""— n Australian Government
`Department of Health and Ageing
`Therapeutic Goods Administration
`
`Australian Public Assessment Report
`for Afibercept
`
`Proprietary Product Name: Eylea
`
`Sponsor: Bayer Australia Limited
`
`‘7
`
`July 2012
`
`Health Safety
`Regulation
`
`NOVITC(CH)00008530
`Regeneron Exhibit 1066.001
`
`Regeneron Exhibit 1066.001
`
`

`

`Therapeutic Goods Administration
`
`About the Therapeutic Goods Administration (TGA)
`
`o The Therapeutic Goods Administration (TBA) is part ofthe Australian Government
`Department of Health and Agei 11g. and is responsible for regulating medicines and
`medical devices.
`
`-
`
`TGA administers the Therapeutic GoodsAct 1989 [the Act), applying a risk
`management approach designed to ensure therapeutic goods supplied in Australia
`meet acceptable standards of quality, safety and efficacy [performance], when
`necessary.
`
`. The work ofthe TGA is based on applying scientific and clinical expertise to decision-
`making. to ensure that the benefits to consumers outweigh any risks associated with
`the use of medicines and medical devices.
`
`0 The TGA relies on the public, healthcare professionals and industry to report problems
`with medicines or medical devices. TGA investigates reports received by it to
`determine any necessary regulatory action.
`
`0
`
`To reporta problem with a medicine or medical device. please see the information on
`the TGA website <wnnv.tga.goy.aii>.
`
`About AusPARs
`
`- An Australian Public Assessment Record [AusPAR] provides information about the
`evaluation ofa prescription medicine and the considerations that led the TGA to
`approve or not approve a prescription medicine submission.
`
`- AusPARs are prepared and published by the TGA.
`
`. An AusPAR is prepared for submissions that relate to new chemical entities, generic
`medicines. major variations. and extensions ofindications.
`
`I An AusPAR is a static document, in that it will provide information that relates to a
`submission at a particular point in time.
`
`I
`
`A new AusPAR will be developed to reflect changes to indications andfor major
`variations to a prescription medicine subject to evaluation by the TGA.
`
`Copyright
`© Commonwealth of Australia 2012
`This work is copyright. You may reproduce the whole or part of this work in unallered form foi' your own pe rsonal
`use or. iiyon are part of an organisation. foi' internal use within yourorganisation. hilt only if you oi' your
`organisation do not use the reproduction for any commercial purpose and retain this copyright notice and all
`disclaimer notICes as part of that reproduction. Apart from rights to use as permitted by the Copyriqiitxict 1968 or
`allowed by this copyright notice, all other rights are reserved and you are not allowed to reproduce the whole or any
`part of this work in any way [electronic or otherwise] without first being given specific written permission From the
`Commonwealth to do so. Requests and inquiries concerning reproduction and rights are to be sent to the TGA
`Copyright GlTicer. Therapeutic Goods Administration. PO Box 100. Worden ACT zmfiorizmailed to
`<l£€h§9iifllglti@ign .RQV-iib>-
`
`
`AusPAR Eytea Aflibercept Bayer Australia Ltci
`PM—ZOiI’J—DSflDZ-B-fi Final 30 JulyI 2012
`
`Page 2 of 93
`
`NOVITC(CH)00008531
`Regeneron Exhibit 1066.002
`
`Regeneron Exhibit 1066.002
`
`

`

`Therapeutic Goods Administration
`
`
`
`
`
`Canton 5
`
`
`I. Introduction to product submission
`
`Submission Details
`
`
`
`5
`5
`
`5
`
`5
`6
`
`6
`
`6
`
`7
`
`8
`
`8
`8
`
`Product background
`
`
`
`
`Regulatory status
`Product information
`
`ll. Quality findings
`
`
`Drug substance [active ingredient]
`
`
`Drug product
`
`Quality summary and conclusions
`
`
`III. Nonclinical findings
`
`Introduction
`
`
`
`
`
`Pharmacology
`Pharmacokinetics
`
`
`
`
`9
`10
`
`Toxicology
`
`Major findings
`
`
`
`1 1
`
`_ 13
`
`
`Nonclinical summary and conclusions
`
`
`1 6
`
`IV. Clinical findings
`
`Introduction
`
`17
`17
`
`Pharmacokinetics
`
`
`
`
`
`Pharmacodynamics
`
`Efficacy
`
`Safety
`
`List of questions
`
`
`
`Clinical summary and conclusions
`
`
`
`V. Pharmacovigilance findings
`
`
`Risk management plan __
`
`VI. Overall conclusion and risk/benefit assessment
`
`
`18
`
`23
`
`27
`
`47
`
`65
`
`65
`
`69
`
`69
`
`72
`
`_ 72
`
`73
`
`Quality
`Nonclinical
`
`
`Clinical
`
`Pliarinacodynamics
`
`
`
`
`
`Risk management plan
`
`
`Risk—benefit analysis
`
`74
`
`78
`
`86
`
`86
`
`OutcomeW
`91
`
`
`AusPAR Eyiea Aflibercepi Bayer Australia Ltd
`PM-2010-03802-3-5 Final 30 July 2012
`
`Page 3 of 93
`
`NOVITC(CH)OOOO8532
`
`Regeneron Exhibit 1066.003
`
`

