`Patent Owner’s Preliminary Response
`
`Filed on behalf of Patent Owner Genentech, Inc. by:
`
`David L. Cavanaugh (Reg. No. 36,476)
`Owen K. Allen (Reg. No. 71,118)
`Robert J. Gunther, Jr. (Pro Hac Vice to be filed)
`Lisa J. Pirozzolo (Pro Hac Vice to be filed)
`Kevin S. Prussia (Pro Hac Vice to be filed)
`Andrew J. Danford (Pro Hac Vice to be filed)
`WILMER CUTLER PICKERING
`HALE AND DORR LLP
`1875 Pennsylvania Ave., NW
`Washington, DC 20006
`
`Adam R. Brausa (Reg. No.
`60,287)
`Daralyn J. Durie (Pro Hac
`Vice to be filed)
`DURIE TANGRI LLP
`217 Leidesdorff Street
`San Francisco, CA 94111
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`____________________________________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`____________________________________________
`
`PFIZER, INC.,
`Petitioner,
`
`v.
`
`GENENTECH, INC.,
`Patent Owner.
`____________________________________________
`
`Case IPR2017-01489
`Patent 6,407,213
`____________________________________________
`
`PATENT OWNER’S PRELIMINARY RESPONSE
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`TABLE OF CONTENTS
`
`Page
`INTRODUCTION ...........................................................................................1
`
`TECHNOLOGY BACKGROUND.................................................................4
`
`I.
`
`II.
`
`A. Antibody “Variable” And “Constant” Domains ........................................4
`
`B.
`
`“Humanized” Antibodies............................................................................5
`
`III. THE ’213 PATENT.........................................................................................8
`
`A.
`
`B.
`
`C.
`
`The Invention..............................................................................................8
`
`Advantages Of The ’213 Invention..........................................................10
`
`Prosecution History ..................................................................................11
`
`IV. PFIZER’S ASSERTED REFERENCES.......................................................12
`
`A. Queen 1989...............................................................................................12
`
`B.
`
`C.
`
`D.
`
`E.
`
`F.
`
`Queen 1990...............................................................................................12
`
`PDB Database...........................................................................................14
`
`Tramontano...............................................................................................15
`
`Kabat 1987................................................................................................16
`
`Hudziak.....................................................................................................16
`
`V.
`
`PERSON OF ORDINARY SKILL ...............................................................17
`
`VI. CLAIM CONSTRUCTION ..........................................................................17
`
`VII. ARGUMENT.................................................................................................19
`
`A.
`
`The Board Should Deny Grounds 2, 3, 4, And 7 Because Neither
`Queen 1990 Nor Tramontano Is Prior Art. ..............................................19
`
`i
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`1.
`
`The inventors produced and tested humanized 4D5 antibodies
`using the ’213 invention before July 26, 1990....................................19
`
`a)
`
`b)
`
`c)
`
`Consensus sequence.......................................................................19
`
`Humanized 4D5 antibody sequences .............................................21
`
`Production and testing of humanized 4D5 antibodies ...................24
`
`(i)
`
`(ii)
`
`First humanized 4D5 variable domain fragment ......................26
`
`First humanized 4D5 full-length antibody................................28
`
`(iii) Other humanized 4D5 variants .................................................29
`
`2.
`
`The challenged claims were reduced to practice before July 26,
`1990.....................................................................................................31
`
`a)
`
`HuMAb4D5-5 and HuMAb4D5-8 embody the challenged
`claims. ............................................................................................31
`
`(i)
`
`Common limitations..................................................................31
`
`(ii) Additional limitations for certain claims ..................................36
`
`b)
`
`c)
`
`The inventors determined that HuMAb4D5-5 and
`HuMAb4D5-8 would work for the intended purpose of the
`challenged claims before July 26, 1990.........................................38
`
`Contemporaneous records from non-inventors corroborate
`the invention of the challenged claims...........................................39
`
`3.
`
`Queen 1990 and Tramontano are not prior art....................................39
`
`a)
`
`b)
`
`Limitations common to all claims..................................................40
`
`Additional limitations for certain claims .......................................41
`
`B.
`
`Pfizer’s Proposed Grounds Fail On The Merits.......................................42
`
`ii
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`1.
