Free Claim Construction Opening Brief - District Court of Delaware - Delaware

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 1 of 44

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE PDL BIOPHARMA, INC. Plaintiff, v. ALEXION PHARMACEUTICALS, INC. Defendant.
) ) ) ) ) ) ) ) ) ) )

Civil Action No. 07-156-JJF

DEFENDANT ALEXION PHARMACEUTICALS, INC.'S OPENING BRIEF IN SUPPORT OF ITS PROPOSED CLAIM CONSTRUCTIONS YOUNG CONAWAY STARGATT & TAYLOR LLP Josy W. Ingersoll (I.D. #1088) John W. Shaw (I.D. #3362) Andrew A. Lundgren (I.D. #4429) The Brandywine Building 1000 West Street, 17th Floor P.O. Box 391 Wilmington, Delaware 19801 Telephone: (302) 571-6600 Attorneys for Defendant Of Counsel: John M. Desmarais Gerald J. Flattmann, Jr. Christine Willgoos KIRKLAND & ELLIS LLP 153 East 53rd Street New York, NY 10022 Telephone: (212) 446-4800 Facsimile: (212) 446-4900 Dated: May 14, 2008

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 2 of 44

TABLE OF CONTENTS NATURE AND STAGE OF THE PROCEEDINGS ......................................................................1 SUMMARY OF THE ARGUMENT ..............................................................................................4 ARGUMENT...................................................................................................................................5 I. II. LEGAL STANDARD FOR CONSTRUING CLAIMS ..........................................5 BACKGROUND OF ANTIBODY TECHNOLOGY .............................................8 A. B. III. Antibody Structure And Function................................................................8 Prior Art Antibody Humanization ...............................................................8

THE PATENTS-IN-SUIT .....................................................................................11 A. B. PDL's CDR And Framework Substitution Limitations.............................11 PDL's Homology Limitation .....................................................................13

IV.

ALEXION'S PROPOSED CONSTRUCTIONS...................................................14 A. "Complementarity Determining Regions" ("CDRs")................................14 1. 2. 3. The Patent Specification Supports Alexion's Construction Of CDR ..........................................................................................15 The File Histories Support Alexion's Definition Of CDR ............16 One Of Ordinary Skill In The Art Would Understand The Claim Term To Include The Hypervariable Regions As Defined By Chothia .......................................................................19

B. C.

"Framework" Terms ..................................................................................21 "Humanized Immunoglobulin"..................................................................22 1. 2. 3. The Specification Of The Patents-In-Suit......................................23 The File Histories Of The Patents-In-Suit .....................................26 One Of Ordinary Skill In The Art..................................................29

D.

"Donor Immunoglobulin" And "Human Acceptor Immunoglobulin".......................................................................................30

i

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 3 of 44

E.

"Humanized Immunoglobulin Heavy Chain Variable Region Framework" And "Humanized Immunoglobulin Light Chain Variable Region Framework" ....................................................................31 "DNA Segment Encoding" Terms.............................................................31 "Hypervariable Regions" ...........................................................................33 "Synthesizing A DNA Segment Encoding" Terms ...................................34 1. 2. The Patent Specification And File Histories Support Alexion's Construction ..................................................................35 One Of Ordinary Skill In The Art..................................................37

F. G. H.

V.

CONCLUSION......................................................................................................38

ii

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 4 of 44

TABLE OF AUTHORITIES Cases Alloc, Inc. v. Int'l Trade Comm'n, 342 F.3d 1361 (Fed. Cir. 2003)........................................................................................... 6 Andersen Corp. v. Fiber Composites, LLC, 474 F.3d 1361 (Fed. Cir. 2007)........................................................................................... 6 ATD Corp. v. Lydall, Inc., 159 F.3d 534 (Fed. Cir. 1998)........................................................................................... 22 CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359 (Fed. Cir. 2002)..................................................................................... 6, 22 Conoco, Int'l, 460 F.3d at 1362 ............................................................................................................. 4, 7 Digital Biometrics, Inc. v. Identix, Inc., 149 F.3d 1335, 1347 (Fed. Cir. 1998)........................................................................... 5, 17 Gentry Gallery, Inc. v. The Berkline Corp., 134 F.3d 1473 (Fed. Cir. 1998)......................................................................................... 22 Gillespie v. Dywidag Systems Int'l, USA, 501 F.3d 1285 (Fed. Cir. 2007)..................................................................................... 5, 17 Hockerson-Halberstadt, Inc. v. Avia Group Int'l, Inc., 222 F.3d 951 (Fed. Cir. 2000)....................................................................................... 6, 20 Johnson Worldwide Assocs., Inc. v. Zebco Corp., 175 F.3d 985 (Fed. Cir. 1999)............................................................................................. 6 Jonsson v. Stanley Works, 903 F.2d 812 (Fed. Cir. 1990)......................................................................................... 5, 6 Liposome Co., Inc. v. Vestar, Inc., Civ. A. No. 92-332-RRM, 1994 WL 738952 (D. Del. Dec. 20, 1994) .............................. 7 Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996)............................................................................................................ 7 MBO Labs., Inc. v. Becton, Dickinson & Co., 474 F.3d 1323 (Fed. Cir. 2007)........................................................................................... 6 Medtronic Navigation, Inc. v. BrainLab Medizinische Computersysteme GmbH, 222 Fed.Appx. 952 (Fed. Cir. 2007)................................................................................. 30

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 5 of 44

Nikon Corp. v. ASM Lithography B.V., 308 F. Supp. 2d 1039 (N.D. Cal. 2004) .............................................................................. 6 Ortho-McNeil Pharm., Inc. v. Caraco Pharm. Labs. Ltd., 476 F.3d 1321 (Fed. Cir. 2007).......................................................................................... 7 Pharmacia & Upjohn Co. v. Mylan Pharmaceuticals, Inc., 170 F.3d 1373 (Fed. Cir. 1999)........................................................................................... 6 Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005)................................................................................. 4, 5, 17 Rheox, Inc. v. Entact, Inc., 276 F.3d 1319 (Fed. Cir. 2002)........................................................................................... 5 Sentry Protection Products, Inc. v. Eagle Mfg. Co., 400 F.3d 910 (Fed. Cir. 2005)........................................................................... 6, 22, 25, 31 Southwall Techs., Inc. v. Cardinal IG Co., 54 F.3d 1570 (Fed. Cir. 1995)............................................................................................. 5 Spectrum Int'l, Inc. v. Sterilite Corp., 164 F.3d 1372 (Fed. Cir. 1998)..................................................................................... 5, 17 Sunrace Roots Enter. Co., Ltd. v. SRAM Corp., 336 F.3d 1298 (Fed. Cir. 2003)......................................................................................... 22 Tanabe Seiyaku Co., Ltd. v. United States Int'l Trade Comm'n, 109 F.3d 726 (Fed. Cir. 1997)............................................................................................. 7 Wang Labs., Inc. v. Am. Online, Inc., 197 F.3d 1377 (Fed. Cir. 1999)........................................................................................... 5

