Amgen, Inc. v. Hoechst Marion Roussel, Inc. , 287 F. Supp. 2d 126 ( 2003 )

287 F.Supp.2d 126 (2003)

AMGEN, INC., Plaintiff,
v.
HOECHST MARION ROUSSEL, INC. and Transkaryotic Therapies, Inc., Defendants.

No. CIV.A. 97-10814-WGY.

United States District Court, D. Massachusetts.

October 30, 2003.

D. Dennis Allegretti, Duane Morris LLP, Boston, MA, Linda A. Baxley, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Michael F. Borun, Marshall, O'Toole, Gerstein, Murray & Borun, Chicago, IL, Craig H. Casebeer, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Jane J. Choi, Marshall, O'Toole, Gerstein, Murray & Borun, Chicago, IL, Lloyd R. Day, Jr., Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Jennifer R. Dupre, Carr & Ferrell, Palo Alto, CA, Deborah E. Fishman, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Kevin M. Flowers, Marshall, O'Toole, Gerstein, Murray & Borun, Chicago, IL, Robert M. Galvin, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Michael R. Gottfried, Duane Morris LLP, Boston, MA, Douglass C. Hochstetler, Marshall, O'Toole, Gerstein, Murray & Borun, Chicago, IL, Ryan M. Kent, Day Casebeer Madrid Batchelder, Cupertino, CA, Patricia L. Leden, Jonathan D.J. Loeb, Day Casebeer Madrid & Batchelder, LLP, Cupertino, CA, David M. Madrid, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Mario Moore, Day Casebeer Madrid & Batchelder, Cupertino, CA, Jackie N. Nakamura, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Edward M. O'Toole, Sandip H. Patel, Marshall, O'Toole, Gerstein, Murray & Bourn, Chicago, IL, Elaine Stracker, Christopher E. Stretch, Day Casebeer Madrid Winters & Batchelder LLP, Cupertino, CA, Terry L. Tang, Courtney Towle, Day, Casebeer Madrid & Batchelder, Cupertino, CA, Richard M. Wong, Scansoft, Inc., Peabody, MA, for Amgen, Inc., Plaintiff.

Padmaja Chinta, Bindu Donovan, Russell W. Faegenburg, Gerald J. Flattmann, Fish & Neave, New York, NY, Robert S. Frank, Jr., Mark S. Freeman, Choate, Hall & Stewart, Boston, MA, Douglas J. Gilbert, James F. Haley, Jr., Kenneth B. Herman, Derek M. Kato, Peter J. Knudsen, Anna A. Kobilansky, Denise L. Loring, Fish & Neave, New York, NY, Peter C. McCabe, III, Winston & Strawn, Chicago, IL, Mark A. Michelson, Choate, Hall & Stewart, Boston, MA, Steven F. Molo, Winston & Strawn, Chicago, IL, Raymond C. Perkins, Abbott Laboratories, Abbott Park, IL, Melanie R. Rupert, Barbara A. Ruskin, Krista M. Rycroft, Herbert F. Schwartz, Ellen A. Scordino, Robert B. Wilson, Keith A. Zullow, Fish & Neave, New York, NY, for Hoechst Marion Roussel, Inc., Transkaryotic Therapies, Inc.

Edward DiLello, Darby & Darby, New York, NY, for Lonza Biologic.

MEMORANDUM AND ORDER

YOUNG, Chief Judge.

INTRODUCTORY NOTE: On September 18, 2003, this Court held that Amgen had successfully rebutted the presumption of prosecution history estoppel but noted that it might have to revisit the issue if the Federal Circuit's Festo opinion was published before this Court's opinion was finished. 9/18/03 Pretrial Conference Tr. at 7, ll. 17-25, and 8, ll. 1-5. This Court had scheduled release of this opinion for early October, 2003, pending final proofreading. The Federal Circuit released its opinion in Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., Ltd., 344 F.3d 1359 (Fed. Cir.2003) ("Festo III") on September 26, 2003. Because Festo III authoritatively interprets the Supreme Court's decision in Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., Ltd., 535 U.S. 722, 122 S.Ct. 1831, 152 L.Ed.2d 944 (2002) ("Festo II"), the recent Federal Circuit decision governs here under familiar stare decisis principles.

As this Court's interpretation of the Supreme Court decision had proceeded along a somewhat different tack, and since the growth of the law in this area "moves at a lightning pace," Festo III, 344 F.3d at 1375-76 (Rader, J., concurring), it seems appropriate to set forth this Court's original analysis of Festo II, noting its confirmation or remediation on a section-by-section basis by the Federal Circuit in Festo III, and identifying and setting forth the revised analysis required by that decision.

As will be discussed in further detail below, this Court reasserts its prior holding—in light of Festo III—that Amgen has successfully rebutted the presumption of prosecution history estoppel. Therefore, its earlier finding that HMR/TKT infringes claims 2 through 4 of U.S. Patent No. 5,621,080 (issued April 15, 1997) (the "``080 patent") is reaffirmed.

I. INTRODUCTION

In light of the recent Supreme Court decision in Festo II, 535 U.S. 722, 122 S.Ct. 1831, 152 L.Ed.2d 944, the Federal Circuit remanded to this Court its ruling in Amgen, Inc. v. Hoechst Marion Roussel, Inc., 126 F.Supp.2d 69 (D.Mass.2001) ("Amgen I"), that Amgen, Inc. ("Amgen") was not estopped by the prosecution history from claiming equivalent infringement by the human erythropoietin ("EPO") product ("HMR 4396") produced by Hoechst Marion Roussel, Inc.^[1] and Transkaryotic Therapies, Inc. (collectively "HMR/TKT"). Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313, 1345 (Fed.Cir.2003) ("Amgen II"). Based upon Festo II, the Federal Circuit ruled that a presumption of prosecution history estoppel applied here. Therefore, it vacated this Court's finding of equivalent infringement and remanded "for an analysis under the narrow ways of rebutting the Supreme Court's presumption of estoppel" as laid out in Festo II. Amgen II, 314 F.3d at 1345. This opinion addresses the discrete issue of whether Amgen has successfully rebutted the Festo II presumption of prosecution history estoppel held to be present in Amgen II. Id.

II. DISCUSSION

A. Background and Procedural History

In Amgen I, this Court construed the amendment "mature erythropoietin amino acid sequence of FIG. 6" in claims 2-4 of U.S. Patent No. 5,621,080 (issued April 15, 1997) (the "'080 patent") as requiring an EPO glycoprotein "comprising the fully realized erythropoietin amino acid sequence of Figure 6, which depicts 166 amino acids." Amgen I, 126 F.Supp.2d at 100. Pursuant to this construction, the Court found that HMR 4396 did not literally infringe claims 2, 3, and 4 of the '080 patent because HMR 4396 comprised only 165 amino acids. Id. at 101.

This Court then turned to the doctrine of equivalents and found that HMR 4396 performed substantially the same function in substantially the same way to obtain substantially the same result as the EPO glycoprotein of claims 2 and 3 of the '080 patent. Id. at 133. Therefore, it ruled that Amgen's '080 patent was equivalently infringed by TKT/HMR's product. Id.

In response to TKT/HMR's argument that Amgen should be estopped from arguing equivalent infringement, this Court ruled that prosecution history estoppel did not apply because Amgen did not add the "mature amino acid sequence of Figure 6" limitation "in an attempt to overcome a rejection, to avoid prior art," but, instead, to "demonstrate that ``same invention' type double patenting did not apply," that is, to distinguish the '080 patent from U.S. Patent No. 5,547,933 (the "'933 patent"). Id. at 134-35.

The Federal Circuit agreed that the amendment was made for this purpose but made clear that under Festo II, "a narrowing amendment to satisfy any requirement of the Patent Act may give rise to an estoppel." Amgen II, 314 F.3d at 1345 (emphasis added). Therefore, it held that the presumption of prosecution history estoppel applied here. The Federal Circuit then vacated this Court's finding of equivalent infringement and remanded for an analysis based on the "narrow ways of rebutting the Supreme Court's presumption of estoppel" outlined in Festo II. Id. at 1345. Because the Court's factual findings (that TKT/HMR's product performed substantially the same function in substantially the same way to obtain substantially the same result) were not disturbed—or even challenged—on appeal, the Court need only analyze whether Amgen can rebut the presumption of prosecution history estoppel and pursue an infringement claim based on the doctrine of equivalents. If Amgen can, the Court's ruling that Amgen's '080 patent was equivalently infringed shall be reaffirmed.

On May 16, 2003, Amgen moved for judgment under the Federal Rule of Civil Procedure 52(c) that claims 2-4 of the '080 patent are equivalently infringed [Document No. 659]. TKT/HMR opposed Amgen's motion and moved for judgment that Amgen is estopped from asserting infringement under the doctrine of equivalents [Document No. 678].^[2] Both parties, however, have agreed that the Court summarily can decide this motion based on the current record without further evidentiary presentation. TKT/HMR's Mem. in Opp'n at 1; Amgen's Mem. in Support at 5-6.