`
`Grounds 1, 3, 5, and 6: Queen 1989 in view of the PDB
`database does not render the challenged claims obvious....................43
`
`a)
`
`b)
`
`c)
`
`Queen 1989 contradicts Pfizer’s obviousness theory. ...................44
`
`Queen 1989 does not disclose or suggest substituting
`residues within about 3 angstroms of a CDR.................................45
`
`Pfizer’s proposed combination of Queen 1989 with the PDB
`database results in a broad genus that would not have led to
`the claimed substitutions................................................................45
`
`2.
`
`Grounds 2, 4, and 7: Queen 1990 in view of the PDB database
`does not render the challenged claims obvious...................................47
`
`3.
`
`4.
`
`5.
`
`a)
`
`b)
`
`a)
`
`b)
`
`Queen 1990 contradicts Pfizer’s obviousness theory. ...................47
`
`Pfizer’s proposed combination of Queen 1990 with the PDB
`database results in a broad genus that would not have led to
`the claimed substitutions................................................................49
`
`Grounds 1 and 2: Queen 1989 and Queen 1990 do not render
`obvious claim 63. ................................................................................50
`
`Grounds 1-4: Queen 1989 and Queen 1990 do not render
`obvious claim 65. ................................................................................51
`
`Grounds 2, 5, and 7: Pfizer’s asserted references do not render
`obvious the “consensus” sequence limitations of claims 4, 33,
`62, 64, and 69. .....................................................................................53
`
`Grounds 2 and 7: Queen 1990 does not render obvious the
`claimed “consensus” sequence.......................................................53
`
`Ground 5: Queen 1989 in view of Kabat 1987 does not
`render obvious the claimed “consensus” sequence........................55
`
`6.
`
`Grounds 3 and 4: Claims 75-77 and 79 would not have been
`obvious in view of Pfizer’s proposed combinations. ..........................57
`
`a)
`
`Claim 75 .........................................................................................57
`
`iii
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`b)
`
`Claims 76-77 and 79 ......................................................................59
`
`Grounds 6 and 7: Claims 30, 31, 33, 42, and 60 would not have
`been obvious in view of Pfizer’s proposed combinations...................60
`
`Objective Indicia Of Non-Obviousness Confirm The Patentability
`Of The Challenged Claims. ......................................................................61
`
`Unexpected results ..............................................................................61
`
`Commercial success ............................................................................64
`
`C.
`
`7.
`
`1.
`
`2.
`
`D.
`
`Inter Partes Review Proceedings Are Unconstitutional. .........................65
`
`VIII. CONCLUSION..............................................................................................66
`
`iv
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`TABLE OF AUTHORITIES
`
`Page(s)
`
`Cases
`
`Amgen, Inc. v. Chugai Pharmaceutical Co.,
`927 F.2d 1200 (Fed. Cir. 1991)................................................................................59
`
`Brown & Williamson Tobacco Corp. v. Philip Morris Inc.,
`229 F.3d 1120 (Fed. Cir. 2000)................................................................................65
`
`In re Clarke,
`356 F.2d 987 (C.C.P.A. 1966) .................................................................................38
`
`Cooper v. Goldfarb,
`154 F.3d 1321 (Fed. Cir. 1998)................................................................................39
`
`Crocs, Inc. v. International Trade Commission,
`598 F.3d 1294 (Fed. Cir. 2010)................................................................................61
`
`In re Cyclobenzaprine Hydrochloride Extended-Release Capsule Patent
`Litigation,
`676 F.3d 1063 (Fed. Cir. 2012)................................................................................61
`
`Innogenetics, N.V. v. Abbott Laboratories,
`512 F.3d 1363 (Fed. Cir. 2008)................................................................................52
`
`Insite Vision, Inc. v. Sandoz, Inc.,
`783 F.3d 853 (Fed. Cir. 2015)......................................................................46, 49, 60
`
`McCormack Harvesting Machine Co. v. C. Aultman & Co.,
`169 U.S. 606 (1898).................................................................................................66
`
`Markman v. Westview Instruments, Inc.,
`517 U.S. 370 (1996).................................................................................................65
`
`In re Merchant,
`575 F.2d 865 (C.C.P.A. 1978) .................................................................................61
`
`v
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`Mikus v. Wachtel,
`504 F.2d 1150 (C.C.P.A. 1974) ...............................................................................35
`
`In re NTP, Inc.,
`654 F.3d 1279 (Fed. Cir. 2011)................................................................................31
`
`Oil States Energy Services, LLC v. Greene’s Energy Group, LLC,
`137 S. Ct. 2239 (2017).............................................................................................66
`
`In re Schaub,
`537 F.2d 509 (C.C.P.A. 1976) .................................................................................36
`
`Sinorgchem Co. v. International Trade Commission,
`511 F.3d 1132 (Fed. Cir. 2007)................................................................................18
`
`In re Soni,
`54 F.3d 746 (Fed. Cir. 1995)....................................................................................62
`
`In re Spiller,
`500 F.2d 1170 (C.C.P.A. 1974) ...............................................................................36
`
`In re Steed,
`802 F.3d 1311 (Fed. Cir. 2015)................................................................................31
`
`In re Taub,
`348 F.2d 556 (C.C.P.A. 1965) .................................................................................35
`
`Tokai Corp. v. Easton Enterprises, Inc.,
`632 F.3d 1358 (Fed. Cir. 2011)................................................................................64
`
`Statutes
`
`35 U.S.C.
