Trustees of Columbia University v. Roche Diagnostics GmbH

150 F. Supp. 2d 191, 2001 U.S. Dist. LEXIS 6383, 2001 WL 498563

• Court: District Court, D. Massachusetts
• Decided: April 27, 2001
• Docket: C.A. 93-11512-NG
• Status: Published

Opinion

MEMORANDUM AND ORDER

GERTNER, District Judge.

I. INTRODUCTION

This case involves an allegation by plaintiff Columbia University (“Columbia”) of patent infringement against defendant Roche Diagnostics GmbH (formerly Boeh-ringer Mannheim, GmbH) (“Roche”), a multinational pharmaceutical corporation having its principal place of business in Mannheim, Germany. In essence, Columbia claims that Roche induced or otherwise collaborated with Genetics Institute (“GI”), a United States company based in Cambridge, Massachusetts, to produce the drug Erythropoietin (“EPO”) 1 using meth ods and products for which Columbia holds the patents.

At issue is the degree of Roche’s involvement in facilitating GI’s infringing conduct. Though typically, United States patent law does not extend extraterritorial protection, the law has carved an exception akin to a common law notion of vicarious liability, whereby a foreign company may be held liable for its complicity in the infringing activities of a domestic actor. Thus, the central question here concerns Roche’s relationship with GI, and in particular, its participation in GI’s allegedly infringing activities.

The dispute revolves around U.S. Patent Nos. 4,399,216 (“the ’216 patent”), 2 4,634,-665 (“the ’665 patent”), 3 and 5,179,017 (“the ’017 patent”) 4 (collectively referred to as the “Axel patents”). 5 The Axel patents cover processes for inserting two genes into a host cell (eotransformation) in which one of the genes encodes a marker protein, and the other gene encodes a protein of interest. 6 The claims also cover the cell lines produced by the process of amplification and cotransformation, variously described hereafter as the EPO generating Chinese Hamster Ovary (“CHO”) host cell, the production clone, or DN2-3PX3. See Trustees of Columbia University in City of New York v. Roche Diagnostics GmbH, 126 F.Supp.2d 16 (D.Mass.2000) (hereinafter “Markman findings”). 7 The claims do not, however, cover the protein of interest that is produced by the cell, EPO.

Columbia alleges direct infringement under 35 U.S.C. § 271(a). 8 It claims that when GI exported production clones to Roche for use in the German manufacture of EPO, GI never relinquished title to the cells. As such, Roche’s extraterritorial use of the cells comprises direct activity in the United States within the meaning' of the statute. Moreover, Roche directly imported into the United States a certain kind of serum-free EPO, which, Columbia argues, is the “fruit of the poisonous tree,” i.e. a compound derived from Roche’s infringement of the Axel patents.

Columbia also alleges vicarious or induced infringement under 35 U.S.C. § 271(b). 9 Columbia contends that Roche was aware of Columbia’s first patent, the ’216 patent, when it gave financial support and encouragement to GI’s infringing acts in the creation of EPO generating CHO host cells. Moreover, Columbia characterizes the agreement between Roche and GI not as a tradi tional buy-sell agreement, but rather as a collaboration, whereby GI was compensated for researching and developing commercially feasibly production clones with both parties jointly owning any intellectual property developed during the course of the project. Arguing largely from that collaboration, Columbia claims that Roche induced GI to (a) produce an infringing cell line, necessary for the EPO manufacture, (b) to export the cell line, (c) to manufacture EPO in the U.S. and in Germany, and (d) to re-import the serum-free EPO into the United States.

Finally, Columbia alleges that the export of the production clone by GI to Roche violates 35 U.S.C. § 271(f). 10 Noting that this provision covers chemical compositions as well as machinery, Columbia contends that Roche incurred liability when GI shipped bulk EPO and EPO-producing cell lines to Germany because these exported components were later assembled with other chemicals into finished products that infringe.

Roche has filed the following three motions for summary judgment in rebuttal: (1) Defendant’s Motion for Summary Judgment of Non-Infringement under 35 U.S.C. § 271 [docket entry # 250]; (2) Defendant’s Motion for Summary Judgment of Non-Infringement of U.S. Patent Number 5,179,017 (“the ’017 patent”) [docket entry # 265]; and, (3) Defendant’s Motion for Summary Judgment of Non-Infringement of Claims to Unlinked DNA Embodiments [docket entry # 268].

Roche’s counter arguments that it did not infringe the Axel patents, either directly or vicariously, summarize as follows. First, it makes the general claim that whatever actions it took or activities it engaged in relative to Columbia’s patents are insulated from liability because they all occurred outside the United States. Moreover, Roche maintains that nothing about its relationship with GI induced GI to engage in infringing conduct because, in contracting with GI to develop commercially feasible means of producing EPO, Roche claims that it had no control over the development processes that GI chose to utilize.

Apart from the inducement theory, Roche maintains that § 271(f), on its face, does not apply to Columbia’s allegations since that provision is not directed to the export of a patented invention — here the EPO generating CHO host cells — but rather to the export of an unpatented component reassembled abroad in a manner that would infringe the patent if such combination occurred within the United States. Roche argues that Columbia’s reliance on a 271(f) theory is misplaced because there is no evidence that Roche supplied or caused to be supplied the components of a patented invention so as to actively induce the combination of such components outside of the United States.

Second, with respect to the later, ’017 patent, Roche argues that the facts indicate no allegedly infringing activities on the part of GI after 1991 (when GI was enjoined from manufacturing and selling recombinantly made EPO). In any event, as the ’017 patent did not issue until 1993, any pre-issuance activities alleged by Columbia are not actionable as direct or induced patent infringement. Furthermore, Roche argues that Columbia has introduced no evidence of infringement of the ’017 patent after 1993, i.e. post-issuance.

Finally, with respect to the two earlier Axel patents, the ’216 and ’665 patents, Roche claims there was no infringement of claims related to unlinked cotransformation processes. In light of this Court’s Markmcm findings, Roche argues that the only claims that could conceivably be infringed by GI’s activities are those restricted to linked cotransformation processes.

Columbia counters that Roche too narrowly characterizes my Markman findings. They contend that nothing in the patent language, this Court’s claim interpretation, or the prosecution history of the patents restricts the claims of the ’216 and ’665 patents to linked cotransformation processes.

This Court entertained oral arguments on the various motions for summary judgment on Tuesday, February 28, 2001. Though the summary judgment motions were filed just prior to trial, and though they might otherwise blend rather neatly into the trial itself, I nevertheless agreed to consider them first, insofar as they might narrow the issues and tailor the scope of the suit. But, while I have found Roche’s challenges to Columbia’s case not insignificant, I cannot conclude on the record before me that summary judgment resolves the dispute in its entirety. Therefore, for the reasons stated herein, the motions are GRANTED in part and DENIED in part.

II. STANDARD FOR SUMMARY JUDGMENT

Fed.R.Civ.P. 56(c) provides, in pertinent part, that a court may grant summary judgment only if “the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law.” See Karlin Tech. Inc. v. Surgical Dynamics, Inc., 177 F.3d 968, 970 (Fed.Cir.1999).