B. Overview Of Festo II As Interpreted By This Court And The Federal Circuit

The doctrine of equivalents was created because language is not always sufficiently precise to capture the essence, nuances, and range of an invention in light of unexpected technological advances. Festo II, 535 U.S. at 731, 122 S.Ct. 1831 ("Things are not made for the sake of words, but words for things." (quoting Autogiro Co. of America v. United States, 181 Ct.Cl. 55, 384 F.2d 391, 397 (1967))). The doctrine of equivalents was designed to provide patent applicants the wiggle room that claim language alone cannot, thus protecting patentees from copyists who make insubstantial changes. Festo II, 535 U.S. at 732-33, 122 S.Ct. 1831; Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 607, 70 S.Ct. 854, 94 L.Ed. 1097 (1950). In other words, it was designed to protect the intended invention, even if not adequately or ably described.

Prosecution history estoppel, on the other hand, arises when an amendment, by its language, discernibly limits the scope of the invention before the Patent and Trademark Office (the "Patent Office"). The purpose of prosecution history estoppel is to "hold the inventor to the representations made during the application process and the inferences that may reasonably be drawn from the amendment." Festo II, 535 U.S. at 737-39, 122 S.Ct. 1831.

The doctrine of equivalents is premised on language's inability to capture the essence of innovation, but a prior application describing the precise element at issue undercuts that premise. In that instance, the prosecution history has established that the inventor turned his attention to the subject matter in question, knew the words for both the broader and narrower claim, and affirmatively chose the latter.

Id. at 734-35, 122 S.Ct. 1831. Prosecution history estoppel, therefore, takes root in the virtues of public notice and the reduction of uncertainty that is sometimes created by the doctrine of equivalents. Tony Caliendo, Foreseeable Trouble: How Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co. Offends Fundamental Policies of the U.S. Patent System by Making Prosecution History Estoppel Depend Upon Foreseeability, 2003 BYU L.Rev. 309, 314 (2003). Competitors "rely on [prosecution history] estoppel to ensure that their own devices will not be found to infringe by equivalence." Festo II, 535 U.S. at 727, 122 S.Ct. 1831.

An obvious tension exists between the doctrine of equivalents and prosecution history estoppel. The doctrine of equivalents provides protection for more than just the literal claims, but this added protection creates uncertainty. Caliendo, supra, at 314. In other words, competitors are not provided notice of the breadth of the patent. Prosecution history estoppel, on the other hand, can provide such certainty and notice. The Supreme Court in Festo II tried to balance the competing interests of public notice and protection of patents by rejecting the Federal Circuit's complete bar approach^[3] and further defining which amendments give rise to a presumption of estoppel, what equivalents are subsequently barred by such estoppel, and how the "presumption" of estoppel can be overcome. Festo II, 535 U.S. at 736-42, 122 S.Ct. 1831.

1. How Does The "Presumption" Work?

a. The Court's Original Analysis: The "presumption" of estoppel is not really a presumption at all, but rather a burden-shifting device similar to the "presumption" of patent validity.

The Supreme Court in Festo II made clear that a "presumption" of prosecution history estoppel arises when an amendment is made to secure the patent and the amendment narrows its scope. Festo II, 535 U.S. at 736, 122 S.Ct. 1831. The "presumption," however, does not completely bar the inventor from asserting infringement against any equivalents to the narrowed element. Id. at 738-39, 122 S.Ct. 1831. The inventor can overcome the "presumption" by showing that the amendment does not surrender the particular equivalent in question. Id. at 740, 122 S.Ct. 1831. Thus, this "presumption" shifts to the inventor not only the burden of production but also the burden of persuasion to get out from under the "presumption" of prosecution history estoppel outlined in Festo II.^[4] Once the patentee meets these burdens, prosecution history estoppel no longer applies and the patentee may allege equivalent infringement of the narrowed element. Accord Festo III, 344 F.3d at 1367-68.

b. The Federal Circuit's Analysis: CONFIRMS THIS COURT'S ANALYSIS.

Neither the Federal Circuit nor the Supreme Court addressed whether the "presumption" outlined in Festo II is a true presumption. The Federal Circuit's description of the "presumption," however, confirms this Court's analysis. Festo III, 344 F.3d at 1367-68.

2. What Quantum Of Evidence Is Necessary To Rebut The Presumption?

a. The Court's Original Analysis: Proof by a fair preponderance of the evidence is sufficient to rebut a "presumption" of prosecution history estoppel.

While the Supreme Court made clear that the burden of overcoming the "presumption" lies with the patentee, it did not define the quantum of evidence necessary to meet such a burden. In other words, the Supreme Court did not lay out whether the patentee must make a showing by a fair preponderance of the evidence, clear and convincing evidence, or some other standard, nor has any other court yet addressed this quandary in a published opinion.^[5]

Important in determining the requisite standard is the surrounding legal landscape. Here, that landscape—patent litigation concerning both the doctrine of prosecution history estoppel and the doctrine of equivalents—is riddled with uncertainty. Festo II, 535 U.S. at 732, 122 S.Ct. 1831 ("It is true that the doctrine of equivalents renders the scope of patents less certain."). Nevertheless, the Supreme Court purposely preserved the doctrine of equivalents in Festo II; it made clear that there was value to the doctrine. Festo II, 535 U.S. at 732, 122 S.Ct. 1831 ("Each time the [Supreme] Court has considered the doctrine of equivalents, it has acknowledged this uncertainty as the price of ensuring the appropriate incentives for innovation ...."); Warner-Jenkinson, 520 U.S. at 28, 117 S.Ct. 1040 (reaffirming that the doctrine is a secure part of a patentee's settled rights and stating that it was up to Congress—not the Supreme Court— to discard the doctrine). With this landscape in mind, the Court holds that the standard for overcoming the "presumption" of prosecution history estoppel is proof by a fair preponderance of the evidence^[6] for the following reasons.

Instituting anything other than the most familiar standard—a fair preponderance of the evidence—would add yet another layer of uncertainty and cause litigation over prosecution history estoppel to proceed inconsistently. See John H. Strong, McCormick on Evidence § 340 (5th ed.1999) (explaining that the clear and convincing standard "has not become as standardized as is the ``preponderance' formula"); see also Kevin Casey, et al., Standards of Appellate Review in the Federal Circuit: Substance and Semantics, 11 Fed. Cir. B.J. 279, 323 (2001) (noting that the clear and convincing standard is "not susceptible to precise definition").^[7] Such a result is contrary to the purpose of prosecution history estoppel and, therefore, should be avoided.

Moreover, a burden of proof that is too exacting could result, de facto, in that which the Supreme Court in Festo II was trying to avoid—death of the doctrine of equivalents. Cf. Note, Estopping the Madness at the PTO: Improving Patent Administration Through Prosecution History Estoppel, 116 Harv. L.Rev. 2164, 2184 (2003) (noting that the Federal Circuit could make the burden of overcoming the presumption more stringent by requiring that patentees produce clear and convincing evidence). Proof by a fair preponderance of the evidence is less exacting than proof by clear and convincing evidence, thereby enabling the "presumption" outlined in Festo II to be one that practically can be rebutted—which in turn preserves the doctrine of equivalents. See Strong, supra, at § 340 (noting that the clear and convincing standard of proof is a "more exacting measure of persuasion" than preponderance of the evidence).^[8]

Furthermore, the fair preponderance of the evidence standard generally is the one that is applied in patent law. David W. Okey, Issued Patents and the Standard of Proof: Evidence Clear and Convincing or Merely Ponderous?, 17 J. Marshall J. Computer & Info. L. 557, 587 (1999). In civil litigation, the clear and convincing standard is saved for special situations in which "a particular type of claim should be disfavored on policy grounds." Strong, supra, at § 340. Accordingly, in patent law, it is applied to challenges regarding validity because of the statutory provision found at 35 U.S.C. § 282, that a patent is "presumed" valid. Casey, supra, at 324. It is also applied when a patentee claims willful infringement of a patent, Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211 (Fed.Cir.1995), or violation of the duty of disclosure. Casey, supra, at 324. While there are indeed public policy issues implicated in the doctrine of prosecution history estoppel, the "presumption" itself vindicates the public notice function of patent law. Thus, as already noted, to apply an overly exacting standard to a patentee's ability to overcome the "presumption" would unravel the balance the Supreme Court sought between protecting the true scope of a patent and honoring a competitor's reliance on the public record.

Based on this reasoning, the Court holds that the burden of proof most appropriate for overcoming the "presumption" of prosecution history estoppel is proof by a fair preponderance of the evidence.^[9]

b. The Federal Circuit's Analysis: DID NOT ADDRESS THIS ISSUE.

Neither the Federal Circuit nor the Supreme Court has discussed the quantum of evidence required to rebut the presumption.

3. What Is The Role Of The Judge And The Jury?

a. The Court's Original Analysis: Anomalous as it may seem, the jury plays no role in any of this.

Although it is clear that the equitable doctrine of prosecution history estoppel is for the court to decide, see, e.g., Pioneer Magnetics, Inc. v. Micro Linear Corp., 330 F.3d 1352, 1356 (Fed.Cir.2003); CAE Screenplates Inc. v. Heinrich Fiedler GmbH & Co. KG, 224 F.3d 1308, 1319 (Fed.Cir.2000), the Supreme Court in Festo II did not provide guidance as to whether the determination that the patentee has rebutted the "presumption" is a question for the court or the jury. Though the bench trial in this case moots the issue here, the question generally remains.

