Hufton et al.v.Hua et al.

Patent Trial and Appeal BoardDec 30, 2013

11593957 (P.T.A.B. Dec. 30, 2013)

Boxinterferences@uspto.gov Paper No. 563 571-272-4683 Filed: December 30, 2013 UNITED STATES PATENT AND TRADEMARK OFFICE ________________ BEFORE THE PATENT TRIAL AND APPEAL BOARD ________________ Patent Interference No. 105,809 (JGN) ADIMAB, LLC Junior Party (Patent No. 7,700,302) v. DYAX CORP. Senior Party (Application 12/625,337) ________________ JUDGMENT Bd.R. 127 Before RICHARD TORCZON, DEBORAH KATZ, and JOHN G. NEW, Administrative Patent Judges NEW, Administrative Patent Judge. In view of the Board’s conclusion that Senior Party Dyax Corp. (“Dyax”) has demonstrated that US Patent No. 6,610,472 does not qualify as prior art against Dyax’s Application Ser. No. 12/625,3371, it is— ORDERED that judgment be entered against Junior Party Adimab, LLC (“Adimab”) for Count 1 (Paper No. 1); FURTHER ORDERED that claims 1-15 of Adimab’s involved U.S. Patent No. 7,700,302 be CANCELED2, 35 U.S.C. 135(a); and FURTHER ORDERED that a copy of this judgment be entered in the administrative records of the involved patent and application. NOTICE: “Any agreement or understanding between parties to an interference, including any collateral agreements referred to therein, made in connection with or in contemplation of the termination of the interference, shall be in writing and a true copy thereof filed in the Patent and Trademark Office before the termination of the interference as between the said parties to the agreement or understanding.” 35 U.S.C. § 135(c); see also Bd.R. 205 (settlement agreements). BRENDA HERSCHBACH JARRELL, FANGLI CHEN, ERIC J. MARANDETT and JUSTIN P. HUDDLESON, Choate, Hall & Stewart LLP, of Boston, Massachusetts. MICHAEL T. SIEKMAN, LAWRENCE M. GREEN and YAHUA J. CHEN, Wolf, Greenfield & Sacks, P.C., of Boston, Massachusetts 1 Paper No. 561 2 Paper No. 386 Boxinterferences@uspto.gov Paper No. 562 571-272-4683 Filed: December 30, 2013 UNITED STATES PATENT AND TRADEMARK OFFICE ________________ BEFORE THE PATENT TRIAL AND APPEAL BOARD ________________ Patent Interference No. 105,809 (JGN) ADIMAB, LLC Junior Party (Patent No. 7,700,302) v. DYAX CORP. Senior Party (Application 12/625,337) ________________ Decision – Dyax Response to Order to Show Cause – Bd.R. 41.125(c) Before RICHARD TORCZON, DEBORAH KATZ, and JOHN G. NEW, Administrative Patent Judges NEW, Administrative Patent Judge. Interference No. 105,809 2 The Board’s Order of November 12, 20121, the accompanying decision2, and the Board’s subsequent clarification3 required Senior Party Dyax Corp. (“Dyax”) to show why U.S. Patent No. 6,610,472 (the “’472 patent”)4 is not available as prior art against Dyax’s involved claims. Because Dyax has demonstrated by a preponderance of the evidence that the Dyax inventors conceived the invention recited in its claims prior to the October 31, 2000 filing date of the ’472 patent and exercised reasonable diligence in the form of continuous, steady, or constant effort from prior to that date through the October 1, 2001 filing date of Dyax’s Provisional Application Ser. No. 60/326,320 (the “’320 application”), we find that the ’472 patent is not available as prior art against Dyax’s Application Ser. No. 12/625,337(the “’337 application”)5 under 35 U.S.C. § 102(e). I. Dyax’s Response to the Order to Show Cause A. Conception of the Invention Dyax argues that its inventors, Drs. Simon E. Hufton (“Hufton”) and Hendricus H. R. J. Hoogenboom (“Hoogenboom”), then employed by TargetQuest, conceived of the invention of constructing yeast libraries that display antibodies (especially multi-chain antibodies) by March 1, 1999, by which date they had prepared a draft grant proposal outlining the project. Dyax Response to 1 Paper No. 387 2 Paper No. 386 3 Paper No. 388 4 Exhibit 1021 5 The ’337 application is a continuation of U.S. Application No. 10/262,646, filed September 30, 2002 (now abandoned), which claims the benefit of the ’320 application, filed October 1, 2001 (Paper No. 1). Interference No. 105,809 3 Show Cause Order6 (“Resp. Br.”) at 4 (citing Ex. 1156 at ¶ 6; Ex. 1233 at ¶ 5; Ex. 1136; see also Ex. 1236). Dyax argues that, in this draft proposal, Drs. Hufton and Hoogenboom memorialized their conception of the development of yeast display vectors, and uses thereof, to construct yeast display libraries, including by homologous recombination, comprising antibody genes, such as mutagenized antibody genes. Resp. Br. at 4 (citing Ex. 1136, Section 3, pp. 1, 5). Dyax contends that this conception document teaches each and every element of Dyax’s involved claim 118. Resp. Br. at 4 (citing Ex. 1156 at ¶ 9). Dyax adds that this proposal was eventually awarded with a grant from BTS (an agency of The Netherlands government that encouraged inter-business cooperation) in June 1999. Resp. Br. at 4 (citing Ex. 1156 at ¶ 6). TargetQuest was acquired by Dyax in August 1999. According to Dyax, on August 20, 1999, Dr. Hufton sent Pamela Hay (Senior Vice-President of Corporate Development at Dyax) via e-mail a description of the ongoing BTS project, including as attachments to his e-mail (1) an early version of the BTS grant proposal, (2) a preliminary text for a possible patent application on yeast/phage transfer, and (3) a summary presentation that was recently given to Dyax officers. Resp. Br. at 5 (citing Exs. 1149-1152; Ex. 1156 at ¶ 7). Ms. Hay forwarded the document describing the early version of the BTS project to Mark Alfenito, Vice- President of Business