Suppression of Intraocular Vascular Endothelial Growth
`Factor During Aflibercept Treatment of Age-Related
`Macular Degeneration
`
`SASCHA FAUSER, VIKTORIA SCHWABECKER, AND PHILIPP S. MUETHER
`
` PURPOSE: To determine the duration of suppression of
`aqueous humor concentrations of vascular endothelial
`growth factor
`(VEGF)
`in eyes with neovascular
`age-related macular degeneration (AMD) treated with
`aflibercept.
` DESIGN: Nonrandomized prospective clinical study.
` METHODS: Twenty-seven eyes of 27 neovascular AMD
`patients receiving intravitreal aflibercept injections on a pro
`re nata regimen driven by spectral-domain optical coher-
`ence tomography (SD OCT) were included in this study.
`A total of 132 aqueous humor specimens were collected
`before intravitreal aflibercept injections and their VEGF-
`A concentrations assayed by multiplex bead analysis.
` RESULTS: Mean aqueous humor VEGF concentrations
`before treatment initiation were 90.6 ± 37.1 pg/mL
`(range 23.4–190.3 pg/mL). Intravitreal injection of afli-
`bercept suppressed the aqueous VEGF concentrations
`to below the lower limit of quantification (<4 pg/mL)
`in all patients. The mean duration of VEGF suppression
`below the lower limit of quantification was >71 ±
`18 days. The earliest time after injection at which the
`VEGF concentration recovered to above the lower limit
`of quantification was 55 days
`in 1 patient and
`>56 days, the recommended aflibercept treatment inter-
`val, in 20 patients. The aqueous VEGF recovery status of
`6 patients was uncertain after 56 days.
` CONCLUSIONS: On average, VEGF concentrations in
`the aqueous humor were suppressed below the lower limit
`of quantification after intravitreal aflibercept injections
`for about 10 weeks. This aqueous suppression time sug-
`gests durable VEGF inhibition for most patients dosed
`with aflibercept every 8 weeks.
`(Am J Ophthalmol
`2014;158:532–536. Ó 2014 by Elsevier Inc. All rights
`reserved.)
`
`A GE-RELATED MACULAR DEGENERATION (AMD) IS
`
`a major cause of vision loss. The neovascular
`variant is characterized by choroidal neovascula-
`rization (CNV), in which formation of blood vessels leads
`
`Accepted for publication May 22, 2014.
`From the Department of Ophthalmology, University Hospital of
`Cologne, Cologne, Germany.
`Inquiries to Philipp S. Muether, Department of Ophthalmology,
`University of Cologne, Kerpener Strasse 62, 50924 Cologne, Germany;
`e-mail: philmuether@mac.com
`
`to sub- and intraretinal macular edema, hemorrhage, fibrosis,
`and visual decay. Effective treatments have been developed
`recently,
`focusing on neutralizing vascular endothelial
`growth factor (VEGF) with antibodies (bevacizumab), anti-
`body fragments (ranibizumab), or fusion proteins (afliber-
`cept). Major clinical
`trials
`found 4-weekly injections
`of ranibizumab to result in best visual outcome for ranibizu-
`mab,1 and found that injections of aflibercept every 8 weeks
`(following a loading phase) provided similar functional ben-
`efits.2 As an alternative to fixed dosing intervals, pro re nata
`(PRN) treatments based on optical coherence tomography
`(OCT)-determined activity achieve similar
`functional
`results.3,4
`Clinical trials with aflibercept suggest a longer duration
`of VEGF suppression than with bevacizumab or ranibizu-
`mab, which is also supported by pharmacokinetic models.5
`We have recently determined the average time for which
`aqueous humor VEGF concentrations are suppressed below
`the lower limit of quanitification of 4 pg/mL following
`intravitreal ranibizumab injections to be 37 days on
`average, with individual VEGF suppression times ranging
`from 26 to 69 days.6,7 Aqueous humor concentrations
`appear suitable for assessing ocular VEGF levels as they
`correlate well with vitreous VEGF concentrations,
`extrapolated from retinal vascular occlusive disease and
`diabetic retinopathy.8,9
`This study aimed to determine intraocular VEGF sup-
`pression duration following aflibercept
`treatment
`for
`neovascular AMD by sampling aqueous humor VEGF
`levels.
`
`MATERIALS AND METHODS
` STUDY POPULATION: This prospective, observational
`study enrolled 27 eyes of 27 patients who were 60 years
`of age or older and had active CNV secondary to AMD.
`All eyes were examined and treated at the Department
`of Ophthalmology, University of Cologne, Germany.