Placing these fact questions before the jury could indeed work against the goals of consistency and predictability since the Federal Circuit would be required to defer to findings of fact made at the trial court level. See Raymond G. Areaux and Theodore S. Owers, III, Festo's Doctrine of Equivalents: Vitality Reaffirmed in View of New and Old Hurdles, 48 Loy. L.Rev. 551, 576-78 (2002). On the other hand, determining whether the presumption has been rebutted is not purely technical in nature and involves the kinds of issues— foreseeability, intent, purpose, and reasonableness —normally reserved for a jury. Cf. Arti K. Rai, Engaging Facts and Policy: A Multi-Institutional Approach to Patent System Reform, 103 Colum. L.Rev. 1035, 1096-97 (2003) (noting that "juries often decide questions of reasonableness even when facts are not in serious dispute."). Thus, unlike the circumstances in Markman v. Westview Instruments, Inc., 517 U.S. 370, 388-90, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996), where the Court concluded, with respect to claim construction, that "history and precedent provide no clear answers," here history and precedent do provide answers. These are matters historically resolved by American juries.

For the reasons already expressed in Control Resources, however, this Court predicts that litigation over the existence of prosecution history estoppel will be taken from the jury. But see Arthur R. Miller, The Pretrial Rush to Judgment: Are the "Litigation Explosion," "Liability Crisis," and Efficiency Cliches Eroding Our Day in Court and Jury Trial Commitments?, 78 N.Y.U. L.Rev. 982, 1087 (2003) (criticizing the decision in Markman v. Westview Instruments, Inc., 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996), which embodies this trend in patent law). For the larger implications of marginalizing the jury's historic role, see William G. Young, An Open Letter to United State District Judges, Fed. Lawyer (2003). See generally Ellen E. Sward, The Decline of the Civil Jury (2003).

b. The Federal Circuit's Analysis: CONFIRMS THIS COURT'S PREDICTION BUT WITH A STRIKINGLY DIFFERENT ANALYSIS.

While the Federal Circuit characterizes the rebuttal question as it relates to the foreseeability criterion as involving mixed fact-law issues, Festo III, 344 F.3d at 1368-69, it treats the issue as one for the district court (not a jury) to decide in the first instance. Id. at 1367-68. This can be interpreted to mean that determining whether the "presumption" has been rebutted should be decided by the court because it arises in the context of an equitable defense and involves nuanced questions of mixed fact and law. This is the reading of Festo III that makes the most sense. It is fully supported by the cases cited by the Federal Circuit in this portion of the Festo III decision. Id. at 1367-68 (citing Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 871 n. 7 (Fed.Cir.1985), overruled on other grounds by Nobelpharma AB v. Implant Innovations, Inc., 141 F.3d 1059, 1068 (Fed.Cir.1998); Black & Decker, Inc. v. Hoover Serv. Ctr., 886 F.2d 1285, 1295 (Fed.Cir.1989); Builders Concrete, Inc. v. Bremerton Concrete Prods. Co., 757 F.2d 255, 258 (Fed.Cir.1985)). In other words, as the entire concept of prosecution history estoppel is a judicially created equitable doctrine, it makes prefect sense for the district court to do equity in the premises and make the necessarily nuanced fact finding subject to deferential appellate review.

The analysis above comports with prior Federal Circuit decisions and, moreover, is logical and practical.^[10] Unfortunately, it appears not to be what the Federal Circuit intends—"[T]he determinations concerning whether the presumption of surrender has arisen and whether it has been rebutted are questions of law for the court, not a jury, to decide." Festo III, 344 F.3d at 1368 (emphasis added). One could hope that the Federal Circuit meant only that the factual issues were reserved for the court given their equitable antecedents and that the phrase "of law" crept in during drafting without much thought. Such comforting yearnings are dashed, however, by the incisive dissent of Judge Newman who points out:

This court now denies Festo the opportunity to present evidence as to the two criteria of tangentialness and some other reason, establishing these criteria as questions of law, not fact, and decides these questions of first impression without development and without evidence.

Id. at 1383 (Newman, J., concurring in part and dissenting in part). In the face of this language, it would strain credulity to conclude that the majority intended something other than pure rulings of law, reviewed de novo on appeal.^[11]

4. Did The Supreme Court Define One Over-Arching Standard Or Three Distinct Criteria For Rebutting The Presumption?

a. The Court's Original Analysis: Festo II provides one broad, overarching standard for rebutting the "presumption" of prosecution history estoppel.

While Amgen and TKT/HMR agree that Amgen bears the burden of showing that the "presumption" of prosecution history estoppel is rebutted by a fair preponderance of the evidence, they disagree about what the evidence must show; that is, they interpret the Supreme Court's decision in Festo II differently. TKT/HMR asserts that Festo II depicts a single rebuttal standard: whether "at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." TKT/HMR's Mem. in Support at 2 (quoting Festo II, 535 U.S. at 741, 122 S.Ct. 1831.) Amgen, on the other hand, argues that Festo II outlines three independent methods (described below) in which to rebut the "presumption" of estoppel. Amgen's Mem. in Opp'n re: '080 [Document No. 716] at 3, 7.

Upon reflection it appears that both parties are partly correct. Amgen is correct in asserting that Festo II lays out more than one way in which to rebut the "presumption" of estoppel. See, e.g., Schwing GmbH v. Putzmeister Aktiengesellschaft, 305 F.3d 1318, 1329 (Fed.Cir.2002) (citing various methods in which to rebut and remanding to the district court); Sound-Tube Entertainment, Inc. v. Brown Innovations, Inc., 233 F.Supp.2d 188, 196-97 (D.Mass.2002) (Saris, J.) (applying the various methods of rebutting the presumption); see also Amgen II, 314 F.3d at 1345 (referring to Festo II's "narrow ways" of rebuttal and citing the page in Festo II that sets out rebutting circumstances) (emphasis added). Indeed, in Festo II, the Supreme Court described three situations in which an amendment cannot reasonably be viewed as surrendering the equivalent in question: (1) when the equivalent was unforeseeable; (2) when the rationale underlying the amendment bears no more than a tangential relation to the equivalent in question; or (3) when some other reason exists that suggests that the patentee could not reasonably be expected to have described the insubstantial substitute in question.^[12]Festo II, 535 U.S. at 740-41, 122 S.Ct. 1831.

TKT/HMR, however, is also correct in asserting that the Supreme Court in Festo II made clear that to rebut the "presumption" of prosecution history estoppel, a patentee must "show that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." Festo II, 535 U.S. at 741, 122 S.Ct. 1831. Upon careful review, it seems that this latter standard is a broad, over-arching one that embraces the three situations described above. When any of the three situations exist, the broad, over-arching standard necessarily is met. In other words, if the equivalent was unforeseeable, it would not have been reasonable to expect the patentee to have drafted the claim otherwise. Likewise, if the reason for the amendment was only tangentially related to the equivalent, it would not have been reasonable to expect the patentee to have drafted the claim so that it literally included the equivalent in question; that is, while the patentee perhaps could have drafted the claim otherwise to include the equivalent, it would not be reasonable to have expected it to have done so if the amendment was not even remotely related to the equivalent in question.^[13] Finally, the third example simply illustrates that the Supreme Court recognized that the first two rebuttal methods were not exhaustive, that is, that there may yet be another reason why it would be unreasonable to expect the patentee to have drafted the claim to have encompassed the alleged equivalent.^[14]

This interpretation is also supported by the language of the opinion itself. The Supreme Court, in describing the third rebuttal standard, used the words "some other reason suggesting that the patentee could not reasonably be expected to have described the insubstantial substitute in question," Festo II, 535 U.S. at 741, 122 S.Ct. 1831 (emphasis added). In using the words "some other reason," the Supreme Court indicated that the first two methods are merely specific ways to rebut the same broad presumption. Thus, the proper interpretation of the Supreme Court's language is that the first two methods are included within the broad, over-arching standard and the third is that Court's way of recognizing that other circumstances may also suffice to rebut the Festo II "presumption." This reading finds further support in the concluding sentence of this section of the opinion, wherein the Supreme Court summarizes the standard: "[t]he patentee must show that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." Id. The fact that the Supreme Court summarized its analysis in this way suggests that one broad standard encompasses the three situations previously outlined.

