Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 1 of 31
`
`IN THE UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF MASSACHUSETTS
`
`
`
`
`TEVA PHARMACEUTICALS
`INTERNATIONAL GMBH and
`TEVA PHARMACEUTICALS
`USA, INC.,
`
`
`
`Civil Action No.
`1:18-cv-12029-ADB
`
`Plaintiffs,
`
`
`v.
`
`ELI LILLY AND COMPANY,
`
`
`Defendant.
`
`
`
`
`
`
`
`
`EXPERT DECLARATION OF JEFFREY V. RAVETCH, M.D., PH.D.
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 2 of 31
`
`TABLE OF CONTENTS
`
`INTRODUCTION ................................................................................................................... 1 
`I. 
`QUALIFICATIONS ............................................................................................................ 1 
`II. 
`TECHNICAL BACKGROUND .......................................................................................... 3 
`III. 
`Calcitonin-Gene Receptor Peptide (“CGRP”) ................................................................. 3 
`A. 
`Background on the Function, Structure, and Development of Antibodies ...................... 4 
`B. 
`1.  Antibodies Generally ........................................................................................................ 4 
`2.  The Structure and Composition of Antibodies ................................................................. 5 
`3.  The Generation and “Humanization” of Therapeutic Antibodies .................................... 9 
`IV.  CLAIM CONSTRUCTION ANALYSIS .......................................................................... 11 
`“anti-CGRP antagonist antibody” / “anti-Calcitonin Gene-Related Peptide (CGRP)
`A. 
`antagonist antibody” .................................................................................................................. 12 
`B. 
`“humanized…antibody” ................................................................................................ 16 
`C. 
`“specific binding” / “preferentially binds” / “wherein the CDRs impart to the antibody
`specific binding to a CGRP consisting of amino acid residues 1 to 37 of SEQ ID No:15 or
`SEQ ID No:43,” ........................................................................................................................ 18 
`1.  “specific binding” / “preferentially binds” ..................................................................... 19 
`2.  “wherein the CDRs impart to the antibody specific binding to a CGRP consisting of
`amino acid residues 1 to 37 of SEQ ID No:15 or SEQ ID No:43,” ...................................... 21 
`D. 
`“human IgG heavy chain” .............................................................................................. 23 
`E. 
`“treating” ........................................................................................................................ 26 
`
`
`
`
`ii
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 3 of 31
`
`
`
`
`I.
`
`INTRODUCTION
`
`
`
`I, Jeffrey V. Ravetch, M.D., Ph.D., have been asked by Plaintiffs Teva
`
`Pharmaceuticals International GmbH and Teva Pharmaceuticals USA, Inc. (collectively, “Teva”)
`
`to review the common specification of nine patents at issue in this lawsuit (the “Patents-in-Suit”),
`
`provide a tutorial of the technical and scientific concepts implicated by these patents, and prepare
`
`an expert declaration addressing the meaning of certain terms used in these patents. In rendering
`
`my opinions, I have reviewed the patents and their prosecution histories and the proposed
`
`constructions set forth by both Teva and Defendant Eli Lilly and Company (“Lilly”) in the Joint
`
`Claim Construction Statement, as well as the materials set out in Exhibit 2.
`
`II.
`
`QUALIFICATIONS
`
`
`
`I am the Theresa and Eugene M. Lang Chair, Professor and Head of the Laboratory
`
`of Molecular Genetics and Immunology at The Rockefeller University. I joined the faculty of The
`
`Rockefeller University in 1996. Prior to that I was a member of the faculty of the Sloan-Kettering
`
`Institute of the Memorial Sloan-Kettering Center (1982-1996) where I was named a full Member
`
`in 1990.
`
`
`
`I received my Bachelor of Science degree in Molecular Biophysics and
`
`Biochemistry from Yale University in 1973. In 1978, I received my Ph.D. in Genetics from The
`
`Rockefeller University, and in 1979, I received my M.D. from Cornell Medical College. From
`
`1979 to 1982, I worked as a post-doctoral researcher at the Laboratory of Molecular Genetics at
`
`the National Institutes of Health, Bethesda, Maryland where I cloned and characterized the genes
`
`for human immunoglobulin heavy chains.
`
`
`
`I have extensive experience in the fields of molecular biology and immunology.
