The AAPS Journal 2006; 8 (3) Article 59 (http://www.aapsj.org).
`Themed Issue: Proceedings of the 2005 AAPS Biotec Open Forum on Aggregation of Protein Therapeutics
`Guest Editor - Steve Shire
` Effects of Protein Aggregates: An Immunologic Perspective
` Submitted: March 3 , 2006 ; Accepted: May 24 , 2006; Published: August 4 , 2006
` Amy S. Rosenberg 1
` 1 Food and Drug Administration, Center for Drug Evaluation and Research, Bethesda, MD
`
` A BSTRACT
` The capacity of protein aggregates to enhance immune
`responses to the monomeric form of the protein has been
`known for over a half-century. Despite the clear connection
`between protein aggregates and antibody mediated adverse
`events in treatment with early therapeutic protein products
`such as intravenous immune globulin (IVIG) and human
`growth hormone, surprisingly little is known about the nature
`of the aggregate species responsible for such effects. This
`review focuses on a framework for understanding how aggre-
`gate species potentially interact with the immune system to
`enhance immune responses, garnered from basic immuno-
`logic research. Thus, protein antigens presented in a highly
`arrayed structure, such as might be found in large nondena-
`tured aggregate species, are highly potent in inducing anti-
`body responses even in the absence of T-cell help. Their
`potency may relate to the ability of multivalent protein spe-
`cies to extensively cross-link B-cell receptor, which (1) acti-
`vates B cells via Bt kinases to proliferate, and (2) targets
`protein to class II major histocompatability complex (MHC)-
`loading compartments, effi ciently eliciting T-cell help for
`antibody responses. The review further focuses on protein
`aggregates as they affect an immunogenicity risk assessment,
`the use of animal models and studies in uncovering effects of
`protein aggregates, and changes in product manufacture and
`packaging that may affect generation of protein aggregates.
`
` K EYWORDS: aggregates , container closure system , immu-
`nogenicity , neutralizing antibody , tolerance
`
`isms depends on complex protein structures critical for
`invasion and propagation such as enzymes (eg, neuramini-
`dase). The immune system has both an innate response
`armamentarium, consisting of multiple cellular and humoral
`pattern receptors, including toll-like receptors (TLRs),
`which bind to conserved molecular patterns and trigger
`rapid defense responses, as well as an adaptive immune
`response arm that responds to unique microbial proteins.
`Such adaptive responses include pathogen-specifi c anti-
`body, which may be generated independent of T-cell help
`(ie, a T helper independent [TI] response) or, less effi ciently,
`may require collaboration with T-helper cells (Th) (ie, a
`T-helper dependent [TD] response).
` An important element of the effi cient response to microbial
`antigens is the rapid production of antibody by B cells that
`do not require T-cell help. The requirements for generation
`of antibody to such determinants have been examined in
`detail for both relatively simple polymers of peptides and
`polysaccharides, as well as for higher order structures such
`as viral capsids, composed of repetitive arrays of multiple
`protein components. Investigating the molecular require-
`ments for TI antibody responses to polysaccharides, Dintzis
`et al 1 hypothesized that the B-cell stimulatory signal is
` “ quantized ” in that a minimum number of cell-surface-
`expressed antigen receptors must be connected together in a
`spatially contiguous cluster that they defi ned as an “ immu-
`non. ” They found that the molecular mass of polymers that
`successfully triggered TI antibody responses exceeded 100
`kDa and that the valence of the hapten moieties exceeded
`10. Successive work by many in the fi eld indicated that the
`generation of a signaling complex leading to B-cell activa-
`tion and antibody production depended on factors in addition
`to molecular mass and hapten valency, including hapten
`affi nity, polymer rigidity, and binding kinetics. 2 More
`recently, Bachmann and Zinkernagel 3 extensively examined
`the effects of antigen organization on antibody responses by
`challenging transgenic mice, tolerant to a viral coat protein
`from vesicular stomatitis virus (VSVgp), with VSVgp in
`progressively more ordered arrays. Despite tolerance to the
`soluble monomeric protein, presentation of VSVgp in a
`loose aggregate was able to generate antibodies in a TD
`fashion, while its presentation in a highly ordered viral cap-
`sid generated antibody in the absence of T-cell help, 3 clearly
`demonstrating the potency of highly ordered structures in
`eliciting rapid and effi cient immune responses.
