Filed: August 8, 2018
`
`Filed on behalf of: Eli Lilly and Company
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`______________________
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________________
`
`ELI LILLY AND COMPANY
`Petitioner
`v.
`TEVA PHARMACEUTICALS INTERNATIONAL GMBH
`Patent Owner
`______________________
`
`Case No. IPR2018-01427
`Patent No. 8,597,649
`______________________
`
`PETITION FOR INTER PARTES REVIEW
`
`
`
`
`
`
`

`

`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`
`
`
`I.
`II.
`
`TABLE OF CONTENTS
`Introduction ...................................................................................................... 1
`Requirements for Inter Partes Review Under 37 C.F.R. § 42.104 ................. 3
`A. Grounds for Standing ............................................................................ 3
`B.
`Identification of Challenge .................................................................... 3
`III. The ’649 Patent and Its Provisional Application ............................................. 4
`A.
`The Challenged Claims ......................................................................... 5
`B.
`Patent Owner Admissions in the Specification ..................................... 6
`1.
`Anti-CGRP Antagonist Antibodies and Methods of
`Making Them, Including Humanization Techniques,
`Were Known ............................................................................... 6
`2. Methods to Obtain and Measure Particular Binding
`Affinities by SPR Were Known .................................................. 8
`Limitations of Dependent Claims Were Also Known ................ 8
`a)
`Anti-CGRP Antagonist Antibodies That Bound to
`the C-Terminus of CGRP Were Known ........................... 9
`Antibody Types, Constant Regions, and Fc
`Regions Were Known ....................................................... 9
`Antibody Formulations and Methods of
`Administering Them Were Known ................................ 10
`Prosecution of the ’649 Patent ............................................................ 10
`C.
`IV. Background and the Asserted Prior Art ......................................................... 12
`A.
`CGRP Structure, Isoforms, and Function ........................................... 12
`B.
`Anti-CGRP Antagonist Antibodies Were Well Known in the
`Art and Had Been Disclosed for Therapeutic Use in Humans ........... 14
`IgG Antibodies .................................................................................... 15
`
`3.
`
`b)
`
`c)
`
`C.
`
`i
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`2.
`
`3.
`
`4.
`
`D. Humanization of Antibodies ............................................................... 18
`E.
`Binding Affinity .................................................................................. 19
`F.
`The Asserted Prior Art ........................................................................ 19
`1.
`Tan 1995 ................................................................................... 19
`2. Wimalawansa ............................................................................ 21
`3.
`Queen ........................................................................................ 22
`Person of Ordinary Skill in the Art ................................................................ 24
`V.
`VI. Claim Construction ........................................................................................ 24
`VII. Claim 1 Is Obvious over Tan 1995, Wimalawansa, and Queen ................... 26
`A. A POSA Would Have Been Motivated to Generate a
`Humanized Anti-CGRP Antagonist Antibody of Claim 1 ................. 27
`1.
`The Prior Art Recommended the Use of Anti-CGRP
`Antagonist Antibodies, Including Humanized Antibodies,
`for Therapeutic Use in Humans ................................................ 27
`The Confirmed In Vivo Effectiveness of Prior Art Anti-
`CGRP Antagonist Antibodies Provided Additional
`Motivation to Prepare a Humanized Antibody ......................... 30
`A POSA Would Have Targeted a Binding Affinity of 50
`nM or Less ................................................................................ 32
`The Prior Art Provided Additional Motivation to Prepare
`a Humanized Antibody ............................................................. 34
`A POSA Would Have Had a Reasonable Expectation of
`Successfully Making a Humanized Anti-CGRP Antagonist
`Antibody of Claim 1 ............................................................................ 36
`1.
`A POSA Would Have Had a Reasonable Expectation of
`Successfully Producing an Antibody Against Human
`αCGRP ...................................................................................... 36
`
`B.
`
`ii
`
`