`The study was performed in accordance with the tenets
`of the Declaration of Helsinki. The protocol was approved
`by the Ethics Committee of the University of Cologne
`(reference number 11–027), and all participants gave
`written informed consent. The study was registered at
`ClinicalTrials.gov (Identifier NCT01213667).
`
`532
`
`Ó 2014 BY ELSEVIER INC. ALL RIGHTS RESERVED.
`
`0002-9394/$36.00
`http://dx.doi.org/10.1016/j.ajo.2014.05.025
`
`Mylan Exhibit 1142
`Mylan v. Regeneron, IPR2021-00881
`Page 1
`
`

`

`TABLE. Clinical Characteristics of the Study Population
`Treated With Aflibercept for Neovascular Age-Related
`Macular Degeneration
`
`Study participants
`
`27 patients
`
`Sex, n (%)
`
`Age at first aflibercept treatment
`(y), mean 6 SD (range)
`Eyes, n (%)
`
`Follow-up time per patient (mo),
`mean 6 SD (range)
`Number of intravitreal injections
`per patient, mean 6 SD
`(range)
`Aqueous VEGF concentration on
`day 0 (pg/mL), mean 6 SD
`(range)
`VEGF suppression time (days),
`mean 6 SD (range)
`Type of choroidal
`neovascularization, n
`
`13 (48%) male
`14 (52%) female
`77.5 6 6.4 (64–90)
`
`27 eyes: 9 (33%) right,
`18 (67%) left
`7.7 6 2.1 (4.1–12.0)
`
`5.5 6 1.3 (3–8)
`
`90.6 6 37.1 (23.4–190.3)
`
`70.5 6 18.0 (41–109)
`
`18 occult
`4 mixed
`2 classic
`3 RAP
`4.5 6 3.4 (0.3–12.7)
`
`59.5 6 17.0 (20–85)
`
`392 6 132 (210–658)
`
`Size of choroidal
`neovascularization (mm2),
`mean 6 SD (range)
`Best-corrected visual acuity on
`day 0 (ETDRS letters),
`mean 6 SD (range)
`Central retinal thickness on
`day 0 (mm), mean 6 SD
`(range)
`ETDRS ¼ Early Treatment Diabetic Retinopathy Study;
`SD ¼ standard deviation; RAP ¼ retinal angiomatous prolifera-
`tion; VEGF ¼ vascular endothelial growth factor.
`
` DIAGNOSTICS AND TREATMENT: Patients
`initially
`received 3 2-mg loading dose injections of aflibercept at
`intervals ranging from 4 to 6 weeks. After this first treat-
`ment phase, patients were monitored monthly by SD
`OCT, ETDRS best-corrected visual acuity tests, and fundus
`examinations. Fluorescein angiography was repeated only
`in unclear cases. CNV persistences or recurrences were
`treated by additional aflibercept injections on a PRN
`regimen mainly driven by morphologic findings in SD
`OCT. Recurrent or persistent CNV activity was detected
`as sub- or intraretinal fluid by SD OCT, leakage in fluores-
`cein angiography, a loss of ETDRS letters if attributable to
`CNV activity, or new sub- or intraretinal macular hemor-
`rhages. Because of variable treatment approval times of
`health insurances, individual disease activities, and patient
`appointment preferences, variable reinjection intervals
`occurred without any experimental study design.
` AQUEOUS HUMOR VASCULAR ENDOTHELIAL GROWTH
`FACTOR MEASUREMENTS: Samples were acquired only
`upon necessary treatment. Prior to each aflibercept injection,
`approximately 0.1 mL of aqueous humor was collected via a
`sterile limbal puncture with a 30 gauge needle connected to
`an insulin syringe. The procedure of sample collection
`immediately followed by aflibercept injection was randomly
`performed by 3 surgeons. No surgeon was assigned to specific
`patients, cancelling out possible dosing variabilities. Samples
`were immediately stored at 80 C in polypropylene tubes
`until they were analyzed on a Luminex xMAP microbead
`multiplex platform (Luminex 200; Luminex Inc, Austin,
`Texas, USA) following the manufacturer’s assay instruc-
`tions (Human Angiogenesis Panel; R&D Systems, Wiesba-
`den, Germany). Standard curves for VEGF were generated
`using the reference standard supplied with the kit. The
`lower limit of quantification for VEGF was 4 pg/mL.