Development for Dyax. Resp. Br. at 5 (citing Exs. 1149, 1232 at ¶ 7; Ex. 1148). Dyax alleges that the English language portion of this early version of the BTS grant proposal provides substantially the same information regarding the yeast display project as the first draft, including the use of homologous recombination. Resp. Br. at 5 (citing Ex. 1156 at ¶ 7; Ex. 1149; Ex. 1136). Dyax asserts that these portions were written by Drs. Hufton and 6 Paper No. 512 Interference No. 105,809 4 Hoogenboom in English and were only later translated into Dutch for submission to BTS. Resp. Br. at 5 (Ex. 1156 at ¶ 7; Ex. 1149). Dyax adduces this evidence as proof that it had conceived of the claimed invention as of March 1, 1999, prior to the October 31, 2000 priority date of the ’472 patent. Resp. Br. at 5. B. Reasonable Diligence Dyax adduces as evidence of its reasonable diligence the declarations and laboratory and work records of Drs. Hufton and Hoogenboom, the inventors, and their colleagues, including Kirsten Teunissen (“Teunissen”), Henk Pieters (“Pieters”), and Twan van den Beucken (“van den Beucken”), as well as the billing statements of Dyax’s patent attorneys, Yankwich & Associates (“Y&A”) who eventually drafted and filed the ’320 application. Review of Dyax’s adduced declarations and records reveals that some or all of the research personnel developing the invention conceived by Drs. Hufton and Hoogenboom were involved in the project on a daily basis, excepting weekends, holidays, reasonable periods of vacation, and a single week (June 22-July 1, 2001) when the laboratory was being moved from Maastricht in the Netherlands to Liege, Belgium. See Resp. Br. at, 28, 91 (App’ces 2, 3); Exs. 1154 (Decl. of van den Beucken); 1156 (Decl. of Hufton); 1131-35 (van den Beucken laboratory notebooks); 1141-1143 (Teunissen laboratory notebooks); 1144-46 (Pieters laboratory notebooks). A review of the record indicates that van den Beucken was the principal investigator active in pursuing reduction of the conceived invention to practice, with somewhat more sporadic contributions from Teunissen and Pieters. Id. Dr. Hufton was the project manager, and Hoogenboom was in overall supervision of the program. Ex. 1154, ¶ 6; Ex. 1156, ¶ 10. Weekly (initially on Mondays, later on Fridays) meetings of approximately 2-3 hours duration were held by the team to discuss project progress, technical problems, and experimental Interference No. 105,809 5 strategies. Ex. 1154, ¶ 7. According to Dr. Hufton, van den Beucken focused on library construction via various approaches (e.g., batch transfer from phage vectors to yeast vectors, error-prone PCR, and mutator strains) and selection of antigen- specific and/or high affinity antibodies; Teunissen assisted van den Beucken in these tasks, and Pieters focused on developing selection and assay methods for yeast display. Ex. 1156, ¶ 10. Another researcher, Lydia Blaise, was subsequently added to the project. Id.; Ex. 1154, ¶ 5. Dyax also adduces the billing records of Y&A associated with the preparation of the ’320 application. Resp. Br. at 91 (App’x 3). Dyax claims that Y&A took up work on the ’320 application in chronological order received unless faced with a critical deadline on another matter, sending 6 draft applications to the client, and spending a total of 444 hours preparing the ’320 application. Resp. Br. at 2 (citing Ex. 1235, at ¶¶ 6, 7, 9, 10, 12-14, 16-19; see also, App’x 3). II. Adimab’s Opposition Junior Party Adimab, LLC (“Adimab”) opposes7 Dyax’s response to the Boards Show Cause Order, arguing that: (1) Dyax provides no proof that it conceived the “homologous recombination” used to incorporate polynucleotide sequences into a yeast expression vector; and (2) the Dyax inventors did not have a definite and permanent idea that yeast cells would actually display recombinant antibodies. Animab Opposition to Dyax Response to Order to Show Cause (“Opp. Br.”) at 1. 7 Paper No. 525 Interference No. 105,809 6 A. Dyax’s conception and “homologous recombination” Adimab admits that both parties agree, and the Board has found, that the ’472 patent discloses the invention defined in Count 1 and hence claims 118-137 of Dyax’s involved ’337 application. Opp. Br. 3. Claim 118 of Dyax’s ’337 application (which was copied from Adimab’s U.S. Patent No. 7,700,302 (the ’302 patent”)) recites, in relevant part: 118. A method for displaying a library of antibodies on the surface of yeast cells, comprising: providing yeast cells transformed with a library of yeast expression vectors encoding a library of antibodies each comprising an antibody heavy chain variable region (VH) and an antibody light chain variable region (VL), wherein at least one yeast expression vector comprises one or more homologous recombination sites and has undergone homologous recombination such that an insert polynucleotide sequence from a first polynucleotide sequence encoding the VH or a second polynucleotide sequence encoding the VL is incorporated into the yeast expression vector, and wherein the VH or VL polynucleotide sequence is fused in-frame with a sequence encoding a yeast surface anchoring protein subunit; and expressing the library of antibodies encoded by the library of yeast expression vectors in the yeast cells, whereby the library of antibodies are displayed on the surface of the yeast cells Ex. 1002. Adimab argues that, although the Board has held that the claim term “homologous recombination” is to be construed broadly to cover any type of homologous