`

`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`
`
`
`2.
`
`3.
`
`C.
`
`A POSA Would Have Had a Reasonable Expectation of
`Success of Producing an Anti-CGRP Antagonist
`Antibody with a Binding Affinity of 50 nM or Less ................ 38
`A POSA Would Have Had a Reasonable Expectation of
`Successfully Producing a Humanized Anti-CGRP
`Antagonist Antibody of Claim 1 ............................................... 41
`The Prior Art Did Not Teach Away from Humanizing Anti-
`CGRP Antagonist Antibodies, as Teva Incorrectly Argued
`During Prosecution .............................................................................. 44
`The Claimed Antibodies Would Have Been Obvious ........................ 49
`D.
`VIII. The Challenged Dependent Claims Would Have Been Obvious .................. 51
`A.
`Claim 3 ................................................................................................ 51
`B.
`Claim 2 ................................................................................................ 53
`C.
`Claim 4 ................................................................................................ 54
`D.
`Claim 5 ................................................................................................ 56
`E.
`Claim 6 ................................................................................................ 57
`F.
`Claim 7 ................................................................................................ 58
`G.
`Claim 8 ................................................................................................ 58
`H.
`Claim 9 ................................................................................................ 59
`IX. Secondary Considerations Do Not Support Nonobviousness ....................... 60
`A.
`Teva Cannot Establish Nexus to the Full Scope of the
`Challenged Claims .............................................................................. 60
`Lilly’s and Other’s Own Near-Simultaneous Development
`Precludes a Holding of Nonobviousness ............................................. 61
`The Evidence Submitted in this Petition Was Not Previously
`Considered by the Office ............................................................................... 63
`
`B.
`
`X.
`
`iii
`
`

`

`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`
`
`
`
`XI. Mandatory Notices Under 37 C.F.R. § 42.8 .................................................. 63
`A.
`Real Parties-in-Interest ........................................................................ 63
`B.
`Related Matters .................................................................................... 64
`C.
`Lead and Backup Counsel .................................................................... 64
`D.
`Service Information ............................................................................. 66
`XII. Payment of Fees ............................................................................................. 66
`XIII. Conclusion ..................................................................................................... 66
`CERTIFICATION OF COMPLIANCE .................................................................... 1
`CERTIFICATE OF SERVICE .................................................................................. 2
`
`
`
`
`
`
`
`
`
`
`iv
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`TABLE OF AUTHORITIES
`
` Page(s)
`
`Cases
`Abbott GmbH & Co., KG v. Centocor Ortho Biotech, Inc.,
`971 F. Supp. 2d 171 (D. Mass. 2013) ........................................................... 36, 42
`AbbVie Deutschland GmbH v. Janssen Biotech, Inc.,
`759 F.3d 1285 (Fed. Cir. 2014) .......................................................................... 60
`Akorn, Inc. v. Senju Pharm. Co.,
`IPR2015-01205, Paper 39 (PTAB Nov. 22, 2016) ............................................. 43
`Allergan, Inc. v. Apotex Inc.,
`754 F.3d 952 (Fed. Cir. 2014) ............................................................................ 60
`Ecolochem, Inc. v. S. Cal. Edison Co.,
`227 F.3d 1361 (Fed. Cir. 2000) .......................................................................... 61
`Geo. M. Martin Co. v. Alliance Mach. Sys. Int’l LLC,
`618 F.3d 1294 (Fed. Cir. 2010) .................................................................... 61, 63
`In re Menig,
`486 F.2d 1066 (C.C.P.A. 1973) .......................................................................... 41
`In re Mouttet,
`686 F.3d 1322 (Fed. Cir. 2012) .................................................................... 45, 49
`KSR Int’l Co. v. Teleflex Inc.,
`550 U.S. 398 (2007) ...................................................................................... 36, 48
`Paice LLC v. Ford Motor Co.,
`881 F.3d 894 (Fed. Cir. 2018) ............................................................................ 42
`PharmaStem Therapeutics, Inc. v. ViaCell, Inc.,
`491 F.3d 1342 (Fed. Cir. 2007) ...................................................................... 6, 38
`Senju Pharm. Co. v. Lupin Ltd.,
`780 F.3d 1337 (Fed. Cir. 2015) .......................................................................... 43
`
`v
`
`