`
` INCLUSION AND EXCLUSION CRITERIA: All included
`patients were suffering from an active sub- or juxtafoveal
`CNV attributable to neovascular AMD. This was
`confirmed by fluorescein angiography and indocyanine
`green angiography as well as spectral-domain optical coher-
`ence tomography (SD OCT) (HRA-2 and Spectralis OCT;
`Heidelberg Engineering, Heidelberg, Germany). CNV size
`(mm2) was determined from fluorescein angiograms using
`the HRA-2 software (Heidelberg Engineering). An addi-
`tional inclusion criterion in the study eye was a best-
`corrected visual acuity >_20 Early Treatment of Diabetic
`Retinopathy Study (ETDRS) letters. Exclusion criteria
`were any previous intraocular surgery (apart from cataract
`removal) or photodynamic therapy; any treatment with
`intraocular steroids; any ranibizumab/bevacizumab/pegap-
`tanib treatment within the previous 90 days; and any pre-
`vious aflibercept treatment.
`
`RESULTS
`
`INTRAOCULAR VEGF CONCENTRATIONS WERE ASSAYED IN
`samples of aqueous humor from 27 patients undergoing
`PRN aflibercept treatment for neovascular AMD. The clin-
`ical characteristics of the study population are listed in the
`Table. We analyzed 132 aqueous humor samples of 149
`intravitreal aflibercept injections administered during the
`study. VEGF levels (y-axis) were plotted in relation to
`the interval
`from the previous aflibercept
`injection
`(x-axis) for each patient; the very first aflibercept injection
`was defined as day 0. Representative examples are depicted
`in Figure 1. Complete aqueous humor VEGF suppression
`was assumed when VEGF levels were below the lower limit
`of quantification of the analytical method (4 pg/mL).
`Aflibercept led to complete suppression of aqueous
`VEGF in all patients at early times after injection. As
`
`VOL. 158, NO. 3
`
`VEGF SUPPRESSION DURING AFLIBERCEPT TREATMENT FOR NEOVASCULAR AMD
`
`533
`
`Mylan Exhibit 1142
`Mylan v. Regeneron, IPR2021-00881
`Page 2
`
`

`

`at which other assays of the same individual showed it to
`be present (data not shown).
`When excluding the upload phase, and excluding
`patients with persistent activity, only 6 injections were
`dictated before 8 weeks based on signs of recurrent activity
`on SD OCT. The mean recurrence interval from the previ-
`ous injection in these 6 cases was 49 6 5 days.
`
`DISCUSSION
`
`IN THIS STUDY WE MEASURED VEGF CONCENTRATIONS IN
`the aqueous humor of neovascular AMD patients undergo-
`ing aflibercept treatment. At early times after injection,
`aflibercept completely suppressed aqueous VEGF below
`the lower limit of quantification of our analytical method
`in all 27 patients. The mean duration of complete aqueous
`VEGF suppression was at least 71 days, 2 weeks longer than
`the recommended dosing interval.
`The aflibercept VEGF suppression time of at least
`71 days is far longer than the previously determined
`VEGF suppression time for
`ranibizumab of around
`for neovascular AMD patients,6,7 and thus
`37 days
`provides a plausible rationale for the 56-day injection inter-
`vals recommended by the treatment label. Of the 27
`patients in this study, only 1 patient had definitely lost
`aqueous VEGF suppression after
`less
`than 56 days
`(55 days), but aqueous VEGF was shown to be suppressed
`for at least 48 days (Patient 25; Figure 2). Six patients
`had an uncertain suppression status at 56 days, owing to
`sample availability, but VEGF suppression times of at least
`41 days (Patients 21–24, 26, 27; Figure 2).
`Individual aflibercept VEGF suppression times were sta-
`ble for up to 12 months; no signs of tachyphylaxis or
`rebound effects were observed. No patient had an assayed
`aqueous VEGF level above detection limit at a time point
`that showed suppression below detection limit for another
`specimen of that same patient at an equal time point. We
`have already shown such stability of VEGF suppression
`times for ranibizumab to be stable in neovascular AMD
`patients7 as well as in diabetic macular edema patients.10
`In both groups, VEGF suppression times vary between
`different patients but are constant for each patient. These
`individual differences in VEGF suppression time may be
`attributable to differences in VEGF production as well as
`aflibercept decay and may support individualized therapy.
`The major limitation of this study was the nonexperi-
`mental design, as varying injection intervals were mainly
`based on disease activity, precluding precise definition of
`the exact time at which aqueous VEGF suppression was
`lost. However, as these uncertain VEGF suppression times
`are longer than the latest sample time for many patients,
`the true mean aqueous VEGF suppression time will be
`longer than our currently determined mean of 71 days.