`
`§ 102.........................................................................................................................39
`
`§ 120.........................................................................................................................40
`
`vi
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`I.
`
`INTRODUCTION
`In the early 1990s, the field of therapeutic antibodies was still in its infancy.
`
`Although scientists had known since the 1970s how to obtain antibodies from
`
`animals (e.g., mice) that would bind to specific targets, those antibodies generally
`
`could not be used in humans because over time the body’s own immune system
`
`would attack and inactivate them (known as an “immunogenic” response).
`
`Beginning in the late 1980s, a few scientists had attempted to create “humanized”
`
`antibodies that incorporated the binding site from a non-human antibody into a
`
`human antibody framework—which they hoped might address the immunogenicity
`
`problem by reducing the amount of non-human amino acid sequences in the
`
`antibody. But those early humanized antibodies suffered from reduced binding
`
`affinity or still produced an immunogenic response. Given those challenges, which
`
`continued throughout the late 1980s, there were no humanized antibodies on the
`
`market, and some doubted it would ever be possible to develop one that could be
`
`used therapeutically.
`
`In the late 1980s, Genentech scientists began developing a new
`
`humanization approach that solved those problems. Rather than starting from an
`
`actual human antibody sequence, they created an artificial “consensus” sequence—
`
`consisting of the most frequently occurring amino acids at each location in all
`
`human antibodies of the same subclass or subunit structure. That novel consensus
`
`1
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`sequence approach—which minimized the prior art immunogenicity problem and
`
`provided a broadly-applicable platform for humanizing antibodies—is protected by
`
`U.S. Patent No. 6,407,213 (“the ’213 patent”). The inventors initially applied their
`
`consensus sequence approach to humanize the murine 4D5 antibody and create the
`
`drug Herceptin®—a lifesaving therapy for an aggressive form of breast cancer.
`
`Their invention was later used to develop numerous other highly successful
`
`therapeutic antibodies for a wide range of diseases.
`
`Pfizer’s petition challenges certain claims of the ’213 patent on seven
`
`different obviousness grounds, but fails to demonstrate a reasonable likelihood of
`
`success for any of them.
`
`As an initial matter, the references underlying Grounds 2, 3, 4, and 7—
`
`Queen 1990 (Ex. 1550) and Tramontano (Ex. 1551)—are not prior art. The ’213
`
`inventors reduced their invention to practice before the publication of Queen 1990
`
`and Tramontano by creating and testing humanized antibodies that embody the
`
`challenged claims. That actual reduction to practice is corroborated by extensive
`
`contemporaneous records from the inventors and several non-inventors.
`
`Even if Pfizer could rely on Queen 1990 or Tramontano, Pfizer has failed to
`
`demonstrate a reasonable likelihood of success for any challenged claim.
`
`First, Pfizer argues for each ground that a skilled artisan would have arrived
`
`at the challenged claims by combining Queen 1989 (Ex. 1534) or Queen 1990 with
`
`2
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`nine different published antibody structures. But the Queen references emphasize
`
`the importance of using a “best-fit” approach starting from the single human
`
`antibody sequence most homologous to the original non-human antibody. A
`
`skilled artisan would not have taken the opposite approach by combining the
`
`Queen references with nine different antibodies—without regard to whether those
`
`antibodies are similar to the original non-human antibody.