2

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 6 of 44

NATURE AND STAGE OF THE PROCEEDINGS Plaintiff PDL Biopharma, Inc. ("PDL") filed its complaint against Alexion Pharmaceuticals, Inc. ("Alexion") alleging infringement of U.S. Patents Nos. 5,693,761 ("the `761 patent"); 5,693,762 ("the `762 patent"); and 6,180,370 ("the `370 patent") on March 16, 2007. D.I. 1. Alexion filed its Answer and Counterclaims on June 4, 2007 and PDL replied on June 25, 2007. D.I. 11, 14. Alexion filed Amended Counterclaims on December 7, 2007 and PDL replied on January 4, 2008. D.I. 45, 46. The parties are currently engaged in fact discovery, with hard copy documents to be exchanged by May 16, 2008, completion of document production by July 10, 2008, and close of fact discovery on October 2, 2008. See D.I. 76. On March 10, 2008, PDL filed a motion to dismiss Alexion's counterclaim of invalidity with respect to certain claims of the patents-in-suit. D.I. 67. Alexion opposed PDL's motion and the Court heard oral argument on April 11, 2008. The Court has not yet decided PDL's motion. PDL has notified Alexion that it intends to provide Alexion with a unilateral covenant not to sue Alexion regarding certain claims, and that, in PDL's view, the covenant moots Alexion's counterclaim and PDL's motion to dismiss. Alexion, however, does not agree that the covenant is sufficient to moot its counterclaim as to the identified claims, and maintains its right to oppose the covenant and dismissal of its counterclaim.1 Accordingly, Alexion seeks

1

For example, PDL's proposed covenant is insufficient to moot Alexion's counterclaim because it narrowly limits the covenant to Alexion's antibody solely as it is used for the current FDA indication of PNH. PDL's patents however, purport to claim compositions and methods of making those compositions, and do not claim any methods of use. Accordingly, PDL has not mooted the claims with respect to Alexion's Soliris® antibody composition or method of manufacture ­ instead, PDL reserves the right to sue Alexion should Alexion obtain approval for the same composition manufactured by the same methods but indicated for the treatment of a different disease. Alexion will fully brief this and other insufficiencies of PDL's covenant if and when PDL argues to the Court that the covenant partially moots Alexion's counterclaim of invalidity.

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 7 of 44

construction of one claim term that is at issue in Alexion's counterclaim, but is identified in PDL's motion to dismiss. The parties exchanged preliminary lists of terms for construction on February 27, 2008. On May 7, 2008, the parties met and conferred and exchanged proposed constructions for terms to be construed by the Court. The parties were able to agree on constructions of some claim terms. However, there are still several terms on which the parties cannot reach agreement. This is Alexion's Opening Brief in support of its proposed claim constructions. SUMMARY OF THE ARGUMENT Two key limitations of the patents-in-suit, "CDR" and "framework substitutions" drive the primary dispute between PDL and Alexion regarding claim construction of PDL's claimed humanized immunoglobulins. Alexion's proposed construction recognizes the true scope of the claims within the context of PDL's statements regarding the criticality of certain claim features vis-à-vis the prior art. Alexion's proposed constructions are supported by the specification and the file histories of the patents in suit. PDL's explicit admissions ­ made to the Patent Office during prosecution of the patents-in-suit and to the European Patent Office during prosecution and opposition of their foreign counterparts ­ compel the claim constructions Alexion proposes. Moreover, as set forth in the accompanying declarations of experts Dr. Michael R. Clark and Dr. Jefferson Foote, Alexion's constructions are the meanings one of skill in the art would ascribe to the claim terms. PDL, in contrast, seeks to reclaim a scope of its claims that it explicitly disclaimed during prosecution of the patents-in-suit. Indeed, PDL's proposed constructions ignore its repeated representations that certain claim limitations were essential to its invention. But PDL is not entitled to constructions that enlarge the scope of its invention to encompass that which it disclaimed during prosecution.

4

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 8 of 44

Accordingly, Alexion requests that the Court adopt its construction of the disputed claim terms. ARGUMENT I. LEGAL STANDARD FOR CONSTRUING CLAIMS Claim construction involves the search for the ordinary and customary meaning of a term to one of ordinary skill in the art. See Conoco, Int'l v. Energy & Envtl. Int'l, L.C., 460 F.3d 1349, 1362 (Fed. Cir. 2006). To determine the meaning of a claim term, "the court starts the decision making process by reviewing the same resources as would that person, viz., the patent specification and the prosecution history." Phillips v. AWH Corp., 415 F.3d 1303, 1313 (Fed. Cir. 2005) (en banc). Thus, claim construction rests primarily on three sources of intrinsic evidence ­ the claims, the specification, and the prosecution history. A patentee may employ the specification to define a patent's use of specific claim terms. See id. at 1316. Federal Circuit precedent "recognize[s] that the specification may reveal a special definition given to a claim term by the patentee that differs from the meaning it would otherwise possess. In such cases, the inventor's lexicography governs." See id. Similarly, the prosecution history of the patent must be considered in defining claim terms. "Like the specification, the prosecution history provides evidence of how the PTO and the inventor understood the patent . . . [and] was created by the patentee in attempting to explain and obtain the patent." Id. at 1317. In particular, "[t]he prosecution history limits the interpretation of claim terms so as to exclude any interpretation that was disclaimed during prosecution." Rheox, Inc. v. Entact, Inc., 276 F.3d 1319, 1325 (Fed. Cir. 2002) (quoting Southwall Techs., Inc. v. Cardinal IG Co., 54 F.3d 1570, 1576 (Fed. Cir. 1995) ("Arguments made during the prosecution of a patent application are given the same weight as claim amendments.") Disclaimers to certain claim meanings may arise from statements made during the prosecution of the patent-at-issue, parent applications, and other related applications. See, e.g., Wang Labs., Inc. v. Am. Online, Inc.,197 F.3d 1377, 1384 (Fed. Cir. 1999); Jonsson v. Stanley 5

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 9 of 44

Works, 903 F.2d 812, 818 (Fed. Cir. 1990). Moreover, statements made in the prosecution history are relevant to the construction of all claims of the patents or related patents. See Digital Biometrics, Inc. v. Identix, Inc., 149 F.3d 1335, 1347 (Fed. Cir. 1998) ("Absent qualifying language in the remarks, arguments made to obtain the allowance of one claim are relevant to interpreting other claims in the same patent."). See also Gillespie v. Dywidag Systems Int'l, USA, 501 F.3d 1285, 1291 (Fed. Cir. 2007) (quoting Phillips, 415 F.3d at 1314) ("[C]laim terms are normally used consistently throughout the patent."). Thus, the law regarding prosecution history disclaimers dictates that the "[c]laims may not be construed one way in order to obtain their allowance and in a different way against accused infringers." Spectrum Int'l, Inc. v. Sterilite Corp., 164 F.3d 1372, 1379 (Fed. Cir. 1998) (internal citations omitted). See also Hockerson-Halberstadt, Inc. v. Avia Group Int'l, Inc., 222 F.3d 951, 957 (Fed. Cir. 2000) (A patentee is not entitled to "a mulligan that would erase from the prosecution history the inventor's disavowal of a particular aspect of a claim term's meaning.") Moreover, when certain claim matter is disclaimed in a parent application, it has also been disclaimed in continuation patents. See Elkay Mfg., 192 F.3d at 980 ("When multiple patents derive from the same initial application, the prosecution history regarding a claim limitation in any patent that has issued applies with equal force to subsequently issued patents that contain the same claim limitation." (quoting Jonsson, 903 F.2d at 817-818)). Further, "[t]he prosecution history modifies the scope of a claim term if `the patentee distinguished that term from prior art on the basis of a particular embodiment, expressly disclaimed subject matter, or described a particular embodiment as important to the invention.'" Sentry Protection Products, Inc. v. Eagle Mfg. Co., 400 F.3d 910, 915 (Fed. Cir. 2005) (quoting CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366-1367 (Fed. Cir. 2002)). Indeed, where the patentees use "language of requirement, not preference" to limit the scope of the invention, they cannot later claim that such limitation is optional and refers only to a preferred embodiment. Andersen Corp. v. Fiber Composites, LLC, 474 F.3d 1361, 1372 (Fed. Cir. 2007) (emphasis added). See also Alloc, Inc. v. Int'l Trade Comm'n, 342 F.3d 1361 (Fed. Cir. 2003) 6