`

`Therapeutic Goods Administration
`
`Attachment 1. Product Information
`
`
`
`92
`
`
`
`AusPAR Eylea Aflibercept Bayer Australia Ltd
`PM-2010-03802—3—5 Final 30 July 2012
`
`Page 4 m 93
`
`NOVITC(CH)00008533
`
`Regeneron Exhibit 1066.004
`
`

`

`Therapeutic Goods Administration
`
`I. Introduction to product submission
`
`Submission Details
`
`Type ofS'uhmiSSi'on
`
`New chemical entity
`
`Decision:
`
`Approved
`
`Date ofDecision:
`
`17 February 2012
`
`Active ingredientjfs):
`
`Aflibercept
`
`Product Name(s}:
`
`Eyiea
`
`Sponsor’s Name
`
`Bayer Australia Limited
`
`Doseformfi}:
`
`Strengthfs}:
`
`Containerfs}:
`
`Pack size{s):
`
`875 Pacific Highway
`
`Pymble NSW 2073
`
`Solution for lntravitreal injection
`
`40 mg/mL
`
`Pre-filled syringe and via]
`
`One unit per package
`
`Approved Therapeutic use:
`
`Eylea [afliberceth is indicated for the treatment of neovascular
`[wet) age-related macular degeneration [AMD].
`
`Routeis} ofadministration:
`
`intravitreal
`
`Dosage:
`
`Injection volume is 50 uL of Eylea [equivalent to 2 mg aflibercept];
`one injection intravitreally monthly for three months followed by
`two monthly injections.
`
`ARTS Number {5}
`
`AUST R 180859 and AUST R 180860
`
`Product background
`
`Aflibercept [VEGF Trap-Eye, also abbreviated to VTE) is a new chemical entity. a biological
`substance that is a recombinant fusion glycoprotein consisting of sequences derived from
`human vascular endotheiia] growth factor (VEGF) receptor extracellular domains 1 and 2 are
`fused to the Fc portion of humanimmunoglobulin subtype G} [[gG 1]. For more information on
`the structure ofaflibercept, please refer to the separate quality findings below.
`
`This AusPAR describes the application by Bayer Australia Ltd to register aflibercept [as Eylea)
`for
`
`The treatmentofrieovoscuior {wet} age-related mocuiur degeneration {11MB}.
`
`The proposed treatment regimen is intravitreai (WT) injection of 50 uL solution [containing
`2 mg afiibercept) once per month for three consecutive months. foliowed by one injection every
`two months.
`
`Reguiatory status
`
`At the time ofapplication, Eylea had also been submitted for registration in the European Union,
`
`Switzerland and the USA. These applications are all currently under evaluation with their
`AusPAR Evlea Afiibercept Bayer Australia Ltd
`Page 5 of 93
`PM—2010-03802—3—5 Final 30 July 2012
`
`NOVITC-{CH}00008534
`
`Regeneron Exhibit 1066.005
`
`