`
`Second, Pfizer has not demonstrated that certain claim limitations would
`
`have been obvious, including (i) “lacks immunogenicity” in claim 63 (Grounds 1-
`
`2); (ii) “up to 3-fold more” binding affinity in claim 65 (Grounds 1-4); and (iii)
`
`“consensus” sequence in claims 4, 33, 62, 64, and 69 (Grounds 1-2, 5, and 7).
`
`Pfizer’s arguments for these claims rest on speculation and are not supported by
`
`the asserted references.
`
`Finally, even under Pfizer’s theory, the proposed obviousness combinations
`
`for each ground would have resulted in numerous possible amino acid
`
`substitutions—including many outside the scope of the challenged claims. Pfizer
`
`has not met its burden to explain why the claimed substitutions would have been
`
`chosen out of the numerous other possibilities that Pfizer admits a skilled artisan
`
`would have had to confront.
`
`The Board should deny institution.
`
`3
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`II.
`
`TECHNOLOGY BACKGROUND
`A.
`Antibody “Variable” And “Constant” Domains
`
`The immune system defends against foreign substances, known as
`
`“antigens” (e.g., viruses or bacteria), by producing antibodies. Antibodies are
`
`proteins that recognize and bind to antigens, which facilitates their removal from
`
`the body. (Ex. 1582 at 1.) A typical antibody (sometimes called an
`
`“immunoglobulin”) consists of four amino acid chains: two identical heavy chains
`
`and two identical light chains, which join to form a “Y” shape, as shown below:
`
`(Ex. 2023 at 10 (annotated); Ex. 1501, 1:17-20.) Each chain contains a “variable”
`
`domain at one end (red box above) and “constant” domains at the other (green box
`
`above). (Ex. 1501, 1:20-27.) The variable domains for the heavy chain (VH) and
`
`light chain (VL) are illustrated above in blue and pink, respectively.
`
`4
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`Variable domains directly bind to the antigen. (Id., 1:35-37.) Each variable
`
`domain contains three “complementarity determining regions,” or “CDRs,” (id.,
`
`1:35-50), shown as CDR1, CDR2, and CDR3 in the enlarged portion above.
`
`Variable domains also contain four “framework regions,” or “FRs”—one on either
`
`side of each CDR—shown as FR1, FR2, FR3, and FR4 in the same enlarged
`
`portion. The framework regions form an immunoglobulin core structure from
`
`which the CDRs extend and form a binding site for interaction with the antigen.
`
`(Id., 1:47-50.) In contrast to the CDRs, which generally contain unique amino
`
`acids (or “residues”) for a particular antigen, the framework regions may have
`
`more amino acid sequences in common (i.e., the same amino acids at the same
`
`positions) across other antibodies. (Id., 1:37-44.)
`
`The constant domains are not directly involved in antigen binding and
`
`typically have similar amino acid sequences across all antibodies within a subclass.
`
`(Ex. 2016, Presta Decl. ¶ 15.)
`
`B.
`
`“Humanized” Antibodies
`
`Before the ’213 patent, antibodies targeting a specific antigen could be
`
`obtained from animals, such as mice. (Ex. 1501, 1:52-58.) Although those non-
`
`human antibodies could bind to a desired target, they had limited use
`
`therapeutically because the human immune system would over time identify them
`
`as antigens and attack them—known as an “antigenic” or “immunogenic”
`
`5
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`response. (Id., 1:55-58.) An immunogenic response had adverse clinical
`
`consequences because it inactivated the antibody and resulted in its premature
`
`removal from the body. (E.g., Ex. 1528 at 3 (noting “large fall in circulating
`
`mouse immunoglobulin” due to immunogenic response and accompanying
`
`“adverse clinical reaction”).)
`
`Scientists developed several techniques trying to address that issue. One
`
`approach used “chimeric” antibodies that combined a non-human variable domain
`
`(e.g., the entire variable domain from a mouse antibody) with a human constant
`
`domain. (Id., 1:59-2:19.) However, because chimeric antibodies retained a
`
`significant portion of the non-human antibody sequence, immunogenicity could
`
`still result. (Id., 2:12-19; Ex. 2022 at 2156.)