`
`FIGURE 1. Scatter diagrams of 2 patients with neovascular age-
`related macular degeneration, showing the time interval since
`the previous intravitreal aflibercept injection and the corre-
`sponding vascular endothelial growth factor concentrations in
`aqueous humor.
`
`patients were seen monthly and only injected with afliber-
`cept PRN, and subsequently were only assayed for aqueous
`humor VEGF levels at that time, it is difficult to measure
`the precise moment at which VEGF suppression is lost.
`Therefore, the latest sampling time of complete aqueous
`VEGF suppression and the earliest sampling time of
`aqueous VEGF suppression loss was determined for each
`patient; the true duration of aqueous VEGF suppression
`lies between these 2 times. The earliest time at which com-
`plete aqueous VEGF suppression was lost could be deter-
`mined for only 7 of the patients (Figure 2), but not for
`the remaining 20 patients, as the latest available sample
`still showed complete suppression.
`After the recommended dosing interval of 56 days,
`aqueous VEGF levels were still completely suppressed in
`20 patients; 1 patient was definitely no longer suppressed;
`and the suppression status of the remaining 6 patients
`was uncertain owing to nonavailability of samples for this
`time point.
`The mean VEGF suppression time was greater than 71 6
`18 days. All VEGF suppression time data are shown in
`Figure 2. Individual VEGF suppression times were appar-
`ently stable because suppression was never lost at a time
`
`534
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`SEPTEMBER 2014
`
`Mylan Exhibit 1142
`Mylan v. Regeneron, IPR2021-00881
`Page 3
`
`

`

`FIGURE 2. Stacked histogram for individual patients with neovascular age-related macular degeneration, showing the duration of
`complete vascular endothelial growth factor suppression as well as the end of suppression whenever definable. The true duration
`of suppression is within the time period during which the suppression status is uncertain owing to unavailable sampling.
`
`For the same reason, we were also unable to determine the
`exact relationship between aqueous aflibercept VEGF sup-
`pression time and the reoccurrence of clinical CNV activ-
`ity. However, for ranibizumab we have shown the sequence
`of events following the loss of aqueous VEGF suppression
`until the reoccurrence of morphologic, and ultimately
`functional, events.6 It seems plausible that aflibercept
`treatments will show a similar correlation. This assumption
`is further supported by the fact that our reinjection regimen
`was based on clinical disease activity and resulted in injec-
`tion intervals of often more than 8 weeks, indicating that
`the observed aflibercept VEGF suppression times correlate
`to clinical findings.
`Our results are in line with the model predictions
`of Stewart and Rosenfeld,5 in which a slight increase
`in elimination half-times from the vitreous, combined with
`a large increase in binding affinity,11 are taken to be the
`likely explanation for differences in functional half-times.
`However, here, as in our previous work with ranizumab,6
`we have assayed VEGF rather than making deductions based
`on the properties of the different anti-VEGF medications,
`
`permitting a more definite correlation to functional effects.
`Concentrations of VEGF in aqueous humor are known to
`be lower than those in vitreous,9 and those in vitreous are
`themselves are presumably only a diluted reflection of those
`within the retina or in the subretinal space, where VEGF is
`postulated to be functionally angiogenic. Nevertheless, it is
`plausible that the VEGF concentrations in these various
`ocular ‘‘compartments’’ are in reasonable equilibrium, so
`that the deductions we make from the aqueous humor con-
`centrations will reflect those relevant to deeper levels of the
`ocular architecture.
`Importantly, the influence of additional growth factors
`and cytokines apart from VEGF may have to be taken
`into account regarding CNV activity. CNV persistence
`in some patients (never drying up completely) points in
`this direction.
`In conclusion, this work provides clinical data support-
`ing the pharmacokinetic rationale for aflibercept injections
`every 8 weeks in patients with neovascular AMD, or at
`least for longer intervals between aflibercept injections
`than are needed for ranibizumab.
`
`ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST.
`Sascha Fauser received grants from the Gilen and Nolting Foundations, Germany; and from Novartis. He is a consultant to Novartis and received payment
`for lectures from Bayer and Novartis. Philipp Muether reveived payment for lectures from Novartis, Heidelberg Engineering, and Bayer. Viktoria Schwa-
`becker has no financial disclosures. This study did not receive any funding or financial support. Contributions of authors: involved in conception and design
`(S.F., V.S., P.S.M.); data analysis and interpretation (S.F., P.S.M); writing the article (S.F., P.S.M.); critical review of the article (S.F., V.S., P.S.M.); final
`approval of the article (S.F., V.S., P.S.M.); data collection (S.F., V.S., P.S.M.); and statistical expertise (S.F., P.S.M.).
