Biogen, Inc. v. Amgen, Inc., 973 F. Supp. 39 (D. Mass. 1997)

*40 John Sylvia, Patrick T. Clendenen, Mintz, Levin, Cohn, Ferris, Glovsky & Popeo, P.C., Boston, MA, James F. Haley, Kenneth B. Herman, Andrew S. Marks, Madge R. Kanter, Jane A. Massaro, Janis P. McLaughlin, Kathleen M. Walker, Fish & Neave, New York City, for Biogen, Inc.

Paul F. Ware, Eileen M. Herlihy, Goodwin, Procter & Hoar, Boston, MA, Karen J. Kramer, Lloyd R. Day, Jr., David M. Madrid, Robert M. Galvin, Ricardo Rodriguez, Gary H. Ritchey, Darren B. Mitchell, Karen A. Gibbs, Cooley Godward LLP, Palo Alto, CA, Steven M. Odre, Karol M. Pessin, Amgen, Inc., Thousand Oaks, CA, Vernon M. Winters, Cooley Godward LLP, San Francisco, CA, for Amgen, Inc.

MEMORANDUM AND ORDER ON AMGEN'S RENEWED MOTION FOR SUMMARY ADJUDICATION OF CLAIM 9 OF THE `702 PATENT

STEARNS, District Judge.

Biogen brought suit alleging among other wrongs that Amgen's product Neupogen® infringes claim 9 of U.S. Patent 4,874,702 (the "`702 patent").<sup>[1]</sup> Amgen struck back with a counterclaim attacking the validity of claim 9. See 35 U.S.C. § 102(b). Amgen maintains that an abstract published by Glen Horn and Dr. Robert Wells in April of 1979 ("Horn & Wells 1979") anticipated the filing of the application for the `702 patent by more than a year. Biogen contends that there is a material dispute of fact whether or not Horn & Wells 1979 would have enabled one skilled in the art to replicate the invention disclosed in claim 9 without undue experimentation.

The `702 patent emanated from U.S. Patent Application Serial No. 921,803, which claims its earliest priority from British Application 8028983, filed on September 8, 1980, by two Belgian scientists, Walter Fiers and Erik Remaut. Biogen is the assignee of all rights under the `702 patent. The `702 patent discloses a method of inducing the production of human proteins in non-human "host" cells through the use of recombinant DNA. Claim 9 of the `702 patent discloses a technique for preparing a plasmid vector, pPLa23, a cloning tool that can be used to transfer genetic information from one organism into another.^[2] Specifically Claim 9 of the `702 patent claims

[a] recombinant DNA molecule comprising a vector according to Claim 1 and 2, and further comprising in one of said endonuclease recognition sites a DNA sequence coding for a eukaryotic, prokaryotic or viral protein, polypeptide, enzyme, hormone or antigen.

Claim 1 claims

[a] plasmid vector comprising at least one DNA sequence comprising the leftward promoter and operator derived from bacteriophage λ, PL OL, said DNA sequence further comprising at least one endonuclease recognition site located less than 300 basepairs downstream from PL OL and located between PL OL and any sequences of λ DNA downstream of the Hae III site at 73.1% of bacteriophage λ in said DNA sequence.

Claim 2 claims: "[t]he vector of claim 1, having no active cro gene and no active N gene."^[3]

On March 1, 1979, Glen Horn, a graduate student at the University of Wisconsin, and his mentor, Dr. Robert Wells, published an abstract entitled "Cloning and Characterization of a 360 B.P. DNA Fragment Containing OL of Phage λ", Federation Proceedings Abstracts, Vol. 38, No. 3, at 298.^[4] Horn & Wells 1979 describes the successful preparation of a plasmid vector, pRW601, which for present purposes is assumed to be identical to the vector disclosed by Fiers and Remaut in the `702 patent.^[5]

Summary judgment is appropriate when, based upon the pleadings, affidavits, and depositions, "there is no genuine issue as to any material fact, and [where] the moving party is entitled to judgment as a matter of law."^[6] Fed.R.Civ.P. 56(c); Gaskell v. Harvard Coop. Society, 3 F.3d 495, 497 (1st. Cir.1993). A dispute of fact is only genuine if there is sufficient evidence to permit a reasonable *42 jury to resolve the point in the nonmoving party's favor. NASCO, Inc. v. Public Storage, Inc., 29 F.3d 28, 32 (1st Cir. 1994). All reasonable inferences must be indulged in favor of the nonmoving party. Oliver v. Digital Equipment Corp., 846 F.2d 103, 105 (1st Cir.1988).