recombination,8 the language of claim 118 is nevertheless very specific as to how homologous recombination must be used. Opp. Br. 4. According to Adimab, claim 118 expressly requires a “yeast expression vector” that “has undergone homologous recombination” such that an insert antibody 8 See Paper No. 386 at 17 Interference No. 105,809 7 polynucleotide sequence “is incorporated” into the yeast expression vector and is “fused in-frame with a sequence encoding a yeast surface anchoring protein subunit ….” Id. Adimab contends that the claim explicitly requires that homologous recombination acts upon a yeast expression vector and accomplishes (“such that”) the incorporation of an insert (“incorporated into the yeast expression vector”), regardless of the specific type of homologous recombination that may be employed. Opp. Br. 4-5. Adimab argues that, even if the claim is not limited to a specific type of homologous recombination, the Dyax inventors did not conceive of any mode of homologous recombination to incorporate inserts into vectors and therefore did not conceive of the claims as construed by the Board. Opp. Br. 5. Adimab argues that, with respect to the limitation of claim 118 reciting “wherein at least one yeast expression vector…has undergone homologous recombination such that an insert polynucleotide sequence…is incorporated into the yeast expression vector….,” Dyax points to a disclosed technique called “DNA shuffling,” by reference to W.P.C. Stemmer, Rapid evolution of a protein in vitro by DNA shuffling, 370 NATURE, 389 (1994) (“Stemmer 1”).9 Id. Adimab observes that both Dyax’s conception document and the Stemmer articles alternatively refer to DNA shuffling as “in vitro homologous recombination” and “homologous recombination of DNA in vitro.” Id at 6. (citing Ex. 1136, p. 5, l. 13; Ex. 1108, p. 10747, left column). Adimab argues that, in its conception document (Ex. 1136), Dyax provides four separate citations in Appendix 4 to DNA shuffling as allegedly supporting the limitation to incorporation of inserts into yeast expression vectors by homologous recombination. Opp. Br. 6. Adimab also contends Dyax also provides three separate citations to DNA shuffling as alleged support for this limitation in the Dutch/English version of the BST grant proposal (Ex. 1149). Id. 9 Paper No. 528 Interference No. 105,809 8 Adimab therefore concludes that these specific references to Dr. Stemmer’s DNA shuffling technique are the only evidence adduced by Dyax identifying its conception of “homologous recombination” as required by claim 118. Id. However, argues Adimab, the inventors conceived of using DNA shuffling not for the purpose recited in claim 118 but, rather, for the completely different purpose of increasing library diversity. Opp. Br. 6 (reciting Ex. 1136, pp. 1, 5). Adimab also argues that DNA shuffling cannot feasibly incorporate polynucleotide sequences into vectors. Opp. Br. 6 (citing Ex. 2092, p. 30, ll. 1-17; Ex. 2109, ¶¶ 26, 31-35). DNA shuffling Adimab points to the Declaration of its expert, Dr. Frances Arnold,10 to support its argument that DNA shuffling—the sole type of “homologous recombination” identified by Dyax’s inventors—“simply does not do the job of incorporating inserts into vectors.” Opp. Br. 6-7. According to Dr. Arnold: DNA shuffling is an in vitro PCR-based process by which genes are diversified in a laboratory. (Ex. 1108, abstract). The process can use different sequence variants of a given gene as a starting material. DNA shuffling recombines these variants at positions that are close or identical in the DNA sequence and creates “shuffled” versions of the starting genes. DNA shuffling can also randomly alter the sequences of the “shuffled” polynucleotides with point mutations. Overall, the process creates a diverse population of new sequences that potentially encode different functions than the starting sequence. The process can, for example, generate genes encoding more effective proteins that bind a broader range of substrates or have substrate-binding affinities different from those of the protein encoded by the parent or input genes. Thus, one of the uses of DNA shuffling during the late 1990s to early 2000s was to increase the 10 Paper No. 536 Interference No. 105,809 9 diversity of antibody libraries. Antibodies with new or improved binding affinities could be discovered by screening these libraries. Decl. of Frances H. Arnold, Ph.D. (“Arnold Decl.”), ¶ 23. Dr. Arnold opines that “Drs. Hufton and Hoogenboom envisioned DNA shuffling (i.e., in vitro recombination) as a means to generate diversity in its antibody libraries. Exhibit 1136, for example, states that ‘DNA shuffling will be used as part of a ‘mutagenesis programme’ for the engineering of antibody libraries.’” Arnold Decl., ¶ 25 (citing Ex. 1136, pages 1 and 5). Dr. Arnold further opines that “those of skill in the art would not use DNA shuffling or in vitro recombination to incorporate antibody-encoding polynucleotide sequences into expression vectors for cell surface display” and that “DNA shuffling is effectively incapable, in and of itself, of incorporating inserts into yeast expression vectors in such a way that the inserts will be efficiently and precisely incorporated, expressed and displayed on the surface of yeast.” Arnold Decl. ¶ 26. Adimab even suggests that Dyax’s own expert, Dr. Natalie Scholler, testified in her Fifth Declaration11 that DNA shuffling cannot effectively be used to incorporate inserts into yeast expression vectors. Opp. Br. 10 (citing Ex. 2092, pp. 26-27, ll. 9-16; p. 27, ll. 2-5, p. 29, ll. 1-5). It is therefore Adimab’s contention that Dyax’s own proffered evidence of conception demonstrates that its inventors conceived only of using DNA shuffling/in vitro homologous recombination to generate diversity before antibody inserts are incorporated (i.e., cloned) into yeast expression vectors, and not to effect the insertion of antibody sequences into a yeast expression vector, as required by claim 118. Opp. Br. 7. 