`

`Therapeutic Goods Administration
`
`respective health authorities, with the exception of the USA, where the product was approved
`by FDA on the 13 November 2011.
`
`Product information
`
`The approved product information [PI] current at the time this AusPAR was prepared can he
`found as Attachment 1.
`
`II. Quality findings
`
`Drug substance (active ingredient}
`
`Aflibercept is a recombinant protein consisting ofsequences derived from human vascular
`endothelial growth factor [VEGF] receptor extracellular domains fused to the Fc portion of
`human lgG 1. The extracellular domain sequences come from two different VEGF receptors.
`VEGFR] [also known as Flt-1] and VEGFRZ [also known as KDR or Flk- 1]. Each ofthe VEGF
`receptors are composed of seven [3 domains in their extracellular regions, with lg domains 2
`and 3 contributing the majority of the binding energy for VEGF. Thus, the amino acid sequence
`ofa single aflibercept subunit comprises lg domain 2. from VEGFRI, fused to lg domain 3 from
`VEGFRZ, which is in turn fused to a Fc domain fragment oflgG]. There are no extraneous linker
`sequences between any oi'the peptide domains. The schematic structure of the drug substance
`is shown below:
`
`Figure 1. Schematic structure
`
`VEGFR‘l VEGFRZ VEGF
`Trap
`
`1. F
`i I961 Fr:
`
`Aflibercept is a dimeric glycoprotein with a protein molecular weight of 96.9 kilo Daltons [kDaj
`[C-i'sleHoi'aaN I 16401304532. 2 x 43 1 amino acids]. It contains approximately 15% glycosylation to
`give a total molecular weight oil 15 kDa. All five putative N-glycosylation sites on each
`polypeptide chain predicted by the primary sequence can be occupied with carbohydrate and
`exhibit some degree of chain heterogeneity, including heterogeneity in terminal sialic acid
`residues, except at the single unsialylated site associated with the Fc domain.
`
`The disulfide bond structure ofaflibercept determined by peptide mapping matches the known
`disulfide patterns of the VEGFRI [lg domain 2), VEGFRZ [lg domain 3] and the lgG Fc domain.
`The C-terminus lacks the predicted lysine residue on the Fc moiety as expected.
`
`
`
`AUSPAR Eylea Ailibercept Bayer Australia Ltd
`Phi—20100380215 Final 30 July 2012
`
`Page E: of 93
`
`NOVITC (C H)0 0008535
`Regeneron Exhibit 1066.006
`
`Regeneron Exhibit 1066.006
`
`

`

`Therapeutic Goods Administration
`
`Manufacture
`
`The manufacturing ofaflibercept drug substance involves growth ofa suspension culture of
`Chinese Hamster Ovary cells [CHO K1] engineered to express aflibercept. The recombinant
`product is secreted into the culture medium and subsequently purified by chromatographic
`[Protein A affinity, cation exchange, anion exchange. hydrophobic interaction and size—exclusion
`chromatography]. virus inactivation/filtration and membrane filtration techniques. Cell banking
`processes are satisfactory.
`
`All viral/prion safety issues have been addressed, including use of animal-derived excipient
`supplements in the fermentation process and in cell banking.
`
`Physical and chemical properties
`
`Product related impurities include aggregates, truncated species, deamidated variants, charged
`variants and oxidised forms. The first four forms ofimpurity are controlled at drug substance
`release. It is well justified to exclude the testing of oxidised form at the drug substance release.
`
`Specifications
`
`Appropriate validation data were submitted in support of the test procedures.
`
`Drug product
`
`The drug product, 40 mg/mL, is formulated in 10 mM sodium phosphate buffer containing 40
`mM NaCI, 0.03% [w/v) polysorbate 20 and 5% [w/v} sucrose, pH 6.2. Two presentations are
`available:
`
`The vial presentation is supplied in a 2R lSO injection via]. The target fill volume for each vial is
`278 [.lL [100 pL extractable volume] to ensure a single 50 uL injectable dose containing 2 mg
`aflibercept. One package includes one vial and one filter needle. The injection needle is not
`supplied.
`
`The syringe presentation is supplied in glass 1 mL syringes. The target fill volume for each
`syringe is 165 uL [90 pL extractable volume). The syringe, when equipped with a 30 gauge, 0.5
`inch needle, can deliver a single 50 Lil. injectable dose containing 2 mg atlibercept. One blister
`package contains one syringe. The injection needle is not supplied.
`
`Manufacture
`
`The drug product is sterilised by filtration.
`
`Blisters containing the syringe are either hydrogen peroxide (H202)'StEl‘iiiSEd or ethylene oxide
`[ETO]—sterilised.
`
`Specifications
`Appropriate validation data have been submitted in support of the test procedures.
`
`Stability
`
`Stability data have been generated under real time/stressed conditions to characterise the
`stability profile ofthe product. Photostahility data shows the product is not photostable and
`should be stored in its original package.
`
`
`
`AusPAR Eylea Afiibercepl Bayer Australia Ltd
`PM—zoio-ossoz-s-s Final 30 July 2012
`
`Page 7 of 93
`
`NOVITC(CH)OOOO8536
`
`Regeneron Exhibit 1066.007
`
`

