D8
`
`Development and
`
`Manufacture of
`
`Protein Pharmaceuticals
`
`Edited by
`
`Steven L. Nail
`Purdue University
`West Lafayette, Indiana
`
`and
`
`Michael J. Akers
`Baxter Pharmaceutical Solutions LLC
`Bloomington, Indiana
`
`Kluwer Academic/ Plenum Publishers
`New York, Boston, Dordrecht, London, Moscow
`
`Kopie Von subito e.V., geliefert for df~-mp (COMMXOMQB)
`
`AMGEN INC.
`
`Exhibit 1016
`
`Ex. 1016 - Page 1 of 82
`
`Ex. 1016 - Page 1 of 82
`
`AMGEN INC.
`Exhibit 1016
`
`

`

`2
`
`Formulation Development of Protein
`Dosage Forms
`
`Michael J. Akers, Vasu Vasudevan, and
`
`Mary Stickelmeyer
`
`1. INTRODUCTION
`
`A formulation scientist assigned the task of developing a stable, elegant, and
`manufacturable dosage form of a therapeutic protein drug has been given a
`significant challenge. Most proteins, as natural physiological molecules, are
`inherently unstable outside the human or animal body. Stability challenges
`in protein formulation development are typically enormous. The instability
`of these reactive and complex molecules must be considered not only in the
`formulation process, but also in development of the packaging system and
`the manufacturing process. These three areas are intimately and inseparably
`connected.
`
`Protein dosage forms are also sterile dosage forms. Sterile dosage forms
`must be essentially free* from microbial contamination (sterile), free from
`pYrogenic (including endotoxin) contamination, and free from particulate
`
`*The term “essentially free" is preferable over the more absolute term “free" when dealing with
`the subject of microbiological contamination. Except for products which can be terminally
`sterilized, which do not include proteins, there is no total and absolute assurance that each unit
`of product is, in fact, sterile.
`
`o Baxter Pharmaceutical Solutions LLC, Bloomington, lndiana 47492.
`Michael J. Akers
`Vasu Vasudevan and Mary Stickelmeyer
`o
`Lilly Research Laboratories. Indianapolis,
`Indiana 46285.
`Developmenl and Manufacture of Protein Pharmaceuticals. edited by Nail and Akers. Kluwer Academic/
`Plenum Publishers, New York, 2002‘
`
`m
`
`47
`
`Kopie von subito e.V., geliefort flit dfmmp (COMO4X01498)
`
`Ex. 1016 - Page 2 of 82
`
`Ex. 1016 - Page 2 of 82
`
`

`

`43
`
`Michael J. Akers et a1.
`
`matter contamination (ready-to-use and reconstitutable solutions). Depend-
`ing on the route of administration, sterile dosage forms must also be
`isotonic. For example, the intravenous route of administration can tolerate
`fairly wide
`ranges of
`“tonicity”
`(osmolality
`or
`osmolarity)
`or
`oncocity“ whereas subcutaneous and intramuscular routes may require
`tighter control of product tonicity. Sterile products administered into spinal
`fluid or topically applied to the eye must be as close to isotonic as possible
`because of the potential of irreparable damage of spinal or corneal cells due
`to extremes in the osmolar concentrations of administered products. In
`addition, products administered by the injectable or ophthalmic topical
`routes should be as close to physiological pH (7.4) as possible to minimize
`pain and tissue irritation or damage.
`This chapter was written to provide the basic approaches and techniques
`used to design and develop dosage forms of proteins. To develop dosage
`forms means not only to generate a viable formulation, but also to identify a
`final packaging system, to design and scale up a quality manufacturing
`method, and to employ valid measurements to assure product quality. In
`addition, in this era of globalization, formulations must be developed that
`are acceptable from a regulatory standpoint throughout the world.
`Since protein stabilization has already been extensively discussed in
`many excellent references (Table I), we also intend to cover other issues
`essential in the complete formulation development of protein products yet
`not covered elsewhere, such as antimicrobial preservation, packaging
`components, container—closure integrity, clinical trial manufacturing, and
`development history reports.
`We have revieWed the literature and have selected the articles which
`provide both intensive analysis and extensive information on solving protein
`formulation and other product development problems. Advanced injectable
`(e.g., controlled release, implantable devices, gene delivery) and noninject-
`able (e.g., pulmonary, oral, buccal) protein formulation research will not be
`covered in this chapter, but other references are available that deal with these
`advances (Baker, 1980; Davis et at, 1986; Senior and Radomsky, 2000;
`Hillery er al., 2001).
`
`2. WHY PROTEINS PRESENT UNIQUE CHALLENGES
`TO THE DEVELOPMENT SCIENTIST
`
`Many texts and articles already discuss the great difficulties scientists
`“whence m protein dosage formulation because of the significant
`. ()nco '
`i
`“c pressure ‘- °Sm°m pressure exerted by colloids (e.g., plasma proteins) in a solution’
`
`Kopie von subito e.V,, geliefert for dfmmp (COMO4XO'I498)
`
`Ex. 1016 - Page 3 of 82
`
`Ex. 1016 - Page 3 of 82
`
`