An invention is not novel and is therefore unpatentable if the invention was "anticipated." See Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 619 (Fed.Cir.1985). Anticipation of an invention is assumed if the invention was "described in a printed publication in this or a foreign country, ... more than one year prior to the date of the application for patent in the United States." 35 U.S.C. § 102(b).^[7] To establish "anticipation" under § 102(b), the proponent must demonstrate by clear and convincing evidence that all of the elements and limitations of the claim are "expressly or inherently described" within a single prior art reference and can therefore be reproduced by one skilled in the art without "undue experimentation."^[8] See Ciba-Geigy Corp. v. Alza Corp., 864 F. Supp. 429, 434 (D.N.J.1994), aff'd in part and vacated in part without op., 68 F.3d 487 (Fed.Cir.1995). See also Tillotson, Ltd. v. Walbro Corp., 831 F.2d 1033, 1036 (Fed.Cir.1987). "Extrinsic evidence may be considered to explain, but not to expand on, the meaning of an anticipatory reference. Specifically, the [c]ourt may look to extrinsic evidence to learn how the person of ordinary skill would interpret an anticipatory reference." Ciba-Geigy, 864 F. Supp. at 436 (citations omitted).

Biogen maintains that Horn & Wells 1979 did not anticipate claim 9 of the `702 patent because it would not have enabled one of ordinary skill in the art to reproduce pRW601 without undue experimentation. Biogen argues that Horn & Wells 1979 is an abstract, not a recipe, and thus lacks the critical information necessary to guide a skilled practitioner working in a 1979 molecular biology laboratory.

The first task is to determine the 1979 genus of one of ordinary skill in the art of plasmid vector preparation and the recombinant DNA technique of promoting expression. In Custom Accessories v. Jeffrey-Allan Industries, 807 F.2d 955 (Fed.Cir.1986), the Federal Circuit taught that

[t]he person of ordinary skill is a hypothetical person who is presumed to be aware of all the pertinent prior art. The actual inventor's skill is not determinative. Factors that may be considered in determining level of skill include: type of problems encountered in art; prior art solutions to those problems; rapidity with which innovations are made; sophistication of the technology; and educational level of active workers in the field. Not all such factors may be present in every case, and one or more of them may predominate.

Id. at 962-963 (citations omitted).

Amgen initially identified a laboratory technician working in the field of plasmid vector preparation as the template for a skilled practitioner in the art of recombinant DNA in 1979. Amgen's Statement of Undisputed Facts ¶ 15, citing Second Chamberlin Decl. ¶ 85. At the hearing on its motion, however, Amgen agreed with the opinion of Biogen's expert, Dr. Nikos Panayotatos, that a laboratory technician in 1979 would not have been able to prepare a pRW601 vector without the benefit of a detailed protocol and knowledgeable supervision. See February 26, 1997 Tr. at 8-11; Panayotatos Decl. ¶ 13-16.

*43 I conclude that Dr. Horn, the co-author of the abstract, is a more appropriate exemplar of one with ordinary skill in the art of recombinant DNA at the relevant time. It is somewhat unlikely that a mere technician in 1979 would have had any extensive familiarity with the literature in what then was an emerging science. Dr. Horn, on the other hand, as a graduate student working under Dr. Wells, would have had a relatively sophisticated grasp of the field. See Wells Decl. ¶ 5. The work underlying his dissertation, which involved the isolation, characterization and analysis of the pRW601 segment of λ phage DNA, would also have tutored him in the techniques of DNA fragment separation and manipulation known to the art in 1979.

As said before, to "anticipate" an invention, the prior art reference must enable one skilled in the art to reproduce the invention described in the patent without "undue experimentation." "Whether making and using the invention would have required undue experimentation, and thus whether the disclosure is enabling, is a legal conclusion based upon several underlying factual inquiries." Genentech, Inc. v. Novo Nordisk, A/S, 108 F.3d 1361, 1365 (Fed.Cir.1997) (citation omitted). The factors to be considered in determining whether or not experimentation is undue are "the quantity of experimentation necessary, the amount of direction or guidance presented, the presence or absence of working examples, the nature of the invention, the state of the prior art, the relative skill of those in that art, [and] the predictability or unpredictability of the art." Ex parte Formal, 230 USPQ 546, 547 (Bd.Pat.App.Inter.1986). "That some experimentation is required is not fatal; the issue is whether the amount of experimentation required is `undue.'" In re Vaeck, 947 F.2d 488, 495 (Fed.Cir.1991).