`

`Therapeutic Goods Administration
`
`The real time stability data support a shelflife of"12 months when stored at 2°C to 8°C,
`protected from light", for both vial and syringe presentation.
`
`Quality summary and conclusions
`
`The draft PI, Consumer Medicine Information [CW] and containerXprimary packaging labels are
`acceptable.
`
`Summary of evaluation
`
`The administrative, product usage, chemical, pharmaceutical and microbiological data
`submitted in support of this application have been evaluated in accordance with the Australian
`legislation, pharmacopoeial standards and relevant technical guidelines adopted by the TGA.
`
`Batch release conditions of registration for clinical delegate
`
`It is a condition of registration that the first five independent batches of Eylea are not released
`for sale until samples and/or the manufacturer’s releaSe data have been assessed and endorsed
`for release by the TGA Office of Laboratories and Scientific Services (01.55).
`
`The sponsor should supply:
`
`1. Certificates ofAnalysis of all active ingredient [drug substance] and final product.
`
`2.
`
`3.
`
`4.
`
`Information on the number ofdoses to be released in Australia with accompanying expiry
`dates for the product and diluents (if included].
`
`Evidence ofthe maintenance ofregistered storage conditions during transport to Australia.
`
`3 vials or 3 syringes of each batch for testing by the Therapeutic Goods Administration
`OLSS together with any necessary standards, impurities and active pharmaceutical
`ingredients [with their Certificates of Analysis] required for method development and
`validation.
`
`These batch release conditions will be reviewed and may be modified on the basis ofactual
`batch quality and consistency.
`
`III. Nonclinical findings
`
`Introduction
`
`General comments
`
`The overall quality ofthe nonclinica] dossier was adequate. All pivotal safety-related studies
`were conducted under Good Laboratoly Practice [GLP] conditions. A safety pharmacology study
`examining effects ofaflibercept on cardiovascular parameters in rodents was non-GLP
`compliant; nevertheless, the study was well documented, and cardiovascular parameters were
`also examined as part of GLP compliant general repeat-dose toxicity studies in monkeys.
`Reports for several non-pivotal, non GLP repeat-dose toxicity studies in mice and rats were of
`poor quality in some respectsmo group means were calculated, clinical signs and the
`histopathologica] findings were not tabulated, nor incidences per dose group calculated; the
`absence ofgroup summary data [such that the results were not presented in a clear and concise
`
`
`
`AusPAR Eylea Afi'lbercepl Bayer Australia Ltd
`PM—2010-03802—3—5 Final 30 July 2012
`
`Page a of 93
`
`NOVITC(CH)OOOO8537
`
`Regeneron Exhibit 1066.008
`
`

`

`manner] is at odds with the TGA adopted EU guideline on repeated dose toxicityl. The pivotal
`toxicology studies were conducted with drug substance manufactured using the commercial
`process.
`
`Therapeuticfioods Administration
`
`Pharmacology
`
`Primary pharmacology
`
`Rationale and mechanism ofoction
`
`Vascular endothelial growth factor [VEGF or VEGF-A] plays a critical role in angiogenesis. In
`age-related macular degeneration. VEGF promotes ocular neovascularisation and excessive
`vascular permeability and oedema. Aflibercept is designed to act as a soluble decoy receptor for
`VEGFR ligands.
`
`Efficacy
`
`in vitro. aflibercept was shown to bind to human VEGF-A with subpicomolar affinity [Kd for
`VEGF—A165,0.497 pM; Kd for VEGF—AIZL 0.360 pM). High affinity was also found for the
`related angiogenic molecule, PiGF—Z (placental growth factor 2; Rd value. 38.8 pM], which acts
`through VEGFR-l. Binding to aflibercept occurs with higher affinity than to the ligands’
`endogenous receptors {compared to respective Kd values for VEGF-A binding to VEGFR~1 and
`VEGFR—Z. 10*30 pM and 75—760 p103 and ~170 pit for PIGF—Z binding to VEGFR—ll. The drug’s
`ability to bind to PIGF may contribute to its pharmacological activity as PIGF is also implicated
`in the development of neovascular AMD‘. Aflibercept also displayed affinity for PlGF—I
`[placental growth factor 1: Kd value, 392 pM]; in this case, though, affinity is below that for the
`endogenous receptor [170 pM for binding to VEGFR~13]. The drug’s affinity was similar for the
`animal and human VEGF isoforms among the species tested (mouse. rat, rabbit]. Binding studies
`were not performed with monkey VEGF as its amino acid sequence is identical to human VEGF.
`In in vitro functional studies. aflibercept blocked VEGF-induced phosphorylation of VEGFR~2
`and the resultant calcium mobilisation in human umbilical vein endothelial cells [HUVECSL
`in viva, intravitreal (WT) injection ofaflibercept inhibited retinal neovascularisation in the
`mouse [oxygen-induced retinopatliy model] and choroidal neovascularisation in the monkey
`{laser-induced], and normalised retinal vascular permeability in the rat [diabetic model].
`
`Secondary pharmacodynamics and safety pharmacology
`
`Aflibercept did not to bind to human VEGF-C or VEGF-D. in an immunohistochemical study
`examining potential cross-reactivity, no specific staining was found for aflibercept [525 tug/ml.)
`against a panel of 33 normal human tissues. The Fc region ofthe aflibercept molecule did not
`mediate any complement-dependent cytotoxicity [CDC] or antibody-dependent cell-mediated
`cytotoxicity (ADCC) in vitro.
`
`
`
`' CPMP/SWPIIMEIBB Rev 1. Guideline on repeated dose toxicity.
`
`httpzflwwwtgagov .aufipdileuguidefswp1042092nrev1pdf
`! Robinson CJ. and Stringer S.E. [2001} The splice variants ofvascuiar endothelial growth factor [VEGFi and their
`receptors. J. Cell Sci. 114: 353~865.
`3 Savvano A., TakahashiT., Yamaguchi 5., Aonuma M. and Shibuya M. (1996} Flt—1 but not KDRi'Fllt—l tyrosine kinase is
`a receptor for placenta growth factor. which is related to vascular endothelial growth factor. Cell Growth Difi‘er.
`7: 213—221i
`
`‘ Rakic J.M., Lambert V., Dew L. Luttun A, Carmeliet P., Claes 0., Nguyen L., Foidart J.M.. Noel A. and Munaut C.
`{2003} Placental growth factor, a member of the VEGF family. contributes to the development of choroidal
`
`neovascularization. invest. Ophthalmot Vis. Sci. 44: 3186—3193.
`
`AusPAR Eylea Aflibercept Bayer Australia Ltd
`Pill—20100300235 Final 30 July 2012
`
`Page 9 of 93
`
`NOVITC (C H)0 0008538
`Regeneron Exhibit 1066.009
`
`Regeneron Exhibit 1066.009
`
`