Most of the cases that have applied the holdings of Festo II also support this interpretation.^[15] In Bosch v. Japan Storage Battery Co., Ltd., 223 F.Supp.2d 1159, 1170 (C.D.Cal.2002), the court listed the three examples that the Supreme Court provided in Festo II and then summarized them by stating "[i]n short, ``[t]he patentee must show that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent.'"<sup>[16]</sup> In Vardon Golf Co., Inc. v. Karsten Mfg. Corp., the court explained that the Supreme Court put the burden of showing that an amendment has not surrendered a particular equivalent on the patentee and implicitly indicated that this burden was "show[ing] at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." Vardon Golf, 2002 WL 1424567, at *4 (N.D.Ill.2002). Later, when describing specific cases in which this burden was met, it stated that the Supreme Court's foreseeability example met the broad over-arching standard: "If for example the equivalent was unforeseeable at the time of the application, then the patentee could not reasonably be expected to have described the insubstantial substitute in question." Id. at *4; see also SoundTube, 233 F.Supp.2d at 197 (summarizing the patentee's argument as "[the patentee] could not have foreseen the SoundTube system during the prosecution of the patent and thus ``one skilled in the art could not reasonably have been expected to draft claim limitations to literally encompass' the SoundTube devices") (emphasis added); Glaxo Wellcome Inc. v. Impax Labs. Inc., 220 F.Supp.2d. 1089, 1093 (N.D.Cal.2002) (summarizing the standard as a single broad, over-arching standard); Conmed Corp. v. Ludlow Corp., 235 F.Supp.2d 109, 122 (N.D.N.Y.2002) (applying the broad, over-arching standard and concluding that the patentee failed to meet its burden); SmithKline Beecham Corp. v. Excel Pharms., Inc., 214 F.Supp.2d 581, 591 (E.D.Va.2002) (stating that, to meet its burden, the patentee must satisfy the broad, over-arching standard); but see Schwing GmbH 305 F.3d at 1329 (stating that the Supreme Court "discussed several ways in which the patentee could overcome [the] presumption" and listing the three examples from Festo II without connecting them to a single, broad, over-arching standard).

Thus, this Court interprets Festo II as providing a single standard—albeit a difficult one to satisfy—to determine whether the "presumption" of estoppel has been overcome and as offering three situations in which the articulated standard would be satisfied.

b. The Federal Circuit's Analysis: DIFFERS FROM THIS COURT'S ANALYSIS BUT DOES NOT CHANGE THIS COURT'S ULTIMATE HOLDING.

The Federal Circuit holds that there are three separate "criteria" which the patentee may meet to rebut the "presumption." Festo III, 344 F.3d at 1365-66, 1368-70 ("Specifically, the Court enumerated the three ways in which the patentee may overcome the presumption."). As mentioned above, whether the standard is viewed as three separate criteria or as one over-arching standard, will not, in most cases, change the end-result because the three criteria work in tandem with what this Court viewed as the over-arching standard. In other words, if the equivalent was unforeseeable, or if the reason for the amendment was no more than tangentially related to the equivalent, it would not have been reasonable to expect the patentee to have drafted the claim to have literally encompassed the equivalent.

The Federal Circuit's interpretation nevertheless may make a significant difference in the outcome of cases because it states that the third—the "some other reason" — criterion is to be narrowly construed. Id. at 1368-69. Aside from stating that "a patentee may not rely on the third rebuttal criterion if the alleged equivalent is in the prior art ...." and that the "shortcomings of language" may satisfy the criterion, the Federal Circuit left to another day further exposition of this third criterion. Id. at 1369-70.

5. What Type Of Evidence May Be Used To Rebut The Presumption?

a. The Court's Original Analysis: Extrinsic evidence may be considered if it does not contradict the intrinsic evidence.

Although the rebuttal standard is narrow, the inquiry and analysis to determine if it has been met is in some respects intricate and, therefore, requires resort to extrinsic evidence. The Supreme Court made clear in Festo II that a court must examine the subject matter surrendered by the amendment and the reasonable inferences that can be drawn from the amendment. Festo II, 535 U.S. at 737-38, 122 S.Ct. 1831; Schwing GmbH, 305 F.3d at 1329; Bosch, 223 F.Supp.2d at 1170. This examination is more than a peripheral one.

There is no reason why a narrowing amendment should be deemed to relinquish equivalents unforeseeable at the time of the amendment and beyond a fair interpretation of what was surrendered. Nor is there any call to foreclose claims of equivalence for aspects of the invention that have only a peripheral relation to the reason the amendment was submitted.

Festo II, 535 U.S. at 738, 122 S.Ct. 1831 (emphasis added). To determine what was fairly surrendered by the amendment, the Court needs to, in many instances, understand the rationale for the amendment. Id. Knowing simply that an amendment was made for reasons related to patentability is not enough. The Court needs to understand more specifically why the exact words or limitations were added or deleted and such an understanding is not always immediately evident in the prosecution history. Often patentees change certain aspects of claims in response to an objection. Other times, such as in the case before the Court, patentees voluntarily make changes even when there is neither a Patent Office objection nor rejection and when no prior art threatens the validity of the invention. Given that the "remarks" that accompany the amendments are cursory at best, in the latter case it is even more difficult to discern the particular rationale for the amendment based on the prosecution history. Amgen I, 126 F.Supp.2d at 135 ("This search for purpose is made more difficult where, as here, there is nothing in the file wrapper as of the date of the amendment indicating that the claims were objectionable to the Patent Office."). Moreover, the "state-of-the-art" at the time of the amendment (a key piece of information necessary to determine foreseeability) generally is not something addressed in the prosecution history—nor are the intricacies of the technology always clear. Therefore, there are instances in which reliance on the prosecution history alone would not inform the court in sufficient detail as to why certain additions and deletions to the claims were made.

Notwithstanding these concerns, TKT/HMR relies on Pioneer Magnetics to argue that only the public prosecution history record can be used to demonstrate the purpose of the amendment and that the Court should not consider any extrinsic evidence in determining whether the rationale for the amendment has no more than a tangential relation to the equivalent in question.

TKT/HMR's argument is unavailing here. First, Pioneer Magnetics is distinguishable from the instant case. In Pioneer Magnetics, when the Federal Circuit stated that only the public record could be used to determine the purpose for the amendment, the court was evaluating whether the amendment was made for patentability reasons (a determination made prior to the assessment of whether the "presumption" of prosecution history estoppel has been overcome). Pioneer Magnetics, 330 F.3d at 1356. Thus, this holding is inapposite to the analysis that occurs after the court has decided that the amendment was a narrowing one made for patentability reasons and that the "presumption" therefore applies.

Second, neither Pioneer nor Festo II foreclose the use of extrinsic evidence once the "presumption" of estoppel has been triggered and the patentee is trying to rebut it. See, e.g., Areaux & Owers, supra, at 574 (noting that the Supreme Court did not offer any guidance as to whether intrinsic or extrinsic evidence can be used). Moreover, in explaining that the interpretation of the patent begins—but does not end—with the literal claims and the prosecution history, Festo II apparently gave the patentee leeway to rely on extrinsic evidence to prove that "one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." Festo II, 535 U.S. at 741, 122 S.Ct. 1831; but see Pavan K. Argarwal, Patenting in Line with the Federal Circuit, 12 Fed. Cir. B.J. 395, 413-14 (2003) ("[W]ith the notice function that the Supreme Court previously announced in Warner-Jenkinson Co. v. Hilton Davis Chemical Co., it would seem appropriate to limit the patentee to the public record of the prosecution as the en banc Federal Circuit decided in Festo.").

Third, the United States Amicus Brief— relied on by the Supreme Court to define the rebuttable "presumption"—argued that extrinsic evidence can be used that explains but does not contradict intrinsic evidence. Brief of Amicus Curiae United States at 23, Festo II, 535 U.S. 722, 122 S.Ct. 1831, 152 L.Ed.2d 944 (2002); Areaux & Owers, supra, at 575-6 (noting that the U.S. Amicus Brief was a clue to what type of evidence the Federal Circuit will allow). Moreover, it specifically urged that facts should be drawn from the specific technology involved. Brief of Amicus Curiae United States at 23 n. 4. Even the Federal Circuit in Festo I (which stated that use of extrinsic evidence would undermine the public notice function) explained that a district court can properly consider any argument made by an attorney that is supported by the prosecution history record. Festo I, 234 F.3d at 586 n. 6.

Lastly, common sense dictates that extrinsic evidence may be necessary to meet the rebuttal requirements outlined in Festo II. For example, the prosecution history alone would not show whether an invention, discovery, or technique was foreseeable, or whether a person skilled in the art could have drafted the claim differently. See, e.g., Areaux & Owens, supra, at 575 ("[I]t seems plain that evidence of foreseeability will not reside in the prosecution history."). Furthermore, it is the prosecution history itself that triggers the "presumption" of estoppel; thus, if the patentee is limited to the prosecution history in its efforts to rebut that "presumption," it would likely never be successful. Like a standard of proof that is too exacting, limiting the patentee to the intrinsic record would in essence create a result similar to the complete bar the Supreme Court overruled in Festo II.