`
`The Laboratory of Molecular Genetics and Immunology, which I direct at The Rockefeller
`
`1
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 4 of 31
`
`
`
`University, focuses its research on cellular and molecular mechanisms governing the generation
`
`of antibody specificity and the translation of that specificity into cellular responses. I am the author
`
`or co-author of over 200 scientific publications in the fields of molecular genetics and
`
`immunology, disciplines which build on the technologies of genetic engineering and cellular
`
`expression. I therefore have extensive knowledge of the key scientific and technological concepts
`
`relevant to the Patents-in-Suit, which are directed to therapeutic antibodies and their use for the
`
`treatment of migraine and other vasomotor conditions. These key concepts include genetic
`
`engineering, cellular expression, and antibody engineering.
`
`
`
`I am a member of a number of professional organizations, including the American
`
`Association of Immunology, the American Association of Allergy and Immunology, the American
`
`Society for Biochemistry and Molecular Biology, and the American Society for Cell Biology,
`
`among others.
`
`
`
`I am an Advisory Editor for the Journal of Experimental Medicine and a
`
`Transmitting Editor for International Immunology. I also serve on the Scientific Advisory Boards
`
`for several research organizations, philanthropic foundations, and corporations.
`
`
`
`I have received many awards and honors for my research efforts in immunology,
`
`including the Pew Scholar Award, the Burroughs-Wellcome Award, the National Institutes of
`
`Health Merit Award, the Lee C. Howley Award, the AAI Huang Award for Meritorious Career,
`
`the Coley Award from the Cancer Research Institute, the Canada Gairdner International Award,
`
`the Sanofi-Pasteur Award, the Wolf Prize in Medicine, the Ross Prize, and the Robert Koch Prize.
`
`I was elected to the National Academy of Sciences of the United States in 2006, the Institute of
`
`Medicine in 2007, and the American Academy of Arts and Sciences in 2008. I am a frequent
`
`lecturer at numerous universities, medical centers, and international symposia.
`
`2
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 5 of 31
`
`
`
`
`
`My experiences and qualifications are further detailed in my curriculum vitae,
`
`which is attached hereto as Exhibit 1.
`
`III. TECHNICAL BACKGROUND
`
`
`
`The nine Patents-in-Suit are generally directed to antibodies that bind to the protein
`
`Calcitonin Gene-Related Peptide (“CGRP”) and the use of such antibodies to treat vasomotor
`
`diseases, such as migraine. Following is a tutorial on the background facts and scientific principles
`
`to assist the Court and explain the meaning of the disputed claim terms.
`
`A.
`
`Calcitonin-Gene Receptor Peptide (“CGRP”)
`
`
`
`Calcitonin-Gene Receptor Peptide or “CGRP” is a small signaling protein made up
`
`of 37 amino acids. See ’045 Patent at 1:25–27.1,2 First identified in 1983, CGRP is widely
`
`distributed in the body and plays a key role in many cardiovascular and neurological processes,
`
`including vasodilation (the expansion of blood vessels). Humans have two main forms
`
`(“isoforms”) of CGRP called α-CGRP and β-CGRP, which only differ in 3 of their 37 amino acids.
`
`Id. at 1:27–31. The primary biological function of CGRP is to seek out and bind to a separate
`
`protein called the CGRP receptor, which is located on the outer surface of certain cells. When
`
`CGRP binds to its receptor, the receptor activates a signaling pathway inside the cell that produces
`
`a downstream physiological response.
`
`
`1 I understand that all of the Patents-in-Suit are part of the same family and share a common
`specification. For consistency, I have cited to the specification of the ’045 Patent throughout my
`declaration. These citations are (apart from minor variations in line and column) equally
`applicable to the other eight Patents-in-Suit.
`
`2 I understand that this patent is attached to the concurrently-filed Declaration of Elaine Herrmann
`Blais in Support of Plaintiffs’ Opening Claim Construction Brief (“Blais Declaration”) as
`Exhibit A.
`
`3
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 6 of 31
`
`
`
` Migraine is a common chronic, neurobiological disorder associated with
`
`dysfunction of the cerebral nerves and blood vessels. Due to its role in vasodilation and pain
`
`signaling, it had been postulated by 2005 that CGRP was involved in the pathology of migraine.
`
`See ’045 Patent at 2:3–13. The Patents-in-Suit describe and claim therapeutic antibodies that target
`
`CGRP. Id. at 3:37–45. This work resulted in the discovery of fremanezumab—referred to as
`
`antibody “G1” in the Patents-in-Suit—which is FDA-approved for the treatment of migraine and
`
`sold by Teva under the brand-name Ajovy®.