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 10 of 44

(interpreting claims as requiring what the specification and prosecution history disclosed as a key feature of the claimed invention); Pharmacia & Upjohn Co. v. Mylan Pharmaceuticals, Inc., 170 F.3d 1373, 1375 (Fed. Cir. 1999) (where patentee "emphasized the `criticality'" of a limitation, it clearly and unmistakably surrendered any embodiments that did not contain said limitation.).2 Extrinsic evidence may also be probative in the claim construction analysis. See Phillips, 415 F.3d at 1317. For example, statements made during the prosecution of foreign counterparts often demonstrate the patentee's understanding of claim terms, and how those of ordinary skill in the art understand the meaning of the claims. See Liposome Co., Inc. v. Vestar, Inc., Civ. A. No. 92-332-RRM, 1994 WL 738952, at *14 (D. Del. Dec. 20, 1994) (patentee's statements before the European Patent Office were relevant evidence of how one of skill in the art would understand the claim terms and how patentee understood the claims outside the litigation); see also Tanabe Seiyaku Co., Ltd. v. United States Int'l Trade Comm'n, 109 F.3d 726, 733 (Fed. Cir. 1997). Finally, expert testimony may serve an important role in claim construction proceedings, and is particularly appropriate when dealing with complex technologies. Markman v. Westview Instruments, Inc., 517 U.S. 370, 387 (1996). See also Conoco, Int'l, 460 F.3d at 1362); see also Ortho-McNeil Pharm., Inc. v. Caraco Pharm. Labs. Ltd., 476 F.3d 1321, 1328 (Fed. Cir. 2007).

2

The rule that claims will be modified by features described as "important" is not absolute. See MBO Labs., Inc. v. Becton, Dickinson & Co., 474 F.3d 1323, 1330 (Fed. Cir. 2007)(declining to read the feature of "immediate needle safety upon removal" into two reissue claims, since these reissue claims, unlike the original claims, did not recite the term "immediately") . Rather, there must be "`a textual reference in the actual language of the claim with which to associate a proffered claim construction.'" See id. at 1331 (quoting Johnson Worldwide Assocs., Inc. v. Zebco Corp., 175 F.3d 985, 990 (Fed. Cir. 1999)). In the present case, the essential feature of "framework substitutions" is a required feature of each and every asserted claim because each and every asserted claim contains the textual reference "humanized immunoglobulin." See, e.g., Nikon Corp. v. ASM Lithography B.V., 308 F. Supp. 2d 1039, 1100 (N.D. Cal. 2004) (internal citations omitted) (emphasis added) ("The Federal Circuit has long noted that `claim language [must be] limited based on a feature that was described as essential to the invention.' Where specification language identifies an essential claim feature, and where the embodiments uniformly disclose that feature, the feature proves a required limitation of all the relevant claims.").

7

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 11 of 44

II.

BACKGROUND OF ANTIBODY TECHNOLOGY A. Antibody Structure And Function Antibodies, also referred to as immunoglobulins (Igs), are Y-shaped proteins of

the immune system that identify and neutralize foreign substances, known as antigens. See Fig. 1. Antibodies are composed of two domains: a variable domain that recognizes specific antigens and a constant domain that plays a role in modulating immune cell activity. Clark Decl. at ¶¶ 25-27; Foote Decl. at ¶¶ 18-19. The variable and constant domains of the antibody are composed of light and heavy protein chains. Within each variable domain there are three regions designated as the complementarityFigure 1 Immunoglobulin Structure

determining regions ("CDRs"). The CDRs are the parts of the antibody that recognize and bind to specific antigens. Clark Decl. at ¶ 26; Foote Decl. at ¶ 26. The CDRs are also called hypervariable regions because there is great variability of the CDR regions between different antibodies ­ this variability results in antibody specificity for a particular antigen. Foote Decl. at ¶ 26. The remaining sections of the variable domains, called framework regions, act as a scaffold that supports the CDRs in the proper position within the three-dimensional structure of the antibody. Unlike the hypervariable CDRs, the framework regions are highly conserved across different antibodies, and even across antibodies of different species. Foote Decl. at ¶ 26. B. Prior Art Antibody Humanization The therapeutic potential of antibodies has long been understood. The specificity of an antibody for its antigen is an ideal means by which to specifically target therapeutic treatment while avoiding harm to healthy parts of the body. Foote Decl. at ¶ 17. Implementation of antibody therapy, however, was initially problematic. Raising antibodies in animal species, particularly in mice and rats, is a routine endeavor. However, these animal antibodies generally cannot be used therapeutically in humans because they are often immunogenic, that is, they 8

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 12 of 44

cause a severe immune reaction that can lead to sickness and death. Clark Decl. at ¶¶ 23-25; Foote Decl. at ¶ 21. Those of skill in the art, therefore, sought a way to combine human and animal antibodies into a single antibody so as to preserve antigen specificity and avoid immunogenicity. In 1986, the first "humanized antibody" was created by Gregory Winter and its invention published by Jones et al. See Ex. 1 ("Jones")3; Ex. 2 ("Winter").4 This humanized antibody was created by transplanting the CDRs of a mouse antibody into the framework region of a human antibody. See Fig. 2. Winter's method is now generally referred to as "CDR grafting." See Clark Decl. at ¶ 26; Foote Decl. at ¶ 27.
Human Variable Region

Humanized Antibody Variable Region CDRs

Mouse CDRs are transferred into human framework Mouse Variable Region Figure 2 Winter Method

Winter's CDR graft resulted in a humanized antibody with a similar ability to bind antigen ("binding affinity") as the original mouse antibody. In addition, Winter's humanized antibody had reduced immunogenicity as compared to the mouse antibody. Clark Decl. at ¶ 26; Foote Decl. at ¶ 27.

3

Peter T. Jones et al., Replacing The Complementarity-Determining Regions In A Human Antibody With Those From A Mouse, 321 Nature 522 (1986). EP 0 239 400 B1, issued August 3, 1994, application published September 30, 1987.