`E501
`
` INTRODUCTION
` The immune system has evolved to respond to invasive
`threats posed by microbial organisms. Such organisms have
`inherent “ signatures ” consisting of repetitive displays of
`proteins, polysaccharides, nucleic acids, or lipids on their
`external surfaces. In addition, pathogenicity of such organ-
`
` Corresponding Author: Amy S. Rosenberg, Director of
`Division of Therapeutic Proteins, Food and Drug
`Administration, Center for Drug Evaluation and Research,
`Building 29A, Room 2D-16, 8800 Rockville Pike,
`Bethesda, MD 20892 . Tel: (301) 827-1790 ; Fax: (301)
`480-3256 ; E-mail: amy.rosenberg@fda.hhs.gov
`
`Novartis Exhibit 2291.001
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`The AAPS Journal 2006; 8 (3) Article 59 (http://www.aapsj.org).
`only linear determinants 8 ; and neutralizing antibody to
` The central conclusion of this body of work is that high
`human thrombopoietin (TPO) was not inhibited by short
`molecular weight (MW) arrays of antigen are highly effi cient
`peptides of TPO that would have interfered with antibody
`in eliciting an antibody response independent of T-cell help,
`binding to linear, but not conformational, determinants. 9
`whereas their less ordered counterparts are less immunogenic
`and may require T-cell help to generate an immune response.
` So, how do protein aggregates relate to this “ immunologic
`imperative ” ? First, we need to defi ne what we mean by pro-
` The activation of antibody-producing B cells via the exten-
`tein aggregates. For the purposes of this review, written
`sive cross-linking of B-cell receptor (BCR) by multivalent
`from an immunologic and not a biochemical perspective,
`ligand involves activation of Bruton ’ s tyrosine kinase/Tec
`protein aggregates are defi ned very broadly as high MW
`kinase. 4 While this signaling apparatus is of itself suffi cient
`proteins composed of multimers of natively conformed or
`to induce proliferation of B cells, antibody production ap -
`denatured monomers. Such species may be soluble or insol-
`pears contingent on delivery of a second signal, mediated
`uble (particulate), and reversible or irreversible, within the
`by Th cells, or potentially by other signaling pathways
`given environment. Moreover, in keeping with an immuno-
`including those mediated via TLR. 2
`logic focus on the consequences of such species, 3 types of
` Microbial pathogens express many enzymes and other
`protein aggregates are considered: the fi rst, an assembly of
`structures necessary for viral invasion and proliferation.
`native proteins in a polymeric structure; the second, an
`The activity of such pathogenic proteins depends critically
`assembly of denatured protein irreversibly associated
`on the 3-dimensional conformation of the active site. To
`(within the given environment) and dependent on hydro-
`protect the host, the immune system makes antibodies that
`phobic interactions; and the third, covalently linked pro-
`are specifi c for the active site of the protein (ie, conforma-
`teins, which could be either in a native or denatured state.
`tion-dependent antibody that neutralizes enzymatic activ-
` The hypothesis is that the potency of protein aggregates
`ity). In fact, this preference for antibody to conformational
`and, particularly, of particulate protein aggregates in elicit-
`rather than linear determinants of proteins appears to be a
`ing immune responses pertains to their resemblance to the
`general property of the immune system. Thus, even when
`microbial “ signatures ” to which the immune system has
`proteins pose no threat, antibodies are preferentially directed
`evolved. Thus, immunity to defi ned proteins can be enhanced
`to the conformationally confi gured portions of the protein.
`by ensuring aggregation of native proteins in a rigid (eg,
`For example, in a recent study by Ito et al, 5 it was found that
`virus capsid) type of presentation, 10 an advantageous
`in mice immunized with isolated hen egg lysozyme (HEL),
`approach for vaccines. In contrast, minimization of im -
`a complex protein with signifi cant tertiary structure, all of
`munogenicity of therapeutic protein products is best
`the mAbs elicited (15/15) reacted to native protein, with
` accomplished by ensuring stability of the native protein
`only 2 of the antibodies also recognizing linear epitopes
`conformation and minimizing formation of high MW spe-
`present in the reduced form, implying that conformational
`cies. Our model, thus, envisions that antibody responses to
`determinants were the binding ligands. The group found
`high MW protein aggregates may be elicited in a similar
`that this was the pattern for 4 different nonmicrobial pro-
`fashion to those elicited by microbial pathogens, (ie, via
`teins that underwent similar scrutiny.