`
`B.
`
`Background on the Function, Structure, and Development of Antibodies
`1.
`
`Antibodies Generally
`
`
`
`Antibodies, also known as immunoglobulins (Ig), are a particular type of protein
`
`that play a vital role in the immune system of animals. Antibodies recognize antigens in the body,
`
`including microbial pathogens (e.g., bacteria, viruses, and other foreign substances), tumor
`
`antigens, and generally substances identified as foreign to the immune system, and facilitate their
`
`removal from the body. Janeway et al., Immunobiology: the immune system in health and disease,
`
`5th Edition, 2001 (“Janeway 2001”), at 93; Pei-Show Juo, Ph.D., Concise Dictionary of
`
`Biomedicine and Molecular Biology, 2nd Edition, 2001 (“Concise Dictionary of Biomedicine
`
`2001”), at 94, 96.3 In general, antibodies work by recognizing and then binding to a region on an
`
`antigen called an “epitope.” See Fig. 1; Concise Dictionary of Biomedicine 2001 at 421. After
`
`binding to an antigen, such as a microbial pathogen, the antibody may be able to neutralize the
`
`antigen itself by preventing its ability to invade a host cell, for example, as well as signal for cells
`
`in the immune system to locate and attack the antigen. Janeway 2001 at 93. In this way, antibodies
`
`
`3 I understand that these references are attached to the Blais Declaration as Exhibits R and U,
`respectively.
`
`4
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 7 of 31
`
`
`
`protect an individual from the potentially harmful effects of antigens. Antibodies are often referred
`
`to in the context of the antigen they target, e.g., an “anti-HIV” antibody targets HIV.
`
`
`
`
`
`
`
`Figure 1: Antibodies recognize antigens.
`
`
`
`There are five main classes or isotypes of antibodies, referred to as IgM, IgD, IgA,
`
`IgG, and IgE, each of which may consist of several subclasses. See ’045 patent, 12:27–37;
`
`Janeway 2001 at 95; Concise Dictionary of Biomedicine 2001 at 579, 583. IgG antibodies are the
`
`most common class of antibody found in the bloodstream and play a significant role in the human
`
`immune response.
`
`2.
`
`The Structure and Composition of Antibodies
`
`
`
`Like all proteins, antibodies are made up of amino acids connected end-to-end in
`
`linear chains called “polypeptides.” Concise Dictionary of Biomedicine 2001 at 73, 881. There
`
`are twenty naturally-occurring amino acids in the body that act as protein building blocks, each of
`
`5
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 8 of 31
`
`
`
`which has its own unique chemical properties. Antibodies (and portions thereof) are therefore
`
`often described in terms of their underlying amino acid sequences.4
`
`
`
`As depicted in Figure 2 below, antibodies typically consist of four separate
`
`polypeptide chains that are linked together with chemical bonds to form a “Y” shape. Janeway
`
`2001 at 94. Each side of the “Y” is formed by a longer “heavy” chain and a shorter “light” chain.
`
`The heavy and light chains on each side of the “Y” are identical. Id. at 95; Concise Dictionary of
`
`Biomedicine 2001 at 534, 648.
`
`Figure 2: Antibodies are comprised of heavy chains and light chains.
`
`
`
`As depicted in Figure 3 below, each antibody chain (whether heavy or light) is
`
`made up of two different “regions” or “domains,” the “variable” regions and the “constant”
`
`
`
`
`4 Individual amino acids in a polypeptide chain are called “residues.” Concise Dictionary of
`Biomedicine 2001 at 936. Each of the 20 amino acid building blocks have been assigned unique
`1-letter and 3-letter abbreviations to make it easier for scientists to describe the sequences of long
`polypeptide chains. For example, the amino acid glycine is abbreviated “gly” or “G.” Individual
`residues in a chain are typically referred to by the combination of their amino acid abbreviation
`and their linear position in the chain, e.g., “G157,” which refers to the glycine that is the 157th
`residue in the chain.