4

9

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 13 of 44

Subsequent prior art methods improved upon Winter's basic CDR-grafting method. In 1988, Riechmann et al. published the results of their research, where they found that grafting a CDR from a rat antibody directed against human lymphocytes onto a human framework did not retain the specificity of the rat antibody. Ex. 3 ("Riechmann").5 Riechmann, therefore, made additional changes in the framework region of the humanized antibody by substituting a rat framework amino acid for a human framework amino acid. See Fig. 3.
Human Variable Region

Humanized Variable Region after CDR graft CDRs

Humanized Variable Region with Framework Substitutions

Rat CDRs

Rat Framework Amino Acids

Rat Variable Region Figure 3 Riechmann Method

Riechmann did so because the amino acid difference between the human and rat amino acids at the particular framework position impacted the three-dimensional structure of the antibody, and thus was likely to affect the binding of an antigen to its antibody. Riechmann concluded that alterations in the framework regions of a human antibody to be more like the rat antibody ("framework substitutions") could enhance binding affinity of the resulting humanized antibody. Clark Decl. at ¶ 27; Foote Decl. at ¶ 28. Riechmann, throughout his published reference reporting this humanized antibody with framework substitutions, relied upon a particular definition of CDR as set forth by the scientist Kabat (hereinafter referred to as a "Kabat CDR"). Ex. 3.

5

Lutz Riechmann et al., Reshaping Human Antibodies For Therapy, 332 Nature 323 (1988).

10

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 14 of 44

III.

THE PATENTS-IN-SUIT The patents-in-suit are directed to specific humanized antibodies and methods for

producing the claimed antibodies. See Ex. 4 (`761 patent); Ex. 5 (`762 patent); Ex. 6 (`370 patent). Each of the three patents-in-suit is a continuation of the same parent application, U.S. Application 634,278, which issued as U.S. Patent 5,530,101 ("the `101 patent"). 6 Accordingly, the specification of each patent-in-suit is essentially the same.7 PDL purports that the novelty of its claimed humanized antibodies lies in two key limitations of the claims of the patents-in-suit: (1) transfer of framework amino acids that can contact the CDRs (i.e. "framework substitutions"); and (2) the use of a human acceptor framework highly homologous (at least 65%) to the mouse antibody. See Ex. 7 at ¶ 11. As shown in detail below, these key limitations of PDL's invention were already known in the art. A. PDL's CDR And Framework Substitution Limitations The purported novelty of PDL's framework substitutions of amino acids that contact the CDRs was already present in the prior art. For example, the prior art Riechmann antibody had framework substitutions of amino acids that "pack against" (i.e., contact) the CDRs. Ex. 3 at 326. Thus, during prosecution, PDL repeatedly distinguished its own purported invention from the Riechmann reference by disclaiming the framework substitutions made by Riechmann as part of PDL's invention. PDL did so by re-defining the term "CDR" as used in the patents so that it was different from the "Kabat CDR" definition used by Riechmann. See infra at Section IV.A. Specifically, PDL characterized Riechmann's humanization method as

6

The `761 patent, the `762 patent, and the `370 patent, are continuations of the same parent application, U.S. Application 07/634,278 ("the `278 application"), which issued as the `101 patent. The `278 application is a continuation-in-part of U.S. Application 07/590,274 ("the `274 application") and U.S. Application 07/310,252 ("the `252 application"), both abandoned. The `252 application is a continuation-in-part of another abandoned application, U.S. Application 07/290,975 ("the `975 application"). Therefore, in addition to the three applications that issued as the patents-in-suit, the file history of each of the patents-in-suit includes the `278 application; the `274 application; the `252 application; and the `975 application. The patent specification of the `370 patent varies to some extent from the specification of the `101 patent. The Patent Office initially classified the `370 patent as a continuation-in-part, but apparently later re-categorized it as a continuation patent. Alexion reserves its right to argue that the `370 patent is continuation-in-part, and not a continuation, of the `101 patent.

7

11

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 15 of 44

"transfer[ing] the Kabat CDRs plus the one Chothia CDR that contains extra amino acids (H1) to a pre-determined framework." See, e.g,, Ex. 7 at ¶ 5. See also Fig. 4 below. By doing so, PDL effectively re-characterized Riechmann's framework substitutions as not substitutions at all, but as transfer of a Kabat CDR plus a Chothia CDR H1 region. In contrast, as will be shown in detail below, PDL described its own inventions as encompassing framework substitutions outside of the Kabat CDRs and Chothia CDR H1. Thus, by distinguishing PDL's purported invention from Riechmann, PDL disclaimed humanized immunoglobulins with framework substitutions based on a Kabat CDR definition and instead limited its claimed invention to humanized immunoglobulins containing the specific definition of "CDR" adopted by PDL, i.e., the Kabat CDR plus Chothia CDR H1. Clark Decl. at ¶¶ 31-34; Foote Decl. at ¶¶ 36, 52.

PDL's "CDR"

PDL's "CDR"

PDL's "CDR"

1

2

...

24 25 26 27 28 29 30 31 32 33 34 35 36 37 . . . 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 . . . 93 94 95 96 97 98 99 00 01 02 03 04

Chothia CDR H1 - framework region amino acids - CDR amino acids as defined by Chothia only - CDR amino acids as defined by Kabat only - overlap of Kabat and Chothia hypervariable regions

Figure 4 Amino Acids of Heavy Chain Variable Region

As will be clear from PDL's opening claim construction brief, it now seeks to reclaim the definition of CDR ­ the Kabat CDR ­ that was very clearly disclaimed during prosecution of the patents-in-suit. PDL does so for one simple reason ­ its foreign counterpart patents were partially revoked in Europe when PDL defined CDR as Kabat plus Chothia. Ex. 8 at 15-16. PDL, however, cannot reclaim what it explicitly gave up during prosecution. Accordingly, the term "CDR" must be given the definition PDL adopted during prosecution, i.e., Kabat CDR plus Chothia CDR.

12

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 16 of 44

B.

PDL's Homology Limitation With respect to the homology limitation, the creation of humanized antibodies

from "highly homologous" human framework and mouse antibodies as claimed by PDL was already known in the art as of the time of PDL's invention.8 Indeed, one of the first published humanized antibodies made by Winter had a percent homology of human framework to mouse antibody of over 65%. Clark Decl. at ¶ 46; Foote Decl. at ¶¶ 42, 55. Moreover, PDL's patent does not demonstrate that its claimed humanized immunoglobulins with simply 65% homology are at all functional. In contrast, the only humanized immunoglobulin disclosed in PDL's patents-in-suit that are functional (i.e. retain the specificity of the mouse antibody) have not only 65% homology of human and mouse frameworks, but also contain framework substitutions outside of the Kabat CDRs plus Chothia CDR H1. See infra at Section IV.C. Thus, PDL's invention embraces humanized immunoglobulins with the claimed 65% homology only if those humanized immunoglobulins also contain framework substitutions. Clark Decl. at ¶ 46; Foote Decl. at ¶¶ 43, 45, 47-49. Indeed, throughout the prosecution of the patents-in-suit, in order to overcome prior art, including the Winter reference, PDL repeatedly represented to the Patent Office that framework substitutions were an "essential" and "critical" element of its invention. See infra at Section IV.C.2. As will be clear from PDL's opening brief, PDL now seeks to define its claims such that they include humanized immunoglobulins with 65% homology but no framework substitutions. PDL's motivation is clear ­ Alexion's accused antibody does not have the framework substitutions that PDL told the Patent Office were essential to its invention. In seeking to ensnare Alexion's antibody, PDL attempts to redefine its claims as including the CDR-grafting method set forth in the prior art by Winter. In order to even possibly avoid anticipation by the Winter patent application, PDL's claimed humanized immunoglobulins
8

The specification of the patents discloses a percent "homology" and the claims of the patents recite a percent "identity." As used herein, "homology" and "identity" are interchangeable.