`
`Attempting to reduce immunogenicity, scientists created “humanized”
`
`antibodies that included a human variable domain substituted with the amino acid
`
`sequence of the non-human CDRs. (Ex. 1501, 2:20-52.) But that approach could
`
`reduce the antibody’s ability to bind to specific antigens. (Ex. 1534 at 5
`
`(“Unfortunately, in some cases the humanized antibody had significantly less
`
`binding affinity for antigen than did the original mouse antibody.”).)1
`
`1
`
`In this proceeding, Patent Owner uses “chimeric” and “humanized” as
`
`defined in the ’213 patent. (Ex. 1501, 1:59-62 (“chimeric” antibodies have “an
`
`6
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`In attempting to address these various shortcomings, scientists pursued
`
`techniques seeking to make humanized antibodies that balanced strong binding
`
`with low immunogenicity. For example, Queen 1989 (Ex. 1534) selected a human
`
`variable domain by comparing a mouse antibody against known human antibody
`
`sequences, and choosing a human framework that was “as homologous as possible
`
`to the original mouse antibody to reduce any deformation of the mouse CDRs.”
`
`(Ex. 1534 at 5.) The humanized sequence was then further refined using computer
`
`modeling “to identify several framework amino acids in the mouse antibody that
`
`might interact with the CDRs or directly with antigen, and these amino acids were
`
`transferred to the human framework along with the CDRs.” (Id.) That technique
`
`became known as the “best-fit” approach because it started from a human sequence
`
`with the closest match to the non-human antibody. (Ex. 2024 at 4184.)
`
`Even using the best-fit approach, however, it still was difficult to produce an
`
`antibody with both strong binding and low immunogenicity. (Ex. 1501, 3:50-52.)
`
`animal antigen-binding variable domain [that] is coupled to a human constant
`
`domain”); id., 8:11-17 (“humanized” antibodies contain a framework region
`
`“having substantially the same amino acid sequence of a human immunoglobulin
`
`and a CDR having substantially the amino acid sequence of a non-human
`
`immunoglobulin”).)
`
`7
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`The best-fit approach also was inefficient because it required a new human
`
`antibody sequence as the starting point for each different humanized antibody.
`
`III. THE ’213 PATENT
`A.
`The Invention
`
`Beginning in the late 1980s, Drs. Paul Carter and Leonard Presta at
`
`Genentech developed a new approach to humanizing antibodies that solved the
`
`prior art binding and immunogenicity problems. Rather than starting from the
`
`most homologous human sequence, Drs. Carter and Presta developed a “consensus
`
`human sequence”—i.e., “an amino acid sequence which comprises the most
`
`frequently occurring amino acid residues at each location in all human
`
`immunoglobulins of any particular subclass or subunit structure.” (Id., 11:32-38.)
`
`That “consensus” sequence provided a single human amino acid sequence that
`
`would be the starting point for any humanized antibody of a particular subclass or
`
`subunit structure (e.g.(cid:15)(cid:3)(cid:79)(cid:76)(cid:74)(cid:75)(cid:87)(cid:3)(cid:70)(cid:75)(cid:68)(cid:76)(cid:81)(cid:3)(cid:539)(cid:20)(cid:12)(cid:17)(cid:3)(cid:3)(cid:11)Id., 54:66-56:57.)
`
`The ’213 inventors developed a multi-step process for their approach. First,
`
`they added the non-human CDRs to the human consensus sequence. (Id., 20:12-
`
`31.) Next, they evaluated the differences between the framework regions of the
`
`non-human antibody and the human consensus sequence to determine whether
`
`further modifications to the consensus sequence were needed. (Id., 20:32-40.)
`
`8
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`For framework positions where the non-human antibody sequence differed
`
`from the human consensus sequence, Drs. Carter and Presta used computer
`
`modeling to identify whether the different non-human amino acid (i) “non-
`
`covalently binds antigen directly”; (ii) “interacts with a CDR”; (iii) “participates in
`
`the VL-VH interface,” i.e., the interface between variable domains of the heavy and
`
`light chains, or (iv) is a glycosylation site outside the CDRs that is likely to affect
`
`“antigen binding and/or biological activity.” (Id., 20:32-21:36, 54:64-56:57.)
`
`They believed that those positions were important to maintaining binding affinity
`
`because they could influence the three-dimensional shape of the CDRs. (Id.,
`
`20:32-35.) If any of those four requirements was met, the amino acid at that
`
`position in the consensus sequence could be substituted with the amino acid that
`
`appears at the same position in the non-human antibody. Otherwise, the amino
`
`acid sequence of the human consensus sequence was retained. (Id., 20:66-21:8.)