In ruling on a summary judgment motion, the Court must view the record and draw inferences in a light most favorable to the non-moving party. 11 Rodime PLC v. Seagate Tech., Inc., 174 F.3d 1294, 1301 (Fed.Cir.1999). “When a party fails to make a showing sufficient to establish the existence of an element essential to that party’s case, and on which that party bears the burden of proof at trial, there can no longer be a genuine issue as to any material fact ... and the moving party is entitled to judgment as a matter of law.” Smith v. Stratus Computer, Inc., 40 F.3d 11, 12 (1st Cir.1994). This Court must, therefore, undertake to determine (1) whether factual disputes exist, (2) whether the factual disputes are genuine (i.e. a reasonable factfinder could return a ver- diet for the nonmoving party on the basis of the evidence), and, (3) whether any fact genuinely in dispute is material (i.e. such that it might affect the outcome of the suit under the applicable substantive law). See Anderson v. Liberty Lobby, 477 U.S. 242, 247-48, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986).

III. FACTS

A. GI’s work and its search for “collaborators”

Beginning in April 1982, GI embarked on a project to isolate the EPO gene for purposes of cloning, insert it into host cells, and express it in those host cells. GI’s business strategy was to conduct research and seek partnerships with large companies who would fund their projects and market their products. See July 12, 1999 Deposition of Robert Kamen [hereinafter “Kamen Depos.”], Defendant’s Exhibit [hereinafter “Def. Exh.”] 3, pp. 23, 24.

In the meanwhile, Columbia’s work was progressing rapidly. Columbia’s first Axel patent, U.S. Patent No. 4,399, 216 (’216 patent) was issued August 16, 1983. It provided a process for inserting DNA molecules into eucaryotic host cells to express a gene of interest, including the EPO gene.

In June 1984, GI entered into agreement with Chugai Pharmaceutical Company, Ltd. (“Chugai”) a Japanese corporation having a principal place of business in Tokyo, Japan, to become GI’s exclusive licensee of its EPO technology for the territories of the United States, Canada, Mexico, Japan and other countries in the Far East. In the GI-Chugai Research and Development Agreement, GI granted Chu-gai the exclusive right to utilize GI’s EPO-related technology and proprietary knowledge to commercialize EPO in these territories, in exchange for Chugai funding the EPO research project and paying royalties to GI. The technology included an EPO-generating production clone that would enable Chugai to manufacture EPO directly in Japan.

By the end of 1984, GI was looking for a licensee of its EPO technology for territories in Europe, South America, and Africa. At the suggestion of German professor, Dr. Gurland, Tuan Ha-Ngoc, GI’s marketing manager, first approached Roche (then BMG). 12 The two entities formally entered into a relationship with the execution of a Confidentiality Agreement on January 2, 1985. After a series of further meetings, GI and Roche drafted a “Development and License Agreement” (the “D & L Agreement”) on October 8, 1985, in which it was contemplated that GI would assume EPO manufacturing responsibility and supply bulk EPO to Roche.

Two issues are contested regarding the scope of the D & L Agreement and GI’s manufacture of EPO: First, Roche suggests that the infringing technology that led to the production of the EPO generating CHO host cell (the cell line for producing EPO on a commercially feasible basis), had already been developed by GI when the D & L Agreement was executed. Thus, Roche maintains that it could not have induced infringement because GI’s direct infringement of the Axel Patents occurred before it entered into a relationship with Roche.

Columbia denies that GI had completed its development of the EPO production clone prior to the formal initiation of the GI-Roche relationship. Rather, Columbia contends that the allegedly infringing production clone used by Roche — the amplified DN2-3PX3 (10 micromolar) — was de veloped after the GI-Roche relationship had begun and after Columbia had patented its EPO-producing technology for commercial EPO.

Second, Roche suggests that its relationship with GI was not collaborative to the extent that Columbia intimates: Roche, beyond the scope of Columbia’s domestic patent protection, simply made international use of products offered by an alleged infringer. Indeed, while Roche’s primary responsibility was for marketing alone, GI exclusively controlled all aspect of cell development, including isolating and expressing the EPO gene, and maintained full ownership and control of the cell lines throughout the GI-Roche enterprise.

Columbia insists that Roche’s role in its relationship with GI was far more proactive, whereby Roche more formally dictated the scope of GI’s EPO production activities (hence amounting to “inducement” in the legal sense). As is described below, the record suggests that there is a material dispute with respect to these issues.

B. When did GI create their infringing production clone?

Roche contends that as early as April of 1982, GI embarked on its project to clone the EPO gene, and that two years later, in August of 1984, Dr. Edward Fritsch of GI successfully cloned the gene. See Def. Exh. 6, May 17, 1999 Deposition of Fritsch [hereinafter “Fritsch Dep.”]. By October of 1984, Dr. Randal Kaufman of GI had allegedly inserted the EPO gene into COS and CHO host cells where it was expressed, and on December 4, 1984, GI filed a patent application which disclosed the cloning, co-transformation, and CHO cell expression system. See Def. Exh. 8, 10. GI’s 1984 Third Quarterly Report purportedly confirms that GI had created an amplified EPO-producing CHO cell by cotransformation before the D & L Agreement was signed.

Columbia argues that the EPO clone developed by Fritsch prior to the D & L Agreement was not the production clone used by GI to make EPO for Roche. See Plaintiffs Exhibit [hereinafter “PI. Exh.”] I, Fritsch Dep., pp. 42-44. Indeed, to the contrary, Fritsch testified in his deposition that the EPO clone that GI had developed at the time it filed for the patent had never been used to produce commercial product for BMG. Id. at 42. Furthermore, Columbia contests that Dr. Kaufman helped develop a CHO expression system at that time, emphasizing that no witness has testified to support Roche’s characterization of GI’s progress. Columbia’s position— that what Fritsch cloned was not the production line at issue in this case — is confirmed, to some extent, by GI’s Quarterly Report for the First Quarter of 1985, in which GI indicated that it “was aggressively developing the CHO expression system for producing EPO.” See Def. Exh. 18. That GI was still “aggressively” attempting to perfect its system of EPO production after the progress of Drs. Fritsch and Kaufman suggests that the clone required further modification (i.e. amplification) and that GI had yet to develop a suitable CHO expression system or cell line for producing EPO. At the very least, it remains an open question on this record as to the stage of GI’s EPO development at the time the GI-Roche alliance started. 13

C. The nature of the GI-Roche Collaboration

Roche claims that it was interested only in the end-product EPO as to which Co lumbia had no patent, and, thus, did not participate in selecting the means by which GI would develop it. However, the record more clearly suggests the nature of Roche’s relationship with GI to be far more interdependent and far less passive than Roche would have this Court accept.

First, as with the GI-Chugai agreement, the GI-Roche D & L Agreement uses language of collaboration. For example, Roche is permitted to inspect and determine if the work was proceeding as contracted. See Agreement, Def. Exh. 20 at 2.2 (“Each Party shall have the right to arrange for its employees and outside consultants involved in the Project to visit the other Party at its offices and laboratories, and to discuss the Project work and its results in detail with the technical personnel and consultants of the other Party”).

Furthermore, the D & L Agreement authorizes the exchange of confidential trade secrets, as GI was to disclose to Roche all “know-how as well as the production clone in confidence to BM ... to enable BM to manufacture and produce Licensed Compound as well as Licensed Products.” See id. at 2.6.