`

`Therapeutic Goods Administration
`
`Specialised safety pharmacology studies were limited in scope. Instead, the sponsor has mostly
`relied on various genera] repeat-dose toxicity studies that incorporated relevant end points.
`This approach is acceptable under the applicable TGA adopted EU guideline-i Aflibercept had no
`effect on respiration in rats following IV administration [5250 mg/kg over 30 min]. There was
`no evidence of particular central nervous system [CNS] toxicity in the repeat-dose toxicity
`studies; lethargy in rats [at 22 mg/kg subcutaneously [SC] administered three times weekly]
`and reduced activity in monkeys [23 mg/kg intravenous [IV] once weekly] were observed, but
`occurred at doses beyond the maximum tolerated dose [MTD: based on body weight loss or
`substantial inhibition of body weight gain]. Intravitreal [[VT] treatment produced no clinical
`signs in monkeys [S4 mg/eye bilateral]. Increases in blood pressure were observed in monkeys
`given aflibercept SC [15w30 mgg'kg, twice weekly, but not IV [530 mg/kg once weekly). In a
`specialised study in mice and rats, SC administration ofaflibercept increased systolic and
`diastolic blood pressure in both species that persisted until plasma concentrations of free
`aflibercept fell below 1 ug/mL. In addition to its function as a vascular growth factor, VEGF is
`involved in the regulation of blood pressure by modulating available nitric oxide and
`prostacyclin levels to promote vasodilatationfi; these results therefore presumably reflect
`inhibition of circulating VEGF by aflibercept. No electrocardiogram [ECG] abnormalities were
`observed in monkeys treated with aflibercept SC or iV. Aflibercept did not affect thrombus
`formation or coagulation parameters in the rabbit [$30 mgfkg IV]. Wound healing was inhibited
`by aflibercept in rabbits at all doses tested [incisional and excisional models; reductions in
`blood vessel density, tensile strength, fibrous response and/or epidermal hyperplasia seen at
`20.3 mgfkg IV]; the finding is consistent with the known role ofVEGF in wound repair
`[reviewed by Bao et at, 2009?].
`
`Pharmacokinetics
`
`Free aflibercept, and sometimes also VEGF—bound aflibercept, were assayed in
`pharmacokineticXtoxicokinetic studies. The bound form is pharmacologically inactive.
`
`Following IVT administration in monkeys, levels of free aflibercept in the vitreous were dose-
`proportiona] and declined with an estimated half-life of40—64h [independent of dose]. Peak
`levels of free aflibercept in plasma were generaliy reached within 1—3 days post-dose and were
`detectable for up to 2 or 3 weeks. Bound aflibercept was detected in plasma with 24b of dosing,
`peaking at 1—3 weeks post-dose; its apparent clearance was much slower compared to free
`aflibercept, remaining detectable in plasma in some animals for up to 18 weeks post-dose. Note,
`however, that the continuous formation ofendogenous VEGF obfuscates the determination of
`the true half-life for bound afiibercept. The proportion of freezbound aflibercept increased with
`dose, consistent with saturable binding of endogenous VEGF. In rabbits given aflibercept IVT,
`half-lives for free atlibercept were determined to be115h in the vitreous and 15711 in plasma;
`peak plasma concentration [CW] and overall exposure [area under the plasma concentration
`versus time response curve from time zero to infinity [AUCU_m]] to free aflibercept in plasma
`was 950- and 310-times lower, respectively. compared to in the vitreous.
`
`Greater than dose-proportional exposure was observed for free aflibercept in serum in rats and
`monkeys following SC administration. This may reflect that clearance comprises a saturable
`component, possibly related to VEGF binding. Again, long half-lives were observed for free
`
`
`
`5 CPMPIICHIS39/00. Note for Guidance on Safetyr Pharmacology Studies for Human Pharmaceuticals,
`http:Nwwwiga.gov.au!pdf/eugulde/ichOSSQDOenpdf
`'5 He H., Venema V.]., (In X.,Venen1a R.C., Marrero MB. and Caldwell RB. [1999] Vascular endotiielial growth factor
`signals elidothelial cell production ofnitric oxide and prostacyclin through flk—l/KDR activation ofc—Sch. Biol.
`Chem. 274:25130-25135.
`
`7 Ban P., Kodra A” Tomic—Canic M, Golinko M.S., Ehrlich I-I.I’. and Brem H. [2009].The role of vascularendothelial
`
`growth factor in wound healing}. Surg. Res. 153:347—358.
`
`AusPAR Eylea Afiibercept Bayer Australia Ltd
`PM—2010-03802-3-5 Final 30 July 2012
`
`Page 10 of 93
`
`NOVITC {C H)0 0008539
`
`Regeneron Exhibit 1066.010
`
`