`
`VOL. 158, NO. 3
`
`VEGF SUPPRESSION DURING AFLIBERCEPT TREATMENT FOR NEOVASCULAR AMD
`
`535
`
`Mylan Exhibit 1142
`Mylan v. Regeneron, IPR2021-00881
`Page 4
`
`

`

`REFERENCES
`
`1. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus
`verteporfin for neovascular age-related macular degeneration.
`New Engl J Med 2006;355(14):1432–1444.
`2. Heier JS, Brown DM, Chong V, et al. Intravitreal aflibercept
`(VEGF Trap-Eye) in wet age-related macular degeneration.
`Ophthalmology 2012;119(12):2537–2548.
`3. Chakravarthy U, Harding SP, Rogers CA, et al. Alternative
`treatments to inhibit VEGF in age-related choroidal neovas-
`cularisation: 2-year findings of
`the IVAN randomised
`controlled trial. Lancet 2013;382(9900):1258–1267.
`4. Martin DF, Maguire MG, Ying GS, et al. Ranibizumab and
`bevacizumab for neovascular age-related macular degenera-
`tion. The CATT Research Group. New Engl J Med 2011;
`364(20):1897–1908.
`5. Stewart MW, Rosenfeld PJ. Predicted biological activity of
`intravitreal VEGF trap. Brit J Ophthalmol 2008;92(5):667–668.
`6. Muether PS, Hermann MM, Viebahn U, Kirchhof B,
`Fauser S. Vascular endothelial growth factor in patients
`with exudative age-related macular degeneration treated
`with ranibizumab. Ophthalmology 2012;119(10):2082–2086.
`
`7. Muether PS, Hermann MM, Droge K, Kirchhof B, Fauser S.
`Long-term stability of vascular endothelial growth factor sup-
`pression time under ranibizumab treatment in age-related
`macular degeneration. Am J Ophthalmol 2013;156(5):
`989–993.
`8. Noma H, Funatsu H, Yamasaki M, et al. Aqueous humour
`levels of cytokines are correlated to vitreous levels and
`severity of macular oedema in branch retinal vein occlusion.
`Eye 2008;22(1):42–48.
`9. Funatsu H, Yamashita H, Noma H, et al. Aqueous humor
`levels of cytokines are related to vitreous levels and progres-
`sion of diabetic retinopathy in diabetic patients. Graefes
`Arch Clin Exp Ophthalmol 2005;243(1):3–8.
`10. Muether PS, Droege KM, Fauser S. Vascular endothelial
`growth factor suppression times in patients with diabetic mac-
`ular oedema treated with ranibizumab. Brit J Ophthalmol 2014;
`98(2):179–181.
`11. Papadopoulos N, Martin J, Ruan Q, et al. Binding
`and neutralization of vascular endothelial growth fac
`tor
`(VEGF)
`and related ligands by VEGF Trap,
`ranibizumab and bevacizumab. Angiogenesis 2012;15(2):
`171–185.
`
`Ophthalmic Injuries at Hiroshima and Nagasaki
`
`B ecause he had studied cataracts induced by radia-
`
`tion exposure in radiologists and cyclotron workers,
`David G. Cogan of the Howe Laboratory in Boston
`was recruited by the Atomic Bomb Casualty Commission
`to go to Japan in 1950 and investigate eye injuries in sur-
`vivors of the atomic bombing of Hiroshima and Nagasaki.
`Japanese ophthalmologists had already reported the imme-
`diate effects of blast and fire injury but the special effects of
`ionizing radiation were just starting to be understood.
`Cogan originally thought that ionizing irradiation of the
`body intense enough to produce eye effects would be
`incompatible with survival, but later admitted that he
`
`had not considered the potential for partial shielding of
`the body that occurred in crowds. The head, however,
`was frequently exposed and Cogan did find victims with
`the after-effects of radiation exposure. Since the develop-
`ment of cataract from ionizing radiation was delayed, often
`for years, clear radiation-related cataracts were not evident
`until sometime after the exposure. David Cogan’s work at
`the Atomic Bomb Casualty Commission helped to set
`safety standards for workers in atomic energy facilities, a
`crying need in the following decades that saw the building
`of atomic energy power plants and nuclear medicine labo-
`ratories
`
`Submitted by Steven A. Newman from the Cogan Ophthalmic History Society.
`
`536
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`SEPTEMBER 2014
`
`Mylan Exhibit 1142
`Mylan v. Regeneron, IPR2021-00881
`Page 5
`
`