`

`Formulation Development of Protein Dosage Forms
`
`49
`
`Table 1
`Major Protein Formulation References
`
`Stability of Protein Pharmaceuticals, Parts A and B. T. J, Ahern and M. C. Manning. (eds).
`Pharmaceutical Biotechnology, vols 2 and 3. Plenum Press. New York l992
`Formulation concerns ofprotein drugs, T. Chen Drug Dev Ind. Pharm. 18:l31 l—l354 (1992)
`The formulation of proteins and peptides, M. .l. Groves. M. H. Alkan. and A.
`.l. Hickey. in:
`Pharmaceutical Biotechnology (M. E. chgerman and M. J. Groves. eds.). lnterpharm Press.
`Englewood, CO, 1992
`Protein stability aml degradation mechanisms, B. L. Currie and M. J. Groves. in: Pharmaceutical
`Biotechnology (M. E. Klegerman and M. J. Groves eds), lntcrphann Press. Englewood. CO.
`l992
`Stability and Characterization of Protein and Peptide Drugs: Case Histories. Y. J. Wang and
`R. Pearlman, Plenum Press, New York, I993
`Factors aflecting short-term and long-term stabilities of proteins, T. Arakawa, S. J. Prestrelski.
`W. C. Kenney, and J. F. Carpenter, Adv. Drug Deliv. Rev. lint—28, I993
`Formulation and Delivery of Proteins and Peptides, Design and Development Strategies.
`J. L. Cleland and R. Langer, (eds). ACS Symposium Series 567, American Chemical Society.
`Washington, DC, 1994
`Formulation, Characterization, and Stability of Protein Pharmaceuticals, R. Pcarlman and
`Y. J. Wang. Plenum Press, New York, 1996
`
`instabilities of these molecules. Depending on the amino acid types and
`sequence, proteins are subject to various types of degradation mechanisms,
`including hydrolysis, oxidation, racemization, and interaction with a variety
`of solutes and surfaces. These mechanisms are especially critical, because
`pharmaceutical proteins are very pure and removed from their natural
`environments where they are most stable (Hanson and Rouan, 1992).
`Dealing with physical
`instability (e.g., denaturation, aggregation, and
`adsorption) often is more a problem with proteins than dealing with their
`chemical
`stabilization. Physical
`instability actually involves
`solubility
`problems with large molecules. Although proteins generally contain many
`polar groups capable of ionization and hydrogen bonding with water, they
`also can contain many hydrophobic amino acids that under various
`conditions will preferentially self-associate,
`leading to aggregation and
`decreased solubility. Therefore, the development scientist nwds to know the
`structure of the protein and its conformation in solution in order to
`anticipate potential chemical and physical stability difficulties and then,
`using principles outlined in this chapter, develop formulation strategies
`which will overcome these instabilities. Protein formulations may also have
`Significant potential
`for supporting microbial growth as compared to
`smaller molecules. The problems associated with protein microbial growth
`promotion properties are covered in Chapter 3. Table ll summarizes some
`of the primary differences one must recognize in developing protein dosage
`forms compared to nonprotein dosage forms.
`
`Kopie von subito o.V,, geliet‘ort fiir din-mp (COMO4X01498)
`
`Ex. 1016 - Page 4 of 82
`
`Ex. 1016 - Page 4 of 82
`
`

`

`50
`
`Michael J. Akers et a1.
`
`Table 1]
`
`Protein versus Small Molecule Comparison from a Stability Standpoint“
`Protein
`Small molecule
`
`Many potential reactive sites
`Many ionizable sites
`Bufl‘er effect usually a unique
`acid/base catalysis
`Secondary/tertiary/quartemary structure
`Disperse (colloidal) aqueous systems
`Temperature effects can be
`discontinuous (denaturation)
`Readily supports microbial growth
`
`Few reactive sites
`Few ionizable sites
`Buffer efl‘ect usually general
`acid/base catalysis
`Lacks “higher order” structure
`Single and continuous phase in solution
`Temperature eflects are continuous
`
`" Courtesy. in part, of Dr. Lee Kirsch, University of Iowa, Iowa City, IA.
`
`3. GENERAL FORMULATION PRINCIPLES FOR PROTEINS
`
`Protein stability, both in the dry state and in solution, is the main
`reason why formulation science has such a presence in the commercial
`development of protein dosage forms. Proteins are complex in size and
`structure and, as macromolecules, contain a large number of functional
`groups. Generally, their biological activity in solution depends on a specific
`three-dimensional conformation. Almost every conceivable environmental
`factor (e.g., temperature, light, water, pH, presence of glass, rubber, or
`plastic, shear, presence of salts and other solutes, both macromolecules and
`low molecular weight compounds, detergents, or sanitizing agents, nature of
`the filling processes, freeze-thawing, freeze-drying) can effect conforma-
`tional changes and lead to denaturation, aggregation, or adsorption to
`surfaces. The challenge to formulation scientists is to develop a stable
`formulation that can be consistently manufactured and is stable in a given
`pacltagrng system over the shelf life of the product. A reasonable target
`expiration dating is 18 months to 2 years at ambient temperature or, failing
`this, at
`refrigerated conditions. Aqueous,
`ready-to-use solutions are
`preferable dosage forms for many reasons (convenience, cost, customer
`acceptance), but most proteins are not sufficiently stable in solution to allow
`practical expiration dating. Therefore, most protein dosage forms are solid
`forms in the commercial package with the solid form being produced by
`m'di‘ymg-‘Stabihty data should include not only the freeze-dried solid,
`but “13° some“, stability after reconstitution with an appropriate vehicle.
`Most additives in protein formulations are needed for stability
`pUIPOSes. These include buffers to enhance stability against specific acid/
`base-catalyzed hydrolysis, antioxidants, chelating agents, and inert gases to
`
`Kopie von subito 63.V., geliefert fiir cit-mp (COMO4X01498)
`
`Ex. 1016 - Page 5 of 82
`
`Ex. 1016 - Page 5 of 82
`
`