`multivalent ligand cross-linking of BCR).
` There are 2 principal pathways by which multivalent ligand
`cross-linking of BCR enhances immune responses: (1)
`acceleration of antigen processing and presentation to Th
`cells, 11 and (2) activation of Bruton ’ s tyrosine kinases me -
`diating B-cell proliferation. 4 Thus, early and rapid IgM
`responses may be generated to aggregated protein in advance
`of the provision of T-cell help, but unlike polysaccharide
`antigens, which are not capable of recruiting T-cell help,
`protein aggregate-mediated BCR cross-linking would be
`expected to accelerate the recruitment of T-helper cells.
`
` Additional evidence of the immune system ’ s proclivity for
`conformation-dependent antibody is found in responses to
`endogenous proteins such as amyloid proteins. Thus, Nath
`et al 6 and O ’ Nuallain and Wetzel 7 found that patients with
`Alzheimer ’ s disease mounted signifi cant antibody responses
`to A b amyloid structures, but that such antibodies did not
`bind to the monomeric A b amyloid precursor protein. Intrigu-
`ingly, O ’ Nuallain and Wetzel found that antibodies to A b
`amyloid structures cross-reacted signifi cantly on amyloid
`structures from other proteins whose primary amino-acid
`sequences differed dramatically from that of the A b amyloid
`precursor. Thus, antibody responses to these proteins are
`specifi c for common higher-order structural elements.
` INDUCTION OF IMMUNE RESPONSES BY PROTEIN
`Finally, neutralizing antibody responses to therapeutic
`AGGREGATES: QUALITATIVE FACTORS
` protein products appear to be dominated by conformation-
`specifi c antibody: serum from 12/13 patients with pure red
` Both qualitative and quantitative factors pertaining to the
`cell aplasia (PRCA) resulting from antibodies to erythropoi-
`aggregate species per se, as well as factors independent of
`etin failed to bind to denatured erythropoietin containing
`protein structure or amount are important in the ability of
`E502
`
`Novartis Exhibit 2291.002
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`The AAPS Journal 2006; 8 (3) Article 59 (http://www.aapsj.org).
` RISK ANALYSIS OF PROTEIN AGGREGATES IN
`any particular aggregate species to induce an immune
`response. Among the qualitative factors critical in inducing
`THERAPEUTIC PROTEIN PRODUCTS
`antibody responses are MW and solubility. While low MW
` Protein aggregates pose risk in terms of generation of
`aggregates such as dimers and trimers appear ineffi cient in
`immune responses to the therapeutic protein product. Of
`inducing immune responses, large multimers whose MW
`principal concern are those immune responses associated
`exceeds 100 kD are effi cient inducers of immune responses.
`with adverse clinical effects: neutralizing antibody that
`However, multimerization is key to immunogenicity, as
`inhibits the effi cacy of the product or, worse, cross-
`larger-sized monomeric proteins are not necessarily more
`reactively neutralizes an endogenous protein counterpart;
`immunogenic than smaller monomeric proteins. Moreover,
`and severe immediate hypersensitivity responses such as
`it has been long known that particulate (insoluble) antigens
`anaphylaxis.
`are very rapidly endocytosed by antigen-presenting cells,
` Regarding immediate hypersensitivity responses, mediated
`which initiate immune responses. 12 More recent work has
`by IgE, the role of protein aggregates in generating such
`shown that blood-borne dendritic cells (CD11c lo Mac1+)
`responses is not known. In fact, speculation by Aalberse
`capture particulate antigens in the periphery, migrate into
`and Platts-Mills 17 is that aggregates, in increasing the
`spleen, and induce TI antibody responses by activating and
` “ strength ” of an immune stimulant, might deviate an IgE
`enhancing the survival of marginal zone (MZ) and B1 B
`response to a far less devastating IgG4 response. Moreover,
`cells. 13 Thus, it is possible that particulate protein aggre-
`a response may be driven from IgE to IgG via TLR- mediated
`gates introduced into the systemic circulation meet a similar
`signaling. 18 Nonetheless, once an IgE response has been
`fate, and traffi c to MZ B cells, where the immune response
`generated, aggregated allergen is highly effi cient at trigger-
`proceeds independent of T-cell help.
