocular pharma-
`cokinetics of bevacizumab after a single intravitreal injection in
`humans. Am J Ophthalmol 2008;146:508–512.
`7. Krohne TU, Liu Z, Holz FG, Meyer CH. Intraocular pharma-
`cokinetics of bevacizumab after a single intravitreal injection in
`humans. Am J Ophthalmol 2012;154:682–686.
`8. Stewart MW, Rosenfeld PJ. Predicted biological activity
`of intravitreal VEGF Trap. Br J Ophthalmol 2008;92:667–
`668.
`
`Copyright © by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
`
`Mylan Exhibit 1143
`Mylan v. Regeneron, IPR2021-00881
`Page 4
`
`

`

`PHARMACOKINETICS OF AFLIBERCEPT  DO ET AL
`
`5
`
`9. Witmer AN, Vrensen GFJM, Van Noorden CJF, Schlingemann
`RO. Vascular endothelial growth factors and angiogenesis in
`eye disease. Prog Retin Eye Res 2003;22:1–29.
`10. Stewart MW. Aflibercept (VEGF Trap-eye): the newest anti-
`VEGF drug. Br J Ophthalmol 2012;96:1157–1158.
`11. Avery RL, Gordon GM. Systemic safety of prolonged monthly
`anti-vascular endothelial growth factor therapy for diabetic
`macular edema: a systematic review and meta-analysis. JAMA
`Ophthalmol 2016;134:21–29.
`
`12. Kitchens JW, Do DV, Boyer DS, et al. Comprehensive review
`of ocular and systemic safety events with intravitreal afliber-
`cept injection in randomized controlled trials. Ophthalmology
`2016;123:1511–1520.
`13. Jampol LM, Glassman AR, Liu D; DRCR Network. Plasma
`vascular endothelial growth factor concentrations after in-
`travitreous anti-vascular endothelial growth factor therapy
`for diabetic macular edema. Ophthalmology 125:1054–
`1063.
`
`Copyright © by Ophthalmic Communications Society, Inc. Unauthorized reproduction of this article is prohibited.
`
`Mylan Exhibit 1143
`Mylan v. Regeneron, IPR2021-00881
`Page 5
`
`