11 Paper No. 553 Interference No. 105,809 10 No evidence of conception of homologous recombination prior to the filing date of the ’472 patent According to Adimab, Claim 118 does not require the use of homologous recombination to generate diversity, rather, it allegedly requires using homologous recombination to incorporate insert polynucleotide sequences into vectors. Id. Adimab argues that the only insertion method referenced in Dyax’s conception document of March 1999 and in its subsequent evidence of diligence prior to the ’472 patent’s October 31, 2000 priority date is the technique of restriction digestion/ligation, which was well known in the art. Id. (citing Ex. 1136). Adimab argues further that Dyax did not conceive the invention recited in claim 118 until May of 2001. Opp. Br. 12. Adimab contends that, on May 9, 2001, Dr. Hufton received from Y&A a draft of what would eventually become the ’320 provisional application. Id. (Ex. 1156, ¶40). According to Adimab, the draft incorporated “additional material” which Dr. Hufton had sent Mr. Yankwich on May 8, 2001, including a concept of transferring antibody-encoding polynucleotide sequences from phage display vector to yeast display vectors. Id. (citing Ex. 1191; Ex. 1178, pp. 1-2). Adimab maintains that this communication is the first instance in the record where Dyax demonstrates any type of homologous recombination used to incorporate polynucleotide inserts into yeast expression vectors. Opp. Br. 12-13. Adimab similarly finds fault with the Declarations of Drs. Hufton and Hoogenboom, arguing that both use similar language in relating how they “contemplated various embodiments for constructing the library of yeast expression vectors used in [their] generic display method, including restriction digestion, PCR amplification, and homologous recombination” without providing Interference No. 105,809 11 documented proof of their conception prior to the filing date of the ’472 patent. Opp. Br. 15-16 (Exs. 1156, ¶ 8; 1233, ¶¶ 15, 17). B. Dyax’s Conception of Cell-Surface Antibody Display by Yeast Cells Adimab next argues that Dyax has not submitted evidence demonstrating that the Dyax inventors had formed “a definite and permanent idea of the complete and operative invention, as it is hereafter to be applied in practice.” Opp. Br. 18 (quoting Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1376 (Fed. Cir. 1986) (citation omitted). Adimab contends that, contrary to Dyax’s assertions, the inventors described their invention as limited to the display of multi-chain antigen-binding fragments (Fab) on yeast, rather than antibodies, as required by the language of claim 118. Opp. Br. 16, 20 (citing, e.g., Ex. 1167). Adimab takes the position that term “antibodies” in the expressing/ displaying step of claim 118 necessarily must be construed to include “multi-chain antibodies,” because completely-assembled antibodies are generally understood to be multi-chain polypeptides that include two heavy chains and two light chains, for a total of four chains. App. Br. 16-17 (citing Ex. 1041, ¶ 51). Somewhat contradictorily, Adimab later states that “[l]ike a fully assembled antibody, a Fab antibody is multichained; it contains part of the heavy chain and part of the light chain.” Opp. Br. 17 (citing Ex. 1041, ¶ 51) (emphasis added). Adimab is does not provide support for thus equating a Fab with an “antibody.” We note our understanding that a Fab is only an antigen binding fragment of an antibody. In addition to stating this position, Adimab points to the language of the specification of the ’337 application, which does not provide an explicit definition for the term “antibody.” Opp. Br. 17. Adimab argues that, because the specification defines the disclosed invention as “multi-chain polypeptide surface display,” the broadest reasonable definition of the term “antibodies” recited Interference No. 105,809 12 in claim 118 must be “multi-chain antibodies.” Adimab contends that the ’337 application is directed to “the successful display of a library of Fab antibodies on the surface of yeast cells using a library of yeast expression vectors.” Id. (quoting Dyax Opposition 1, p. 17, lines 10-1212). Adimab argues that, like a fully-assembled antibody, a Fab “antibody” is multichained; it contains part of the heavy chain and part of the light chain. Opp. Br. 17 (citing Ex. 1041, ¶ 51). Adimab contends that the evidence indicates that the Dyax inventors conceived of (1) displaying of Fab fragments on yeast, and (2) a combinatorial yeast display library taking advantage of the yeast mating cycle to form libraries of yeast displaying antibody Fab fragments. Opp. Br. 20 (citing Ex. 1167). Adimab argues that nothing in the record suggests the inventors had a reasonable expectation that yeast would yield successful assembly and surface display of multi-chain antibodies. Opp. Br. 20. On the contrary, maintains Adimab, the record reflects uncertainty on the part of the inventors as to whether Fabs could be displayed on the yeast cell membrane surface, as opposed to being secreted by yeast cells. Opp. Br. 20-21 (citing Exs. 2051, 1136, 1156, 1154, 1233). Adimab points, inter alia, to the statements of Dr. Hufton that “[i]n the late nineties and early 2000’s, yeast display technology, particularly its application to antibody selection and engineering, was only just emerging” and that “[w]e must develop (sic) yeast display know-how from scratch….” in support if this contention. Opp. Br. 21 (quoting Ex. 1156, ¶ 16). Adimab argues further that the Dyax inventor’s uncertainty was reflective of the state of the contemporary art, in which it was not known whether yeast cells could be made to express cell-surface antibodies. Opp. Br. 22. 