The D & L Agreement included also a provision for joint ownership: “the Parties shall own jointly the entire right, title and interest in and to all patent and other rights in any product, method or apparatus conceived, reduced to practice or developed jointly by GI and BM in the course of the Project;” at the very least, the provision supports an inference that the parties anticipated a relationship more cohesive and reciprocal than an ordinary contract between an independent manufacturer and purchaser. See id. at 5.3

Further supporting Columbia’s characterization of the GI-Roche relationship as a concerted partnership, the terms of the D & L Agreement specify that Roche would fund a research project to be conducted by GI in Massachusetts, “utilizing recombinant DNA technology for producing [EPO] on a commercially feasible basis for use in humans.” The Agreement further provided that Roche would pay GI to try to “scale up” certain manufacturing processes involving GI’s patented technology, in order to determine if these technologies could be used to make EPO on a commercially feasible basis. 14

Roche disputes this characterization, suggesting instead that GI simply agreed to produce EPO at its facility for sale to Roche, until Roche developed the infrastructure to assume the full scale of EPO production. Roche was also to pay royalties to GI in exchange for, what Roche says, was a license to sell GI manufactured EPO in Europe and Africa, but not the U.S. because of Chugai’s exclusive license. See Roche’s Statement of Undisputed Facts ¶ 18.

Columbia disputes that the royalties were exclusively for “GI manufactured EPO” sold outside the United States. Rather, argues Columbia, the language of the agreement suggests that Roche was to pay royalties on all net sales of EPO— whether manufactured by GI or by Roche employing GI’s cells and expertise.

D. The D & L Agreement in Operation

In Novemher of 1985, GI shipped bulk EPO to Roche, produced from allegedly infringing EPO generating CHO host ceils. By March of the following year, GI transferred a batch of the EPO-generating CHO host cells directly to Roche. Though Roche designates it a “one-shot deal,” from which Roche manufactured bulk EPO, the record is ambiguous. 15

In October 1987, Amgen, Inc. sued GI (along with Chugai Pharmaceutical Co.), alleging that GI’s process for producing EPO infringed certain patents which it owned. See Amgen, Inc. v. Chugai Pharmaceutical Co., Ltd., 706 F.Supp. 94, (D.Mass.1989). Amgen sought an order enjoining GI from making EPO.

In response to the Amgen litigation and in anticipation of an injunction on the further manufacture of EPO, GI and Roche agreed to a “bailment” of EPO producing host cells to Roche in Germany. 16 The vials of “bailed cells” containing the production clone DN2-3 alpha, 10 micro molar, were in addition to the initial “one-shot” batch which had been shipped to Roche two years earlier. 17 The parties took these steps apparently to insure their continued collaboration in Germany, notwithstanding an adverse ruling by a U.S. Court. Towards this end, GI was to provide detailed technical instructions to Roche regarding operating procedures, parameters, production data, records, and production plant descriptions and specifications.

As it turns out, however, though finding a reasonable likelihood of success on the merits of the validity of Amgen’s patent and Chugai’s infringement, the Federal District Court refused to enjoin the further production of EPO because the Court was wary to inhibit production of a drug that might otherwise benefit thousands of patients simply to protect one party’s vested property interest. Instead, it ordered all profits from the production of EPO to be held in escrow pending the outcome of the lawsuit. See Amgen, Inc. v. Chugai Pharmaceutical Co., Ltd., 1989 WL 169006, *4 (D.Mass.).

On January 11, 1988, Roche and GI supplemented their original D & L Agreement with a Supply Agreement concerning the production of EPO. Under the modified agreement, Roche agreed to purchase 130 grams of EPO produced by GI at a price of $200,000 per gram. Roche also indemnified GI for part of the damages arising out of the Amgen patent dispute. Furthermore, in 1988, Roche agreed to import to GI in Massachusetts EPO reformulated in Germany for use in a study of Jehovah’s witnesses suffering from end-stage renal disease.

There is some dispute between the parties surrounding the ultimate fate of the bailed vials of DN2-3P23. Columbia contends that Roche retained some portion of the bailed vials, but the November 2, 1999 Affidavit of Barbara O’Connell and accompanying Shipment Log Summary [hereinafter “Shipment Summary”] documenting the transfers of the EPO cells lines be tween GI and Roche belies this contention. According to the Shipment Summary, three vials of the cell line were sent on September 30,1987 to Roche, 18 followed by three more vials on October 5, 1987. However, these vials were later returned by Roche in 1989. 19 See Def. Exh. 25.

In 1990 Columbia brought a patent infringement action against GI in the U.S. District Court in Delaware, alleging that GI’s EPO production infringed the Axel patents. Columbia later withdrew its action in light of a March 5, 1991 ruling by the Federal Circuit enjoining GI from further EPO production, use, or sales in the United States. Amgen, Inc. v. Chugai Pharmaceutical Co., Ltd., 927 F.2d 1200 (Fed.Cir.1991).

Roche maintains that GI stopped all manufacturing, research, and development of EPO after this date. See September 29, 1994 Affidavit of Bruce A. Leicher, Associate General Counsel of GI, Def. Exh. 31. However, this does not put to rest all issues with respect to Roche’s activities and potential liability for infringement post-injunction. Indeed, a number of questions remain open which, at this early stage in the trial process, I cannot dismiss as insubstantial: (1) what Roche did with the original batch of production clone cells shipped from GI in 1986 and whether that batch was capable of producing bulk EPO or required GI-Roche collaboration; (2) whether GI has continued to maintain the original cell-batch in the United States, and, if it has, whether those actions violate the Amgen injunction and infringe the Columbia patents; 20 and, (3) the significance, if any, of the purportedly missing documents and whether those documents bear on the extent of GI and Roche’s EPO-related activities after 1991. 21 (I am certainly not persuaded at this juncture that it is reasonable to infer from the destruction of documents by Roche — apparently a unilateral act of one employee unrelated to this case — that GI and Roche must have engaged in activities of direct infringement after 1991.)

Finally, questions have been raised with respect to the importation of serum-free EPO into the United States. 22 There is evidence to support Columbia’s contention that Roche used the original cell line to reformulate, in Germany, a limited amount of EPO which it then shipped to GI in the United States. Specifically, Roche import ed into the United States small shipments of serum-free EPO for use by Jehovah’s Witnesses who eschew, for religious reasons, medicines derived from a base of animal blood (i.e. serum-based). In fact, in May of 1991, GI and Roche modified their earlier D & L Agreement, agreeing specifically to collaborate on the development of serum-free EPO. GI was to be paid in excess of $5 million for their participation in this venture.

On July 12, 1993, Columbia brought the present action against Roche claiming infringement of the Axel patents. 23 Columbia claims that Roche has sold more than $1.5 billion of EPO since 1990 and paid to GI for research, development, and bulk production of EPO more than $110 million in royalties. 24 In the meantime, Columbia has licensed the Axel patents to pharmaceutical companies worldwide, generating in excess of $450 million in royalties that have been channeled to further education and research for a wide of range of other projects. To date, Roche has not compensated Columbia. 25

IV. LEGAL DISCUSSION

A. Infringement of the Three Patents 26

1. Direct Infringement under 35 U.S.C. § 271(a)

To prevail on its claim of direct infringement under 35 U.S.C. § 271(a), Columbia must demonstrate that Roche made, used, offered to sell, or sold within the United States any patented invention during the term of the patent. Liability for direct infringement can also accrue from the importation of a patented invention into the United States.