I propose to evaluate each of the disclosures of Horn & Wells 1979 in light of the standards described above, keeping in mind that the issue is the enablement of one skilled in the art in 1979.

"A Hae III digest of a segment of λ DNA was first fractionated by RPC-5 column chromatography."

That the theory and practice of accomplishing a digest using Hae III restriction enzyme were a matter of common knowledge to a practitioner of the art in 1979 is not disputed. Because the details of RPC-5 column chromatography were published in 1976, a person skilled in the art (as defined) would have been familiar with the column chromatography method of DNA fragment separation in 1979. See Hardies & Wells, Proceedings from the National Academy of Sciences USA, Vol. 73, at 3117-3121.

What Horn & Wells 1979 does not, however, disclose is the identity of the "segment" of λ phage DNA that was subjected to the Hae III digest. Thus, a skilled practitioner in 1979 would have had to isolate the desired 352 basepair fragment^[9] from the entire 48,000 basepairs of λ phage DNA. The magnitude of this task is made clear by Dr. Horn in his Ph.D. dissertation.

The cloning [of the 352 basepair Hae III fragment containing the PL OL sequence] required the initial partial purification of the PL fragment away from most other λ Hae III fragments. This was accomplished by RPC-5 fractionation of the Hae III digest of a plasmid containing only a portion of the lambda genome. Thus, the frequency of insertion of the desired PL fragment was high enough to allow direct screening of all inserted plasmids. One-fourth of all cloned plasmids had picked up an inserted fragment, and approximately one-third of these had inserted the PL fragment. The PL fragment was identified by restriction mapping, sequencing, operator function, and promoter location.

Horn Ph.D. dissertation, at 11-15, ¶ 1.

Since the desired fragment represents less than 1% of the lambda genome, the insertion frequency of 352-PL was increased by *44 first partially purifying it away from the other lambda Hae III fragments.

Id. at 11-10, ¶ 1.

It follows that the probability of isolating and cloning the desired 352 basepair fragment on any given attempt using the entire λ Phage genome would have been considerably less than 0.1%, even if all of the technical procedures were 100% successful.^[10]

"The partially purified OL fragment was then ligated into the Eco RI site of pBR322 and cloned in E. coli C600 (R- M- recBC +) using a technique which converts the Hae III ends of the fragment into Eco RI sites."

It is unclear what Horn & Wells 1979 meant by a "partially purified" OL fragment. (How mary other Hae III fragments in addition to the 352 basepair fragment of interest were present in the DNA mixture used for ligation into pBR322?). How Horn and Wells determined that the fragment used for ligation into pBR322 contained the OL sequence is also unclear.

Assuming the successful isolation of the desired fragment, a skilled practitioner in the art in 1979 could have prepared an Eco RI-digested pBR322 plasmid for ligation, as it is undisputed that the technique of blunt-end DNA ligation was well known to molecular biologists in 1979. However, there is no description of the specific "technique" used by Horn and Wells to convert the Hae III ends to Eco RI sites prior to ligation. Moreover, a practitioner might have found the wording of the abstract confusing because the modification used by Horn to allow bluntend ligation of the Hae III fragments within an Eco RI site in pBR322 requires the enzymatic alteration of the Eco RI sites, not the Hae III sites as Horn & Wells 1979 implies.

"At least six single inserts and one double insert of this fragment were characterized."

The method used by Horn and Wells to "characterize" the Hae III fragment is not disclosed. Assuming that the clones were analyzed by Eco RI digestion and partial sequence analysis, a skilled practitioner in 1979 would have had no guidance in distinguishing the clone containing the 352 basepair fragment of interest from the remaining clones possessing incorrect or "contaminating" Hae III fragments.

"All have two EcoRI sites and have the PL promoter(s) pointing toward the tetr genes of the vector."

Assuming that the practitioner was able to accomplish separation analysis of the fragments resulting from the Eco RI digestion, he or she could have performed an Eco RI digestion of the clone(s) to determine the presence of the two Eco RI sites. There is, however, no indication of how the presence and direction of the PL promoter(s) was determined.