`

`Therapeutic Goods Administration
`
`aflibercept in serum following 1V and SC dosing [~40—50h in the mouse and rat, and up to ~50—
`1DOh in the monkey]. Bioavailability by the SC route was high in mice [94%] and monkeys
`[85%] and moderate in rats [33%]. No sex differences in pharmacokinetic profiles were
`observed for any route/species.
`
`Distribution to the retina and choroid after lV'I‘ administration was shown in the rabbit, with
`peak and overall exposure to free aflibercept in these tissues ~5% ofthe corresponding values
`for the vitreous; half—lives were comparable for all three matrices (115—132 h]. With IV dosing
`in mice, rats and monkeys, steady—state volumes ofdistribution were only slightly greater than
`the whole blood volume. consistent with limited distribution outside ofthe central
`compartment [as is typical for large molecular weight, protein-based drugs]. Results from a
`tissue distribution study in rats with radioactively labelled (”SD—aflibercept, administered IV,
`support this. Highest tissue concentrations ofradioactivity were found in the liver. followed by
`other highly perfused tissues. The liver (and not the kidney] was identified as having the major
`role in the clearance of aflibercept. Consistent with this, functional nephrectomy did not
`significantly affect the serum kinetics of aflibercept in rats. Given aflibercept‘s protein nature, no
`classical biotransformation studies were conducted; this is in accordance with the relevant TGA
`adopted EU guidelineB.
`
`Anti-aflibercept antibodies were formed in mice, rats and rabbits. and less commonly in
`monkeys. Their development was associated with decreased drug exposure in rabbits and the
`rodent species but rarely in monkeys. The aflibercept molecule contains multiple N—linked
`glycosylation sites. Differences in the extent of sialic acid occupancy were found to affect the
`drug's serum kinetics [in rats] but not its potency {assessed in in vitro binding and functional
`assays].
`
`Pharmacokinetic drug interactions
`
`No nonclinical studies were performed.
`
`Toxicology
`
`Acute toxicity
`
`Single~dose toxicity studies, performed by the [V route in rats, revealed a low order of acute
`toxicity for aflibercept. with no deaths observed up to the highest dose tested (500 mg/kg].
`
`Repeat-dose toxicity
`
`Repeat—dose toxicity studies by the clinical route [lVT] were conducted in the cynomolgus
`monkey only [up to 8 months duration]. To better characterise the systemic toxicological
`profile, SC studies were performed in mice [up to 8 weeks duration], rats [up to 13 weeks] and
`monkeys [up to 13 weeks], and 1V studies were performed in the rabbit [2 weeks; in non
`pregnant animals as a pilot study for reproductive toxicity] and monkey [up to 6 months].
`Aflibercept is pharmacologically active in all ofthese species. IVT dosing was once per 4 weeks
`in the pivotal and most other studies [consistent with the initial phase of the clinical treatment
`regimen]. or else once per 2 weeks or 6 weeks. The proposed clinical formulation was used in
`the pivotal lVT study; the strength ofaflibercept varied with dose though, being the same as that
`for Eylea at the mid-dose level and double it at the high-dose level. SC and 1‘! doses were
`administered more frequently than is proposed clinically for lVT administration. ranging from
`once per 2 weeks to up to 3 times weekly.
`
`
`8 CPMPfICH/BDZJQS Note for Guidance on Preclinical Safetyr Evaluation of Biotechnologyr Derived Pharmaceuticals.
`http:f/www.tga.gov.au{pdffeuguide/ich030295enpdf
`AusPAR Eylea Aflibercept Bayer Australia Ltd
`Page 11 of 93
`PM—2010-03802-3-5 Final 30 July 2012
`
`NOVITCfCH)00008540
`
`Regeneron Exhibit 1066.011
`
`