`

`Formulation Development of Protein Dosage Forms
`
`5!
`
`enhance stability against oxidative degradation, cryoprotectants/lyopro-
`tectants to enhance stability during freeze-drying of the protein product,
`surface-active agents to minimize interfacial dcnaturation, and excipients
`(e.g., albumin) to minimize protein adsorption to inert surfaces such as
`glass. The best formulation strategy is to keep the formulation as simple as
`possible,
`to have a clear reason for including each additive, and,
`if
`possible,
`to use excipients that have previously been used in Food and
`Drug Administration (FDA)-approved formulations. Hundreds of articles
`have appeared in the literature just in the last 20 or so years that report
`the stabilizing effects of additives on various proteins. We will reference
`those we feel have the most
`relevance to the industrial
`formulation
`scientist.
`
`It is also important in this age of globalization that the formulation
`scientist develops a formulation that is acceptable worldwide. This is not an
`easy assignment, because there are many commonly used additives
`acceptable in one country, but not another. For example, disodium
`ethylenediaminetetraacetic acid (DSEDTA) is acceptable for use in inject-
`able products in the United States and Europe, but is not acceptable in
`Japan. Levels of antimicrobial preservative agent(s) needed to pass the
`United States Pharmacopeia (USP) preservative eliicacy test are much lower
`than levels required to pass the European Pharmacopoeia (EP) test.
`Other additives in protein products (not including controlled drug
`delivery systems) serve one or more of the following functions:
`
`0 Agents for antimicrobial preservation
`0 Agents for solubility enhancement
`0 Bulking agents for freeze-dried products
`0 Agents for achieving isotonicity
`Most proteins alone and in final product formulations support the
`growth of microorganisms. The microbial growth properties of proteins
`alone and in the final product formulation should be well known and steps
`should be taken to assure that the antimicrobial properties of the final
`formulation meet the appropriate acceptance criteria. For multiple-dose
`PTOdUCtS, the addition of an antimicrobial preservative system is reunited to
`provide antimicrobial properties to the final product. Although including
`an antimicrobial preservative in a single-dose product has the advantage of
`Providing additional assurance against
`introduction of microorganisms
`during manufacturing, this practice is generally frowned upon by regulatory
`a8encies. Therefore, strict microbial control during manufacturing and the
`integfl't)’ of the packaging system all must be optimized in order to
`minimize the risk of inadvertent microbial contamination of the final
`Product.
`
`Kopio von subito e,V., goliefort for clf~mp (COMO4X01498)
`
`Ex. 1016 - Page 6 of 82
`
`Ex. 1016 - Page 6 of 82
`
`

`

`52
`
`Michael J. Akers e! a].
`
`Common Stability and Compatibility Problems with Proteins and Possible Solutions
`
`Stability problem
`
`Possible solutions
`
`Table III
`
`pH control, buffers, low ionic strength
`
`Hydrolysis, deamidation
`(e.g., asparagine deamidation)
`Oxidation (c.g., methionine
`oxidation)
`
`fi-Eliminaiion
`Transpeptidation
`Racemization
`Disultide exchange
`Denaturation during freezea‘lrying
`Aggregation, precipitation
`
`Antioxidants, chelating agents,
`low pH, oxygen-free processing
`and packaging
`Low pH. chelating agents
`pH control, lower concentratioa
`pH control, bufi'ers
`Thiol scavengers (e.g.. cysteine)
`Cryo~, lyoprotectants
`pH control, surface-active agents, minimize
`mechanical stress
`Adsorption to surfaces Surface—active agents, albumin, presaturationM
`
`
`
`The formulation scientist must be aware of the potential for adverse
`effects of low-level impurities in formulation components and packaging
`materials on physical and chemical stability of proteins. Impurities such as
`peroxides from surface-active agents and other polymeric agents, aldehydes
`from polymer synthesis and degradation, and extractables from rubber
`closures must be known and controlled to avoid both short-term and long-
`tenn adverse effects on product quality.
`Table III summarizes common stability and/or compatibility issues
`with protein dosage forms and suggested approaches for solving these
`issues.
`ese approaches will be covered in more detail
`later in this
`
`4. WHY PACKAGING, PROCESSING, AND FORMULATION
`ARE INTERRELATED
`
`the three are
`aspects of the manufacturing process or packaging. Yet
`interrelated. A formulation is not stable unless the product can be
`manufactured consistently at a large scale and packaged in a container/
`closure system that can maintain sterility and stability for a relatively long
`period of time. Packaging of proteins is especially challenging because 0f
`the inherent interactive nature of proteins with inert surfaces such as glaSS,
`rubber, and plastic. For many proteins, adsorption at
`these surfaces
`
`Kopie von subito ev.‘ gelicfert fUr cit-«mp (COMO4X01498)
`
`Ex. 1016 - Page 7 of 82
`
`Ex. 1016 - Page 7 of 82
`
`