`
`6
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 9 of 31
`
`
`
`regions. Janeway 2001 at 96; Concise Dictionary of Biomedicine 2001 at 301–02, 1119. The
`
`variable regions of an antibody’s heavy and light chains are the portions of the antibody that bind
`
`to the antigen. Janeway 2001 at 100. Accordingly, antibodies with different amino acid sequences
`
`in their variable regions will bind to different antigens. The constant regions of an antibody’s
`
`heavy and light chains provide structure to the antibody and play a key role in signaling the
`
`immune system to mount a response when the antibody binds to an antigen. Id. at 100–01. The
`
`amino acid sequences of the constant regions are, unlike those of the variable regions, generally
`
`much less diverse between different antibodies of the same class. Id. at 93. For this reason,
`
`scientists use the constant regions of an antibody’s heavy and light chains to determine the class
`
`or subclass to which it belongs. The amino acid sequence of a constant region is often unique to
`
`the animal species in which the antibody was generated, meaning that, e.g., the constant region of
`
`a human IgG heavy chain will be different than the constant region of a mouse IgG heavy chain.
`
`Figure 3: The chains of an antibody have variable and constant regions.
`
`
`
`
`7
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 10 of 31
`
`
`
`
`
` The variable regions of an antibody’s heavy and light chains are further subdivided
`
`into “complementarity determining regions” (or “CDRs”) and “framework regions.” Janeway
`
`2001 at 100–01. As shown in Fig. 4, below, each heavy and light chain variable region of an
`
`antibody contains three CDRs, each comprised of a unique amino acid sequence, interspersed
`
`between four framework regions. See Janeway 2001 at 100–02; ’045 Patent at 15:33–40. The
`
`CDRs play a key role in antigen binding as they form the primary binding interface between the
`
`antibody and the antigen. Janeway 2001 at 101–02. In general, the framework regions primarily
`
`act as structural support for the CDRs, although they can contribute to antigen binding in some
`
`circumstances. Id.
`
`Figure 4: Each variable region contains three CDRs interspersed with four framework regions5
`
`
`
`
`
`5 This figure is a simplified depiction of how the
`CDRs and framework regions are arranged in nature.
`Specifically, the CDRs and framework regions are
`folded together in a complex bundle from which the
`CDRs protrude as discrete “loops.” See Janeway
`2001 at 102, Fig. 3.7. This concept is depicted in the
`exemplary three-dimensional rendering at right.
`Here, the light chain and heavy chain framework
`regions are shown
`in
`light blue and green,
`respectively, while the CDRs are shown in dark blue,
`red, and pink.
`
`
`8
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 11 of 31
`
`
`
`3.
`
`The Generation and “Humanization” of Therapeutic Antibodies
`
`
`
`As described above, the antibody response evolved to identify and bind to
`
`potentially harmful foreign antigens and then to trigger an immune response to eliminate the threat.
`
`The role of therapeutic antibodies (i.e., antibody drugs), however, is somewhat different.
`
`Antibodies used in human therapeutics are most commonly designed to target a naturally-
`
`occurring antigen within the human body itself. Among the most common targets for antibody
`
`drugs are human proteins that play a key role in cellular processes that can contribute to disease.
`
`Antibodies that disrupt or shut-down the natural function of their target antigen are referred to as
`
`“inhibitors” or “antagonists.” Concise Dictionary of Biomedicine 2001 at 92, 591.
`
`
`
`Broadly speaking, to generate an IgG antibody in a laboratory or clinical setting,
`
`the antigen of interest is injected into a subject, which will recognize the antigen as foreign and
`
`subsequently produce or “raise” antibodies against it. Vaccines against microbial pathogens, in
`
`which a subject is injected with a viral or bacterial antigen, result in the generations of antibodies
`
`(and hopefully immunity) against that antigen. This same basic approach can also be used to
`
`generate and subsequently isolate antibodies for use as medicines.
`
`
`
`For practical and safety reasons, one approach to produce antibodies for therapeutic
`
`use in humans is to immunize non-human mammals such as mice. For example, the Patents-in-
`
`Suit describe injecting mice with human isoforms of CGRP for purposes of generating the claimed
`
`“anti-CGRP” antibodies. ’045 Patent at 49:10–53:35. This approach is particularly useful when
`
`the antigen of interest is of human origin because, unlike a healthy human immune system, a mouse
`
`immune system will treat the human antigen as a foreign substance and generate antibodies to
`
`attack it.
`
`
`
`Dosing non-human antibodies directly to a human patient can be problematic
`
`because they may provoke an immune response in the patient (called a “Human Anti-Mouse
`9
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 12 of 31
`
`
`
`Antibody” or “HAMA” response) which significantly reduces the intended therapeutic effect and
`
`can result in adverse reactions. Antibody Engineering: Methods and Protocols, Methods in
`
`Molecular Biology™, Edited by Benny K.C. Lo, 2004 (“Lo 2004”) at 135.6 In other words, the
`
`patient’s immune system may recognize and subsequently attack the therapeutic antibodies as
`
`foreign substances due to their non-human amino acid sequences. Id.; cf. ’045 Patent at 29:15–20.