`

`Therapeuticfioods Administration
`
`Based on their short duration [s3 months] and non-ocular route ofadministration. no rodent
`study sufficient to be regarded as pivotal has been provided with the current Submission (this is
`not to say that they provided no useful information). A 6—month study in rodents was found not
`to be feasible due to the development of anti-aflibercept antibodies; repeated intravitreai
`administration is largely impractical in mice and rats. Considering this, the pre-eminence of the
`primate over the rodent as a relevant and feasible model for the assessment of the toxicity of the
`proposed product, and that there is existing experience with the pharmacological class. the
`reliance on the cynomolgus monkey as a single species for which there is a pivotal study is
`deemed to be acceptable. Group sizes were adequate; the small group size used in the monkey
`studies is typical but does limit their predictive value.
`
`Relative exposure
`
`Relative systemic exposure in selected toxicity studies has been calculated based on
`animal:human Cam and AUC for free aflibercept in plasma/serum [see Table 1 below]. The
`human reference values used are from Clinical Study VGFT~OD-0702, obtained following lVT
`administration of2 mg aflibercept [clinical formulation] to one eye of patients. The values have
`been doubled for the calculation here to reflect that Eylea may be administered to both eyes in
`clinical use.
`
`Relative ocular exposure is considered based on dose adjusted for species differences in
`vitreous volume‘3:tlie IVT doses used in the pivotal monkey study (0.5, 2 and 4 mg/eye) are 0.3.
`1.25 and 2.5-times the proposed human dose [2 mgfeye).
`
`
`
`" Values for vitreous volumes of 3.2 ml in cynornolgus monkeys and 4.0 mt in humans have been used for the
`calculation here. in accordance with the approach to safety evaluation of Short {2008}. Less conservative values [1.5
`mL and 4.5 mL, respectively] were used by the author of the Nonclinical Expert Report.
`
`AusPAR Eylea Ailibercept Bayer Australia Ltd
`Phi—20100380235 Final 30 July 2012
`
`Page 12 of 93
`
`NOVITC(CH)00008541
`Regeneron Exhibit 1066.012
`
`Regeneron Exhibit 1066.012
`
`