12 Paper No. 74 Interference No. 105,809 13 According to Adimab, Dyax must be able to demonstrate that it possessed, at the time of conception, a definite and permanent understanding that yeast, once transformed with yeast expression vectors encoding antibody domains would not only express domains, but would also assemble and display them. Opp. Br. 22 (citing Hitzeman v. Rutter, 243 F.3d 1345, 1357 (Fed. Cir. 2001)). Adimab argues that, by the Dyax inventors’ own statements, they lacked the requisite definite and permanent understanding as of their alleged date of conception. Opp. Br. 22. Adimab maintains that conception is not complete if the subsequent course of experimentation, especially experimental failures, reveals uncertainty that so undermines the specificity of the inventor’s idea that it is not yet a definite and permanent reflection of the complete invention as it will be used in practice. Opp. Br. 24 (citing Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d 1223, 1229-1230 (Fed. Cir. 1994); Amgen, Inc. v. Chugai Pharm. Co., 927 F.2d 1200, 1206 (Fed. Cir. 1991)). Adimab points to the Dyax inventors statement that “many of the experiments conducted by the inventors were research in its true sense—time intensive, meticulous scientific groundwork, which yielded many failed experiments.” Opp. Br. 24 (quoting Paper No. 510, p. 10, lines 4-8 (citing Ex. 1154, ¶¶ 9- 10; Ex. 1156, ¶ 16)). Adimab argues that Dyax has thus failed to demonstrate that it held a definite and permanent idea of the complete and operative invention, as it was thereafter to be applied in practice at the time of the alleged invention, or at any time prior to the filing date of the ’472 patent. Opp. Br. 25. Interference No. 105,809 14 III. Dyax’s Reply A. Conception and DNA Shuffling Dyax first argues, as a threshold matter, that Adimab errs by applying the wrong standard in their argument. Reply Br. 4. Dyax contends that Adimab asserts that the Rule 131 showing by Dyax must be commensurate in scope with Count I, but argues that this is not the standard against which Dyax is held in order to antedate the ’472 patent. Id. According to Dyax, in a 37 C.F.R. § 1.131 context, an “applicant need show priority with respect to only so much of the claimed invention as the reference[ ] disclose[s].” Id. (citing Application of Scheiber, 587 F.2d 59, 61-62 (C.C.P.A. 1978). Dyax contrasts this with the requirement of 35 U.S.C. § 120, which it argues “operates independently of the prior art” and would require Dyax to establish disclosure of “the claimed subject matter in compliance with 35 U.S.C. § 112.” Reply Br. 4 (quoting Scheiber, 587 F.2d at 62). Consequently, Dyax argues, Adimab’s reliance on Hitzeman is inapposite and Adimab’s assertion — that a conception document must enable the invention — is legally incorrect. Id.; see also Burroughs Wellcome, Co. v. Barr Labs., 40 F.3d 1223, 1231 (Fed. Cir. 1994). With respect to the date of conception, Dyax replies by noting that the conception document cites Stemmer 1 as teaching DNA shuffling (or in vitro homologous recombination). Reply Br. 8. Dyax argues that Stemmer 1 teaches two methods of DNA shuffling, the second of which comprises a one-step approach in which a whole plasmid is shuffled/ reassembled, resulting in the production of whole plasmids that carry mutated DNA fragments (LacZ genes). Id. Dyax contends that this second option contemplates sequence insertion by regions of homology — a technique allegedly used to incorporate inserts into DNA recipient molecules via in vitro homologous recombination. Id. (citing (Ex. 1108, Interference No. 105,809 15 Abstr.; Ex. 1111, p.89, ll. 6-18; Ex. 1243, ¶ 15). Therefore, argues Dyax, Stemmer 1 teaches that DNA shuffling is a feasible technique for sequence insert incorporation in vectors, in-frame. Reply Br. 8 (Ex. 1243, ¶¶ 14, 22). Dyax also adduces the teachings of A. Crameri, G. Dawes, E. Rodriguez, Jr., S. Silver, and W.P.C. Stemmer, Molecular evolution of an arsenate detoxification pathway by DNA shuffling, 15 NATURE BIOTECHNOLOGY, 436 (1997) (“Crameri”)13 which it alleges confirms the teachings of Stemmer 1. Dyax further argues that Adimab misapprehends the statements provided by Dr. Scholler in her Fifth Declaration. Reply Br. 8. Dyax points to the repeated testimony of Dr. Scholler in which she explains that Stemmer 1 also teaches “whole plasmid reassembly.” Reply Br. 8-9 (citing Ex. 1111, pp. 90-91, ll. 15-15). Dyax argues that Dr. Scholler explains that Stemmer 1 teaches shuffling whole plasmids to produce expression vectors in which a target gene is inserted and expressed, and that Stemmer 1 is not limited to shuffling inserts. Id. (also citing Ex. 1243, pp. 5-8, ¶¶ 6-15). Dyax argues, therefore, that, based upon the teachings of the contemporary prior art and