`
`
`
`One solution to this problem is to “humanize” the non-human antibody.
`
`Humanized antibodies and the basic techniques for making them were known as of 2005. The first
`
`humanized antibodies were prepared in the late 1980s by a British research group headed by Sir
`
`Greg Winter. Lo 2004 at 137. Since this work was first published in an early 1988 Nature paper,
`
`the methods and technology for humanizing antibodies have advanced substantially. Riechmann
`
`et al., Reshaping human antibodies for therapy, Nature, 332, 323–27, 1988.7 By 2005, at least 7
`
`humanized antibodies were already being sold as FDA-approved drugs.8
`
`
`
`As the Patents-in-Suit explain, antibodies generated by non-human species (e.g.,
`
`mice) are “humanized” by replacing the CDRs of a human antibody with those of an antibody
`
`generated against a particular antigen in a non-human animal such as a mouse. See, e.g., ’045
`
`Patent at 12:66–13:5. The resulting “humanized” hybrid antibody has the same activity against
`
`the chosen antigen as its non-human counterpart but has fewer therapeutic limitations because the
`
`vast majority of its amino acid sequence is human and thus protects it from attack by the patient’s
`
`immune system. The Patents-in-Suit describe this process and explain, “antibodies that are
`
`
`6 I understand that this reference is attached to the Blais Declaration as Exhibit S.
`
`7 I understand that this reference is attached to the Blais Declaration as Exhibit T.
`
`8 See e.g., Trastuzumab, palivizumab, omalizumab, natalizumab, dacilizumab, bevacizumab, and
`alemtuzumab.
`
`10
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 13 of 31
`
`
`
`compatible with the human immune system, such as humanized antibodies or fully human
`
`antibodies, may be used to prolong half-life of the antibody and to prevent the antibody being
`
`attacked by the host’s immune system.” ’045 Patent at 22:30–35; 29:15–20. As depicted below
`
`in Figure 5, a humanized antibody classically consists of: (1) CDRs from a donor non-human
`
`antibody; (2) human (variable) framework regions; and (3) human constant regions.9 See, e.g., Lo
`
`2004 at 135–37; ’045 Patent at 12:61–13:25.
`
`
`
`Figure 5: Humanized antibodies contain non-human CDRs
`
`IV. CLAIM CONSTRUCTION ANALYSIS
`
`
`
`I understand that claim construction and patent validity are evaluated from the
`
`perspective of a person of ordinary skill in the art (“POSA”) to which the patent pertains as of the
`
`priority date of the patents. I understand that the POSA is a hypothetical person who is presumed
`
`to have a comprehensive knowledge of the available prior art and a level of skill commensurate
`
`with a typical practitioner in the relevant field as of the patent’s priority date. For purposes of this
`
`
`9 In some instances, antibody engineers may also choose to transplant non-human sequences into
`one or more of the framework regions in order to optimize binding affinity. Lo 2004 at 135–37.
`
`11
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 14 of 31
`
`
`
`declaration, I have been asked to assume the priority date for the Patents-in-Suit is November 14,
`
`2005.10 I am informed by counsel that the Patent Trial and Appeal Board (PTAB) of the U.S.
`
`Patent and Trademark Office adopted the following definition of a POSA in a prior proceeding
`
`involving the Patents in Suit:
`
`(1) a Ph.D. in a relevant field, such as immunology, biochemistry, or pharmacology,
`with several years of post-doctoral experience
`in antibody engineering,
`pharmacokinetics, and pharmacodynamics, or (2) an M.D. with a residency or
`specialty in neurology, and several years of experience studying CGRP or treating
`patients with migraine headaches.
`
`I agree with and have accepted the PTAB’s definition of a POSA for the Patents-in-Suit for
`
`purposes of considering my opinions. A POSA would also be able to draw upon the knowledge
`
`and experience of a multi-disciplinary antibody development team comprising individuals with
`
`expertise outside the POSA’s primary training. These individuals could include immunologists,
`
`biochemists, antibody engineers, pharmacologists, pharmacists, and medical doctors.
`
`A.