`

`Formulation Development of Protein Dosage Forms
`
`53
`
`sometimes results in the surface denaturation and subsequent aggregation
`of the protein (Cleland et al., 1993). This includes interfacial denaturation
`at
`the air—water interface (e.g.,
`the headspace in a vial containing a
`protein solution). Minimizing foaming caused by agitation during
`manufacture, as well as during use of the product, may be critical
`in
`order to avoid significant loss of protein activity or generation of visible
`particulate matter.
`It is well known now that processing of protein formulations can affect
`protein stability. Examples include adverse effects of freezing and/or drying
`that occur during the lyophilization process, mixing/agitation processes, the
`filtration process, complicated manufacturing procedures requiring longer
`filling or hold times, and the movement of intermediate product from one
`location or site to another. In all these examples, the protein formulation
`must be designed to resist changes in potency, purity, and other physical—
`chemical characteristics of the protein itself and the finished formulation.
`The bottom line message here is simple: A formulation scientist
`developing .a protein (or, for that matter, any) dosage form must consider
`the formulation, process, and package together, not focus on one aspect
`exclusive of others. The smart formulation scientist, in fact, not only will
`consider all aspects ofthe formula, process, and package, but also will develop
`close interactions with packaging engineers, polymer scientists, manufactur-
`ing experts, and other experts in areas outside of the formulation scientist’s
`direct expertise. (“None of us is as smart as all of us"-—Satchel Paige.)
`
`5. COMMERCIALLY AVAILABLE PROTEIN DOSAGE FORMS
`
`Table IV summarizes U.S. marketed protein dosage forms approved by
`the FDA through 2000. The table contains information from the Physicians'
`Desk Reference (2001) on the dosage form, route of administration, and
`types and quantities of additives. Although preferential interaction experi~
`ments (e.g., Arakawa and Timasheff, 1982) can predict which solutes can
`serve as protein stabilizers, the majority of protein formulation research and
`development requires a great amount of trial and error to finalize the type
`and amount of formulation components. Prior “art,” in the sense of
`kUOWing what has worked before and, particularly for injectable formula-
`tions, which additives have a history of safety and regulatory acceptance,
`greatly assists the protein formulation scientist in developing stable, elegant,
`and manufacturable dosage forms. Characterization of protein structure, as
`well as collecting prefonnulation data as described in Chapter 1, Will
`provide supporting data for stabilizers and other additives that are most
`
`Kopie von subito e.\/., gellefert fU'r cit-mp (COMO/M01498)
`
`Ex. 1016 - Page 8 of 82
`
`Ex. 1016 - Page 8 of 82
`
`

`

`re
`
`«5.5.6M22.3.
`
`
`
`EEO.“380530.5888
`
`.6850%
`
`if};
`
`
`
`318.88388583%353%
`
`
`
`
`
`02.830533mhasnezsuaui§2u§~5¢
`
`
`
`03.38%8:683newswom“8850Snoodpan—33¢
`
`
`
`
`
`
`
`
`
`nfluoafixm$235??va$th.83an“85835225:03:.ans“:urn—BO
`
`
`
`
`
`
`
`Afiamm8ow“Maugham—omQ3:magma.833%:83050
`
`
`
`
`
`mo>um>uomoa5529:3885
`
`
`
`3..a5.8cmRagga—om£3588$SSEVEQ:33qu
`
`
`
`Eonuuocqmoam8.2Skid“5.43%
`
`
`
`
`
`94no.8oEobOS3.2393ExamBucfifiooom
`
`
`
`“Bumcows0mm2votVouoEm3:2mafia—wow~88.“cannofiuawg
`
`
`
`
`
`
`
`05534.;>~notvénoem£025508w>no<own—n82
`
`
`
`
`
`
`
`
`
`no}G5—5.5?8335:55Auguugv.fifiaubv
`
`
`
`“Eng3Smaumoza33-?35
`
`
`
`:28my028306328
`
`came0:5:32
`
`.m28%EEvom~33.QSEVx0383
`
`M§§El«SEQM3289aCS:3.5:E§§§«5.55:
`
`
`
`
`
`..MwQEEENV822wonoafiobomZ35.33amEazH5E8=<“ummawmooMBEEEén<
`
`
`
`
`
`38:0Eavomnovafiamoguom08590emu—.503.98333.
`
`
`
`
`
`I.Sagas3393m028:05:68
`
`
`
`
`
`
`
`
`
`
`
`2220wuESEu52030cowflmnfiflames/w58550053955325Sad—on
`
`Kopie von subito e.V., geliefert fUr df~mp (COMO4XO1498)
`
`Ex. 1016 - Page 9 of 82
`
`Ex. 1016 - Page 9 of 82
`
`
`