`

`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`
`
`
`
`Statutes
`35 U.S.C. § 102(b) ..................................................................................................... 4
`35 U.S.C. § 103(a) ..................................................................................................... 3
`35 U.S.C. § 311 .......................................................................................................... 3
`35 U.S.C. § 325(d) ................................................................................................... 63
`Regulations
`37 C.F.R. § 42.8 ....................................................................................................... 63
`37 C.F.R. § 42.15(a) ................................................................................................. 66
`37 C.F.R. § 42.100(b) .............................................................................................. 24
`37 C.F.R. § 42.103(a) ............................................................................................... 66
`37 C.F.R. § 42.104 ..................................................................................................... 3
`Other Authorities
`Notice of Proposed Rulemaking (May 3, 2018) ...................................................... 25
`
`
`
`
`
`
`vi
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`GLOSSARY
`
`America Invents Act
`AIA
`Broadest reasonable interpretation
`BRI
`Complementarity-determining region
`CDR
`Calcitonin gene-related peptide
`CGRP
`European Patent Office
`EPO
`Fragment antigen binding
`Fab
`U.S. Food and Drug Administration
`FDA
`Framework region
`FR
`Inter partes review
`IPR
`Emphasis added unless otherwise indicated
`Italicized text
`Eli Lilly and Company
`Lilly or Petitioner
`Monoclonal antibody
`MAb
`Person of ordinary skill in the art
`POSA
`provisional application U.S. Provisional App. No. 60/736,623
`RIA
`Radioimmunoassay
`SPR
`Surface plasmon resonance
`Teva or Patent Owner Teva Pharmaceuticals International GmbH
`USPTO or Office
`U.S. Patent and Trademark Office
`’649 patent
`U.S. Patent No. 8,597,649
`’438 patent
`U.S. Patent No. 6,344,438
`
`vii
`
`

`

`
`
`I.
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`Introduction
`Teva’s ’649 patent broadly claims any isolated human or humanized anti-
`
`CGRP antagonist antibody that binds to human αCGRP with a particular binding
`
`affinity as measured by a well-known prior art technique. The claims recite an
`
`antibody that binds to human αCGRP with an unexceptional binding affinity, traits
`
`possessed by many prior art antibodies, as measured by the well-known technique
`
`of surface plasmon resonance (“SPR”).
`
`Teva’s claims are obvious over the prior art. By Teva’s earliest possible filing
`
`date of November 14, 2005, the prior art had already singled out anti-CGRP
`
`antagonist antibodies, including humanized antibodies, for use as therapeutic agents
`
`for human diseases and conditions. Multiple prior art references explicitly identified
`
`anti-CGRP antagonist antibodies to treat or prevent a variety of clinical conditions.
`
`Several research groups had also prepared and tested the antagonistic activity of anti-
`
`CGRP antibodies in vitro and in vivo, including antibodies that specifically bound
`
`to human αCGRP with binding affinities within the range Teva has attempted to
`
`claim. Thus, the prior art provided express motivation to prepare a humanized anti-
`
`CGRP antagonist antibody that binds human αCGRP.
`
`Moreover, as Teva’s ’649 patent admits, humanization was a well-established
`
`and routine procedure by 2005. Indeed, researchers had long understood that
`
`humanized antibodies advantageously avoided immunogenic reactions caused by
`
`1
`
`