`
`“anti-CGRP antagonist antibody” / “anti-Calcitonin Gene-Related Peptide
`(CGRP) antagonist antibody”
`
`
`
`I understand that the Parties dispute the meaning of the terms “anti-CGRP
`
`antagonist antibody” and “anti-calcitonin gene-related peptide (CGRP) antagonist antibody” and
`
`that Lilly contends the terms are indefinite. I have been asked by counsel to offer my opinion
`
`concerning the meaning of these terms to a POSA. I have also been asked by counsel to evaluate
`
`whether those terms, read in light of the specification and the prosecution history, fail to inform a
`
`POSA, with reasonable certainty of the scope of the invention.
`
`
`
`As I explain in more detail below, the specification expressly defines the terms
`
`“anti-CGRP antagonist antibody” and “anti-calcitonin gene-related peptide (CGRP) antagonist
`
`
`10 My opinions would not change if the Court were to find that the priority date of the Patents-in-
`Suit is November 2, 2006 (the filing date of the earliest non-provisional application).
`
`12
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 15 of 31
`
`
`
`antibody” in a manner that would be easily understood by a POSA. The specification defines
`
`“anti-CGRP antagonist antibody” as follows:
`
`As used herein, an “anti-CGRP antagonist antibody” (interchangeably termed
`“anti-CGRP antibody”) refers to an antibody that is able to bind to CGRP and
`inhibit CGRP biological activity and/or downstream pathway(s) mediated by
`CGRP signaling. An anti-CGRP antagonist antibody encompasses antibodies that
`block, antagonize, suppress or reduce (including significantly) CGRP biological
`activity, including downstream pathways mediated by CGRP signaling, such as
`receptor binding and/or elicitation of a cellular response to CGRP.
`
`
`’045 Patent at 13:62–14:4.
`
`
`
`This express definition is consistent with how a POSA would understand the term
`
`“anti-CGRP antagonist antibody” in the context of the asserted claims. As explained above, the
`
`standard convention for referring to antibodies is by identifying the antigen the antibody targets—
`
`in this case, CGRP. The term describes the antibody as an “antagonist” of CGRP, which means it
`
`binds to CGRP and inhibits the CGRP from fulfilling its natural signaling function. As described
`
`above, a POSA would have understood that inhibition of CGRP signaling by, e.g., preventing
`
`CGRP from binding to its receptor, led to the suppression of a downstream biological response.
`
`
`
`In addition to the explicit definition, the specification provides detailed descriptions
`
`of multiple tests that were used to measure various antibodies’ antagonism of CGRP, which would
`
`be easily understood to a POSA. For example, the patents describe the use of Surface Plasmon
`
`Resonance (“SPR”), which was a common method used to measure the “binding affinity” or
`
`overall strength of binding between the anti-CGRP antibodies and CGRP. See ’045 Patent at 50:6–
`
`37, 58:53–60:19. In this SPR assay, the target antigen (CGRP) is attached to a surface and a
`
`solution of an antibody is passed over the surface. As the antibody binds to the antigen, an increase
`
`in SPR signal is observed. An antibody-free solution is then injected over the surface, leading to
`
`the detachment of any bound antibodies and a corresponding decrease in SPR signal. The
`
`13
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 16 of 31
`
`
`
`experimenter can compare the respective association and dissociation data to determine a “binding
`
`constant” or “KD.” A lower KD value indicates a stronger binding affinity.
`
`
`
`The specification provides a POSA with specific guidance of potential KD values
`
`that an anti-CGRP antibody should exhibit:
`
`The binding affinity (KD) of an anti-CGRP antagonist antibody to CGRP (such as
`human α-CGRP as measured by surface plasmon resonance at an appropriate
`temperature, such as 25 or 37o C) can be about 0.02 to about 200 nM. In some
`embodiments, the binding affinity is any of about 200 nM, about 100 nM, about
`50 nM, about 10 nM, about 1 nM, about 500 pM, about 100 pM, about 60 pM,
`about 50 pM, about 20 pM, about 15 pM, about 10 pM, about 5 pM, or about 2
`pM. In some embodiments, the binding affinity is less than any of about 250 nM,
`about 200 nM, about 100 nM, about 50 nM, about 10 nM, about 1 nM, about 500
`pM, about 100 pM, or about 50 pM.
`
`’045 patent, 5:35–46.