`

`Formulation Development of Protein Dosage Forms
`
`
`
`wmomIn8#:5quEEvom
`
`
`
`$03.8£83?53%52:3.»
`
`
`
`02830Eamoow
`
`2.8%828m
`
`Umno>~
`
`nowfiom
`
`580$055
`
`coma—2
`
`E853:33
`
`
`
`nowoqmcsfionokfibm
`
`
`
`KEEEatfiefi
`
`CSNBNEMK?
`
`
`
`
`
`wfioxmrb.«08:98.233
`
`38256
`
`>~
`
`aougom
`
`~00
`
`£5258:353
`
`@0399cm33858
`
`3::52
`
`E83.82::qu
`
`
`
`BSUUNEnmuom
`
`Um.5>H
`
`cows—om
`
`uowE<
`
`
`
`comanozEmma—E
`
`MEm6:0398mgmoEcm
`
`S:
`
`.5323
`
`M9530
`
`
`
`mkuowamv58?m333»:
`
`wEWeno338%“woe$6
`
`5255:?
`
`Rmobfimf.
`
`£253GB:58;323
`
`€2.53$88.825
`
`8083
`
`
`
`28.38:.“EEvom
`
`
`
`magmas—E828m
`
`
`
`02520Eavow
`
`82:82“
`
`#882559
`
`
`
`
`
`95sz02820Eavow
`
`
`
`
`
`cfiéesome35%323
`
`
`
`953333.9.38..“:23
`
`
`
`2233—3Eavow
`
`33:52
`
`S:
`
`S:
`
`s:
`
`S:
`
`5359.5
`
`x962
`
`$5988on
`
`
`
`935.?38me
`
`~33duauwé
`
`953%39.6%
`
`5.?.580va
`
`Buchanan
`
`
`
`5.5..233:
`
`
`
`95:.33
`
`Ecvaam
`
`ozohum
`
`Macaw
`
`:05ka
`
`
`
`Samoan922.63058::
`
`cacaovucow
`
`Enohovucow
`
`notYuNookm
`
`.&u><.fiob$
`
`Addemactonu55::
`
`55
`
`SwaiEBV
`
`ill-ii:
`
`Kopie von subito o.V., geliefert fUr (If-«mp (COMMXOMQB)
`
`Ex. 1016 - Page 10 of 82
`
`Ex. 1016 - Page 10 of 82
`
`

`

`Michael J. Akers et a1.
`
`SE283:0Ezmuom
`
`§§vafikogom
`
`
`
`02530528m
`
`Sane65$
`
`
`
`Game336%
`
`€820
`
`£29666-:
`
`
`
`AXQQOV355
`
`
`
`SmanmonmEnmuow
`
`oEESoE
`
`
`
`31.8nonflmfimfioz
`
`”€258:68
`
`238mEivom
`
`oExooEN
`
`
`
`02.550EEvow
`
`0&0me
`
`AfiLodVowBunuombom
`
`9:253388685so:
`
`€253€820
`
`
`
`oamnnmonnEivom
`
`3::qu
`
`ofiobO
`
`
`
`muumnamonmEafiom
`
`3:522
`
`
`
`
`
`2253Some20%353
`
`
`
`9%:me38R».
`
`5.5.538$
`
`Ow
`
`85802.,”
`
`.35cactus—i
`
`532m:35:
`
`one—Eon
`
`Um
`
`Um
`
`Um
`
`Um
`
`0w
`
`
`
`C5:.23Sh:583.5
`
`3.5.5505
`
`
`
`£02550590532
`
`cow—.58
`
`30255.0OZ:EESZ
`
`gozfiswm
`
`conaom
`
`:3a£25m
`
`
`
`£33:583m
`
`202.23%
`
`commcommnw
`
`:5z535.5
`
`£35583m
`
`802-9335
`
`momwcummam
`
`35._538.5
`
`.5ng555m
`
`S:.5>H
`
`€38quEamon306
`
`«53:35:83
`
`85.53.,”
`
`~326:8va
`
`szmam225.3
`
`63£33»u§§9
`
`
`
`5.15onusage”
`
`1111':
`
`$3255632snub
`
`iii!
`
`nflaoimoxm
`
`92235383
`
`goSnow
`
`Q58ESE
`
`
`
`
`
`housuuflncmEcan:own;
`
`
`
`a3.3:0.550
`
`Kopie von subito eV geliefert fUr de—mp (COMMXOMQB)
`
`Ex. 1016 - Page 11 of 82
`
`Ex. 1016 - Page 11 of 82
`
`