`

`
`
`administering murine or chimeric antibodies to humans. By 2005, half of the FDA-
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`approved antibodies were humanized antibodies, and most antibodies in phase 2 and
`
`3 clinical trials were humanized.
`
`Binding affinities falling below the recited binding affinity of the claimed
`
`antibodies had also already been disclosed in the prior art as a desirable property for
`
`both anti-CGRP antagonist antibodies specifically, and for therapeutic antibodies as
`
`a class generally. By November 2005, nearly all FDA-approved antibodies were
`
`reported to have such binding affinities. A POSA would therefore have been
`
`motivated to make a humanized anti-CGRP antagonist antibody as claimed, and
`
`would have reasonably expected to succeed in doing so.
`
`As explained below and in the Expert Declarations of Dr. Andrew Charles, a
`
`neurologist who specializes in the treatment of migraine and long-time CGRP
`
`researcher, and Dr. Alain Vasserot, an antibody engineer with expertise in antibody
`
`humanization, there is a reasonable likelihood that Petitioner will prevail on all
`
`challenged claims, and that the prior art renders the claims obvious by at least a
`
`preponderance of the evidence. (Ex. 1002 ¶¶1-17; Ex. 1003 ¶¶1-15.) Notably, in
`
`parallel proceedings at the EPO, Teva did not appeal the EPO’s revocation of broad
`
`antibody claims similar to those challenged here. Similarly, in parallel UK
`
`proceedings, Teva abandoned broad antibody and composition claims challenged by
`
`2
`
`

`

`
`
`Lilly to pursue only claims to the prevention and treatment of headaches.
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`Accordingly, Lilly requests inter partes review of claims 1-9 of the ’649 patent.
`
`II. Requirements for Inter Partes Review Under 37 C.F.R. § 42.104
`A. Grounds for Standing
`Petitioner certifies that the ’649 patent is available for IPR based on Teva’s
`
`assertions to the Office that it is entitled to claim priority to a pre-AIA effective filing
`
`date of November 14, 2005, and that Petitioner is not barred or estopped from
`
`requesting review on the ground identified. (Ex. 1131, 3; Ex. 1132, 1-2 (listing
`
`priority chain and declining to designate as a transition application); Ex. 1001, 1:8-
`
`17, title page, item (60).)1
`
`Identification of Challenge
`B.
`Lilly respectfully requests review under 35 U.S.C. § 311 of claims 1-9 of
`
`the ’649 patent. In the sole ground presented in this Petition, Lilly requests that the
`
`Board find these claims unpatentable as obvious under 35 U.S.C. § 103(a) in view
`
`of the following combination of references:
`
`
`1 Citations refer to the original page numbering of each exhibit except for
`
`references that do not have any pagination in their original form. Citations to such
`
`references refer to the stamped-on page numbers.
`
`3
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`Reference 1: K.K.C. Tan et al., Clin. Sci. 89:565-573 (1995) (“Tan 1995”)
`
`(Ex. 1022), published on December 1, 1995. Tan 1995 is prior art to the ’649 patent
`
`under 35 U.S.C. § 102(b).
`
`Reference 2: S.J. Wimalawansa, Endocrine Reviews 17(5):533-585 (1996)
`
`(“Wimalawansa”) (Ex. 1096), published in 1996. Wimalawansa is prior art to the
`
`’649 patent under 35 U.S.C. § 102(b).
`
`Reference 3: U.S. Patent No. 6,180,370 (“Queen”) (Ex. 1023) issued on
`
`January 30, 2001. Queen is prior art to the ’649 patent under 35 U.S.C. § 102(b).
`
`III. The ’649 Patent and Its Provisional Application
`The ’649 patent is entitled “Antagonist Antibodies Directed Against
`
`Calcitonin Gene-Related Peptide and Methods [of] Using Same.” (Ex. 1001, title
`
`page, item (54).) It states that the alleged invention “relates to the use of anti-CGRP
`
`antagonist antibodies for the prevention, amelioration, or treatment of vasomotor
`
`symptoms, such as CGRP related headaches (e.g., migraine) and hot flushes.” (Id.,
`
`1:29-32; Ex. 1002 ¶81.) The ’649 patent discloses a sole humanized antagonist
`
`antibody (G1) and its purported derivatives. (E.g., id., Abstract, Example 4.) The
`
`’649 patent does not include any clinical or other human data.
`
`The ’649 patent belongs to a family of fifteen patents and applications, all of
`
`which purport to claim priority to U.S. Provisional Application No. 60/736,623 (“the
`
`provisional application”), filed on November 14, 2005. (Ex. 1002 ¶82.) Lilly has
`
`4
`
`