`ing degranulation of mast cells via aggregating IgE
` Other critical factors bearing on immunogenicity of
`receptors.
` aggregates pertain to both product and host: product origin
`(foreign versus endogenous), the presence of product con-
` In considering the mechanism by which protein aggregates
`taminants with immunomodulatory activity, the presence of
`might generate neutralizing antibody responses in a highly
`neoepitopes (as may be created in fusion proteins), and gly-
`effi cient manner, it is clear that preservation of the confor-
`cosylation/pegylation may exert profound infl uences on
`mation of the active site of the protein or proteins within the
`either the formation of aggregates, or the generation of
`aggregate is required. This strategy is potentially useful for
`immune responses to aggregates. Host and protocol factors
`vaccines, where it is critical to neutralize the natively con-
`important in determining the potency of aggregate species
`formed structures responsible for invasion or toxicity. In
`in triggering immune responses include the frequency of
`contrast, this confi guration might prove devastating for
`administration, the route of administration, the immuno-
`therapeutic proteins. Of great importance for avoiding the
`modulatory activity of the product itself, the host immune
`formation of natively conformed aggregates in therapeutic
`protein products is product formulation. 19-22 Thus, “ struc-
`status, the activity of concomitant immunomodulators, and
`tural ” formulations that form a bilayer into which proteins
`for therapeutic versions of endogenous proteins, the robust-
`can potentially insert, such as nonionic and ionic detergents
`ness of immunologic tolerance to the endogenous protein.
`and liposomes, are of interest. Insertion of proteins into
` Thus, small amounts of aggregates would be expected to
`such bilayers is feasible for membrane proteins, which con-
`generate a robust response to foreign proteins, but not nec-
`tain clustered hydrophobic residues, whereas such a confi g-
`essarily for therapeutic counterparts of endogenous proteins
`uration is not common for soluble therapeutic proteins.
`to which the immune system has been tolerized. However,
`Moreover, micelles of the nonionic detergent, polysorbate,
`the mammalian immune system is not equally tolerant to all
`are relatively small and likely not capable of incorporating
`endogenous proteins. Most notably, the works of Weigle 14
`more than a few therapeutic protein molecules, whereas
`and of Goodnow 15 have demonstrated that the relative abun-
`the extended bilayers of ionic detergents, such as sodium
`dance and manner of presentation of the endogenous pro-
`dodecylsulfate (SDS), may contain substantial numbers
`tein are essential in determining the extent to which the
`of proteins capable of insinuating themselves into the
`immune system is tolerized to that protein. Moreover, T
`structure.
`cells are tolerized more readily by proteins at low abundance
` More likely to be present in therapeutic protein products are
`than are B cells, perhaps because of the active expression of
`denatured protein aggregates that are generated via unfold-
`tissue-specifi c antigens in the thymus by promiscuous gene
`ing of the native conformation of the protein. While aggre-
`expression. 16 Thus, aggregates of relatively low abundance
`gates of denatured protein may potently induce antibody,
`proteins, to which we are not robustly tolerant, might
`they would not be expected to induce neutralizing antibody
`potently trigger immune responses, whereas aggregates of
`because protein conformation was lost on denaturation.
`high abundance proteins may have a limited ability to induce
`However, antibodies directed to linear epitopes within the
`immune responses.
`E503
`
`Novartis Exhibit 2291.003
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`The AAPS Journal 2006; 8 (3) Article 59 (http://www.aapsj.org).
`binding site could effectively neutralize if they bound to a
`In this case, simultaneous ligation of BCR and complement
`linear determinant critical in ligand-receptor binding. Such
`receptor (CR2) on the B-cell surface profoundly enhances
`antibody production. 27 Aggregated IgG and pentameric IgM
`occurrences appear to be rare. Moreover, through epitope
`bound to antigen can directly fi x and activate complement.
`spreading, antibodies that bind to the protein but do not
`Whether such factors infl uenced immunogenicity of the
` neutralize its activity may facilitate the generation of neu-
`aggregated species in these studies in mice is not certain, as
`tralizing antibodies. However, this is not a rapid and highly
`the affi nity and binding of human Ig with rodent comple-
`effi cient process. Thus, for many therapeutic protein products,
`ment, complement receptors, and FcRs is not clear.