`

`Qwaztcsy
`
`
`
`
`
`magmfivfla5m?viva—:85
`
`Formulation Development of Protein Dosage Forms
`
`$.3835.x
`
`$562th
`
`
`
`EEBEBaum555$
`
`«€25eavom
`
`£38BEE
`
`Om.8S:
`
`noun—owDEM
`
`31.85833SEE:
`
`
`
`0282“.5:333
`
`
`
`moumnamonm:56an
`
`
`
`unto—noEnmuom
`
`£8.58E335E5
`
`
`
`mouwnumonnEavom
`
`$3.96:83{Sam
`
`2:220
`
`
`
`5E3?820m5352
`
`
`
`moumnmwonn8.:va
`
`2530
`
`838mm
`
`
`
`EEBEEEum5E5:
`
`155.8cmowmnuombom
`
`22%5:68.
`
`3mm46onE83:.
`
`
`
`Eocaémcoofi“2.2
`
`
`
`oUtoEoEauow
`
`33:52
`
`
`
`03583.EEvow
`
`Eta—EX
`
`=28EE~
`
`Um.8S:
`
`Um
`
`Um
`
`Snaga
`
`no_mo_~.:5
`
`Um8«5
`
`5.3..ESQEV
`
`Illllllllllllllllllillll'
`
`
`
`5...:.588:
`
`
`
`Q5:~33».
`
`
`
`:8Ex!ESEEXV
`
`CESSEE
`
`2:33“.
`
`=25
`
`Egan—2m
`
`23.5.5.5
`
`newECVoNuEm
`
`unuox
`
`(éohzom
`
`nouaom
`
`cokutouE
`
`zGOuDM~<
`
`Essia:8be»:
`
`
`
`«NacohotSE
`
`
`
`Qtui3:5va
`
`2782283.5
`
`
`
`2%::3
`
`85-325
`
`waEfim
`
`<855
`
`32§£:855:
`
`
`
`3N8centen—
`
`nous—om
`
`metasom
`
`<:82:
`
`5535=85.65
`
`
`
`ouécohotoufi
`
`vutvénufim
`
`3:3
`
`8&6
`
`cohomfiom
`
`
`
`3.3%?”£5.33:mTn965:8:—
`
`:ociom
`
`£92550
`
`acne—52x
`
`
`
`m1.cohotsfi
`
`Kopie von subito e.\/., gefiefert far df~mp (COMO4XO1498)
`
`Ex. 1016 - Page 12 of 82
`
`Ex. 1016 - Page 12 of 82
`
`

`

`Michael J. Akers et al.
`
`
`
`owSachembomJozcamfid
`
`.2568020«55880
`
`E926~23vBEcmaoooM
`
`
`
`«Eonozoubflocgqia
`
`33:56.9
`
`2%38¢.
`
`
`
`.55quS:vocvénooi«839$52.Saun—uoaaqo§8<3:05:04
`
`
`
`
`
`02530£38
`
`31.8on33.83an2aoaiom580593033ququnebaaofiofiz
`
`
`
`mozfiumoaaEivomMEMO0.382%?»235:5
`
`
`
`3:83“Sum
`
`
`
`
`
`
`
`
`
`02.830EEuomUmaovaom«chasmmad.moan—A888<unmanned
`
`>—~33.
`
`Qussutgv
`
`..
`
`
`
`
`
`
`
`umunoaaxm«o8.8M$58Rumba“839$:quuEun38H:25:0:250
`
`E“nozmbmfimfica
`
`N.ma83&3“assume—ESE.2.ch23.8.?§h§§$
`
`
`
`
`029:0Umvofluéuooum8332.:8350umofinuz2.535%:
`
`
`
`
`
`
`
`
`
`CBS.3.28.4.SrmeAueaESS
`
`
`
`“2&«x\w§h3355‘
`
`S\~$~:Miogfi
`
`Kopie von subito eV geliefert f0!" df~mp (COMO4XO1498)
`
`Ex. 1016 - Page 13 of 82
`
`Ex. 1016 - Page 13 of 82
`
`

`

`Formulation Development of Protein Dosage Forms
`
`
`
`
`mgtmtomoafir:usgafioofi0:35.55
`3:5“:nose“:2EEYuNuEmxo=:EE~2533Eznofifimuwm
`
`
`30.6%$3.28kammuietufiakaaxanfiuv
`mm”:9.3»3.5:.E53:«:3855.9
`
`
`
`
`
`
`
`
`
`
`
`5:6853>@83vamgSE..ESEV53.5.3ESNESQV
`
`6&3Banamona«5.935835.5%333..E3.3»
`
`
`
`
`
`
`3:532Um355N005oafiomwince“2008385£225“
`
`.0333»
`
`Eamon$96«2828:
`
`
`
`“$8.8ouaonombomAuficfisbBimgtég
`38:0EnmwomOnIn,:3i.3EwESEb
`
`
`
`02.62“.EavomZcocsmom53550555%nmfimxsgx
`
`
`
`
`
`
`
`£233E38:25:comma“:Q5:3.5:.
`
`
`
`
`
`
`
`
`
`03830Eumvombungee43.:.3335
`
`5%
`
`
`
`
`
`6:552.32:.233532»scant.unnamxopp<82395
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`38amass—ES"Duo—cu.umo£2.23...1.2335.”2”
`
`
`
`
`
`.8:o.33..3.5m€5.23.83£028?.5.“.833E3.3520E293<3.om2.533E(5.39:50.BAD.1833592..uco§
`
`
`
`
`
`
`
`3382555«55.883353Um.3825":.>—..
`
`
`
`.82:=8.»5!§~3~3USU:
`
`Kopie von subito e.V., geliefert (Ur df~mp (COMO4XO1498)
`
`Ex. 1016 - Page 14 of 82
`
`Ex. 1016 - Page 14 of 82
`
`