`

`
`
`also filed IPR petitions for related U.S. Patent Nos. 9,266,951; 9,340,614; 9,346,881;
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`9,890,210; and 9,890,211. These patents are similarly directed to anti-CGRP
`
`antagonist antibodies.
`
`The provisional application, like the ’649 patent, identifies only one
`
`humanized anti-CGRP antagonist antibody, G1, as well as its variants with minor
`
`sequence differences. (E.g., Ex. 1019, Example 4; Ex. 1001, Abstract, Example 4.)
`
`The only in vivo data disclosed in the provisional application was generated using a
`
`well-known assay—the rat saphenous nerve assay—used in the prior art for the
`
`specific purpose of evaluating anti-CGRP antagonist antibodies. (Compare
`
`Ex. 1019, ¶[0244] (citing Ex. 1052), with Ex. 1022, 572 (citing Ex. 1052 as
`
`reference 9); Ex. 1002 ¶¶88-89.) When filing its PCT application a year later, Teva
`
`only added additional animal study results, and not any clinical data, to the
`
`disclosure of its provisional application. (Ex. 1020, 66-68 (adding Examples 6-8).)
`
`A. The Challenged Claims
`Claim 1 of the ’649 patent, the only independent claim, recites:
`
`An isolated human or humanized anti-CGRP antagonist
`antibody with a binding affinity (KD) to human α-CGRP
`of 50 nM or less as measured by surface plasmon
`resonance at 37° C.
`
`(Ex. 1001, claim 1; Ex. 1002 ¶¶83-84; Ex. 1003 ¶70.)
`
`5
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`The challenged dependent claims 2-9 additionally require:
`
`• binding to a particular fragment or epitope of CGRP (claims 2 and 3);
`
`• an IgG, an IgM, an IgE, and IgA, or an IgD antibody (claim 4);
`
`• an Fc region, including one with an impaired effector function (claims
`
`5 and 6);
`
`• a humanized antibody (claim 7);
`
`• a pharmaceutical composition with a pharmaceutically acceptable
`
`excipient (claim 8); and
`
`• a formulation for systemic administration (claim 9).
`
`(Ex. 1001, claim 2-9; Ex. 1002 ¶¶85-86; Ex. 1003 ¶71.)
`
`Patent Owner Admissions in the Specification
`B.
`The ’649 patent itself expressly discloses that multiple claim limitations were
`
`known in the art. “Admissions in the specification regarding the prior art are binding
`
`on the patentee for the purposes of a later inquiry into obviousness.” PharmaStem
`
`Therapeutics, Inc. v. ViaCell, Inc., 491 F.3d 1342, 1362 (Fed. Cir. 2007).
`
`1.
`
`Anti-CGRP Antagonist Antibodies and Methods of Making
`Them, Including Humanization Techniques, Were Known
`The ’649 patent states that “[a]nti-CGRP antagonist antibodies are known in
`
`the art.” (Ex. 1001, 25:66-26:3 (citing Ex. 1022).) It confirms that anti-CGRP
`
`antibodies were commercially available, such as antibody #4901 from Sigma
`
`6
`
`