`
`
`
`The patents also describe five additional in vitro (conducted outside of an animal)
`
`and in vivo (conducted in an animal) functional tests that measure the biological effects of
`
`antibody-CGRP binding. For example, the patents describe the use of an in vitro radioligand
`
`binding assay, id. at 54:65–55:25, one of the most elementary assays to study the interaction
`
`between a signaling protein and its receptor. This particular assay involves incubating cell
`
`membranes containing the CGRP receptor with a solution of radioactively-labeled CGRP and the
`
`antibody to be evaluated. After incubation, the cell membranes are extracted and measured for
`
`radioactivity. If the antibody has negligible affinity for CGRP, the extracted membranes will have
`
`high levels of radioactivity because there was nothing to stop the radioactive CGRP from binding
`
`to the receptors on the membranes. However, if the antibody has substantial affinity for CGRP,
`
`some of the radioactive CGRP will bind to the antibody rather than the receptors on the membrane,
`
`meaning that the amount of radioactivity associated with the membrane will diminish. The
`
`specification further instructs the POSA to perform this assay several times with varying amounts
`
`of antibody in order to determine an “IC50” value which corresponds to the amount of antibody
`14
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 17 of 31
`
`
`
`needed to inhibit the binding between the receptor and ligand by 50%. A lower IC50 indicates a
`
`more effective antibody.
`
`
`
` Further, the specification also provides descriptions of (1) an in vitro functional
`
`assay measuring the inhibition of cAMP activation in cells (a downstream biological effect of
`
`CGRP antagonism), id. at 53:36–54:64; (2) an in vivo rat model measuring skin vasodilation (e.g.,
`
`the “saphenous nerve” model), id. at 55:26–57:12, 67:52–68:57; (3) an in vivo “closed cranial
`
`window” model in rats, id. at 68:58–69:67; and (4) an in vivo morphine withdrawal model in rats,
`
`id. at 70:1–40. In each instance, the specification provides a detailed description of the
`
`experimental conditions and procedure used by the inventors as well as any resulting data that was
`
`generated. A POSA would have been familiar with these tests and could interpret them to
`
`determine with reasonable certainty whether a given antibody was an “anti-CGRP antagonist
`
`antibody,” as described in the asserted claims.
`
`
`
`I understand that Lilly claims that the specification fails to define “CGRP biological
`
`activity” or provide specific criteria by which to assess CGRP biological activity or its inhibition.
`
`See Defendant’s Preliminary Disclosures Pursuant to Rule 16.6(d)(4) (July 10, 2020) (“Lilly
`
`Disclosures”) at 27.11 I disagree. As described above, the specification provides a detailed
`
`description of “CGRP biological activity” that includes several examples and would have allowed
`
`a POSA to understand the meaning of the term “anti-CGRP antagonist antibody.” Moreover, the
`
`terms “inhibition” and “antagonism” are foundational concepts in drug development, and POSA
`
`would know what it means to inhibit the natural function of CGRP.
`
`
`
`I understand that Lilly also claims that the specification “provides no statement as
`
`to which one or more assays to use and how to measure inhibition of CGRP biological activity as
`
`
`11 I understand that this document is attached to the Blais Declaration as Exhibit H.
`
`15
`
`
`

`

`Case 1:18-cv-12029-ADB Document 70 Filed 09/11/20 Page 18 of 31
`
`
`
`recited in the claims” and states that “different assays can provide contradictory readouts of
`
`whether an antibody is an anti-CGRP antagonist antibody as defined by Teva’s specification.”
`
`Lilly Contentions at 27. Neither of these statements is correct in my opinion. First, as described
`
`above, the specification provides the POSA with at least six different methods to assess an
`
`antibody’s CGRP antagonism and examples of positive results in each. A POSA would be able to
`
`successfully perform any of these assays and interpret the resulting data. Second, Lilly’s assertion
`
`regarding “contradictory readouts” incorrectly assumes that an antibody must return a positive
`
`result in every assay in which it is tested to be considered an “anti-CGRP antagonist.” Lilly cites
`
`as an example the scenario where an antibody shows affinity for CGRP in vitro but subsequently
`
`lacks activity when tested in an in vivo animal model. However, a POSA would understand this
`
`type of result to simply reflect the complexity of in vivo assays, where considerations such as the
`
`biophysical properties of an antibody (e.g., its stability, half-life, or tissue penetration) may impact
`
`in vivo response regardless of whether the antibody exhibits CGRP antagonism. A POSA would
`
`know from both the specification and his or her own experience that the baseline measure