the supporting testimony of Dr. Scholler, Adimab is simply incorrect in contending that DNA shuffling could not be used to incorporate inserts into yeast expression vectors in-frame as required by claim 118. B. Possession of the Idea of the Cell-Surface Expression of Multi-Chain Antibodies by Yeast Cells In response to Adimab’s argument that the Dyax inventors did not have adequate possession of the technique for displaying a library of multi-chain antibodies on the surface of yeast cells required by Claim 118, Dyax replies that Adimab applies the incorrect standard, and that to antedate a reference under 13 Exhibit 1238 Interference No. 105,809 16 Rule 131, an applicant must demonstrate prior invention of only so much as the allegedly prior art reference teaches. Reply Br. 9 (citing Scheiber, 587 F.2d at 61- 62). Dyax argues that the ’472 patent discloses display of antibodies by yeast cells as an alternative method for antibody selection using the yeast two-hybrid system. Reply Br. 10 (citing Ex. 1021, col 16, ll. 51-55). According to Dyax, the ’472 patent suggests the construction of yeast libraries for displaying single-chain and multi-chain antibodies but, as Dr. Scholler admits, there is nothing in the ’472 patent that supports an expectation of success in displaying multi-chain antibodies. Reply Br. 10 (citing Ex. 1243 at ¶ 23-24; Paper No. 98, p. 3, citing Ex. 1091, ¶ 69; Ex. 1110, pp. 74-75, ll. 23- 22). Furthermore, argues Dyax, because the ’472 patent does not provide experimental data showing the successful display of single- chain or multi-chain antibodies on yeast cells, the descriptions of yeast display antibody libraries are entirely prophetic. Reply Br. 10 (citing Ex. 1243, ¶ 24; see, e.g., Ex. 1021, col. 62, ll. 36-39 (“can be constructed”)). Similarly, contends Dyax, their conception document of March, 1999, prophetically contemplates the construction of single-chain or multi-chain antibody yeast display libraries. Reply Br. 10 (citing Ex. 1136, p. 4; Ex. 1243, ¶ 25). Therefore, argues Dyax, the conception document’s disclosure is of the same scope as the ’472 patent with respect to the construction of multi-chain display antibody in yeast and eliminates the ’472 patent as prior art. Reply Br. 10 (citing Ex. 1243 at ¶ 25). Interference No. 105,809 17 IV Analysis A. Legal Standard As an initial matter, we agree with Dyax with respect to the legal burden it must meet in order to demonstrate that the ’472 patent is not available as prior art against claim 118. Rule 131 states, in relevant part, that: (a) When any claim of an application . . . is rejected, the applicant . . . may submit an appropriate oath or declaration to establish invention of the subject matter of the rejected claim prior to the effective date of the reference or activity on which the rejection is based. * * * (b) The showing of facts for an oath or declaration under paragraph (a) of this section shall be such, in character and weight, as to establish reduction to practice prior to the effective date of the reference, or conception of the invention prior to the effective date of the reference coupled with due diligence from prior to said date to a subsequent reduction to practice or to the filing of the application. * * * 37 C.F.R. §1.131. Section 120 recites, also in relevant part: An application for patent for an invention disclosed in the manner provided by section 112(a) (other than the requirement to disclose the best mode) in an application previously filed in the United States, or as provided by section 363, which names an inventor or joint inventor in the previously filed application shall have the same effect, as to such invention, as though filed on the date of the prior application, if filed before the patenting or abandonment of or termination of proceedings on the first application or on an application similarly entitled to the benefit of the filing date of the first application …. 35 U.S.C. § 120. The predecessor court to the Federal Circuit recognized the difference in the burden imposed by these two rules, holding that: The operation of [Section] 120 differs from the operation of Patent and Trademark Office Rule 131 (37 CFR [§] 1.131). The latter provides an applicant a mechanism for overcoming specific prior art references predating his effective filing date. The applicant need Interference No. 105,809 18 show priority with respect to only so much of the claimed invention as the references disclose, or only so much as to render the claimed invention obvious. Section 120, on the other hand, concerns only an applicant’s effective filing date. Unlike Rule 131, s[ection] 120 operates independently of the prior art, of which it makes no mention, and it expressly requires an earlier application to disclose the claimed subject matter in compliance with 35 U.S.C. [§] 112, first paragraph. Thus it is entirely appropriate that the showing required under s[ection] 120 differs from that required under Rule 131. Scheiber, 587 F.2d at 61-62 (internal citations omitted). The Board’s Order to Show Cause required Dyax to “show no later than 30 November 2012, why its claims are not unpatentable over the prior art asserted in Dyax motion 1.” Consequently, we agree with Dyax that the burden established by Rule 131 applies to the Dyax’s Response to the Order to Show Cause, and that Dyax need show priority with respect to only so much of the claimed invention as the references as the ’472 patent discloses. We find that the ’472 patent provides but a limited