`development of neutralizing antibody is preceded by the
`existence of a high titered and prolonged binding antibody
` Although ordinary protein antigens cannot simultaneously
`response. Moreover, other immunologically provocative
`bind to BCR and FcR or CR, immune responses to proteins
`factors may also be required. In such cases preventing a
`may be potentiated by the formation of protein:antibody
`neutralizing antibody response may be possible by limiting
`complexes. For example, IgM potently enhances antibody
`the titer and duration of the primary response, which may
`responses to particulate antigen, while IgG enhances
`limit epitope spread.
`responses to soluble proteins. Reciprocally, IgM does not
` Although antibodies that bind but fail to neutralize product
`enhance responses to small soluble proteins and IgG sup-
`are of somewhat lesser concern than neutralizing antibody,
`presses responses to particulate antigens. The boost in anti-
`in some circumstances such antibodies have been shown to
`body response resulting from the binding of IgM to antigen
`(1) affect the pharmacokinetics of product, necessitating
`requires complement fi xation and activation, whereas
`more frequent dosing, 23 (2) mediate anaphylactoid reac-
`enhancement through IgG is mediated by engagement of
`tions, 24 and (3) potentially facilitate epitope spreading.
`activating FcR on non-B cell antigen presenting cells, such
`as macrophages. 27 Thus, IgM bound to large protein aggre-
`gates may well enhance the antibody response through
` EVIDENCE FOR THE ROLE OF PRODUCT
`direct uptake by complement receptors, or via the simulta-
`AGGREGATES IN INDUCTION OF IMMUNE
`neous ligation of BCR by the aggregated protein and CR
`RESPONSES TO THERAPEUTIC PROTEIN PRODUCT:
`through complement fragments bound to Fc. IgG binding to
`ANIMAL MODELS
`smaller soluble aggregates may theoretically also enhance
`antibody responses via activating FcR.
` Much of the information framing our understanding of the
`immunogenicity of protein aggregates and the relative
` The ability of protein aggregates to elicit antibody responses
`tolerogenicity of soluble protein monomers comes from
`to a self-protein was most objectively and thoroughly evalu-
`studies performed in the 1950s and 1960s in which human
`ated by Braun et al, who employed human interferon alpha
`immunoglobulin (Ig) products were injected into experi-
`(huIFN- a ) transgenic mice to evaluate the capacity of
`mental rodents. Such studies found that immune responses
`IFN- a product aggregates to break tolerance. 28 They clearly
`to human Ig preparations could be eliminated by techniques
`demonstrated that homogenous aggregates of huIFN- a and
`that removed high MW material from the preparation. 25 In
`composite aggregates of mouse serum albumin and huIFN- a
`Gamble ’ s studies in 1966, aggregates generated by heat
`potently induced antibody to huIFN- a , whereas IFN- a
`treatment and added back to deaggregated tolerogenic Ig
`monomer failed to do so (within the time frame of the study).
`preparations were shown to induce an immune response
`Unfortunately, for the preparation containing homogenous
`that, in a dose-dependent fashion, decreased the time to
`huIFN- a aggregates, the characteristics of the aggregate
`appearance of antibody and increased the antibody titer. 26
`species responsible for the effi ciency of response induc-
` The use of Ig as a model antigen for studying the effects of
`tion — the large amount of dimer in the product versus the
`aggregates on the immune response was, in retrospect, an
` “ very small amount ” of high MW aggregate — were not
`interesting one, as these proteins differ from most other pro-
`explored. Also critical in this study was the elucidation of
`teins in terms of complex immune system interactions.
`the role of dosing frequency, route of administration, and
`Thus, Ig can modify activation of the immune response to
`immunomodulation on generation of antibodies, albeit in
`itself through the binding and coligation of multiple
`response to nonaggregated IFN- a monomer. However,
`re ceptors on the B-cell surface. While the antigen binding
`these factors may also be critical in the ability of protein
`fragment (Fab) of an antibody molecule binds to BCR, the
`aggregates to trigger immune responses, particularly to
`complement fi xing fragment, Fc, can simultaneously bind
`antigens to which the immune system is tolerant. Thus,
`to Fc receptors on the B-cell surface. Coligation of BCR
`more frequent dosing, a subcutaneous (SC) or intraperito-
`with FcR (Fc g RII), the only (and inhibitory) FcR expressed
`neal (IP) route of administration rather than an intravenous
`on B cells, limits antibody responses. 27 In contrast, comple-
`(IV) route, and administration of concomitant immunomod-
`ment bound to the Fc moiety of an antibody molecule can
`ulators would be expected to facilitate the ability of small
`bind to complement receptors (CR2) on the B-cell surface.