13

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 17 of 44

cannot be limited only by 65% homology. Instead, as PDL argues during prosecution, the claims drafted to humanized antibodies with 65% homology may possibly avoid anticipation by Winter only if those humanized immunoglobulins also contain framework substitutions outside of the Kabat CDRs plus Chothia CDR H1. See infra at Section IV.C.2. The two key limitations of the patents-in-suit, "CDR" and "framework substitution" drive the primary dispute between PDL and Alexion regarding claim construction of PDL's claimed humanized immunoglobulins. As will be set forth in detail below, Alexion's construction of the claim terms at issue recognizes the true scope of PDL's claims within the context of PDL's own statements regarding the criticality of certain claim features vis-à-vis the prior art. Accordingly, the Court should construe PDL's claim terms such that the term "CDR" is construed consistently with PDL's prosecution history disclaimers to mean "Kabat CDR plus Chothia CDR." The Court should also construe the claim term "humanized immunoglobulin" to encompass the framework substitutions that PDL characterized as an "essential" element of its invention. IV. ALEXION'S PROPOSED CONSTRUCTIONS A. "Complementarity Determining Regions" ("CDRs") Alexion proposes that the claim limitation "complementarity determining regions (CDRs)" in the claims of the patents-in-suit be construed to have the same meaning that PDL disclosed in its specification and used to overcome the prior art during prosecution ­ "a hypervariable region as defined by Kabat9 together with a hypervariable region as defined by Chothia10." PDL, in contrast, seeks to reclaim a definition of CDR that it very clearly disclaimed during prosecution, i.e., a CDR defined only by reference to Kabat.

9

See Elvin A. Kabat et al., Sequences of Proteins of Immunological Interest, U.S. Dept. of Health and Human Services, Public Health Service, National Institutes of Health (1983). Ex. 9. Cyrus Chothia & Arthur M. Lesk, Canonical Structures for the Hypervariable Regions of Immunoglobulins, 196 J. Mol. Biol. 901 (1987). Ex. 10.

10

14

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 18 of 44

1.

The Patent Specification Supports Alexion's Construction Of CDR

The specification of the patents-in-suit recognizes that there is no single definition of "CDR" to one of skill in the art. Clark Decl. at ¶¶ 29-30; Foote Decl. at ¶¶ 31, 33. Accordingly, the specification teaches that the term "CDR" as used in the patent may be found in the art in the teaching of Kabat et al., Sequence of Proteins of Immunological Interest, US Department of Health and Human Services (1983). Ex. 5 at Col. 11:39-41. The specification also teaches that CDRs may be defined by reference to an article authored by Chothia & Lesk: "Chothia and Lesk (op. cit.) define the CDRs differently from Kabat et al. (op. cit.). Notably, CDR1 is defined as including residues 26-32." Ex. 5 at Col. 15:37-39. One of ordinary skill in the art at the time of the invention would have been familiar with both the Kabat and Chothia references, and would have known that there is significant overlap in the CDRs as defined by Kabat and those as defined by Chothia. Foote Decl. at ¶ 31. See also Fig. 4, supra. The patent specification adopts the definition of CDR as that of Kabat together with that of Chothia: The chains all exhibit the same general structure of relatively conserved framework regions joined by three hypervariable regions, also called Complementarity Determining Regions or CDR's (see, `Sequence of Proteins of Immunological Interest,' Kabat, E. et al., US Department of Health and Human Services, (1983); and Chothia and Lesk, J. Mol. Biol., 196, 901-917 (1987), which are incorporated herein by reference). Ex. 5 at Col. 23:8-15 (emphasis added). See also Clark Decl. at ¶ 30; Foote Decl. at ¶¶ 31-34. In the prosecution of its related European patents, PDL took the position that the identical paragraph in the specification of the European patent supported a construction of the term CDRs as being Kabat together with Chothia ­ the exact opposite construction that it proposes here. See Ex. 11 at 375; Clark Decl. at ¶¶ 36-40; Foote Decl. at ¶ 37. PDL represented to the European Patent Office that this passage "is also found in the first application to which the Patent claims priority, USSN 290,975," which is also a priority application to the patents-in-suit. Based on this passage in the specification, PDL represented that "a skilled person approaching 15

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 19 of 44

the original [European] specification . . . is taught by the specification how to read the term CDRs," i.e., as Kabat taken together with Chothia. Ex. 11 at 376. PDL emphasized the significance of the conjunction "and" in the corresponding passage of the European specification as support that CDR meant Kabat CDR taken together with the Chothia definition of CDR. Ex. 11 at 376 ("[I]t is not correct to ignore how skilled persons would actually understand the teachings of the patent, or to construe the word `and' to mean the quite different word `or' ... [I]t is not legally appropriate to ignore the passage which equates hypervariable regions and CDRs, or the simple fact that hypervariable regions (and hence CDRs) cannot be interpreted without including a structurally-based approach (Chothia).") Moreover, PDL specifically incorporated the proceedings in the EPO as part of the U.S. patent filing, and represented to the Patent Office that the inventions of the European and U.S. patents were the same: Applicants would like to direct the [United States patent] Examiner's attention to the "Patentee's Observations on the Arguments of the Opponents," which was filed in the previously discussed Opposition at the European Patent Office in the corresponding EP patent . . . The Observations . . . restate[] the contributions underlying the present invention. Ex. 12 at 14. Just as PDL concluded that CDR meant Kabat plus Chothia CDR regions, one of skill in the art reading the patent specification of the patents-in-suit at the time of invention would have concluded that the term CDR, as used in the claims, means the Kabat CDR regions plus the Chothia CDR regions. Clark Decl. at ¶¶ 31, 35, 42; Foote Decl. at ¶¶ 34, 39. 2. The File Histories Support Alexion's Definition Of CDR

The file histories of the patents-in-suit also support Alexion's proposed definition of the claim term CDR as including the Chothia H1 regions. In particular, PDL repeatedly represented to the Patent Office that the claim term CDR included both the Kabat CDRs plus CDRs as defined by Chothia, including the Chothia H1 region. Clark Decl. at ¶¶ 31-35; Foote 16

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 20 of 44

Decl. at ¶¶ 35-36. PDL did so to get over the prior art and allow issuance of the claims. PDL, therefore, cannot regain through claim construction that which it disclaimed during prosecution. Spectrum Int'l, 164 F.3d at 1379 (internal citations omitted) ("Claims may not be construed one way in order to obtain their allowance and in a different way against accused infringers."). For example, during the prosecution of the `101 patent, the parent patent of the patents-in-suit, PDL argued that "CDR" means Kabat CDRs plus Chothia CDR H1. PDL adopted this definition in order to distinguish the prior art Riechmann reference. In contrast to its own alleged invention, PDL stated: The position 27 and 30 modifications of Riechmann et al. were in fact within the Chothia-Lesk H1 CDR, rather than outside this CDR. Moreover, Riechmann et al. provided no suggestion or guidance as to which if any non-CDR positions to substitute, or which amino acids should be put in those positions (see paragraph 26 of accompanying Queen declaration). Ex. 13 at 17 (emphasis in original). Thus, in order to overcome a prior art rejection, PDL told the Patent Office that the Chothia H1 region should not be considered as part of the framework, but rather should be considered part of the CDR. 11 Clark Decl. at ¶ 31-33; Foote Decl. at ¶ 36. In support of its argument that Chothia H1 region was part of the CDR, PDL also submitted a declaration by the inventor of the patents-in-suit and PDL founder Cary L. Queen.12 Dr. Queen represented to the Patent Office that, in contrast to PDL's invention, Riechmann did not teach framework substitutions outside of the CDR: [T]he final humanized antibody of Riechmann et al. contains the Kabat CDRs and the Chothia CDR H1 from the mouse antibody, but no other mouse amino acids. Therefore the method of humanization actually utilized in Riechmann et al. can be stated as: transfer the Kabat CDRs plus the one Chothia CDR that contains extra amino acids (H1) to a pre-determined human framework.