`
`The ’213 challenged claims reflect the inventors’ novel consensus sequence
`
`approach. Each challenged claim requires a “humanized” antibody or variable
`
`domain that contains non-human CDRs and one or more specified framework
`
`amino acid substitutions. As explained below, the claimed framework
`
`substitutions are the amino acid positions that the inventors determined were
`
`important to antibody binding.
`
`9
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`B.
`
`Advantages Of The ’213 Invention
`
`The ’213 patent’s consensus sequence approach was a significant advance
`
`over the prior art.
`
`First, using a consensus sequence minimized the immunogenicity problems
`
`that plagued other humanization techniques. (Ex. 1502 at 3439-41, ¶¶ 2-9.) At the
`
`same time, humanized antibodies made according to the ’213 invention retain
`
`strong binding for the targeted antigen, or even have improved binding over the
`
`original non-human antibody. (Ex. 1501, 4:24-28, 51:50-53.)
`
`Second, under the best-fit approach, the most homologous human sequence
`
`itself may be a rare antibody sequence that would trigger an immunogenic
`
`response—for example, due to unique variations in individual patients. (Ex. 2020,
`
`Presta Decl. ¶ 24.) The ’213 patent avoids that problem by starting from a
`
`consensus sequence comprising only the most frequently occurring amino acids at
`
`each position. (Ex. 1501, 11:32-38.)
`
`Third, unlike the prior art best-fit approach—that required identifying the
`
`most homologous human antibody sequence for each antibody to be humanized—
`
`the ’213 patent provided a single human antibody sequence as a starting point that
`
`could be applied to a wide variety of antibodies. (Ex. 1502 at 3439-41, ¶¶ 2-9.)
`
`Genentech has used the ’213 invention to develop numerous drugs for a wide
`
`variety of diseases, such as Herceptin® (breast and gastric cancer), Perjeta® (breast
`
`10
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`cancer), Avastin® (colon, lung, ovarian, cervical, kidney, and brain cancer),
`
`Lucentis® (macular degeneration), and Xolair® (asthma). (Ex. 2017, Carter Decl. ¶
`
`4; Ex. 2016, Presta Decl. ¶ 5.)
`
`C.
`
`Prosecution History
`
`The ’213 patent is a continuation-in-part of an application filed on June 14,
`
`1991. (Ex. 1501, coversheet.) The challenged claims issued over hundreds of
`
`references considered during prosecution, including every reference underlying
`
`Pfizer’s proposed grounds. (Ex. 1501 at 1-6.) Pfizer asserts that the PDB database
`
`was not considered during prosecution. (Paper 1 at 14.) But that is incorrect.
`
`Indeed, the ’213 specification itself repeatedly cites the PDB database. (Ex. 1501,
`
`16:31-34, 19:35-41, 48:13-17.)
`
`During prosecution, the applicants submitted a joint affidavit from Drs.
`
`Carter and Presta to antedate U.S. Patent No. 5,693,762, which had a filing date of
`
`September 28, 1990. (Ex. 1502 at 4432-33.) The examiner allowed the claims
`
`after accepting that antedation evidence. (Id. at 4443.) As detailed below, the
`
`record in this proceeding further confirms that the ’213 invention was also
`
`conceived and reduced to practice before the publication of either Queen 1990
`
`(July 26, 1990) or Tramontano (September 5, 1990).
`
`11
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`IV. PFIZER’S ASSERTED REFERENCES
`A.
`Queen 1989
`
`Queen 1989 describes the humanization of a murine anti-TAC antibody.
`
`(Ex. 1534 at 1 (abstract).) Unlike the ’213 patent, Queen 1989 does not disclose
`
`the use of a generalized “consensus” sequence. Instead, as discussed above, Queen
`
`1989 used a best-fit approach, which involved (i) searching a database of antibody
`
`sequences to identify a human framework “as homologous as possible to the
`
`original mouse antibody to reduce any deformation of the mouse CDRs” (id. at 5);
`
`and (ii) incorporating the murine CDRs into that human sequence (id. at 3).