`amounts of aggregates to enhance the immune response.
`E504
`
`Novartis Exhibit 2291.004
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`The AAPS Journal 2006; 8 (3) Article 59 (http://www.aapsj.org).
` EVIDENCE FOR THE ROLE OF PRODUCT
`Thus, a considerable amount of the product has a MW >100
`kD as well as having >20 ligands per aggregate, conditions
`AGGREGATES IN INDUCTION OF IMMUNE
`that were shown to optimize immunogenicity. Indeed, this
`RESPONSES TO THERAPEUTIC PROTEIN PRODUCT:
`product is highly immunogenic, inducing binding antibody
`CLINICAL STUDIES
`responses in 80% to 100% of patients. 33 Of interest, neutral-
` Clinically, evidence that aggregates in therapeutic protein
`izing antibody to IL-2 largely arises in a group of patients
`products had profound effects on immunogenicity was
`treated with IL-2 in an immunologically provocative fash-
`apparent from very early studies. In the 1950s and early
`ion (ie, SC administration or concomitant treatment with
`1960s intravenous immune globulin (IVIG) preparations
`IFN- a , a known immunomodulator, and extensive treat-
`contained substantial aggregated material that triggered
`ment over a prolonged time period). 33 Also intriguing is the
`severe hypersensitivity (anaphylactoid) responses due to
`time course in which the neutralizing response arises, as it
`fi xation and activation of the complement cascade and also
`is detected several months following detection of binding
`potentially to release of histamine. 29 , 30 Antibody responses
`antibody, implicating epitope spreading as the likely route
`to such aggregates appeared to be specifi c for novel deter-
`of development. Moreover, it is not clear whether T-cell
`minants present only on the aggregated species (ie, for
`help is required for either binding or neutralizing antibody.
`higher order structures) or for normally cryptic determi-
` In fact, these few examples illustrate a more general princi-
`nants, which are exposed on aggregates of denatured pro-
`ple, which is that for most therapeutic proteins, product
`tein. 26 Of importance, responses were far more prevalent
`neutralizing antibody is much less common than product
`and potent in the antibody-defi cient patient population than
`binding antibody. Neutralizing antibody arises more fre-
`in those without antibody defi ciency, suggesting that anti-
`quently with the following risk factors: generation of a high
`bodies generated to human Ig in nontolerant patients
`titer binding antibody response, sustained treatment, and
`increased the frequency and severity of these reactions. 29
`treatment by a more immunogenic route (ie, SC), concomi-
`These types of adverse clinical responses were also noted
`tant treatment with immune modulators, and genetic defi -
`in patients treated with early commercial preparations of
`ciency of the factor. There are some very notable exceptions
`human serum albumin (HSA) and pasteurized plasma
`to this rule, one being the case of immune responses to the
`solutions. 24
`TPO congener, pegylated megakaryocyte growth and devel-
` A further example of the potency of large amounts of aggre-
`opment factor (PEG-MGDF), in which as few as 2 doses
`gates on generation of immune responses is that of the clinical
`were suffi cient to induce a vigorous neutralizing antibody
`experience with human growth hormone (hGH). Originally
`response. 9 However, the very low abundance of the endoge-
`purifi ed from formalin-fi xed pituitary glands, hGH con-
`nous protein counterpart, TPO, in normal physiology 34
`tained substantial amounts of aggregates (50%-70%) that
` predicts that tolerance to this protein would be minimal, a
`induced immune responses. Although hGH prepared by an
`situation somewhat akin to individuals with genetic defi -
`improved method contained substantially less aggregated
`ciencies of particular proteins (such as hGH) who are highly
`material (10%), antibody responses were nevertheless still
`prone to formation of neutralizing antibody. The SC route
`substantial. Intriguingly, the level of aggregates appeared to
`of administration was an additional provocative factor in
`determine not whether there would be an antibody response,
`triggering the TPO neutralizing response, as was the immune
`but rather the nature of the antibody response. Patients
`competency of the patient population. Moreover, whether
`treated with the heavily aggregated product demonstrated
`immunogenicity was triggered by truncation of the full
`persistent antibodies to hGH, whereas those treated with the
`length TPO, thereby exposing the normally protected active
`lesser aggregated product developed a transient antibody
`site to the immune system, is not clear. Certainly the
`response. 31 Despite the frequency of appearance of binding
`pegylation of the truncated TPO would have been expected
`antibody in response to hGH therapy, the development of
`to confer protection against immune response generation.