11

12

PDL is likely to argue that these statements were made only in the context of certain claims of the patent that explicitly require framework substitutions. A claim term, however, must be used consistently, and cannot have different meanings in different claims. See Digital Biometrics, 149 F.3d at 1347 ("Absent qualifying language in the remarks, arguments made to obtain the allowance of one claim are relevant to interpreting other claims in the same patent."). See also Gillespie, 501 F.3d at 1291 ("claim terms are normally used consistently throughout the patent" (quoting Phillips, 415 F.3d at 1314)). Thus, PDL's prosecution disclaimer of the Kabatonly definition of CDR applies to all claims of the patents-in-suit. This declaration was submitted not only in support of the parent application, but was specifically resubmitted to the Patent Office in support of the applications that issued as the `761 and the `762 patents.

17

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 21 of 44

Ex. 7 at ¶ 9 (emphasis in original). See also Clark Decl. at ¶ 32; Foote Decl. at ¶ 36. During the Appeal of this application to the Board Of Appeals, PDL again defined its own invention as requiring framework substitutions outside of the CDR including both Kabat and Chothia CDRs: Riechmann et al. made changes only within the first heavy chain CDR loop (Chothia CDR H1), and did not modify or replace any framework residues outside the loop. . . Riechmann et al. shows only the CDR loop itself, and does not show any so-called `packing' interactions between the loop and framework residues outside the loop. Thus, Riechmann et al. does not teach the importance, as Appellants have invented, of changing framework residues outside the CDR loops in order to maintain the shape of the CDRs and thus their binding affinity. Ex. 14 at 19 (underlining in original, italics added). In doing so, PDL disclaimed its own use of framework substitutions within the Chothia H1 region and thereby defined the CDR of its invention as including both Kabat CDRs and the Chothia H1 region. PDL made these same representations and disclaimers repeatedly during the prosecution of the patents-in-suit. For example, during the prosecution of the `761 and `762 patents, PDL re-submitted the Queen Declaration distinguishing the PDL invention from Riechmann and adopting the CDR definition of Kabat CDR plus Chothia CDR H1. In addition, the file histories of the patents-in-suit contain additional representations that the invention was directed to framework substitutions outside of the framework regions as defined by Kabat and Chothia. See, e.g., Ex. 15 at 6-7 (emphasis added) ("Riechmann did not teach the importance of changing framework residues outside of the CDR loops . . . as is clearly provided in Applicants' presently pending claims."); Ex. 12 at 15 ("[T]he contribution [of the invention] involves one or more amino acid substitutions in framework amino acid positions from the nonhuman donor antibody, which positions are outside the CDRs as defined by Kabat and Chothia."). See also Clark Decl. at ¶ 34; Foote Decl. at ¶¶ 35-36. Thus, the file histories of the patents-in-suit and the parent patent applications support Alexion's proposed construction of CDR as including Kabat CDRs and Chothia CDR H1. 18

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 22 of 44

3.

One Of Ordinary Skill In The Art Would Understand The Claim Term To Include The Hypervariable Regions As Defined By Chothia

One of skill in the art, reading the patents-in-suit and their associated file histories, would have understood the claim term "CDR" to encompass both the Kabat and Chothia definitions. Clark Decl. at ¶¶ 31, 35, 42; Foote Decl. at ¶¶ 34, 39. Indeed, PDL has itself admitted that one of skill in the art would understand the claim term CDR to mean the Kabat definition plus the Chothia definition. PDL has several European patent applications that claim priority from the patents-in-suit. At least two of these patents have been involved in European Opposition proceedings.13 Notably, in those proceedings ­ despite the substantially similar disclosure of the European and U.S. patent specifications and the near identity of the pending EP and U.S. claims ­ PDL took the position that CDR should be construed as Kabat CDRs taken together with the Chothia regions, the exact opposite construction as it proposes here. See, e.g., Ex. 11 at 374-75 (PDL "define[s] CDRs in terms of both a sequence-based approach (Kabat) and a structurally-based approach (Chothia) . . . This definition is consistent with how the CDR terminology was and is used by skilled scientists."). In support of its construction, PDL told the European Patent Office ("EPO") that "eminent scientists accept [PDL's] definition of CDRs [as Kabat plus Chothia CDRs]" Ex. 11 at 376. See also Clark Decl. at ¶¶ 39-40; Foote Decl. at ¶ 37. For example, PDL represented to the EPO that contemporaneous literature reflected the understanding of one of ordinary skill in the art and supported its definition of "CDR" as hypervariable regions defined by Kabat and Chothia. PDL stated: Cheetham . . . explicitly described CDRs as being defined by both Kabat and Chothia: `Existing structural data shows that in CDR1 of the antibody heavy chain, which extends from residue 31 to 35 by sequence (Kabat et al., 1987) and from residue 26-32 in structural terms (Chothia and Lesk 1987).

13

EP 0 451 216 issued from application PCT/US89/05857, published as WO 90/07861. The application claims priority to two separate U.S. applications: U.S App. 07/290/975, filed on December 28, 1988, and U.S. App. 07/310,252, filed on February 13, 1989. EP 682 040, is a divisional of parent patent EP 0 451 216, that claims priority dates on two U.S. patents ­ U.S. App. 07/290975 and U.S. App. 07/310252.

19

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 23 of 44

Ex. 23 at 55 (citing James Cheetham, Reshaping the antibody combining site by CDR replacement--tailoring or tinkering to fit?, 2 Protein Eng. 170 (1988)). See also Clark Decl. at ¶ 38; Foote Decl. at ¶ 37. In support of its construction of "CDR" in the European Opposition, PDL submitted the declaration of no less than four experts who opined that, as used in PDL's patents, the term CDR meant Kabat and Chothia hypervariable regions. PDL further explained: Dr. Kornberg has no doubt as to the message conveyed by the definition [Kabat plus Chothia] adopted by [PDL] . . . Like Dr. Kornberg, Dr. Edmundson is very clear on the meaning of [PDL's] definition of CDRs that refers both to the work of Kabat and Chothia. Going even further, Dr. Edmundson [states:] `[T]he many crystallographic studies of antibodies [demonstrate] that the Kabat and Chothia H1 CDRs behave as a single structural unit, which extends into the solvent to make contact.' Dr. Edmundson concludes that he would therefore have considered `the composite Kabat and Chothia CDR H1 to be the appropriate unit to consider when humanizing antibodies' and indeed, `a key biophysical concept.' Dr. Michael Levitt also remarks that `like many scientists in the field, I often use the term CDR to refer to the loops L1-L3 and H1-H3 described by Chothia et al., as well as to the Kabat CDRs.' Dr. Poljak does not even find the matter worthy of comment, using the term `Kabat and Chothia CDRs' naturally. Other declarants for [PDL] also freely use the term `Chothia CDRs' or `Kabat and Chothia CDRs.' Ex. 11 at 376-377 (internal citations omitted, emphasis added). See also Clark Decl. at ¶ 39-40; Foote Decl. at ¶ 37. PDL's current reversal of position as to the construction of CDR to mean a CDR as defined by Kabat, rather than Kabat plus Chothia, is transparent. Its European counterparts to the patents-in-suit were partially revoked because the Kabat plus Chothia definition of CDR was considered to be "new matter" that rendered the applicable claims unpatentable. Ex. 8 at 15-19. PDL wants to avoid a similar fate here. PDL's repeated representations that CDR means Kabat plus Chothia, made to both the U.S. Patent Office and the EPO, however, render such an about face impossible. See Hockerson-Halberstadt, 222 F.3d at 957 (patentee is not entitled to "a 20