It is clear from the statutory language of 35 U.S.C. § 271(a) that a finding of direct patent infringement requires sufficient connection with U.S. territory such that the infringement could be construed as having occurred within the United States. With this in mind, I must establish whether Roche’s activities in the United States are sufficient to establish a pattern of direct infringement — either through its domestic activities (manufacture or sale) or its foreign ones (importation into U.S. domestic market of inventions protected under American patent law).

To begin with, the record before me is devoid of any indication that Roche ever made, sold, or offered to sell in the United States cotransformed cells, methods of cotransformation (including amplifying DNA), or methods of obtaining protein expressed in cotransformed cells within the United States. 27 To the extent that these events did in fact occur, they occurred exclusively in Germany (the appropriate situs of jurisdiction under governing German patent laws), and, thus, are outside the scope of American patent protection. E.g. Dowagiac Mfg. Co. v. Minnesota Moline Plow Co., 235 U.S. 641, 650, 35 S.Ct. 221, 59 L.Ed. 398 (1915) (“[t]he right conferred by a patent under our law is confined to the United States and its territories, and infringement of this right cannot be predicated of acts wholly done in a foreign country”). Plainly, if there was any direct infringement of the Axel patents in the U.S. under § 271(a) it was committed exclusively by GI, the only actor alleged to have made, used, offered to sell, or sold products that infringe the Axel patents within the United States.

Nevertheless, Columbia maintains that Roche is liable under section 271(a) because (1) an American company (GI) owns the cell lines used by Roche in Germany, and, (2) Roche imported into the United States serum-free EPO, a byproduct of the Axel patents. However, I am not persuaded that Columbia may prevail under § 271(a) on these novel and abstract theories of liability — one suggesting that ownership status transcends geographical boundaries, the other crafting a “fruit of the poisonous tree” analog in patent law.

Columbia’s first argument proceeds as follows:

(1) The cell line in question was physically located in Germany and used exclusively in Germany to generate EPO;

(2) GI owns this cell line;

(3) Owned by an American company, the cells should be considered “American” in the abstract; 28

(4) Roche pays “rent” for its possession of the “American” cell line in Germany in the form of royalties; therefore,

(5) Roche’s use of the “American” cell lines to generate EPO directly infringes the Axel patents in violation of § 271(a) because the cells are “in the United States as a matter of law.” 29

In essence, Columbia’s conception of liability equates infringing activities that occur abroad with infringing activities that occur within the U.S. However, this reading simply does not accord with the plain reading of the statute. Section 271(a) refers explicitly to infringing activities that occur within the United States. The mere ownership by a domestic actor of the products used to infringe abroad does not relocate the infringing activities to U.S. soil. 30 See Deepsouth Packing Co. v. Laitram Corp., 406 U.S. 518, 531, 92 S.Ct. 1700, 32 L.Ed.2d 273 (1972) (“Our patent system makes no claim to extraterritorial effect ... to the degree that the inventor needs protection in markets other than those of this country, the wording of 35 U.S.C. §§ 154 and 271 reveals a congressional intent to have him seek it abroad through patents secured in countries where his goods are being used.”); Johns Hopkins University v. CellPro, Inc., 152 F.3d 1342, 1366 (Fed.Cir.1998) (neither export from the United States nor use in a foreign country of a product covered by a United States patent constitutes infringement).

Likewise, Columbia’s suggestion that section 271(a) liability extends to “fruits of the poisonous tree” simply does not comport with the unequivocal terms of the statute. 31 Section 271(a) makes reference to patented invention, not to by-products that derive from that invention. Thus, even were I to accept that Roche did in fact import into the U.S. serum-free EPO, Roche would not be liable under § 271(a) because the EPO itself is not the “patented invention.”

In any event, to some extent a different statutory provision already incorporates a version of Columbia’s “fruit of the poisonous tree.” As a result of the Process Patent Amendments Act of 1988, Publ. L. No. 100-418, § 9006, 102 Stat 1156 (1988) it is an infringement to import into the United States, a product which is made outside the country by a process patented in the United States. 32 Codified at 35 U.S.C. § 271(g), the provision states:

Whoever without authority imports into the United States or offers to sell, sells, or uses within the United States a product which is made by a process patented in the United States shall be liable as an infringer, if the importation, offer to sell, sale, or use of the product occurs during the term of such process patent ... (emphasis added).

If Columbia can successfully demonstrate that Roche imported into the U.S. the serum-free product that was made by a process protected under the Axel patents, they may prevail under this theory of liability, but not under 35 U.S.C. § 271(a).

The record before me does not resolve the issue one way or the other. It does not make clear whether the EPO generating CHO host cells, produced by GI and shipped to Roche in March of 1986, were used to produce the serum-free EPO that was imported into the United States, thus violating § 271(g). 33 All that is clear is that there is no colorable theory of liability under 35 U.S.C. § 271(a). Accordingly, I GRANT summary judgment in favor of Roche on the issue of liability under § 271(a).

2. Infringement under 35 U.S.C. § 271(f)

As a result of the Patent Law Amendments Act of 1984, Pub.L. No.98-622 § 101(a), 98 Stat. 3383, it is an infringement to supply within the United States, or to export from the United States, all or a substantial portion of the uncombined components of a patented invention in such a manner as to actively induce a combination of such components outside the United States that would infringe the patent if such combination occurred within the United States. 35 U.S.C. § 271(f). 34

Section 271(f) was enacted in response to the Deepsouth case, 406 U.S. 518, 92 S.Ct. 1700, 32 L.Ed.2d 273, in which an accused infringer exported the unpatented components of a shrimp de-veining machine that were then assembled outside the country into an allegedly infringing device. The Supreme Court held that these actions did not constitute an infringement because the patent did not eover the components but rather the fully assembled combination machine which was never within U.S. territory. § 271(f) was enacted by Congress to close this loophole.

Columbia claims the export of the bulk EPO and the EPO-producing cell lines from GI to Roche in Germany violated § 271(f). Columbia maintains that the EPO and the cell line are the functional equivalent of components for purposes of liability because they each contain substantially all the components needed to manufacture the EPO protein and create new cells and products.

I disagree. Section 271(f) prevents companies from circumventing the U.S. patent laws by exporting non infringing components to be assembled abroad into an infringing final product. Here, the EPO and EPO-generating cell line are not themselves components of any patented invention or process at issue. Moreover, Roche did not incorporate the EPO and EPO-producing cells into a finished assembly that itself would infringe the Axel patents but for extraterritoriality. 35

Thus, to accept Columbia’s position in this case would be to equate a subpart (the component) with the whole (the complete patented invention), and redefine “assem bly” to cover “any use” of the component. In other words, Columbia would have this Court accept the proposition that the exported product infringes both prior to assembly (as patented inventions that infringe independently under § 271(a)) and after assembly (as separate parts that infringe once incorporated into the ultimate product). This viewpoint, however, misinterprets the scope of the statute which expressly distinguishes between the constituent elements (i.e. building blocks) that do not infringe and the finished, assembled product that does.