For these reasons, in determining whether Amgen has rebutted the "presumption" of estoppel, the Court will examine the prosecution history in light of the presentations made during the trial and in the parties' submissions. Cf. Canton Bio-Medical, Inc. v. Integrated Liner Tech., Inc., 19 F.Supp.2d 22, 32 (N.D.N.Y. 1998) (relying in part on expert opinion to determine what was surrendered with the amendment and holding that "[a]fter considering the arguments of the parties, their briefing and submissions, the entire relevant record, and the applicable law, the [c]ourt finds that the Defendant's accused process does not infringe the Plaintiff's [] patent, as a matter of law, based on the application of prosecution history estoppel ....").^[17]

In other words, the Court will consider extrinsic evidence. It will do so, however, only when that evidence does not contradict the intrinsic evidence—the patent, the specification, and the prosecution history. Such an examination thus cannot dissuade the Court from giving the public record its due weight—a significant policy on which patent law is based. Moreover, the Court will not go so far as to consider the subjective intent of the patentees when such intent is not expressed in or supported by the patent or file wrapper. "Such testimony cannot guide the court to the proper interpretation when the patent documents themselves do so clearly." Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1584 (Fed.Cir.1996). When, however, such intent is supported by the prosecution history, the Court will consider it because "a full understanding of what the inventors actually invented and intended to envelop with the claim" guides a court to the proper interpretation. Renishaw PLC. v. Marposs Societa' per Azioni, 158 F.3d 1243, 1250 (Fed.Cir.1998) (discussing claim construction). These limitations on the Court's consideration of extrinsic evidence strike the proper balance between supporting the notice functions of the patenting process and protecting the true scope of an inventor's patent despite the ineptness of language to do so on its own.^[18]

b. The Federal Circuit's Analysis: RESTRICTS THIS COURT'S APPROACH

The Federal Circuit adopts a more restrictive approach to extrinsic evidence than that set out above. Festo III, 344 F.3d at 1368-70. Both this Court and the Federal Circuit interpret Festo II as allowing consideration of extrinsic evidence when determining whether the first criterion, foreseeability, has been met. Id. The Federal Circuit, however, clearly limits the determination under the second criterion, tangentiality, to intrinsic evidence and strongly suggests that the third criterion should be limited in the same way.^[19]Id. This is the crucial difference between its analysis and that of this Court. Notwithstanding the revisions necessary to comport with the Federal Circuit's analysis, the Court's holding that Amgen has rebutted the presumption has not changed. The basis for this holding, however, rests on different criteria, as will be explained below.

6. Should Foreseeability Be Analyzed At The Time Of The Amendment Or The Time Of The Original Patent Application?^[20]

a. The Court's Original Analysis: Foreseeability Should Be Analyzed At The Time Of The Amendment

Some scholars and courts have indicated that the Supreme Court was not clear whether foreseeability should be analyzed at the time of the amendment or at the time of the application. See, e.g., Caliendo, supra, at 321; Fu, supra, at 134 (noting that the Supreme Court was unclear with respect to when the foreseeability test should be applied).

This Court, however, holds that Festo II defined the relevant time period as the time of the amendment, not the time of the patent application. First, in Festo II the Supreme Court explicitly stated twice that the relevant time period is the time of the amendment. When it first explains why a narrowing amendment should not be deemed to relinquish all equivalents, it considers foreseeability based upon what was known at the time of the amendment. Festo II, 535 U.S. at 738, 122 S.Ct. 1831. Later, when it summarizes the holding of Festo II, the Federal Circuit again states that the "presumption" may be rebutted when the patentee "show[s] that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." Id. at 741, 122 S.Ct. 1831 (emphasis added). Although the Supreme Court used the word "application" in describing the foreseeability test at one point, see Festo II, 535 U.S. at 740, 122 S.Ct. 1831 ("The equivalent may have been unforeseeable at the time of the application ...."), the Court was referring to the application process, not the time at which the application was originally filed. Cf. Johnson & Johnston Assoc. Inc. v. R.E. Serv. Co., 285 F.3d 1046, 1059 (Fed.Cir.2002) (Rader, J., concurring) ("In reaching this result, the court suggested that if variation had been unforeseeable during the application process, the doctrine of equivalents would still have been available to the patentee.").

Second, the Federal Circuit and other federal courts have also interpreted the relevant time reference to be the time of the amendment (not the time of the application). See, e.g., Pioneer Magnetics, 330 F.3d at 1357 (analyzing foreseeability based on what was known at the time of the amendment); SmithKline, 214 F.Supp.2d at 592 (same); Masco Corp. v. United States, 56 Fed. Cl. 400, 412 n. 11 (2003) (noting that there is some support for interpreting Festo II as requiring unforeseeability at the time of the amendment).

Third and finally, common sense dictates the same interpretation. Prosecution history estoppel only arises when an amendment has been made that potentially surrenders an equivalent. See Fu, supra, at 134 (noting that the more logical time to apply the foreseeability test is at the time of the amendment). The key inquiry is what was known at the time the patentee added the words now being used to estop it from arguing the equivalent.

Amgen strenuously argues that the relevant time period is the time of the application because when amending a patent application, an applicant is limited to the written description and drawings that were in the application at the original filing date and cannot add new matter to an application, even to correct an error in the specification or claims. Amgen's Post-Hr'g Mem. [Document No. 742] at 5. This argument is unavailing because a patentee certainly can—instead of amending—file a continuation-in-part application if the original written description and drawings do not lend themselves to the desired amendment. See 35 U.S.C. § 120, 112.^[21] In other words, if, as Amgen suggests, there is new matter to add that is necessary to complete or correct the patent description and protect the invention, the patentee can and should re-file the application instead of creating a make-shift amendment to patch up the patent's deficiencies.^[22] Therefore, the Court interprets the Supreme Court's holding in Festo II as requiring foreseeability to be adjudged relevant to the time the amendment was made.

b. The Federal Circuit's Analysis: SUPPORTS THIS COURT'S CONCLUSION

Although the Federal Circuit did not explain its conclusion, the recent Federal Circuit opinion in Festo III makes clear that foreseeability should be analyzed at the time of the amendment. Festo III, 344 F.3d at 1369 (stating repeatedly that this criterion asks "whether the alleged equivalent would have been unforeseeable to one of ordinary skill in the art at the time of the amendment") (emphasis added).

C. Has Amgen Rebutted The "Presumption" Of Estoppel?^[23]

The following analysis is conducted in light of the Federal Circuit's recent decision in Festo III.

Applying the rules laid out in Festo II as interpreted by the Federal Circuit in Festo III, the Court begins with the presumption that Amgen has surrendered all the equivalents that fall within the territory between the original claims of the application for the '080 patent and the subsequently narrowed claims of the '080 patent. See Festo II, 535 U.S. at 733-34, 122 S.Ct. 1831; Festo III, 344 F.3d at 1367-68. When the application for the '080 patent was originally filed, it contained 60 claims. These claims did not limit the invention to human EPO nor did they limit the claims to the amino acid sequence of Figure 6. In the third preliminary amendment, Amgen added the following limitation: the "mature human erythropoietin sequence of Figure 6." Third Preliminary Amendment, Trial Ex. 2005, TKT/HMR's App. [Document No. 708], Tab E at 7.

As explained in Amgen I, Figure 6 displays the nucleotide series or DNA sequence of human EPO including both exons (the portions of the sequence that code for the desired protein) and introns (the portions of the sequence that do not code for the protein and are spliced out during transcription). Amgen I, 126 F.Supp.2d at 100. Figure 6 also shows a series of codons, sets of three adjacent nucleotides that determine which of the twenty amino acids are incorporated into a protein at a particular location. Id. Each amino acid is identified by its three letter abbreviation representing a codon. Id. Figure 6, therefore, demonstrates the deduced amino acid sequence that is arrived at by reading the codons of the DNA encoding the protein. Id. Figure 6 numbers the amino acids from -27 to a final position of 166, which is labeled arginine. Because the original claims covered all human EPO regardless of the number of amino acids, but the amended claims limited the EPO to Figure 6 (depicting a 166 amino acid sequence), Amgen is presumptively barred from claiming equivalent infringement by a human EPO product containing anything other than that depicted in Figure 6, such as HMR/TKT's EPO which comprises a 165 amino acid sequence.

Amgen attempts to overcome the "presumption" by arguing (1) that HMR 4396, that is, a mature human EPO with a 165 amino acid sequence, was not foreseeable at the time of the application; (2) that the rationale for the amendment bears "no more than a tangential relation to the equivalent in question," Festo II, 535 U.S. at 740, 122 S.Ct. 1831; and (3) that there is another reason why Amgen could not reasonably be expected to have drafted a claim that encompassed the alleged equivalent.

TKT/HMR argues that Amgen cannot meet the requirements of Festo II to rebut the presumption of estoppel because (1) Amgen could have made its claim without specifying the amino acid sequence (Figure 6) in its entirety or at all; and (2) the claims of the original '080 patent covered all EPOs (including the TKT/HMR product) and by amending the claims to include a reference to Figure 6, Amgen surrendered this broad coverage. TKT/HMR's Mem. in Support re: '080 [Document No. 679] at 2.

Before addressing these arguments, a review of the prosecution history is necessary.

1. The Prosecution History Of The '080 Patent

In the first preliminary amendment, the original 60 claims were canceled and new claims 61-67 added. June 6, 1995 Amendment, TKT/HMR's App., Tab E. Of importance to this case are claims 61, 62, 63, and 64.

61. An isolated human erythropoietin glycoprotein product not being isolated from human urinary sources having glycosylation which differs from that of human urinary erythropoietin.

62. An isolated human erythropoietin glycoprotein product not being isolated from human urinary sources having a higher molecular weight than human urinary erythropoietin as measured by SDS-PAGE.