`

`60
`
`Michael J. Akers et a].
`
`likely to be effective. For example, prolubility/stability studies will
`provide direction on what type of buffers to use, if any. Having knowledge
`of the structural conformation of a protein in order to predict which amino
`acids in the protein sequence may be particularly vulnerable to degradation
`because of exposure to the environment may give the formulation scientist
`some direction on the stabilizers required. The final selection of excipients,
`unfortunately, must be a result of much empirical evaluation. However,
`information such as that given in Table IV summarizes what others have
`done with their protein products, and thus can give significant guidance to
`formulation scientists facing the development and stabilization of new
`peptide and protein dosage forms.
`Note that protein dosage forms primarily are divided into three types
`
`1. Ready-to-use solutions
`2. Freeze—dried powders that are reconstituted into solutions immedi-
`ately before administration
`3. Ready-to-use suspensions
`
`Proteins are commonly formulated at very low doses (very dilute
`solutions), although there are examples of relatively high dose protein
`products, such as formulations of immunoglobulin G (IgG) at 50 mg/ml. 1“
`general, dilute solutions are less physically stable than more concentrated
`solutions (Hanson and Rouan, 1992) and adsorption to surfaces will result
`in a higher fractional loss of protein. However, in the case of the Neutral
`Protamine Hagedorn (NPH) formulation of insulin, the rate of formation of
`higher molecular weight polymers increases as a function of concentration
`(Brange et
`(11-, 1992b). Also, for interleukin 1,8 (IL-1,8), aggregation/
`precipitation was shown to demonstrate biphasic kinetics (slower rate
`followed by a more rapid rate) at temperatures lower than 55°C and to be
`dependent on concentration. When the concentration was increased from
`100 to 500 mg/ml, the slower rate was observed to be suppressed and a more
`rapid degradation was observed (Gu et al., 1991). In general, however, the”
`are surprisingly few literature reports of protein stability as a function of
`concentration.
`
`6. CHEMICAL STABILIZATION
`
`6.1. pH, Hydrolysis, and Buffers
`
`The effect of solution pH on stability is probably the most impo11am
`factor to study in early protein dosage form development. Figure l
`
`Kopie von subito e.V., geliefert for df~~mp (COM04X01498)
`
`Ex. 1016 - Page 15 of 82
`
`Ex. 1016 - Page 15 of 82
`
`

`

`Formulation Development of Protein Dosage Forms
`
`61
`
`schematically depicts expected stability problems of proteins as a function of
`pH. pH~stability studies are conducted very early to understand relative
`protein stability over a pH range, typically from about pH 3 to about pH 10.
`The relationship of stability and solubility to pH usually follows a pattern of
`higher solubility resulting in lower chemical stability and lower solubility
`resulting in lower physical stability. Protein solubility is usually at a minimum
`at its isoelectric point. Insulin, for example, has an isoelectric point of 5.4, and
`at this pH it is quite insoluble in water ( < 0.1 mg/ml). Adjusting the solution
`pH to less than 4 or greater than 7 greatly increases insulin solubility (> 30
`mg/ml, depending on zinc concentration and species source of insulin), but
`also increases the rate of deamidation at these pH ranges (Brange, 1992). An
`example of the effect of pH on deamidation and polymerization of insulin is
`shown in Fig. 2 (Brange and Langkjaer, 1993). In dosage form development,
`the scientist must first determine what pH range provides acceptable
`solubility of the protein for proper dosage, then determine whether this pH
`range also provides acceptable stability. There is usually a trade—oil" between
`solubility and stability and it is up to the scientist to identify what pH is
`optimal for both. When an acceptable trade-off does not exist for a solution
`formulation, a freeze-dried formulation is usually indicated.
`Hydrolysis or deamidation occurs with proteins containing susceptible
`Asn and Gln amino acids, the only two amino acids that are primary
`amides. The side-chain amide linkage in a Gln or an Asn residue has been
`shown to undergo deamidation to form free carboxylic acid. Deamidation
`
`WW...
`
`Selectively Oxidized
`Under Acidic Conditions
`
`.
`
`.
`
`Peptide cleavage
`
`Un-ionizcd Terminal
`
`Carboxylic Acid
`C-terminal
`Carbonyl—arm
`
`deamidation
`reactions
`
`
`Un~ionized Side Chain
`
`
`Carboxylic Acids
`
`
`
`
`
`.
`Cleavage of Dissacharide
`
`ng l
`3.2!“
`Disaccharide browning
`to Reactive Species
`
`
`
`
`
`
`
`V .
`oxidations
`
`Un-ionized Protein
`Amides
`
`Base-Caulyzed
`
`Ionized Cysteine
`
`Acid
`
`Neutnl
`
`3‘59
`
`Kopie von subito e.V., geiiofert for cit-«mp (COMOIIXOMQB)
`
`Ex. 1016 - Page 16 of 82
`
`Ex. 1016 - Page 16 of 82
`
`