Accordingly, summary judgment is GRANTED in favor of Roche on the issue of liability under § 271(f). See Standard Havens Products, Inc. v. Gencor Indus., Inc., 953 F.2d 1360, 1374 (Fed.Cir.1991) (foreign sales of a machine which used a patented asphalt-making process did not implicate § 271(f) because no components were involved); Aerogroup Int’l, Inc. v. Marlboro Footworks, Ltd., 955 F.Supp. 220, 231 (S.D.N.Y.1997) (a design patent for a shoe sole had no component parts to assemble, and therefore was beyond the scope of § 271(f)).

3. Induced Infringement under 35 U.S.C. § 271(b)

Unlike the statutory sections defining direct infringement, § 271(b) does not limit its scope to activity within the United States; rather, extraterritorial activity that induces direct infringement within the United States may incur liability. 36 Hauni Werke Koerber & Co. v. Molins, Ltd., 183 U.S.P.Q. (BNA) 168, 170 (E.D.Va.1974). The linchpin for § 271(b) liability is an actor’s specific intent to induce — i.e. the knowledge that the product in question infringes on the patent of another — and the steps undertaken actively to encourage or advance the infringement. 37 See Hewlett-Packard Co. v. Bausch & Lomb, Inc., 909 F.2d 1464, 1469 (Fed.Cir.1990) (to sustain a finding of induced infringement, “proof of actual intent to cause the acts which constitute the infringement is a necessary predicate”). Direct evidence is not required to prove actual intent; circumstantial evidence may suffice. Water Technologies Corp. v. Calco. Ltd., 850 F.2d 660, 668 (Fed.Cir.1988), (citing, Moleculon Research Corp. v. CBS Inc., 793 F.2d 1261, 1272 (Fed.Cir.1986)).

However, if a party exercises no meaningful control over the production of the infringing item, it can support a finding that it lacked actual intent to induce infringement. See Ardco, Inc. v. Page, Ricker, Felson Marketing, Inc., 25 U.S.P.Q.2d 1382, 1385 (N.D.Ill.1992) (where defendant did not exercise any control over the creation, the manufacture, or the sale of the patented device, the relationship is too peripheral to sustain a finding of inducement); Hewlett-Packard Co., 909 F.2d at 1470 (finding no intent to induce where the seller of infringing assets had “no interest or control” over how the buyer used the assets after sale); Water Technologies, 850 F.2d at 668 (holding that a defendant who provided an infringing component, helped the manufacturer of the final product, and conditioned the sale of the final product on his approval, did induce direct infringement). Thus, the relevant issue with regard to Roche’s liability for active inducement centers on two inquiries: Roche’s cognizance of Columbia’s patent ownership and the nature of Roche’s relationship with GI.

Roche represents that GI exerted 100% control over the making of its EPO production system and had completed developing the processes of cotransformation and amplification before Roche had ever received the DN2-3alpha3 clone in March of 1986. Furthermore, Roche characterizes the D & L Agreement merely as a traditional sale and licensing arrangement for a commodity.

In support of their argument, Roche places particular emphasis on Amgen, Inc. v. Elanex Pharmaceuticals, Inc., 1996 WL 84590 (W.D.Wash.1996). In that case, Elanex was in possession of an EPO-producing clone and was seeking partners to market EPO outside the United States. Elanex entered into an agreement with Merckle, a German corporation, under which Merckle would market the EPO produced by Elanex in European markets.

The district court found that Elanex had directly infringed Amgen’s patents, but found that Merckle did not induce infringement because Elanex retained control over the technology and had the technology in place prior to negotiating with Merckle. The court reasoned that Merck-le’s actions:

show a lack of specific intent to induce infringement. Prior to negotiating its distribution agreement with Merckle, Elanex had obtained its technology, had filed for various patent applications, and had developed a plan to commercialize and market its EPO. Elanex solicited Merckle to market EPO in three countries in Europe. The contract provided that the EPO supplied would be manufactured in Germany by an independent supplier. Under its agreement with Merckle, Elanex retained control over the technology, production of EPO, and its contract supplier. Id. at *6.

The facts of this case are distinguishable at this stage for a number of reasons. The language of the D & L Agreement suggests that the parties negotiated a highly integrated collaboration. 38 As previously discussed, provisions of the Agreement provide that both parties were to share in the development of the technology, jointly own the fruits of the relationship, and retain comprehensive rights of inspection and access to confidential information. 39 As this Court previously noted:

... BMG’s D & L Agreement was not merely a contract to purchase goods in Massachusetts for delivery outside of Massachusetts. Rather it created, by its own wording, a “collaboration” between the two companies. Although it appears that only GI personnel actually performed experimental or production work in Massachusetts, BMG’s connection with that work was more intimate than that of mere customer. BMG was the principal underwriter of the research in question. If the research produced valuable technology, BMG was to have an exclusive license to use the technology outside the United States. Moreover, BMG retained, under the D & L agreement, the right to prosecute foreign patent applications on any technology developed by GI which GI failed to prosecute itself. Trustees of Columbia University, 35 U.S.P.Q.2d at 1368.

Moreover, the terms of the D & L Agreement specifically entertain the possibility that Roche’s use of GI’s manufacturing process might infringe certain third-party patents. 40 A provision of the D & L Agreement permits Roche to deduct from its royalty obligations to GI the cost of licensing payments that Roche might indemnify to third parties in order to legally produce or sell EPO. The provision, in relevant part, reads:

The Parties recognize that the undertaking of the Project and the marketing of Licensed Products involves some degree of risks of patent infringement ... BM may deduct from royalties ... payments (including without limitation royalties, option fees or license fees) to one or more third parties to obtain a license or similar right in the absence of which the Licensed Product or Licensed Compound could not legally be manufactured or sold. See Def. Exh. 20 at 7.6.

Furthermore, there is a genuine issue of material fact concerning whether the EPO clone developed by GI prior to the D & L Agreement was the same production clone used by GI to make EPO for Roche. The deposition of Dr. Fritsch (suggesting that GI had not developed an EPO clone to produce commercial product prior to the D & L Agreement), GI’s Quarterly Report for the First Quarter of 1985 (suggesting that GI had yet to perfect a suitable CHO expression system or cell line for producing EPO), and the fact that no witness has testified to support Roche’s characterization of GI’s progresses in expression, co-transformation, and amplification by the time the D & L Agreement was executed all support Columbia’s claim that GI’s infringing processes were not completed before Roche received the DN2-3P23, 10 E2M MTX production clone in March of 1986. 41 At the very least, there remain open questions on this record as to the stage of GI’s EPO development at the time the GI-Roche alliance started and the extent of.Roche’s knowing participation in facilitating GI’s allegedly, infringing progress.

However, whether triable issues of fact remain with respect to induced infringement of the ’017 patent is another matter. I am persuaded by Roche’s argument that Roche could not have induced infringement as a matter of law because GI was enjoined from making, using, or selling EPO in 1991, two years before the issuance of the ’017 patent in 1993. As liability under 35 U.S.C. § 271 may be predicated only on post-issuance activities, 42 National Presto Industries, Inc. v. West Bend Co., 76 F.3d 1185, 1196 (Fed.Cir.1996), Roche cannot be liable for inducing infringement of an as yet “non-existent” patent. Micro Chemical, Inc. v. Great Plains Chemical Co., Inc., 103 F.3d 1538, 1549 (Fed.Cir.1997).