63. An isolated human erythropoietin glycoprotein product not being isolated from human urinary sources and free of other human proteins.

64. The in vivo biologically active erythropoietin product of the process comprising the steps of:

(a) growing, under suitable nutrient conditions, host cells transformed or transfected with an isolated DNA sequence selected from the group consisting of (1) the DNA sequences set out in FIGS 5 and 6, (2) the protein coding sequences set out in FIGS 5 and 6, and (3) DNA sequences which hybridize under stringent conditions to the DNA sequences defined in (1) and (2) or their complementary strands; and

(b) isolating said erythropoietin product therefrom.

Id.

Thereafter, in the Second Preliminary Amendment, Amgen canceled product claims 61-63, amended process claim 64 to specify that it was a non-naturally occurring in vivo biologically active EPO product, and added new "product-by-process" claim 68. Id.; Amgen's Post-Hr'g Mem. in Support at 12, n. 12; December 20, 1995, Second Preliminary Amendment, Trial Ex. 3, Tab 3 at 119-120.^[24]

68. A non-naturally occurring erythropoietin product of the process comprising the steps of:

a) growing, under suitable nutrient conditions, host cells transformed or transfected with an isolated DNA sequence encoding the human erythropoietin amino acid sequence set out in FIG. 6 or a fragment thereof; and b) isolating an erythropoietin product therefrom.

Id.

Amgen then amended the claims again. In this third preliminary amendment, Amgen cancelled claims 64-68 and inserted new claims 69-75. Third Preliminary Amendment, Trial Ex. 2005, TKT/HMR's App., Tab E. Claims 69, 70, and 71 became claims 1, 2, and 3 of the '080 patent and are, therefore, the claims of interest here.

69. An isolated erythropoietin glycoprotein having the in vivo biological activity of causing bone marrow cells to increase production of reticulocytes and red blood cells, wherein said erythropoietin glycoprotein comprises the mature erythropoietin amino acid sequence of Figure 6 and has glycosylation which differs from that of human urinary erythropoietin.

Claim 70. An isolated erythropoietin glycoprotein having the in vivo biological activity of causing bone marrow cells to increase production of reticulocytes and red blood cells, wherein said erythropoietin glycoprotein comprises the mature erythropoietin amino acid sequence of Figure 6 and is not isolated from human urine.

Claim 71. A non-naturally occurring erythropoietin glycoprotein having the in vivo biological activity of causing bone marrow cells to increase production of reticulocytes and red blood cells, wherein said erythropoietin glycoprotein comprises the mature erythropoietin amino acid sequence of Figure 6.

Third Preliminary Amendment, Trial Ex. 2005, TKT/HMR's App., Tab E at 8.

2. Was The Equivalent Foreseeable?

As discussed supra, foreseeability is one of the bases on which to determine whether the standard for rebutting the "presumption" is met. If the equivalent in question was not foreseeable at the time of the amendment, then the patentee can overcome the "presumption."

Unfortunately for Amgen, it supports its foreseeability argument primarily with expert testimony that addresses knowledge of those skilled in the art at the time of the application, as opposed to the time of the amendment (the relevant time period). Festo II, 535 U.S. at 741, 122 S.Ct. 1831. Amgen claims (and TKT/HMR does not dispute) that it was unknown, at the time of the patent application, that the 166th amino acid cleaved off to leave a 165 amino acid sequence. Accordingly, Amgen argues, it was unforeseeable (at the time of the application) that Figure 6's depiction of a sequence of 166 amino acids did not cover all mature human EPOs (including TKT/HMR's HMR 4396). Moreover, it argues that it was unforeseeable that the mature human EPO glycoprotein produced in Example 10 would contain only the 1-165 amino acid sequence.

Crucially, Amgen made admissions in its submissions and oral argument that it—as well as those skilled in the art—knew at the time of the amendment that human urinary EPO had a 165 amino acid sequence. Amgen's Mem. in Support at 10; 7/28/03 Mot. Hr'g Tr. at 71, ll. 11-20; 7/31/03 Mot. Hr'g Tr. at 77, ll. 12-18.^[25] In fact, on October 30, 1987, Amgen explained in its FDA product license application that the 166th arginine (depicted in Figure 6) cleaves off leaving 165 amino acids (instead of the originally believed 166):

The Arg 166, coded for by the erythropoietin gene (Section 2.B) appears to be cleaved post-translationally, resulting in a C-terminus residue of Asp 165. The detailed analysis of the C-terminus of erythropoietin is described later in this Section.

Trial Ex. 2024 at AM 57 007747. Therefore, Amgen cannot meet its burden here. Amgen's argument that this admission does not show that it was foreseeable that the amendment (that incorporated the Figure 6 limitation) would be construed by this Court to exclude human EPO with a 165 amino acid sequence is unavailing at this juncture in the analysis.^[26] Simply because the Court's construction (that resulted in a complete surrender of an equivalent human EPO with a 165 amino acid sequence) was not foreseeable at the time of the amendment does not mean that the equivalent was not itself foreseeable— which is the correct inquiry. See Masco, 56 Fed. Cl. at 413 (finding unavailing patentee's argument that surrender was not foreseeable because it did not have the Court's claim construction); SmithKline, 214 F.Supp.2d at 592 n. 23 ("[T]he issue before this court is whether the patentee had the scientific ability to redraft the claims more broadly ....").^[27] Although this places a "premium on forethought of patent drafting," this premium is not new to the doctrine. Sage Prods., Inc. v. Devon Indus., Inc., 126 F.3d 1420, 1425 (Fed. Cir.1997); Johnson & Johnston, 285 F.3d at 1057 (Rader, J., concurring) ("A foreseeability bar thus places a premium on claim drafting and enhances the notice function of claims. To restate, if one of ordinary skill in the relevant art would reasonably anticipate ways to evade the literal claim language, the patent applicant has an obligation to cast its claims to provide notice of that coverage. In other words, the patentee has an obligation to draft claims that capture all reasonably foreseeable ways to practice the invention."). Here, given its knowledge of human EPO having 165 amino acids, Amgen had "every incentive to redraft the claims as broadly as possible in an effort [to] surrender no more subject matter than necessary to secure the patent." SmithKline, 214 F.Supp.2d at 592. For example, Amgen might have deleted the reference to Figure 6 altogether and claimed "mature human EPO" instead of limiting it to a sequence.^[28] Indeed, Amgen admits that it "could have sought broader claims that literally encompassed human EPO with a 165 amino acid sequence, and did in fact do so in the '422 patent [U.S. Patent No. 5,955,422]."^[29] Therefore, Amgen cannot prevail under this prong of the analysis. The determinative factor here is that the alleged equivalent was foreseeable. In fact, it was known and could easily have been encompassed in the amendment. Thus, Amgen has not met its burden with respect to the foreseeability criterion.

3. Was The Rationale Underlying The Amendment No More Than Tangentially Related To The Equivalent?

Although the Court has determined that the 1-165 amino acid human EPO equivalent was indeed foreseeable, Amgen can rebut the "presumption" of prosecution history estoppel if it can show by a preponderance of the evidence that the rationale underlying the amendment was only tangentially related to the equivalent. Festo II, 535 U.S. at 740, 122 S.Ct. 1831.^[30] With respect to this standard, more than the others, the purpose of the amendment comes to the forefront of the analysis. Contrary to some scholars' views, the correct inquiry is not whether the amendment itself narrows the scope of a claim in a way that affects the equivalent in question.^[31] If this were the test, it would be an impassable one—the only reason why the dispute arises is because the equivalent is related to the amendment and thereby affected. The correct inquiry is whether the rationale underlying the amendment, the "reason the amendment was submitted"— not the amendment itself—is more than peripherally related to the equivalent in question. Festo II, 535 U.S. at 738, 740, 122 S.Ct. 1831 ("Nor is there any call to foreclose claims of equivalence for aspects of the invention that have only a peripheral relation to the reason the amendment was submitted .... [T]he rationale underlying the amendment may bear no more than a tangential relation to the equivalent in question ....") (emphasis added). Thus, the Court must delve deeper to determine whether the rationale underlying the amendment to the '080 patent was tangential to the equivalent in question. While it is undisputed that the amendment was made to avoid a double patenting objection,^[32] the parties disagree about why the amendment was so constructed, that is, why Amgen used the exact language it did to distinguish the '080 patent from Amgen's '933 patent. Moreover, in light of the Federal Circuit's mandate that the Court rely only on intrinsic evidence in this criterion, determining the purpose of the amendment is made even more "thorny" because the amendment was not made to address a rejection or objection actually made or described by the Patent Office. Amgen I, 126 F.Supp.2d at 135.

Amgen argues primarily that it amended its claim to specify that the '080 patent referred to the mature human EPO (which can be described as having the amino acid sequence of Figure 6) and not to distinguish or exclude a 165 amino acid human EPO such as TKT/HMR's HMR 4396. Amgen's Mem. in Opp'n at 12.