disclosure, and Dyax argues that its March 1999 conception document discloses as much of the claimed invention as does the ’472 patent. Reply Br. 6. The ’472 patent provides a flow chart in Figure 1B showing the construction of a yeast display library using in vivo homologous recombination in yeast. See, e.g., Ex. 1021, col. 56, col. 62, subsection 5. The ’472 patent likewise discloses use of only a single species (in vivo) of the homologous recombination genus to create yeast libraries. We agree with Dyax that the March 1999 conception document similarly discloses a single species of homologous recombination, viz., DNA shuffling. We therefore next address the issue of whether the disclosure of DNA shuffling, as disclosed by the inventor’s conception document and the supporting references and expert witness testimony as adduced by Dyax, discloses as much of the claimed invention as does the ’472 patent. Interference No. 105,809 19 B. DNA Shuffling and Homologous Recombination We are not persuaded by Adimab’s argument that Dyax has failed to demonstrate conception of the limitation of claim 118 reciting: wherein at least one yeast expression vector comprises one or more homologous recombination sites and has undergone homologous recombination such that an insert polynucleotide sequence from a first polynucleotide sequence encoding the VH or a second polynucleotide sequence encoding the VL is incorporated into the yeast expression vector, and wherein the VH or VL polynucleotide sequence is fused in- frame with a sequence encoding a yeast surface anchoring protein subunit. Claim 118.14 Adimab’s contention that Dyax’s March 1999 conception document and the Stemmer 1 reference teach only in vitro DNA shuffling as a means of driving increased library diversity, and that DNA shuffling cannot feasibly incorporate polynucleotide sequences into vectors, is not persuasive when viewed in light of the references cited by Dyax and by the several Declarations of Dyax’s witness, Dr. Scholler. Specifically, Stemmer 1 teaches, inter alia, that “[f]or comparison with published data, several combinations of mutations were constructed into the wild- type p 182Sfi vector (Table 1). All contained the g4205a promoter mutation, and were confirmed by partial sequencing. The known clinical TEM-1-derivatives (TEM-1-19) all contain up to four of a set of eight dispersed mutations.” Ex. 2102, p. 390. Furthermore, Crameri teaches that: A single sequence is randomly fragmented by DNaseI, and the fragments are reassembled based on sequence similarity by primerless PCR. The diversity consists of the variable level of mutations that is introduced by the PCR reaction. After the first selection cycle, the mutations in the pool of selected sequences are partially unlinked by 14 Paper No. 11 Interference No. 105,809 20 random fragmentation of the DNA. Following reassembly of the fragments, new combinations of mutations are formed within a conserved framework sequence, and a controlled level of new point mutations is formed as well. When coupled with selection or screening, this process allows rapid accumulation of useful combinations of mutations (green) from multiple parental sequences, while at the same time removing detrimental combinations of mutations (red). Crameri, p. 437 (emphasis added). This is reflected in Dr. Scholler’s testimony that: [T]he teachings in Stemmer [1] (regarding “whole plasmid reassembly”) and Crameri, 1997 demonstrate that whole plasmids were successfully reassembled in DNA shuffling processes and DNA inserts (i.e., a LacZ gene and an arsenic resistance operon) were incorporated in the plasmid in-frame such that functional proteins were expressed from the resultant plasmids. (Ex. 1108, at page 10749, 2nd col., 2nd paragraph; and Ex. 1238, at p. 438, right col., subsection titled “DNA shuffling” and p. 436, right col., 2nd paragraph.) Contrary to the assertion in Dr. Arnold’s declaration and Adimab’s Opposition noted above, neither were the insert DNA sequences destroyed nor were they separately reassembled from the vector; nor did the vectors lose their functionality after the DNA shuffling. Scholler Decl., ¶ 22. Dr. Scholler’s statements regarding the teachings of Stemmer 1 and Crameri are thus supported by the references. We accord due weight to the testimony of Adimab’s witness Dr. Arnold and her characterization of DNA shuffling as “an in vitro PCR-based process by which genes are diversified in a laboratory.” Arnold Decl., ¶ 23. But we find that her assertion that: If one took a polynucleotide sequence encoding an antibody VH or VL domain and a vector, digested them with DNaseI, and then shuffled them together in a primerless PCR, the insert polynucleotide sequence would either be destroyed or reassemble separate from the vector. The Interference No. 105,809 21 vector could conceivably reassemble with pieces of the polynucleotide insert somewhere inside, but the chances of the insert being incorporated in any useful manner are infinitesimally low is contradicted by the teachings of Stemmer 1 and Crameri, as described by Dr. Scholler, supra. Moreover, we find it significant that Dr. Arnold does not mention in her Declaration the teaching of Stemmer 1 quoted supra, or any reference to the teaching of Crameri. Consequently, we find that the evidence adduced by Dyax has demonstrated that the Dyax inventors had conceived of the disputed limitation as of the March 1999 conception document. Moreover, we find that the extensive documentation provided by Dyax