`

`
`
`Aldrich. (Id.; Ex. 1051, 350.) The ’649 patent also expressly discloses that the
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`claimed anti-CGRP antagonistic antibodies may be made using prior art methods:
`
`The anti-CGRP antagonist antibodies may be made by
`any method known in the art. The route and schedule of
`immunization of the host animal are generally in keeping
`with established and conventional techniques for
`antibody stimulation and production, as further described
`herein.
`
`(Ex. 1001, 27:47-51, 31:34-38 (“Anti-CGRP antagonist antibodies and polypeptides
`
`derived from antibodies can be identified or characterized using methods known in
`
`the art . . . .”), 27:51-53 (“General techniques for production of human and mouse
`
`antibodies are known in the art and are described herein.”).)
`
`The ’649 patent states that preparing human and humanized antibodies from
`
`non-human antibodies, such as murine antibodies, was “known” and “conventional.”
`
`(Id., 27:47-53, 31:34-38, 32:14-15, 35:52-53, 37:34-39; see also id., cols. 27-30;
`
`Ex. 1003 ¶72.) According to the ’649 patent, the prior art taught methods to
`
`humanize a monoclonal antibody:
`
`(1) determining the nucleotide and predicted amino acid
`sequence of the starting antibody light and heavy variable
`domains[;] (2) designing the humanized antibody, i.e.,
`deciding which antibody framework region to use during
`
`7
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`the humanizing process[;] (3) the actual humanizing
`methodologies/techniques[;] and (4) the transfection and
`expression of the humanized antibody.
`
`(Id., 28:61-29:18 (citing Queen (Ex. 1023) and other prior art patents); Ex. 1003
`
`¶73.)
`
`The ’649 patent also states that humanized anti-CGRP antagonist antibodies
`
`“are designed to minimize unwanted immunological response toward rodent anti-
`
`human antibody molecules.” (Ex. 1001, 29:21-26.)
`
`2. Methods to Obtain and Measure Particular Binding
`Affinities by SPR Were Known
`The ’649 patent states that “[m]odification of polypeptides is routine practice
`
`in the art and need not be described in detail herein,” including methods to obtain a
`
`desired binding affinity. (Ex. 1001, 37:44-48 (“[amino acid sequence] may be
`
`mutated to obtain an antibody with the desired binding affinity to CGRP”); Ex. 1003
`
`¶74.) The ’649 patent also states that “[m]ethods for determining binding affinity
`
`are well-known in the art,” and specifically lists SPR as a suitable, known method.
`
`(Ex. 1001, 41:28-40; see also id., 27:9-46 (stating that SPR may be used to obtain
`
`binding affinity); Ex. 1003 ¶74).
`
`Limitations of Dependent Claims Were Also Known
`3.
`The ’649 patent also acknowledges that many of the limitations recited in its
`
`dependent claims were known, conventional, or routine.
`
`8
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`a)
`
`Anti-CGRP Antagonist Antibodies That Bound to the
`C-Terminus of CGRP Were Known
`The ’649 patent confirms that prior art anti-CGRP antagonist antibody #4901
`
`binds to the C-terminal fragment of CGRP (e.g., amino acids 25-37), just as the
`
`literature had earlier reported. (Ex. 1001, 25:66-26:3 (identifying #4901 as
`
`commercially available from Sigma), 50:50-57 (“The data indicate that these anti-
`
`CGRP antibodies generally bind to the C-terminal end of CGRP.”), Fig. 1; see
`
`Ex. 1033, 104, 102 (“The monoclonal antibody recognizes an epitome [sic] in the
`
`10 amino acid C-terminal region of the rat α-CGRP.”).)
`
`b) Antibody Types, Constant Regions, and Fc Regions
`Were Known
`The ’649 patent states that anti-CGRP antagonist antibodies may include a
`
`heavy chain constant region derived from human IgG, IgM, IgE, IgA, or IgD
`
`constant regions, and that such regions were known in the art. (Ex. 1001, 12:34-43,
`
`5:6-34, 12:45-47 (“The subunit structures and three-dimensional configurations of
`
`different classes of immunoglobulins are well known.”); Ex. 1003 ¶75.)
`
`Ex. Ex. The ’649 patent acknowledges that antibodies with Fc regions,
`
`including modified Fc regions, were known. (Ex. 1001, 40:13-67; see also id., 44:4-
`
`8.)
`
`9
`
`