"EcoRI digestion of the single insert plasmid (pRW601) followed by sucrose gradient centrifugation separated the 360 b.p. fragment from the vector"

That a person of ordinary skill in the art in 1979 could have performed an Eco RI digestion of pRW601 is not disputed. The technical details of sucrose gradient centrifugation were known to the art in 1979 and thus available to a technician with the requisite skill. See Hardies & Wells, Gene, Vol. 7, at 1-14.

"DNA sequencing on the half of the fragment `upstream' from the PL transcript reveals an unusually large A-T rich segment"

The technical details of DNA sequencing were in the public domain in 1979, and thus a person of ordinary skill in the art would have been able to partially sequence the cloned Hae III fragment. See Maxam & Gilbert, Proceedings from the National Academy of Sciences USA, Vol. 74, at 560-564.

It follows that most of the steps set out in Horn & Wells 1979 were within the capacity of a person skilled in the art in 1979. However, the critical missing element is the (probably deliberate) failure of Horn and Wells to disclose their starting segment of λ DNA. The λ phage genome has 148 Hae III *45 sites. When digested with Hae III, the 48,000 basepair λ DNA string is cut at each of those sites. Thus, the digestion produces 149 separate fragments. Of these 149 fragments, seventeen are in the 310-410 basepair range. Because the RPC-5 column chromatography method identified in Horn & Wells 1979 was incapable of separating these seventeen fragments, a person of ordinary skill in the art in 1979 would have faced the intimidating task of seeking the basepair fragment described in the abstract by trial and error. See Panayotatos Decl. ¶¶ 19-22.^[11] Equally daunting, Horn and Wells sequenced only half of the cloned DNA fragment. Consequently, a skilled practitioner, even if he or she believed that the correct starting segment of λ phage DNA had been isolated, would have had difficulty discerning that the completed vector lacked the cro gene and the N gene as required by claim 2.

Amgen more or less concedes the omissions in Horn & Wells 1979,^[12] but argues that a skilled practitioner would have compensated for the lack of concrete direction by resorting to his or her knowledge of the art. For example, Amgen maintains that a practitioner could have used gel electrophoresis to isolate a Hae III fragment containing a PL OL promoter by relying on the "teaching" of an article published by Zafri Humayan and his colleagues in 1977, "Completed DNA Sequences and Organization of Repressor-binding Sites in the Operators of Phage Lambda," Journal of Molecular Biology, Vol. 112, at 265-277. Amgen contends that Humayan isolated a precursor of the Horn & Wells 1979 Hae III segment using gel electrophoresis. Putting aside Biogen's argument that Humayan's "Hae 340" preparation most likely contained all of the seventeen Doppelgänger fragments, see Biogen's Response to Amgen's Supplemental Submission, at 9, Amgen's attempt to bridge the critical gap in Horn & Wells 1979 unacceptably expands on the "known in the art" doctrine. The Federal Circuit's discussion in Genentech, 108 F.3d at 1366, is worth quoting at length.

It is true, as Genentech argues, that a specification need not disclose what is well known in the art. See, e.g., Hybritech, Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1385, 231 USPQ 81, 94 (Fed.Cir. 1986). However, that general, oft-repeated statement is merely a rule of supplementation, not a substitute for a basic enabling disclosure. It means that the omission of minor details does not cause a specification to fail to meet the enablement requirement. However, when there is no disclosure of any specific starting material or of any of the conditions under which a process can be carried out, undue experimentation is required; there is a failure to meet the enablement requirement that cannot be rectified by asserting that all the disclosure related to the process is within the skill of the art. It is the specification, not the knowledge of one skilled in the art, that must supply the novel aspects of an invention in order to constitute adequate enablement. This specification provides only a starting point, a direction for further research.

While I conclude that a practitioner skilled in the art could have prepared the pRW601 vector by bridging the gaps in Horn & Wells 1979 with inspired recourse to the existing literature,^[13] for the reasons stated, I also conclude that Amgen has failed to meet its *46 burden of showing by clear and convincing evidence that this feat could have been achieved without undue experimentation.

For the foregoing reasons, Amgen's Motion for Summary Adjudication of Claim 9 of the `702 patent is DENIED.

SO ORDERED.