demonstrating the activities of its research team and Y&A has sufficiently demonstrated continuous diligence from prior to the filing date of the ’472 patent (October 30, 2000) through the date of filing of the ’320 application (October 1, 2001). See Ex Parte Ovshinsky, 10 U.S.P.Q.2d 1075, 1076 (B.P.A.I. 1989) (Requiring a showing of “due diligence from prior to said date [the effective date of the reference]…to the filing of the application”) ((quoting Rule 131). C. Possession of the Idea of the Cell-Surface Expression of Multi-Chain Antibodies by Yeast Cells As related supra, Dyax bears the same legal burden under Rule 131 with respect to demonstrating that the Dyax inventors possessed the idea of cell-surface expression by yeast cells of multi-chain antibodies. See 37 C.F.R. § 1.131. Specifically, Dyax must demonstrate that it possessed the idea of surface display of multi-chain antibody fragments to the extent disclosed by the ’472 patent. We find that the ’472 patent discloses display of multi-chain antibodies in a general schematic format without producing any experimental results. See, e.g., Ex. 1021, Fig. 1B. Furthermore, the language employed by the ’472 is prophetic Interference No. 105,809 22 rather than demonstrative of an accomplished method. See, e.g., Ex. 1021, col. 16, ll. 51-55 (“FIG. 1B depicts a general scheme for this alternative method of selection of antibodies displayed on the surface of yeast cells”); col. 62, ll. 36-39 (“A library of yeast surface display vectors for expressing an antibody library can be constructed by following similar protocols as described above for construction of the library of two-hybrid expression vectors”) (emphasis added). The ’472 patent elaborates that Figure 12D of the ’472 patent displays a vector map in which the PYD1 vector encodes the Aga2 subunit of the yeast cell wall protein a- agglutinin: “The protein complex formed between Aga1 and Aga2 subunits binds to the a-agglutinin yeast adhesion receptor on the yeast cell wall, thus being displayed on the surface of yeast cells.” Ex. 1021, col. 62, ll. 45-52. However the ’472 patent does not disclose that this procedure was ever performed by the inventors; it certainly provides no experimental data disclosing that the method was ever performed to the point of expressing recombinant cell surface antibodies in yeast cells. The March 1999 conception document states that: Using the alpha agglutinin cell wall protein we will display a ScFv and a Fab fragment repertoire on the yeast cell surface. This system has been developed by Unilever Research/ Vlaardingen. By way of a model system we have a phage displayed selected repertoire of ScFv antibodies to streptavidin. This repertoire would be transferred into pUR4174 (yeast alpha agglutinin display vector) and we would test for display using fluorescent microscopy. We would select this repertoire first for high level display using the epitope tag (9e10) present in the ScFv antibodies. We would also test the correlation between high level display and high level expression by assessing the intracellular level of proteins in relation to that displayed and to that in the culture supernatant. The clone which displays to the highest level will then be used as a control antibody to establish the mutagenesis and affinity maturation program. Interference No. 105,809 23 Ex. 1136, p. 4. We find that this passage expresses the inventors possession of the idea of cell surface expression of Fabs prophetically, and employing a similar mechanism (using yeast alpha-agglutinin) in a manner that is commensurate in scope with the disclosure of the ’472 patent. “[A]n inventor need not know that his invention will work for conception to be complete. He need only show that he had the idea; the discovery that an invention actually works is part of its reduction to practice.” Burroughs Wellcome, 40 F.3d at 1228 (citing Applegate v. Scherer, 332 F.2d 571, 573 (C.C.P.A. 1964); see also Oka v. Youssefyeh, 849 F.2d 581, 584 n.1 (Fed. Cir. 1988)) (internal citations omitted). We are not persuaded by Adimab’s argument that Dyax lacked a definite and permanent idea of the complete and operative invention, as it was thereafter to be applied in practice at the time of the alleged invention, or at any time prior to the filing date of the ’472 patent. Adimab’s arguments notwithstanding, the conception document need not be enabling in isolation, rather, that its enablement can draw on unstated techniques already familiar in the art. We have related supra how the teachings of Stemmer 1 and Crameri, which were known in the art at the time of conception, supply the required enablement. We find that Dyax has adduced sufficient proof to demonstrate that, at the time of conception document, the Dyax inventors possessed the idea of cell-surface expression by yeast cells of multi-chain antibodies to the extent disclosed by the ’472 patent. V. Conclusion We find that Dyax has provided sufficient proof to show cause why the ’472 patent should not be available for use as prior art against Interference No. 105,809 24 Dyax’s ’320 provisional application. We consequently disqualify the ’472 patent as a prior art reference against the claims of Dyax’s ’337 application. cc: BRENDA HERSCHBACH JARRELL, FANGLI CHEN, ERIC J. MARANDETT and JUSTIN P. HUDDLESON, Choate, Hall & Stewart LLP, of Boston, Massachusetts. MICHAEL T. SIEKMAN, LAWRENCE M. GREEN and YAHUA J. CHEN, Wolf, Greenfield & Sacks, P.C., of Boston, Massachusetts.

Hufton et al. V. Hua et al.