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 24 of 44

mulligan that would erase from the prosecution history the inventor's disavowal of a particular aspect of a claim term's meaning"). Thus, by PDL's own admissions, supported by its own experts, one of ordinary skill in the art would understand the term CDR as used in the patents-in-suit to mean a CDR as defined by Kabat plus the Chothia CDR H1 region. The expert declarations submitted in support of Alexion's claim constructions further demonstrate that Alexion's proposed construction of the claim term CDR as "a hypervariable region as defined by Kabat together with a hypervariable region as defined by Chothia" reflects the understanding of the claim terms by one of skill in the art. Clark Decl. at ¶¶ 31, 35, 42; Foote Decl. at ¶ 39. B. "Framework" Terms The claim terms "framework", "framework region" and "variable region frameworks" are used interchangeably in the claims of the patents-in-suit to mean "those portions of the variable region of an immunoglobulin that are not a CDR (where the term `CDR' is used as construed by the Court)" Although PDL and Alexion dispute the meaning of the claim term "CDR," they do not dispute that the term "framework" means whatever portion of the variable region is not a CDR. Foote Decl. at ¶ 26. Indeed, within the variable region of an immunoglobulin, a particular amino acid is part of the framework or it is part of a CDR; it cannot be both. Thus, the parties jointly ask the Court to construe the claim terms "framework," "framework region," and "variable region frameworks" to mean "those portions of the variable region of an immunoglobulin that are not a CDR (where the term `CDR' is used as construed by the Court)." Similarly, the parties jointly request that the Court construe the claim terms "heavy and light chain variable region frameworks" and "heavy and light chain frameworks" as interchangeable terms meaning "those portions of the heavy chain and light chain variable regions of an immunoglobulin that are not a CDR (where the term `CDR' is used as construed by the Court)." Alexion, as set forth above, seeks a definition of CDR that includes the term as 21

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 25 of 44

defined by both Kabat and Chothia. Thus, as used herein, Alexion uses the claim term "framework" to mean "those portions of the variable region of an immunoglobulin that are not a CDR, wherein CDR includes Kabat CDRs plus Chothia CDR H1." C. "Humanized Immunoglobulin" Alexion requests that the Court construe the term "humanized immunoglobulin" as set forth in the claims of patents-in-suit to mean "an immunoglobulin comprising: a human acceptor framework in which one or more amino acids have been replaced by the corresponding amino acid(s) of the donor immunoglobulin; and one or more CDRs (as defined herein) from the donor immunoglobulin." Alexion's construction of the term "humanized immunoglobulin" is compelled by the specification and file histories of the patents-in-suit and is necessary for PDL's invention to avoid the prior art. Clark Decl. at ¶¶ 46, 48-58; Foote Decl. at ¶¶ 41-52. In particular, PDL repeatedly emphasized the criticality of framework substitutions in its humanized immunoglobulins, and represented that such framework substitutions were essential to the functionality of the claimed invention. Ex. 5 at Col. 43:16-18 ("[W]e conclude that at least one of these mouse framework residues are essential for high affinity binding."). Such representations are dispositive and may drive claim construction. See Sentry Protection, 400 F.3d at 915 (quoting CCS Fitness, 288 F.3d at 1366-1367) (emphasis added) ("The prosecution history modifies the scope of a claim term if `the patentee distinguished that term from prior art on the basis of a particular embodiment, expressly disclaimed subject matter, or described a particular embodiment as important to the invention.'" ).14 Moreover, one of ordinary skill in the art would understand that PDL's claimed humanized immunoglobulin must include a framework substitution outside of the CDRs, even where that limitation is not explicitly set forth in the claims, because humanized antibodies
14

"The Federal Circuit has long noted that `claim language [must be] limited based on a feature that was described as essential to the invention.' See Sunrace Roots Enter. Co., Ltd. v. SRAM Corp., 336 F.3d 1298, 1305 (Fed. Cir. 2003). Where specification language identifies an essential claim feature, and where the embodiments uniformly disclose that feature, the feature proves a required limitation of all the relevant claims. See, e.g., ATD Corp. v. Lydall, Inc., 159 F.3d 534, 542 (Fed. Cir. 1998); Gentry Gallery, Inc. v. The Berkline Corp., 134 F.3d 1473, 1478-80 (Fed. Cir. 1998)." Nikon Corp. v. ASM Lithography B.V., 308 F.Supp.2d 1039, 1100 (N.D.Cal. 2004) (emphasis added).

22

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 26 of 44

without such framework substitutions were well known in the art prior to PDL's patent applications. Clark Decl. at ¶ 46; Foote Decl. at ¶¶ 41-43. 1. The Specification Of The Patents-In-Suit

The specification of the patents-in-suit makes clear that the humanized immunoglobulins of the invention are required to have framework substitutions outside of the CDRs. Indeed, the patent specification teaches that framework substitutions are critical and even "essential" for the creation of a functional humanized immunoglobulin. For example, the patent specification teaches that the prior art Winter CDRgrafting method has a "major problem" because it results in a large loss of affinity of the humanized antibody as compared to the mouse donor antibody. See Ex. 5 at Cols. 1:63-2:25. The purported purpose of the invention was to solve the prior art problems and create humanized antibodies with high specificity by (1) creating a humanized antibody with high affinity by starting with human and mouse sequences with high homology (i.e., 65-70% or more) in the framework regions, and (2) creating a humanized antibody with high affinity by making framework substitutions of the human acceptor framework with the corresponding donor amino acid. See Ex. 5 at Cols. 2:40-3:32. The patents-in-suit purport to encompass humanized immunoglobulin compositions as well as methods to make the claimed compositions. Humanized antibodies with a 65% homology between the human and mouse framework regions, however, were already known in the art at the time of PDL's filing of its patent application. See, e.g., Ex. 1; Ex. 2; Ex. 16 ("Verhoeyen").15 Indeed, one of the first humanized antibodies reported in the art by Winter meets the exact homology limitation that PDL purports to be its invention. Clark Decl. at ¶ 46; Foote Decl. at ¶ 42. Winter's humanized antibody ­ like PDL's humanized immunoglobulins described in its patent specification ­ is a humanized antibody wherein the framework regions of the human and mouse antibodies used to make the humanized antibody are at least 65% homologous (i.e., have 65% percent identity).
15

Martine Verhoeyen et al., Reshaping Human Antibodies: Grafting an Antilysozyme Activity, 239 Science 1534 (1988).