Nevertheless, while pre-issuance activities are not relevant here, I find that there is some contested evidence of direct infringement by GI after 1993. The record suggests that, though having been enjoined from EPO production before the ’017 patent issued, GI continued to maintain the cell-bank on Roche’s behalf. By analogy, where use of a patented device constitutes an infringement under § 271(a), courts have found that the maintenance or repair of the device likewise constitutes an infringement. E.g. Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U.S. 476, 484-85, 84 S.Ct. 1526, 12 L.Ed.2d 457 (1964) (the replacement of worn-out fabric components in unlicensed General Motors cars was impermissible because “[w]here use infringes, repair does also, for it perpetuates the infringing use”); Hughes Aircraft Co. v. United States, 29 Fed.Cl. 197, 29 U.S.P.Q.2d 1974 (1993) (launch of spacecraft that embodied patented device was an infringing use even though device was not activated until after spacecraft was in space); Olsson v. United States, 87 Ct.Cl. 642, 25 F.Supp. 495, 498 (1938) (possession and maintenance of disassembled howitzers was infringing use); Remington Rand Business Serv., Inc. v. ACME Card System Co., 71 F.2d 628, 630 (C.C.A.4 1934) (finding patent infringement where the defendant kept certain protected frames in good repair and aided in the use of the index strips which they were especially designed to hold).

While these post-issuance activities are not overwhelming, they are enough to de feat summary judgment. Accordingly, I DENY Roche’s motions for summary judgment on the question of liability under 35 U.S.C. § 271(b).

B. Non-itifi'ingement of Claims to Unlinked DNA

In a separate Motion for Summary Judgnent of Non-Infringement of Claims to Unlinked DNA Embodiments, Roche argues that claims 1, 2, 9, 10, 11, 14, 15, 18-22, 24-28, 30-32, 39, 41, 42, 44-51, and 53 of ’216 patent, and claims 1, 3-5, 7, 10-18, and 21-23 of the ’665 patent should be construed to preclude a finding of infringement because they pertain only to the use of cotransformed cells with “unlinked” DNA. GI’s activities, Roche maintains, involved only “linked” DNA. Specifically, Roche maintains: (1) that liability for the infringement of claims involving “unlinked” DNA is foreclosed by this Court’s Markman findings, and, (2) there should be no infringement of claims to “unlinked” DNA under the doctrine of equivalents because that subject matter was surrendered by Columbia during its prosecution of the Axel patents.

In suggesting that cotransformation claims involving unlinked DNA and claims that recite cotransformation involving linked DNA are mutually exclusive, Roche overstates my Markman findings. Indeed, nothing in the language of the opinion was intended to suggest that claims which refer to linked DNA per se exclude unlinked DNA. 43

The infringing acts alleged in this dispute include the use of two unlinked DNAs — a DNA I encoding EPO and a DNA II encoding DHFR — that were joined into the same piece of contiguous DNA by cotransformation prior to their insertion into the eucaryotic cell. As an initial step in interpreting the scope of Columbia’s claims, this Court defined, literally, “linked” as “physically and chemically joining DNA I and DNA II into the same piece of contiguous DNA prior to their insertion into the eucaryotic Mammalian Cell” and “unlinked” as “not physically or chemically linked on the same piece of contiguous DNA.” See Trustees of Columbia University in City of New York, 126 F.Supp.2d at 32-33.

Conceptually, however, both “linked co-transformation claims” and “unlinked co-transformation claims” are directed, in essence, to the same invention: the insertion of DNA I and DNA II (linked or unlinked) into the host cell, with DNA II coding for a selectable phenotype. The thrust of the Markman finding was that the principal difference between the two claims lies not in the linkage issue but rather in the role of amplification. See Trustees of Columbia University in City of New York, 126 F.Supp.2d at 32-33.

For example, claim 1 of the ’216 patent covers two cases: DNA I and DNA II unlinked prior to cotransformation and subsequently either (1) joined prior to insertion or (2) not joined prior to insertion. 44 Either could then be amplified un der the patent. In this sense, claim 54 of the ’216 patent may be viewed as a specific subcase of the more generic claim 1 variety, involving DNA I and II linked prior to insertion, followed by an amplification step. 45

Because I find that subject matter that is alleged to infringe claims to linked DNA may literally infringe claims drawn to unlinked DNA, I need not resolve whether Columbia is estopped from making an argument under the “doctrine of equivalents,” 46 because of its activities during prosecution of the patents. 47 Accordingly, having determined that there are disputed issues of material fact with respect to whether GI’s infringing acts literally infringe claims that recite unlinked DNA, I DENY Roche’s Motion for Summary Judgment of Non-Infringement of Claims to Unlinked DNA Embodiments.

V. CONCLUSION

To summarize, because the record before me indicates no evidence of direct infringement on the part of Roche (discounting vicariously through GI) within the United States, summary judgment is GRANTED in favor of Roche on the issue of liability under 35 U.S.C. § 271(a). Furthermore, because the record before indicates no evidence of Roche exporting non-infringing components to be assembled abroad into an infringing final product, summary judgment is GRANTED in favor of Roche on the issue of liability under 35 U.S.C. § 271(f).

However, insofar as Columbia has raised disputed issues of material fact regarding the scope of Roche’s involvement in GI’s pattern of direct infringement, summary judgment is DENIED on the issue of liability under 35 U.S.C. § 271(b). Finally, in light of the language of the Axel patents and this Court’s Markman decision, subject matter that is alleged to infringe claims to linked DNA can literally infringe claims drawn to unlinked DNA; accordingly, summary judgment is DENIED on the issue of non-infringement of claims to unlinked DNA embodiments.

SO ORDERED.

1 . EPO is useful in treating end stage renal disease.

2 . The '216 patent was issued August 16, 1983.

3 . The '665 patent was issued January 6, 1987.

4 . The '017 patent was issued January 12, 1993.

5 . Columbia initially brought this action against Roche for infringement of the '216 patent and the '665 patent, both entitled "Processes for Inserting DNA into Eucaryotic cells and for Producing Proteinaceous Materials.” Columbia added a claim of infringement of the third patent with the same title, the '017 patent, in its Second Amended complaint.

6 . The inventors named on all three patents are Drs. Richard Axel, Saul Silverstein, and Michael Wigler (hence "the Axel patents”).

7 . From Markman v. Westview Instruments, Inc., 52 F.3d 967, 979 (Fed.Cir.1995) (en banc), aff’d, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996).

8 . § 271(a) provides:

Except as otherwise provided in this title, whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United Stales or imports into the U.S. any patented invention during the term of the patent therefor, infringes the patent.

9 . § 271(b) provides: "Whoever actively induces infringement of a patent shall be liable as an infringer.”