The prosecution history supports Amgen's contention. Amgen's '933 patent issued just before this amendment was made. It claimed human and non-human EPO. When the application for the '080 patent was filed, it contained 60 claims, none of which limited the EPO protein products to "human" EPO. Amgen's Post-Hr'g Mem. at 11. As described in detail above, in a preliminary amendment, those original 60 claims were canceled and new claims 61-67 were added. June 6, 1995 Preliminary Amendment, TKT/HMR's App., Tab E. Claims 61-63, product claims, were indeed limited to a human EPO glycoprotein product. Id. In a second preliminary amendment, however, all but two claims (which were amended) were cancelled, and a new "product-by-process" claim was added. Amgen's Post-Hr'g Mem. at 12; Second Preliminary Amendment, Trial Ex. 3, Tab 3 at 119-120. At this point, there is no dispute that the claims, as depicted in the second preliminary amendment were not limited to "human" EPO.^[33] In a third preliminary amendment (the amendment at issue in this case), Amgen changed the product-by-process claims back to product claims adding the limitation at issue here: "the mature erythropoietin amino acid sequence of Figure 6." Third Preliminary Amendment, Trial Ex. 2005, TKT/HMR's App., Tab E (canceling claims 64-68 and adding new claims 69-75).

TKT/HMR's main counter-argument is that Amgen had previously referred to an "isolated human erythropoietin glycoprotein" and, therefore, could not have been amending in order to distinguish this claim on the basis of human EPO versus nonhuman EPO. TKT/HMR's Mem., Tab E (emphasis added). TKT/HMR, however, compares the third (and final) preliminary amendment with the first preliminary amendment, failing to account for the fact that the second preliminary amendment did not so limit the claims to human EPO. Neither Amgen nor TKT/HMR provide any evidence as to why Amgen changed the claims from product claims to product-by-process claims and then back again, nor do they provide evidence as to why the claims were (in the first amendment) limited to human EPO and then not so limited (in the second amendment) and then finally limited again (in the third amendment). See, e.g., 7/31/2003 Mot. Hr'g Tr. at 76, ll. 1-25 (describing the chronology of amendments and stating that Amgen does not know why the limitation to "human" was dropped out when the claim was put into a product-by-process claim). Regardless of the reason why Amgen switched back and forth, the chronology and language of the amendments support Amgen's position that it added the amendment in question (the third amendment) to distinguish the '080 patent from the '933 patent on the basis that '080 was limited to human EPO.^[34]

This position is also supported by Amgen's statements to the Patent Office. In its remarks accompanying the third preliminary amendment, Amgen told the Patent Office that it distinguished the '080 patent from the '933 patent by "specifying that the claimed subject matter [of the '080 claims] comprises the mature human erythropoietin sequence of Figure 6." Third Preliminary Amendment, Trial Ex. 2005, TKT/HMR's App., Tab E at 9 (emphasis added).^[35]

In further support of Amgen's argument that it did not amend to distinguish the '080 patent from human urinary EPO that has a 165 amino acid sequence, it points to the fact that it told the Patent Office that glycosylation and source limitations — not the Figure 6 limitation — distinguished the EPO of its amended '080 claims from human urinary EPO. Id.^[36] This indicated to the Patent Office that Amgen was not differentiating its product from prior art based on the Figure 6 limitation but instead based on glycosylation and source limitations. This comports with the state of the art at the time. There was no prior art disclosing a 1-165 sequence for human EPO and thus Amgen had no need to distinguish its claims over any such equivalent. To the extent that it needed to distinguish prior-art human urinary EPO, Amgen did so by including the limitations noted above regarding differences in glycosylation and isolation from human urine.

Further, Amgen points out that at the time this amendment was made, Amgen had already disclosed to the Patent Office the reference that reveals that human EPO is missing the COOH-terminal Arg 166 amino acid residue. Amgen's Post-Hr'g Mem. at 11; Second Preliminary Amendment, Trial Ex. 3, Tab 3 at 140 (disclosing to the Patent Office that the 166 arginine cleaved off by referencing Recny).^[37] Given that Amgen and the Patent Office knew that human EPO (and, therefore, example 10 of the '080 patent) actually had a 1-165 amino acid sequence, a "reasonable inference" is that the amendment was not made to surrender human EPO with a 1-165 amino acid sequence or distinguish 1-166 from 1-165 because that would have illogically resulted in either Amgen surrendering its preferred embodiment or pursuing a patent that it and the Patent Office knew to be incorrect. Festo II, 535 U.S. at 724, 122 S.Ct. 1831 (noting that the purpose of applying estoppel is "to hold the inventor to the representations made during the application process and the inferences that may reasonably be drawn from the amendment") (emphasis added).^[38] To infer such a chain of events would not be reasonable.

TKT/HMR counters, in essence, with the argument that all of this history is irrelevant. It contends that any amendment made to avoid double patenting cannot be tangential "because it goes to the heart of the reasons why the claims were ultimately allowed." TKT/HMR's Post-Hr'g Mem. at 35. This argument fails, however, because it conflates the inquiry regarding whether the amendment was made for patentability reasons, that is, whether the presumption of prosecution history estoppel applies in the first place, with the inquiry into the rationale for the amendment at hand. The rationale cannot be more than tangentially related to the equivalent in question — but it can be more than tangentially related to patentability issues — if this were not so, this tangentiality standard would be circular and insurmountable because prosecution history estoppel applies when a narrowing amendment is made for patentability reasons. If HMR/TKT's interpretation were correct, arguably the Supreme Court would not have held that tangentiality was one way of overcoming the "presumption;" and the Federal Circuit, which agreed with this Court that the amendment was made for patentability reasons, Amgen II, 314 F.3d at 1345, would likely have noted on remand that, of the potential ways to rebut the "presumption," tangentiality was foreclosed.

In sum, Amgen has shown by a fair preponderance of the evidence that the amendment at issue was made for a reason only tangentially related to the equivalent in question. Therefore, the Court holds that Amgen has satisfied the "tangential" criterion as defined by the Federal Circuit in Festo III.^[39]

4. Some Other Reason?

As discussed above, the Federal Circuit stated that the third criterion should be a narrow one and that courts should consider only the prosecution history record "[w]hen at all possible" in conducting its analysis. Festo III, 344 F.3d at 1369-70. That being said, the Court did not detail an exhaustive list of the type of circumstances that satisfy the criterion or decide when resort to extrinsic evidence would be deemed necessary. Nor did it describe "what evidence outside the prosecution history record, if any, should be considered in determining if a patentee has met its burden under this third rebuttal criterion." Festo III, 344 F.3d at 1370. Therefore, the boundaries of analysis under this criterion will be refined by courts, such as this one, faithfully interpreting Festo II, Festo III, and any future relevant decisions by the Supreme Court and the Federal Circuit.

While resort to extrinsic evidence will not always be necessary to determine if this third criterion has been met, objectively determining whether "some other reason" reasonably prevented the patentee from describing the alleged equivalent in its narrowed claim, may — like the foreseeability analysis — "depend on underlying factual issues relating to ... the state of the art and the understanding of a hypothetical person of ordinary skill in the art at the time of the amendment." Id. at 1369. This is so even if the "other reason" is the shortcomings of language — the one reason the Federal Circuit suggested would meet this criterion — because, without a thorough understanding of the state of the technology and how certain language would be interpreted by one skilled in the art at the time of the amendment, the Court would not be able to determine if adequate language existed that should have been used by the patentee.

While it is true that courts regularly — and in fact ought — construe claims without resort to expert testimony regarding how one skilled in the art would interpret the claims, a court's claim construction often narrows the claim in a way that may have been unintended by the patentee. Such narrowing is the reason why the doctrine of equivalents was created — to capture, despite the inadequacies of language, the essence of the invention in light of technological advances. See generally Festo II, 535 U.S. at 731-32, 122 S.Ct. 1831; Graver Tank, 339 U.S. at 607, 70 S.Ct. 854; Autogiro Co., 384 F.2d at 397. Undeniably, there is a difference between how a court approaches claim construction and how it approaches the doctrine of equivalents and prosecution history estoppel. The policy of avoiding all extrinsic evidence except when necessary in claim construction is to ensure that a court is determining solely the literal terms of the claim, that is, those claims of which the public was notified. The policy behind the doctrine of equivalents, on the other hand, is more equitable in nature; it seeks to determine what the patentee was surrendering and any "reasonable inferences" that can be drawn from the patentee's representations. Festo II, 535 U.S. at 737-39, 122 S.Ct. 1831; see also Festo III, 344 F.3d at 1367 ("Prosecution history estoppel has traditionally been viewed as equitable in nature, its application being ``guided by equitable and public policy principles.'") (citations omitted). Thus, reasonableness, although not a part of claim construction, is central to the doctrine of equivalents.

The idea of reasonableness was reinforced by the Supreme Court in Festo II. It held, generally, that to rebut the presumption of surrender "the patentee must show that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent." Festo II, 535 U.S. at 741, 122 S.Ct. 1831 (emphasis added). Moreover, the Supreme Court also specifically included an analysis of reasonable expectations of the patentee in the third criterion to rebut the presumption. Id. Thus, the Court's analysis at this stage — as opposed to that during claim construction — includes a reasonableness inquiry that involves questions of equity, even after Festo II.