`

`Therapeutic Goods Administration
`
`Table 1. Relative exposure to free aflibercept in selected toxicity studies
`
`Route;
`frequen
`CV
`
`Exposure ratio"
`
`AUC0_
`25,,
`lug-hi
`mL)
`
`
`
`10
`5C:
`4- weeks
`Mouse
`[CD-l]
`PKUIUIT'
`thm‘
`ms/ks
`times
`1.5
`82.2
`—
`213
`-
`
`weekly
`tug/kg
`U
`{1.1
`??.t
`—
`199
`_
`SE:
`
`mg/kg
`5
`three
`times
`0.5
`C109
`weekly
`mgfkg
`U
`1 mg/kg
`1?01
`—
`44-0
`—
`
`65
`Zing/kg
`915
`—
`23?
`—
`
`05
`
`33.8
`
`235
`
`—
`
`
`
`Rat [SD]
`
`13 weeks
`vorr-rx-
`02006"
`
`Monkey
`[Crnnmol
`gus]
`
`3 months
`VGF'I'v'I‘X-
`@203?
`
`6 months
`VGFT-Tx-
`05009.
`
`\IGFT—OD-fli'OZ
`
`
`
`
`
`SE:
`“Vice
`weekly
`
`IV:
`once
`weekly
`
`{WM
`
`
`
`
`
`
`
`
`
`2.856
`
`—
`
`1.5
`31.9
`—
`825
`—
`
`"13/ k8
`5 mg/kg
`109
`—
`282
`—
`
`5
`15
`286
`—
`T41
`—
`
`mg/kg
`U
`30
`721
`—
`186
`—
`
`mg/kg
`80
`3 monll‘ls,
`0.5
`9.54
`”3'6
`245
`310
`
`juvenile
`tug/kg
`VGFT—TX-
`3 mg/kg
`73.3
`19296
`191
`338
`
`05mm
`0
`0
`30
`830
`16382
`215
`286
`
`ring/kg
`4
`05
`so
`IV:
`3 mg/kg
`93.2
`8332
`241
`154
`
`0110? Per
`5
`S
`1—
`10
`305
`35952
`390
`629
`
`Z/weeks
`mg/kg
`0
`5
`30
`7’30
`72336
`189
`126
`
`mg/kg
`10
`65
`3 months
`[VT Ill;
`0.5
`0.936
`157.6
`24
`28
`
`"lg/ere
`vorr-rx-
`once/’1 W
`2
`6.9?
`1394
`180
`245
`05011r
`eeks
`
`ling/eye
`4
`16.9
`3360
`440
`590
`
`ling/eye
`0.019
`2
`Human
`3
`Ins/ere
`[AMD paii
`
`enls]
`
`
`
`—
`
`—
`
`
`
`.. a mlt‘lllatee] as animalzliil man values, Inllnw‘ing (louliling ol' the clinical rel'erenre values to rcllect bilateral use; exposure ratios greater than 100 are
`rounded tn the nearest 5: — = IHJ{liltilfl'lt’rHE‘I‘IlI'illllllj
`"-|= :milat
`idliiiuis
`ition: data are the meaiisol male and female values:
` -I = parameters obtained on day 22; '-= parameters olitainet! on day 83; i = parameters obtained alter dosing in Week Iii:
`.i : parameters obtained after (losing In week 13 {Mlljn '.J.§I|| Is Iiiilltlplled by '1. art'ountllig for [losing frequency}
`- = parameters obtained after dosing in week 21: dosing was once per week to week ‘1 5, then once per 2 weeks:
`[N |I."I.-.
`l..<:|. is; mull iplit'd by 2 In :ll'lilJll ”I Ior dosing frequent'y};
`'- = parameters obtained after the 3"-" dose.
`
`Major findings
`
`IVT administration of aflihercept was associated with an anterior segment/vitreous
`inflammatory response in monkeys. This was generally mild, usually peaked at 2 days post close
`and was completely (anterior) or mostly (vitreous) reversed by 4 weeks post dose.
`
`AusPAR Eylea Afiibercept Bayer Australia Ltd
`PM-2010-03802—3—5 Final 30 July 2012
`
`Page 13 of 93
`
`NOVITC(CH)OOOO8542
`
`Regeneron Exhibit 1066.013
`
`

`

`Therapeutic Goods Administration
`
`Administration ofthe vehicle alone also produced some inflammation. No angiographic or
`electro-retinographic changes were found in treated monkeys, nor were any ocular
`abnormalities observed in imaging and microscopic evaluations. lntraocular pressure was
`unaffected by the drug. Several-fold increases in intra ocular pressure [IOP] occurred
`immediately post dose, though, including following administration of the vehicle only,
`consistent with injection ofa fluid bolus. No animal developed glaucomatous optic nerve head
`cupping in response to these lOP spikes. The only findings oftoxicological significance in the IVT
`studies involved the nasal turbinates, with microscopic erosion and ulceration ofthe respiratory
`epithelium, often accompanied by chronic-active inflammation, seen at 2 and 4 mgXeye in the
`pivotal study. These lesi0ns were generally mild and demonstrated to be reversible. Based on
`the absence of effects on other tissues, the nasal turbinate findings are considered more likely to
`result from local rather than systemic exposure [that is, by way of anastamotic connections
`between the ophthalmic and nasal venous plexuses or leakage into the nasal lacrimal duct].
`Cross-species exposure comparisons for such an effect are probably best made on a mg/kg
`basis: assuming 4 kg body weight for a monkey and 50 kg for a human {as a conservative
`measure], the lowest-observable-effect level (LOEL; 2 mgfeye] is more than 6 times the human
`dose and the no-observable-effect level (NOEL; 0.5 mg/eye] is >1.5 times the human dose. More
`pronounced effects on the nasal cavity were seen with systemic administration in monkeys,
`along with changes in numerous additional tissues, consistent with the very much higher
`exposure levels achieved. The nasal cavity findings included atrophy/loss of the septum and/or
`turbinates associated with necrotising inflammation. The other principal organs targeted were
`bone [such as osteocartilaginous exostoses of vertebrae: interference with growth plate
`maturation], kidney (increased glomerular mesangial matrix; glomerulopathy with tubular
`dilatation and cast fo