`

`
`
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`c)
`
`Antibody Formulations and Methods of
`Administering Them Were Known
`The ’649 patent indicates that pharmaceutical formulations, including
`
`pharmaceutically acceptable excipients, of antibodies are well known. (Ex. 1001,
`
`19:32-47, 21:1-6, 21:39-54; Ex. 1003 ¶76.)
`
`The ’649 patent states that the anti-CGRP antagonist antibody is administered
`
`in accord with known methods, such as by intravenous and subcutaneous
`
`administration. (Ex. 1001, 21:13-20, 21:63-65; Ex. 1003 ¶76.) The patent also
`
`acknowledges that intravenous administration is a type of systemic administration.
`
`(Ex. 1001, 21:20-21; see also id., 5:66-67; Ex. 1003 ¶76.)
`
`Prosecution of the ’649 Patent
`C.
`In a non-final Office Action of June 19, 2013, the Office rejected claims 1-9,
`
`which are identical to those that ultimately issued, as obvious over Frobert
`
`(Ex. 1032) in view of Pisegna (Ex. 1134) and Capon (Ex. 1135). (Ex. 1133, 3.)
`
`According to the Office, Frobert discloses CGRP antagonists such as monoclonal
`
`antibodies that bind to human αCGRP, including residues 25-37 of CGRP. (Id., 4.)
`
`The Office noted that Frobert’s monoclonal antibodies are able to recognize rat and
`
`human CGRP because of peptide sequence conservation. (Id.) In addition, the
`
`Office found that Capon teaches “fusion proteins” comprising immunoglobulin (Ig)
`
`polypeptides fused to “ligand binding partners.” (Id.) According to the Office, such
`
`10
`
`

`

`
`
`fusions “combine the adhesive and targeting characteristics of a ligand binding
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`partner with immunoglobulin effector functions such as complement binding and
`
`cell receptor binding.” (Id.) The Office also stated that Pisegna teaches using SPR
`
`to measure the affinity of CGRP antagonists such as anti-CGRP antagonist
`
`antibodies. (Id.)
`
`According to the Office, a POSA would have been able to discern and isolate
`
`an anti-CGRP antagonist antibody from the antibodies disclosed in Frobert with an
`
`affinity of 50 nM using SPR. (Ex. 1133, 4.) Recognizing that Frobert does not teach
`
`SPR, the Office stated that “the binding determination method of Frobert would have
`
`been reasonably interchangeable with the SPR method of Pisegna” and that a POSA
`
`would have had a reasonable expectation of success because the art recognized the
`
`utility of SPR in characterizing anti-human CGRP antibody binding properties. (Id.,
`
`4-5.)
`
`The Office also rejected the pending claims based on statutory obviousness-
`
`type double patenting over U.S. Patent No. 8,007,794, which issued from the same
`
`provisional application. (Ex. 1133, 5-6.)
`
`In response, Teva argued that none of the references taught or suggested an
`
`anti-CGRP antagonist antibody that was human or humanized. (Ex. 1136, 3.)
`
`Although Teva acknowledged Pisegna’s statement that humanized antibodies are
`
`desirable for therapeutic applications, Teva relied on Tan 1995, as discussed in
`
`11
`
`

`

`
`
`§ VII.C below, to argue that a POSA would not have been motivated to humanize
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`an anti-CGRP antagonist antibody and would not have had a reasonable expectation
`
`that such a humanized antibody would have a therapeutic application.
`
`Notably, Teva did not challenge the Office’s assertion that a POSA would
`
`have been able to isolate an anti-CGRP antagonist antibody and measure its affinity
`
`based on the teachings of Frobert with a binding affinity of 50 nM for human αCGRP
`
`as measured by SPR at 37° C. (Ex. 1136, 3-4.) Thus, Teva implicitly admits that
`
`the prior art provided the requisite tools to make and isolate the claimed anti-CGRP
`
`antagonist antibodies.
`
`Teva did not substantively dispute the obviousness-type double patenting
`
`rejection, but instead filed a terminal disclaimer. (Ex. 1136, 5; Ex. 1137; Ex. 1138.)
`
`The Office found Teva’s statements regarding humanization sufficient to
`
`overcome the § 103 rejection, but as addressed below these statements are inaccurate
`
`and incomplete. (Ex. 1140; infra § VII.C.)
`
`IV. Background and the Asserted Prior Art
`A. CGRP Structure, Isoforms, and Function
`By 2005, the neuropeptide CGRP had been identified and extensively studied.
`
`(Ex. 1002 ¶¶18-25.) Human CGRP is expressed in two closely related isoforms,
`
`αCGRP and βCGRP, both 37 amino acids in length. (Id. ¶18; Ex. 1032, 275;
`
`Ex. 1096, 534.) Human αCGRP and βCGRP differ by only three amino acids.
`
`12
`
`