`

`62
`
`Michael J. Akers et a1.
`
`%
`
`Hydrolysis
`
`Monodesamido
`
`m Didesamido
`Splil producl
`
`
`
`
`3.5
`
`4.0
`
`4.5
`
`5.0
`
`5.5
`
`6.0
`
`6.5
`
`7.0
`
`7.5
`
`8.0
`
`pH
`
`Polymerization
`
`3.5
`
`4.0
`
`4.5
`
`5.0
`
`5.5
`
`6.5
`
`7.0
`
`7.5
`
`8.0
`
`6.0
`pH
`
`m 2. 31mm! transformation of insulin during storage or rhombohedral insulin crystals
`f
`l 2no ms
`In crystals: 0.? /0 NaCl, 0.2% phenol) as a function of pH during storage at 25°C
`.or
`_ morxths. (A) Formation of the hydrolysis products mono and didesamido insulins and the
`insulin 5'91" product (AB-A9). (B) Formation of covalent dimers and oligomets. Reprinted with
`permission from Brange and Langkjaer (1993). Copyright I993 Plenum PMS.
`
`Kopie von subito e.V., geliefert fUr df-ump (COMO4X01498)
`
`Ex. 1016 - Page 17 of 82
`
`Ex. 1016 - Page 17 of 82
`
`

`

`Formulation Development of Protein Dosage Forms
`
`63
`
`including extremes in pH,
`can be promoted by a variety of factors,
`temperature, and ionic strength. Many investigators have observed altered
`forms of proteins which have been attributed to this deamidation process.
`Robinson and Rudd (1974), Geiger and Clarke (1987), and Clarke et al.
`(1992) reviewed the chemical conditions necessary for hydrolysis of Asn and
`Gln residues for many proteins. It was shown that neutral or alkaline
`conditions enhanced the rate of deamidation of proteins mainly at the Asn~
`Gly sequence. This rate was also found to be higher than the hydrolysis of
`the amino acid Asn itself. This deamidation proceeds through a five-
`membered cyclic imide intermediate formed by intramolecular attack of the
`succeeding peptide nitrogen at the side-chain carbonyl carbon of the Asn
`residue. The cyclic imide then spontaneously hydrolyzes to give a mixture of
`peptides in which the backbone is attached via either an Asp (at-carboxyl) or
`an iso-Asp (fi-carhoxyl) linkage. Haley e! a]. (1966) showed that Gln also
`undergoes a similar deamidation reaction via the formation of a six-
`membered ring. Peptide backbone hydrolysis also has been shown to occur
`from this cyclic imide (Tyler-Cross and Schirch, I991).
`One approach that has been used to lower the rate of deamidation is to
`reduce pH, because in general, deamidation is slower at acidic pH than at
`neutral or alkaline pH. However, caution must be exercised here, because a
`reduction in pH may lead to cleavage or cyclization at Asp-X residues, where
`X is usually a residue with a small side chain, such as Gly or Ser. Proteins with
`Asp~X degradation must be formulated at a higher pH to avoid cyclization
`(Manning et (11., 1991). However, higher pH conditions (i.e.,greater than pH 8)
`may catalyze oxidation, thiol—disulfide exchange, and [i-elimination reactions.
`In the case of insulin, multiple deamidation sites are observed, where
`deamidation at the A21 position predominates at acidic pH and deamidation
`at B3 predominates at neutral pH (Brange, I992).
`Buflers. Buffers are used to prevent small changes in solution pH which
`can affect protein solubility and stability. Buffers are composed of salts of
`ionic compounds,
`the most common of which are acetate, citrate, and
`phosphate. Buffer systems acceptable for use in parenteral solutions are
`listed in Table V.
`
`The proper selection of buffer type and concentration is done by
`Performing solubility and stability studies as a function of pH and butter
`species. In general, it is good practice to keep the buffer concentration as low
`as practical.
`.
`'
`Potential problems associated with using buffers include the followmg.
`
`1. It may be difficult to meet the pH target with a bufi‘er system while
`preparing solutions during sealeup and full-scale manufactunng.
`Dilute solutions of strong acids (hydrochloric acid) or bases (sodium
`hydroxide) usually are required, which may alter the buffer capacrty
`
`Kopie von subito e.V., goliefort ftir dt'mmp (COMO/M01498)
`
`Ex. 1016 - Page 18 o