10 . § 271(f) provides:

(1) Whoever without authority supplies or causes to be supplied in or from the United states all or a substantial portion of the components of a patented invention, where such components are uncombined in whole or in part, in such manner as to actively induce the combination of such components outside the U.S. in a manner that would infringe the patent if such combination occurred within the U.S., shall be liable as an infringer.;

(2) Whoever without authority supplies or causes to be supplied in or from the United States any component of a patented invention that is especially made or especially adapted for use in tire invention and not a staple article or commodity of commerce suitable for substantial non infringing use, where such component is uncombined in whole or in part, knowing that such component is so made or adapted and intending that such component will be combined outside of the U.S. in a manner that would infringe the patent if such combination occurred within the U.S., shall be liable as an infringer.

11 . I emphasize that summary judgment is no less appropriate in a patent case than in any other action. See Avia Group Int’l, Inc. v. L.A. Gear Calif., Inc., 853 F.2d 1557, 1561 (Fed. Cir.1988)

12 . The parties concede that prior to being solicited by GI, Roche had shown no interest in EPO.

13 . While Columbia claims that GI was generating EPO only from unamplified CHO host cells prior to the its collaboration with Roche, Roche counters that GI had been producing the amplified host cells before the collaboration and simply forgot to include the ''alpha” symbol for amplification in its patent applica tion. See PL Exh. H, pp. 58-60; Def. Exh. 6, May 19 & 20 Deposition of Dr. Kaufman.

14 . Columbia maintains that the fact that under the D & L Agreement GI was first to deliver non-commercial materials to Roche, and then, at a later date, to deliver commercially feasible materials, suggests that when the D & L Agreement was executed, a commercially feasible clone had not yet been developed by GI.

15 .Columbia contests that the shipped cells were amplified and capable of producing EPO. Katherine Evans, the GI Project Manager of EPO development, testified that she was not certain if the cell banks shipped to Roche were capable of producing EPO. See March 29, 1999 Deposition of Katherine Evans, Def. Exh. 25, p. 32. The implication is that the development of amplified cells must have occurred after the D & L Agreement and as a result of a joint GI-Roche venture.

16 . The bailed cells were made in the United States by GI after the issuance of Columbia’s '216 and '665 patents.

17 . It is not disputed that GI retained ownership over the additional bailed production clone cells. See Plaintiff's Response to Roche's Statement of Undisputed Facts, ¶ 25.

18 . After this shipment, an additional six vials remained in the GI freezer in Andover, Massachusetts, the central storage place for vials of the DN2-3alpha3 cell line.

19 . In addition, GI sent two separate shipments of vials to Roche in March of 1986. Though there is no record of these vials having ever been returned to GI, I need not address the earlier shipment any further because there is no indication that they contained the DN2-3P23 production clone at issue.

20 . Columbia contends that affirmative steps must be taken to maintain cells in a viable state, such as monitoring temperatures and replenishing liquid nitrogen.

21 . Doctor Lothar Wieczorek, Roche’s Vice President of International Project Management since 1993, destroyed a series of correspondences between GI and Roche concerning its EPO projects. See August 26, 1996 Affidavit of Dr. Wieczorek, PI. Exh. M I wish to emphasize that Columbia is not alleging— nor is there any indication — of bad faith on the part of Dr. Wieczorek. The relevant documents, along with roughly thirty boxes of files relating to "dead” projects, were discarded during a routine move between offices in February of 1995. Roche, however, claims that by he end of 1996, Columbia was given access to roughly 140,000 documents, including originals or copies of the allegedly missing documents pertinent to the GI/Roche EPO related activities.

22 .Serum-free EPO refers to erythropoietin produced out of a process which did not utilize serum in its manufacturing process. Serum-free EPO was developed for European markets due to concerns related to mad cow disease that might be connected with a serum-containing process. See March 3, 1999 Deposition of Tuan Ha-Ngoc, DeL Exh. J at 111-113.

23 . Unlike Columbia’s original suit against GI, this action includes the '017 patent, issued in January of 1993.

24 . See also Amgen, Inc. v. Chugai Pharmaceutical Co., Ltd., 808 F.Supp. 894, 897 (D.Mass.1992) (estimating GI's sales of EPO to Roche/ BMG to be in excess of $34,000,000)

25 . Columbia estimates that it lost over $90 million in royalties from Roche's infringing activities.

26 . Roche has filed a motion for summary judgment alleging non-infringement of all ihree Axel Patents [docket entry # 250] and a separate motion for summary judgment that focuses specifically on Roche's non-infringement of the '017 patent [docket entry #265]. As the arguments raised overlap quite substantially, I consider them together for purposes of clarity and simplicity.

27 . As per my Markman findings, this summarizes broadly the coverage of claims related to the Axel patents.

28 . Columbia has actually dubbed the cells used by Roche “American cells.” See Plaintiffs Memorandum in Opposition to Defendant's Motion for Summary Judgment of Non-Infringement Under 35 U.S.C. § 271 at 17.

29 . See id.

30 . To be honest, I cannot foresee any finite limitation to the application of Columbia’s logic that nationality is somehow intrinsic and transcends physical boundaries. For example, one could imagine a scenario where a state might invoke this theory to exert personal jurisdiction over a foreign corporation or citizen with no sustained American contacts simply for having committed a tortious act abroad with a car owned by an American, or a gun manufactured by an American corporation, or a newspaper printed in the American tongue, i.e. instrumentalities with a nebulously "American” imprimatur.

31 . The notion being that liability extends to the byproducts of infringing acts much as the taint of evidence discovered as a result of an unconstitutional search or seizure extends to future evidentiary yields.

32 . Prior to the enactment of the 1988 statute, a patentee holding a process patent could sue for infringement if others used the process in this country, but had no cause of action if such persons used the patented process abroad to manufacture and then import, use, or sell its products in this country. Congress redressed this gap by making it an act of infringement to import into the United States, or to sell or use within the United States "a product which is made by a process patented in the United States ... if the importation, sale, or use of the product occurs during the term of such process patent.” 35 U.S.C. § 271(g).

33 .Columbia implies in its Plaintiff's Memorandum in Opposition to Defendant’s Motion for Summary Judgment of Non-Infringement Under 35 U.S.C. § 271 at 18 that this Court previously determined that Roche imported into the United States the serum-free EPO at issue. However, Columbia misreads my previous decision in which I acknowledged that "in 1988, BMG agreed to supply GI in Massachusetts with EPO which it had reformulated in Germany” and that "BMG and GI agreed to collaborate on the development of ‘serum-free’ EPO.” Trustees of Columbia University in City of New York v. Mannheim, 35 U.S.P.Q.2d 1364, 1365 (D.Mass.1995) (emphasis added). However, I never made factual findings to the extent that Roche did in fact import the serum-free EPO into the U.S.

34 . 35 U.S.C. § 271(f)(1) applies to the supply of "all or a substantial portion of the components of a patented invention,” while § 271(f)(2) applies to the supply of "any component of a patented invention that is especially made or adapted for use in the invention and not a staple article or commodity of commerce suitable for substantial noninfring-ing use." The latter provision requires the infringer to have knowledge that the component is specially made or adopted and an intent that the components will be combined outside of the United States in a manner that would infringe if the combination occurred within the United States.