`

`
`
`(Ex. 1002 ¶18; Ex. 1032, 275; Ex. 1096, 534.) Rat CGRP is also expressed in α and
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`β isoforms, and they differ by only one amino acid. (Ex. 1002 ¶18; Ex. 1032, 275;
`
`Ex. 1096, 534.) CGRP shows significant homology across species: human αCGRP
`
`and βCGRP differ from their rat counterparts by only four and three variations,
`
`respectively. (Ex. 1002 ¶18; Ex. 1033, 93-94; Ex. 1096, 534.) Whereas human
`
`βCGRP is predominantly expressed in the enteric nervous system and pituitary
`
`gland, αCGRP was known to be expressed in sensory neurons, suggesting that
`
`αCGRP had an important role in diseases such as migraine. (Ex. 1002 ¶23;
`
`Ex. 1031, 317 (citing Ex. 1096).)
`
`CGRP has powerful vasodilatory effects that, by 2005, had been directly
`
`linked to various human diseases, including migraine, neurogenic pain, and
`
`psoriasis. (Ex. 1002 ¶¶35-47; Ex. 1026, 7:5-12, 7:19-24, 10:25-30, claims 2, 7;
`
`Ex. 1025, 1105; Ex. 1040, 182-83; Ex. 1096, 533, 567-70.) Moreover, in 2004,
`
`Olesen reported successful human clinical trials demonstrating proof of concept that
`
`blocking the CGRP pathway leads to migraine relief. (Ex. 1002 ¶¶40-43; Ex. 1025,
`
`1104, 1108-09.)
`
`Researchers had also investigated the biological functions of CGRP by using
`
`monoclonal antibodies that bound to CGRP and prevented it from binding to its
`
`receptors. (Ex. 1002 ¶¶48-59.) This is known as “immunoblockade.” (Id.;
`
`Ex. 1022, 566.; Ex. 1002 ¶48.) By 2005, immunoblockade with monoclonal anti-
`
`13
`
`

`

`
`
`CGRP antibodies was shown to inhibit the effects of CGRP in vivo and was
`
`
`
`Petition for Inter Partes Review
`U.S. Patent No. 8,597,649
`
`recognized to have “inherent advantages of defined specificity, known affinity,
`
`reproducibility, and unlimited availability.” (Ex. 1002 ¶59; Ex. 1022, 572.)
`
`B. Anti-CGRP Antagonist Antibodies Were Well Known in the Art
`and Had Been Disclosed for Therapeutic Use in Humans
`By 2005, several publications had described anti-CGRP antagonist antibodies
`
`and methods of making them. (Ex. 1021; Ex. 1022; Ex. 1032; Ex. 1033; Ex. 1055.)
`
`These well-established prior art methods generally involve immunizing mice with
`
`αCGRP, collecting serum, screening for antibodies that exhibit anti-CGRP activity,
`
`culturing hybridoma cells, and producing the monoclonal antibodies in bulk. (See,
`
`e.g., Ex. 1021, 704; Ex. 1033, 95-96; Ex. 1003 ¶¶28-30.) Anti-CGRP antagonist
`
`antibodies had been prepared against both human and rat αCGRP (Ex. 1021, 704;
`
`Ex. 1033, 95-96, 102; Ex. 1055, 88). Because of their sequence homology,
`
`antibodies raised against rat CGRP were known to bind both human and rat CGRP.
`
`(Ex. 1003 ¶105; Ex. 1033, 94, 102; Ex. 1021, 704, 706.)
`
`The prior art specifically identified anti-CGRP antagonist antibodies,
`
`including humanized antibodies, to treat a