[1] Biogen's Complaint alleges that Amgen infringed on all claims of two of Biogen's patents, U.S. Patents 5,401,642 and 5,401,658, and claims 3, 7-9, 13 and 17 of the `702 patent. In a January 19, 1996 decision, the court dismissed all claims and counterclaims relating to the `702 patent except those involving claims 9 and 17.

[2] A vector is a strand of DNA that can be inserted into a host cell where it will replicate itself. The "parent vector" identified by Fiers and Remaut in the `702 patent is pPLa23. Amgen asserts that for present purposes pPLa23 "is identical in all relevant respects" to the Horn & Wells 1979 vector, pRW601. Amgen's Memorandum in Support of Its Renewed Motion for Summary Adjudication, at 13. Biogen contends that claim 9 requires that the PL promoter be capable of driving expression. Biogen also maintains that pRW601 is incapable of expression. Whether this is true or not (or is in fact a limitation of claim 9) is unnecessary to resolve given my ultimate conclusion that Horn & Wells 1979 is not enabling.

[3] Claim 9 depends from claims 1 and 2. The parties agree that a complete claims construction is unnecessary to resolve the present motion.

[4] Horn & Wells 1979 is a nine sentence abstract. Biogen claims that of the nine sentences, only two actually describe how Horn and Wells constructed plasmid vector pRW601.

[5] Amgen claims that like the vector described in the `702 patent, the plasmid vector pRW601 described in Horn & Wells 1979:(1) contains the leftward promoter and operator region of phage, PL OL; (2) contains an Eco RI site converted from the Hae III site at 73.1% of the DNA phage λ; (3) contains a DNA sequence of phage λ DNA which extends no further than the Hae III site at 73.1% phage λ and terminates at the same Eco RI site converted from a Hae III site disclosed in the Fiers and Remaut plasmid vector pPla23 and Figure 6 of the `702 patent; (4) lacks the cro gene; (5) lacks the N gene; (6) includes a DNA sequence known as "tet" downstream from PL OL; and (7) contains two Taq I endonuclease recognition sites surrounding the tetr gene sequence. As in the `702 patent, one of the Taq I sites is located 26 basepairs downstream from the Eco RI site (converted from the Hae III site at 73.1% phage λ). The second Taq I site is located 1270 basepairs downstream from the Eco RI site.

[6] A fact is considered material only when it has the "potential to affect the outcome of the suit under applicable law." Nereida-Gonzalez v. Tirado-Delgado, 990 F.2d 701, 703 (1st Cir.1993).

[7] To be a "printed publication" under § 102, a prior art reference must have been sufficiently accessible to the public interested in the art. Constant v. Advanced Micro-Devices, Inc., 848 F.2d 1560, 1568 (Fed.Cir.1988). Biogen does not dispute that Horn & Wells 1979 is a "printed publication."

[8] I agree with Biogen that In re Sasse, 629 F.2d 675, 681 (C.C.P.A.1980), to the extent that it can be read to imply that prior art references are presumed to be enabling, has no relevance to the facts of this case.

[9] The parties agree that Horn & Wells 1979's identification of the desired fragment as containing 360 rather than 352 basepairs is immaterial.

[10] There is evidence that Dr. Horn required nineteen months to achieve the result reported in Horn & Wells 1979.

[11] Amgen argues that in 1979 there were two "common knowledge" techniques for isolating DNA fragments for cloning: (1) nitrocellulose filter binding; and (2) gel electrophoresis. Biogen counters that nitrocellulose filter binding was not a known technique in 1979 for isolating DNA fragments and that gel electrophoresis cannot isolate the Horn & Wells 1979 Hae III fragment from the sixteen other similarly sized Hae III fragments in λ phage DNA.

[12] Amgen specifically concedes that a person skilled in the art would not have known the segment of λ DNA with which Horn and Wells had started and would as a result have had to digest the entire phage λ genome with Hae III. Amgen also agrees that the practitioner would have faced the problem of separating the desired Hae III fragment from the 149 Hae III fragments produced by the digest. Amgen's Reply Memorandum, at 14-15.

[13] The Declaration of Dr. Keith Backman is the most persuasive of Amgen's submissions on this point. I do note that Dr. Backman was forced to turn to Dr. Horn's later published Ph.D. thesis to confirm his deduction as to how Horn was able to verify that the Hae III fragment contained the leftward promoter and operator region of λ DNA. Backman Decl. ¶ 18.