35 . It is true, as Columbia argues, that courts have extended section 271(f) beyond machines and other structural combinations to cover chemical components of chemical compositions. E.g. W.R. Grace & Co.-Conn. v. Intercat, Inc., 60 F.Supp.2d 316, 319 (D.Del.1999); Lubrizol Corp. v. Exxon Corp., 696 F.Supp. 302, 325 (N.D.Ohio 1988). Nevertheless, though I agree with the rationale behind extending § 271(f) to chemical compounds (the statutory language does not limit coverage to components of machines, nor would it make sense to have done so in light of the Deepsouth loophole), I do not see how the chemicals in this case — the bulk EPO and the EPO cell lines — are to be considered non-infringing constituents of a greater, infringing compound.

36 . Prior to the enactment of 35 U.S.C. § 271 (the Patent Act of 1952), there was no statute that defined what constituted infringement. Infringement was divided under the common law into "direct infringement” (the unauthorized making, using or selling of the patented invention) and “contributory infringement” (a theory of joint tortfeasance, where an actor, though not technically making, using or selling a patented invention, nevertheless displayed sufficient culpability to be held liable as an infringer). See, e.g., Henry v. A.B. Dick Co., 224 U.S. 1, 33-34, 32 S.Ct. 364, 56 L.Ed. 645 (1912); Tubular Rivet & Stud Co. v. O'Brien, 93 F. 200, 202-05 (C.C.D.Mass.1898). However, 35 U.S.C. § 271 divided the generic concept of “contributory infringement” between section 271(b) (active inducement) and section 271(c) (contributory infringement). See S.Rep. No. 1979, 82d Cong., 2d Sess. 8, 28 (1952), U.S.Code Cong. & Admin. News 1952, p. 2394.

37 . There is an analogous theory of liability in copyright law: One who knowingly aids, induces, or contributes to a copyright infringement by another, but who has not himself committed or participated in the infringing acts, may be held liable as a "contributory” infringer without violating the doctrine of no-nextraterritoriality. Armstrong v. Virgin Records, Ltd., 91 F.Supp.2d 628, 635-636 (S.D.N.Y.2000) (A foreign defendant in an infringement action may be held liable for domestic acts of infringement if that defendant is either contributorily or vicariously liable for another’s direct act of infringement); Blue Ribbon Pet Prods., Inc. v. Rolf C. Hagen (USA) Corp., 66 F.Supp.2d 454, 461-63 (E.D.N.Y. 1999) (holding that acts in Canada could provide the basis for liability under U.S. copyright law because “a defendant can be liable for contributory infringement, even for acts committed outside the United States, by inducing or contributing to another's infringement occurring in the United States ... provided the defendant knew or should have known that the other would or could reasonably be expected to commit the infringement”); ITSI T.V. Prods., Inc. v. California Auth. of Racing Fairs, 785 F.Supp. 854, 864 (E.D.Cal.1992) (noting that “it is possible for a defendant to commit acts outside the United States sufficient to find it contributorily or vicariously liable for acts of infringement committed by others within the United States”), rev'd on other grounds, ITSI T.V. Prods., Inc. v. Agrie. Ass'n, 3 F.3d 1289 (9th Cir.1993); GB Marketing USA Inc. v. Gerolsteiner Brunnen GmbH & Co., 782 F.Supp. 763, 773 (W.D.N.Y.1991) (assuming jurisdiction on a theory of contributory infringement over acts of copying in Germany that were undertaken with the "knowledge and intent” that the infringing products would be exported to the United States by the direct infringer in violation of the copyright owner's importation right). See also Metzke v. May Dept. Stores Co., 878 F.Supp. 756 (W.D.Pa.1995) (holding that defendant could be liable where it supplied samples of the product at issue to copier in Taiwan if it knew or should have known that unauthorized copies made in Taiwan would be distributed in the United States).

38 . Notably, in its Amgen opinion, the district court did not cite specific provisions of the Elanex-Merclde agreement.

39 . By its own terms, the D & L Agreement uses language of collaboration:

BM desires that GI, on behalf of and in collaboration with BM, undertake a research and development project utilizing recombinant DNA technology for producing erythropoietin on a commercially feasible basis for use in humans. In return for certain rights under the patents and know-how developed by GI, BM will financially support the research and development activities of GI and will pay GI the royalties provided herein (emphasis added).

40 . This fact is particularly significant to Columbia's inducement theory of liability under § 271(b), because it tends to corroborate that Roche had actual knowledge that GI’s activities might infringe the patented invention of another.

41 . Furthermore, one of the benchmarks of the D & L Agreement provides that GI was initially to deliver non-commercial materials to Roche and later to deliver commercially feasible ones (presumably only after developing an effective piocess).

42 . To be sure, pre-issuance activities may be actionable under other legal theories of the common law, such as fraud or unfair competition; however, liability for induced infringement exists only as a matter of statute under 35 U.S.C. § 271(b).

43 . In fact, the nature of this dispute suggests that conducting Markman hearings out of the context of specific summary judgment allegations is not at all advisable.

44 . Claim 1 of the '216 patent states:

A process for inserting foreign DNA I into a suitable eucaryotic cell which comprises co-transforming said eucaryotic cell with said foreign DNA I and with unlinked foreign DNA II which codes for a selectable phenotype not expressed by said eucaryotic cell, said cotransformation being carried out under suitable conditions permitting survival or identification of eucaryotic cells which have acquired said selectable phenotype, said foreign DNA I being incorporated into the chromosomal DNA of said eucaryotic cell. .

45 . Claim 54 of the '216 patent states:

A process for generating a multiplicity of foreign DNA I molecules corresponding to multiple copies of a gene in a cell with a molecule which comprises transforming said eucaryotic cell with a molecule which is formed by linking one of said DNA I molecules to a DNA II molecule corresponding to an amplifiable gene for a dominant selectable phenotype not expressed by said eucaryotic cell, and culturing the transformed eucaryotic cells in the presence of successively elevated concentrations of an agent permitting survival or identification of eucaryotic cells which have acquired multiple copies of said amplifiable gene, said transformation and culturing being carried out under suitable conditions.

46 . Under the doctrine of equivalents, a paten-tee may prevail in an infringement action against a producer of a device that performs the equivalent function as a protected device, even though the product does not literally infringe the patented invention. In other words, the theory prevents an accused in-fringer from avoiding liability simply by changing minor details of a claimed invention. See Royal Typewriter Co. v. Remington Rand, Inc., 168 F.2d 691, 692 (2d Cir.1948) (L.Hand, J.) (Patents are “entitled to its benefit to an extent, measured on the one hand by their contribution to the art, and on the other by the degree to which it is necessary to depart from the meaning to reach a just result”). By the same token, where a device is substantially changed to perform the same or a similar function in a very different way, but nevertheless falls within the literal words of the claim, the doctrine of equivalents may be used to restrict the claim and defeat the pat-entee's action for infringement. See Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 608, 70 S.Ct. 854, 94 L.Ed. 1097 (1950)

47 . Prosecution history estoppel precludes a patentee from obtaining protection under the doctrine of equivalents after surrendering, during prosecution, the right to claim particular matter as within the reach of the patent. See Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., Ltd., 234 F.3d 558, 564-65 (Fed.Cir.2000) (en banc). Roche argues that Columbia, during the prosecution of its patent application to the U.S. Patent and Trademark Office, clearly distinguished between claimed subject matter (directed to linked DNA) and unclaimed subject matter (directed to unlinked DNA).