`
`Queen 1989 then identified additional locations in the human framework to
`
`substitute with murine residues. If the human framework contained “atypical”
`
`residues, Queen 1989 substituted them with more commonly-occurring amino
`
`acids from the murine antibody. (Id. at 4.) Queen 1989 also used a computer
`
`model of the murine antibody “to identify several amino acids which, while outside
`
`the CDRs, are likely to interact with the CDRs or antigen.” (Id. at 1 (abstract).)
`
`Using those techniques, Queen 1989 identified nine substitutions. (Id. at 3.) None
`
`of those substitutions, however, fall within the scope of the challenged claims.
`
`B.
`
`Queen 1990
`
`Queen 1990 is a PCT application published July 26, 1990. It is not prior art.
`
`(See infra pp. 19-41.)
`
`12
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`Like Queen 1989, Queen 1990 used a best-fit approach to produce a
`
`humanized antibody by starting from a human sequence most homologous to the
`
`mouse antibody. (Ex. 1550, 26:5-33:25.) Queen also identified four general
`
`criteria for designing humanized antibodies.
`
`Criterion I: As a starting point, Queen 1990 emphasized the importance of
`
`choosing the human sequence most similar to the non-human antibody to reduce
`
`the possibility of distorting the binding site formed by the CDRs. (Id., 12:17-35.)
`
`Queen 1990 mentioned “a consensus framework” (id., 12:19-20), but included no
`
`details of what that “consensus framework” might be or how it might be used to
`
`make a humanized antibody.
`
`Criterion II: After selecting a best-fit human framework sequence, Queen
`
`1990 provided that “unusual” or “rare” amino acids could be replaced with more
`
`common amino acids from the non-human sequence. (Id., 13:22-32.) This step
`
`was intended to eliminate residues from the selected human framework that may
`
`“disrupt the antibody structure” by replacing them with non-human residues
`
`commonly found in other human antibody sequences. (Id., 13:32-37.)
`
`Criterion III: Queen 1990 disclosed that non-human residues may be used
`
`immediately adjacent to CDRs because “[t]hese amino acids are particularly likely
`
`to interact with the amino acids in the CDR’s [sic]” or “interact directly with the
`
`13
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`antigen.” (Id., 14:1-12.) Accordingly, Queen 1990 hypothesized that using non-
`
`human residues at those positions may help maintain strong binding. (Id.)
`
`Criterion IV: Queen 1990 used computer modeling, “typically of the
`
`original donor antibody,” to identify other residues that “have a good probability of
`
`interacting with amino acids in the CDR’s [sic] by hydrogen bonding, Van der
`
`Waals forces, hydrophobic interactions, etc.” (Id., 14:14-19.) Non-human
`
`residues may be substituted at those positions that may interact with CDRs. (Id.,
`
`14:19-21.) Amino acids satisfying this criterion “generally have a side chain atom
`
`within about 3 angstrom units of some site in the CDR’s [sic].” (Id., 14:22-25.)
`
`Queen 1990 disclosed the sequence of an anti-TAC antibody produced using
`
`its humanization technique. (Id., Fig. 2.) However, Pfizer does not contend that
`
`any antibody sequence disclosed in Queen 1990 anticipates or renders obvious the
`
`challenged ’213 claims. Instead, Pfizer argues that Queen 1990’s four general
`
`criteria would have led a skilled artisan to the specific residue substitutions
`
`identified in the challenged claims. (Paper 1 at 36-38.)
`
`C.
`
`PDB Database
`
`The Protein Data Bank (“PDB”) “was established in 1971 as a computer-
`
`based archival file for macromolecular structures” that could “collect, standardize,
`
`and distribute atomic co-ordinates and other data from crystallographic studies.”
`
`(Ex. 1580 at 535.)
`
`14
`
`
`
`IPR2017-01489
`Patent Owner’s Preliminary Response
`
`Pfizer cites data from nine antibody crystal structures available in the PDB
`
`database prior to August 1989. (Ex. 1503C, Foote Exs. F-N.) As discussed below,
`
`Pfizer contends that those crystal structures would have supposedly led to
`
`numerous possible framework substitutions—only a fraction of which correspond
`
`with the challenged claims.
`
`D.
`
`Tramontano
`
`Tramontano (Ex. 1551) was published on September 5, 1990. (Ex. 2027
`
`(showing date).) Tramontano therefore is not prior art. (See infra pp. 19-41.)
`
`Tramontano analyzed several antibody structures and found that “the major
`
`determinant” of the position of one of the CDRs “is the size of the residue at
`
`[heavy cha