`neutralizing antibody was uncommon. However, neutraliz-
` Why the immune system commonly makes robust responses
`ing antibody occurred much more commonly in patients
`to nonneutralizing determinants yet has to be mightily pro-
`with severe congenital isolated GH defi ciency (ie, in protein
`voked to make responses to neutralizing determinants, sug-
` “ knock out ” children in whom the lack of hGH protein con-
`gests that the immune system is more tolerant (or ignorant)
`ferred a lack of immune tolerance to the hormone).
`of the active site than of other epitopes, which raises several
` The therapeutic protein IL-2 can be viewed as a good test of
`interesting possibilities: the active sites of self-proteins are
`the general principles governing immune response genera-
`normally very well shielded from antigen specifi c T- and/or
`tion to large MW aggregates expressing multivalent anti-
`B-cells 35 ; the active sites of proteins lack primary sequence
`gens. Indeed, although the monomeric protein is ~15 kD,
`for complexation with self Class II MHC and thus fail to be
`the recombinant human product is formulated as an aggre-
`T-cell immunogens; or the active sites of proteins may be
`gate with an average size of 27 molecules per aggregate. 32
`exposed to the immune system in such a way as to induce a
`E505
`
`Novartis Exhibit 2291.005
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`The AAPS Journal 2006; 8 (3) Article 59 (http://www.aapsj.org).
`higher level of tolerance than other portions of the protein
`high-risk changes include changes in formulation and in
`(ie, through binding of the active site to a cell surface or
`source materials or cell lines.
`internal receptor with resultant altered processing and pre-
`sentation or via proximity of the active site to disulfi de
`bonds, which may be processed and presented differently
`than other epitopes). 36 Substantial work needs to be done to
`investigate such issues.
`
` CONCLUSION
` In summary, protein product aggregates are potent inducers
`of immune responses to therapeutic protein products. The
`extent to which these responses impact on therapy is deter-
`mined by multiple factors. Manufacturers of therapeutic
`protein products should ensure that their products contain
`minimal product aggregates, that they employ several or -
`thogonal methods for assessment of product aggregates,
`and that manufacturing changes or container closure changes
`prompt vigorous investigation of changes in levels of prod-
`uct aggregates.
`
` MEASUREMENT OF PRODUCT AGGREGATES/HIGH
`RISK CHANGES FOR AGGREGATE GENERATION
` Protein aggregates in therapeutic protein products are almost
`exclusively assessed by a size exclusion chromatography
`(SEC) method. While sensitive, this method has critical
`fl aws, including failure of some high MW species to pene-
`trate the gel. Thus, methods orthogonal to SEC need to be
`employed to assess product aggregates, particularly in the
` REFERENCES
`setting of comparability assessments but potentially also for
`routine lot release. 37-42 Moreover, in routine testing for
` 1 . Dintzis R , Okajima M , Middleton M , Greene G , Dintzis H . The
`immunogenicity of soluble haptenated polymers is determined by
`product aggregates, an inherent assumption is that very few
`molecular mass and hapten valence. J Immunol . 1989 ; 143 : 1239 - 1244 .
`samples are required to be tested for any given lot, as justi-
` 2 . Vos Q , Lees A , Wu ZQ , Snapper CM , Mond JJ . B-cell activation by
`fi ed by the demonstrated homogeneity of the bulk fi ll and
`T-cell-independent type 2 antigens as an integral part of the humoral
`the presumed homogeneity of the container closure.
`immune response to pathogenic microorganisms. Immunol Rev .
` 2000 ; 176 : 154 - 170 .
` The Food and Drug Administration (FDA) is particularly
`aware of container closures that are not homogeneou