23

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 27 of 44

See Clark Decl. at ¶ 46; Foote Decl. at ¶ 42. Winter's humanized antibody, like the claims of PDL's patents, also has at least a 65% homology between the humanized antibody and the donor antibody. See Clark Decl. at ¶ 46; Foote Decl. at ¶ 42. Thus, despite the overbroad and generalized description of the first purported method of PDL's invention, such claims could not survive over the prior art. Accordingly, the patent specification makes clear that the second method claimed in PDL's patents ­ framework substitution of amino acids in the human framework region ­ was the crux of PDL's purported invention. Clark Decl. at ¶¶ 48-49; Foote Decl. at ¶¶ 41-45. PDL taught that the prior art humanized antibodies used only the framework of the human acceptor immunoglobulin, without any framework substitutions. PDL also taught that such CDR-grafted antibodies did not always work. Ex. 5 at Col. 2:10-25. In contrast, PDL's purported invention was that framework substitutions increased the affinity of the resulting humanized antibody: Most humanized immunoglubulins that have been previously described . . . have comprised a framework that is identical to the to the framework of a particular human immunoglobulin chain, the acceptor, and three CDR's from a non-human donor immunoglobulin chain . . . The present invention includes criteria by which a limited number of amino acids in the framework of a humanized immunoglobulin chain are chosen to be the same as the amino acids at those positions in the donor rather than in the acceptor, in order to increase the affinity of an antibody comprising the humanized immunoglobulin chain. Ex. 5 at Col. 12:29-43 (emphasis added). See Clark Decl. at ¶ 48; Foote Decl. at ¶¶ 44-45. PDL's alleged invention is supported by the examples of the patent specification, wherein each functional humanized immunoglobulin produced had, at a minimum, four framework substitutions. Ex. 5 at Examples 1-9. The Examples of the patent make clear that the patent only describes and claims functional humanized immunoglobulins that contain framework substitutions. Indeed, the patent explicitly taught that humanized antibodies created by CDR-grafting (without framework substitutions) did not work. See e.g. Ex. 5 at Col. 43:6-18. Clark Decl. at ¶ 49; Foote Decl. at ¶¶ 44-45.

24

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 28 of 44

For instance, Example 2 of the patent describes the creation of two humanized anti-Tac antibodies, one of which contained framework substitutions (the "PDL humanized antibody"), and one which was a "CDR-only humanized antibody" that did not have any framework substitutions. See Ex. 5 at Cols. 41:35-43:17. Notably the "CDR-only humanized antibody" had a homology between the human acceptor framework and mouse donor framework of over 65%, as well as over 65% homology of humanized framework to donor framework. See Foote Decl. at ¶ 45. Example 2 demonstrates that the humanized immunoglobulin with framework substitutions retained binding affinity (i.e. was functional), while the "CDR-only humanized immunoglobulin" did not. Ex. 5 at Col. 42:54-59 ("[S]taining by the CDR-only antibody was indistinguishable from the negative control antibody...the CDR-only humanized antibody does not detectably bind the [target antigen]."). See also Clark Decl. at ¶¶ 49; Foote Decl. at ¶¶ 44-45. The patent thus teaches that the framework substitutions are essential to maintain the binding affinity of the mouse antibody: Because the sequences of the PDL and CDR humanized anti-Tac antibodies differ only at positions where mouse framework residues . . . were used in the PDL molecule, we conclude that at least one of these mouse framework residues are essential for high affinity binding. Ex. 5 at Col. 43:13-17 (emphasis added). This "essential" component of the humanized immunoglobulins of the invention, therefore, is required in each and every claim. See Sentry Protection, 400 F.3d at 915 (internal citation omitted) (emphasis added) ("The prosecution history modifies the scope of a claim term if `the patentee . . .described a particular embodiment as important to the invention.'"). The remainder of the patent supports PDL's conclusion that framework substitutions are essential for the functionality of the claimed humanized immunoglobulins. No other examples of CDR-grafted antibodies ­ other than the failed CDR-only humanized antibody of Example 2 ­ are set forth in the patent. Indeed, the patent describes at least six other humanized immunoglobulins, each of which contains one or more framework substitutions. See, 25

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 29 of 44

e.g., Ex. 5 at Examples 1, 3. Accordingly, PDL's patents-in-suit fail to teach any humanized antibodies with 65% homology but no framework substitutions. Foote Decl. at ¶¶ 44-45. Indeed, the patent specification demonstrates the "major problem" of the prior art that PDL explicitly set forth: that CDR-only humanized antibodies ­ even those with the claimed 65% homology between acceptor and donor antibodies or humanized and donor antibodies ­ do not maintain binding affinity for the antigen. Thus, the patent specification supports Alexion's proposed construction of the claimed "humanized immunoglobulin" of the inventions as "an immunoglobulin comprising: a human acceptor framework in which one or more amino acids have been replaced by the corresponding amino acid(s) of the donor immunoglobulin; and one or more CDRs (as defined herein) from the donor immunoglobulin." Clark Decl. at ¶¶ 47-49; Foote Decl. at ¶¶ 44-45. 2. The File Histories Of The Patents-In-Suit

The file histories of each of the patents-in-suit, as well as the file history of the parent `101 patent, further support Alexion's proposed claim construction of the claim term "humanized immunoglobulin" by reiterating the conclusion of the specification that framework substitutions are essential for functional humanized immunoglobulins with high binding affinity. For example, during the prosecution of the parent applications of the patents-insuit, PDL represented to the Patent Office that framework substitutions of the human acceptor immunoglobulin were "essential" to its invention because without such framework substitutions the binding affinity of the donor could not be maintained in the humanized immunoglobulin. See Ex. 17 at ¶ 7 (emphasis added) ("[W]e conclude that at least one of [the] mouse framework residues are essential for high affinity binding."). See also Clark Decl. at ¶ 51; Foote Decl. at ¶ 47. PDL further stated the criticality of framework substitutions: "Applicants here have defined essential structural rules . . . for the method of Winter to be used successfully by one of ordinary skill in the art to construct a functional human-like immunoglobulins to the human IL-2 receptor." Ex. 18 at 17 (emphasis added). See also Clark Decl. at ¶ 52; Foote Decl. 26

Case 1:07-cv-00156-JJF

Document 96

Filed 05/14/2008

Page 30 of 44

at ¶ 50. Based on PDL's repeated representations, the Patent Office examiner understood the criticality of such framework substitutions, and issued the claims over the Winter reference because of the required framework substitutions: "The declaration of the applicant establishes that retention of the residues immediately adjacent to the CDR regions of the murine source immunoglobulin are also required to ensure that the antigen binding function is transferred." Ex. 19 at 2 (emphasis added). See also Clark Decl. at ¶ 53; Foote Decl. at ¶ 51. During the prosecution of the parent `101 patent, PDL described its invention16 as requiring framework substitutions: The claimed invention is directed to humanized immunoglobulins, inter alia, having: (1) light and heavy chain CDRs which correspond to (that is, are from) a donor immunoglobulin, (2) variable region frameworks which correspond to human acceptor immunoglobulin framework sequences, and (3) at least one amino acid position in the