The reasonableness inquiry as defined by Festo II, however, is now different than it used to be. Before Festo II, the reach of prosecution history estoppel was decided by considering what a competitor or one skilled in the art reasonably would have determined was surrendered by the amendment. Modine Mfg. Co. v. U.S. Intn'l Trade Comm'n. 75 F.3d 1545, 1555 (Fed.Cir.1996); Zenith Labs. v. Bristol-Myers Squibb Co., 19 F.3d 1418, 1424 (Fed.Cir.1994); Hoganas AB v. Dresser Indus., Inc., 9 F.3d 948, 952 & n. 15 (Fed. Cir.1993); Haynes Intn'l, Inc. v. Jessop Steel Co., 8 F.3d 1573, 1578 & n. 4 (Fed. Cir.1993). After Festo II, however, the focus is not on what those skilled in the art reasonably would have concluded was relinquished but on what subject matter the patentee itself reasonably should have included in the amendment. Leading Cases, supra, at 409 ("The Court's new test shifts the estoppel inquiry from the ex post perspective of a reasonable competitor to the ex ante perspective of the person drafting the patent."); Golian, supra, at 599. Under the pre-Festo II approach, a patentee could expect to retain all but those equivalents surrendered in the amendment. Now, the patentee is on notice that all equivalents that reasonably could be expected to be included are surrendered. In other words, a patentee survives prosecution history estoppel if it can show that drafting the claim to include the equivalent was not a reasonable expectation. Thus, even after Festo II, when analyzing the third criterion, the Court must make a reasonableness inquiry, but now it must ask what a reasonable patentee could be expected to have done. See, e.g., Fu, supra, at 137 (noting that Festo II includes a reasonableness test, that is, an equitable consideration). This reasonableness inquiry requires the Court to look into the circumstances and in some cases, like the one at hand, to consider extrinsic evidence.

Amgen's main argument under this third rebuttal criterion is that it could not reasonably be expected to have described the equivalent in question because competitors and those skilled in the art would have interpreted the amendment, at the time it was made, actually to cover the equivalent. In support, Amgen submits extrinsic evidence in the form of expert testimony, suggesting that the drafter of the amendment in question — and those skilled in the art — interpreted the amendment to cover all human EPO whether it had 165 amino acids or 166. Borun Trial Test., Amgen's App., Tab D at 2885.^[40]

There is also intrinsic evidence supporting this position. As explained earlier, the prosecution history shows that the Patent Office Examiner was informed by Amgen that human urinary EPO, such as that produced by the preferred embodiment depicted in Example 10, has the 1-165 amino acid sequence of Figure 6. Recny, supra, Trial Ex. 53 at 17156; Amgen's Post-Hr'g Mem. in Support at 11. In other words, the Patent Office examiner had the information that human urinary EPO has the exact amino acid sequence shown in Figure 6, except for the 166th arginine. Moreover, the reference that discloses how the 166th arginine is cleaved off was revealed to the Patent Office during the prosecution and cited as a reference in the '080 patent disclosure. Trial Ex. 3 ('080 Patent), TKT/HMR's App., Tab A at 8; Amgen's Post-Hr'g Mem. in Support at 11 n. 26. Thus, given Example 10, either the Patent Office was knowingly enabling the prosecution of an incorrect patent, or it too interpreted the amendment to cover human urinary EPO with a 1-165 amino acid sequence.^[41]

Overall then, the evidence supports the contention that Amgen, the Patent Office, and, importantly, those skilled in the art interpreted the claims as amended literally to encompass the particular equivalent in question here, an EPO with a 165 amino acid sequence.

While the correct inquiry, after Festo II, is not what those skilled in the art or a reasonable competitor would conclude was surrendered by the amendment, a competitor's reasonable interpretation of the amendment is necessarily relevant in determining what the patentee reasonably can be expected to have drafted.^[42] Claims are drafted for a particular audience — those skilled in the art. Given that there is evidence that those skilled in the art would have interpreted the amendment to include human EPO with a 165 amino acid sequence — the equivalent in question — it would not be reasonable to expect Amgen to have drafted the claim differently. While the Court did indeed construe the claim differently,^[43] such construction should not prevent the Court from resorting to this extrinsic evidence — evidence it did not consider during claim construction^[44] — to understand what can be reasonably expected of the patentee. While such extrinsic evidence may be inappropriate in claim construction, as discussed earlier, it appropriately is considered in the context of this criterion's reasonableness inquiry.

It is this part of the criterion, the reasonableness element, that TKT/HMR completely ignores and that is determinative here. TKT/HMR focuses its counter-argument on what Amgen, the patentee, could have drafted. TKT/HMR argues, as it did under the foreseeability prong, that Amgen could have easily drafted a claim that literally encompassed the equivalent. While the Court finds this assertion to be true, as discussed supra, Amgen has shown by a fair preponderance of the evidence that those skilled in the art (and the Patent Office) interpreted the amendment to encompass the equivalent.^[45] The fact that the Court later determined — under a pre-Festo II analytic framework — that this amendment did not encompass all theoretically possible human EPO does not justify holding Amgen to the literal terms of the amended claim. The Supreme Court made clear that Festo II does not go that far and that, if it did, it would defeat the purpose of the doctrine of equivalents:

The amendment does not show that the inventor suddenly had more foresight in drafting of claims than the inventor whose application was granted without amendments having been submitted. It shows only that he was familiar with the broader text and with the difference between the two. As a result there is no more reason for holding the patentee to the literal terms of an amended claim than there is for abolishing the doctrine of equivalents altogether and holding the patentee to the literal terms of the patent.

Festo II, 535 U.S. at 738, 122 S.Ct. 1831. Here, the evidence shows that Amgen was familiar with the broader text, human and non-human. It amended its claims to specify human EPO and believed that its patent and this amendment encompassed the alleged equivalent. It is not reasonable to expect one skilled in the art to have used different claim language when the language that was used was interpreted by those skilled in the art to be sufficiently descriptive.^[46] This is just the sort of "shortcoming[] of language" or "linguistic" limitation to which the Federal Circuit referred in its opinion. Festo III, 344 F.3d at 1369-70, 1371-72. Here, the amendment made clear that the patent did not cover non-human but failed (according to the Court's claim construction) to make clear that it covered human EPO with a 165 amino acid sequence. While this drafting error was fatal to Amgen's literal infringement argument, it is not fatal to Amgen's ability to rebut the "presumption" under this prong of the analysis. As the Supreme Court explained, "language is an imperfect fit for invention," the "narrowing amendment may demonstrate what the claim is not; but it may still fail to capture precisely what the claim is." Festo II, 535 U.S. at 739, 122 S.Ct. 1831; Lang, supra, at 478-86 (commenting that language limitations or an art's unpredictability should work to overcome the presumption.) A patentee should not be able to attain through litigation that which was surrendered during the prosecution history. Here, however, Amgen has shown by a fair preponderance of the evidence that, while the amendment failed to capture precisely what the claim is, the 165 amino acid equivalent was not surrendered.

Although this analysis is not directly contrary to the holdings of Festo III, the Federal Circuit strongly indicated that it will — when the issue is ripe before it — cabin in the third criterion with respect to the use of extrinsic evidence and the circumstances that satisfy the criterion. Thus, this Court rests its decision on the tangential criterion but holds in the alternative that Amgen has also met the third criterion, that is, Amgen has shown by a fair preponderance of the evidence that some other reason exists that suggests that at the time of the amendment it could not reasonably be expected to have drafted the claim to describe the insubstantial substitute in question.

III. CONCLUSION

In light of the Federal Circuit's narrow interpretation of the third criterion and admonition to rely only on intrinsic evidence whenever possible, the Court's analysis under the "some other reason" category and its resort to extrinsic evidence is less certain. Ironically, however, the "tangential" criterion — originally rejected on the basis of extrinsic evidence — is now persuasive once the focus narrows to the prosecution history alone. Frankly, a holding based on the "tangential" criterion seems less satisfactory and more artificial than the Court's original holding based on "some other reason," yet I deem it compelled by the Federal Circuit's less than generous reading of the Supreme Court's decision in Festo II. Perhaps the disjunction I perceive arises from the incredibly complex nature of recombinant DNA technology compared to the mechanical technology present in the Festo line of cases. Future decisions will doubtless refine these matters. In any event, either way Amgen "wins," that is, rebuts the presumption of prosecution history estoppel.

Accordingly, upon further consideration in light of Festo III, HMR/TKT's motion for judgment that Amgen is estopped from asserting equivalent infringement of the '080 patent [Doc. No. 678] is DENIED and Amgen's motion for judgment that claims 2-4 of the '080 are infringed under the Doctrine of Equivalents [Doc. No. 659] is ALLOWED.

SO ORDERED.

NOTES

[1] Hoechst Marion Roussel, Inc. is now known as Aventis Pharmaceuticals, Inc.

[2] TKT/HMR argues that Amgen should not be allowed to pursue a Rule 52(c) motion because Amgen has already presented evidence on the issue of infringement of claims 2-4 of the '080 patent and a Rule 52(c) motion is a vehicle for a party that has not yet presented its evidence. TKT/HMR's Opp'n [Document No. 703] at 1. Here, when TKT/HMR moved for judgment of non-infringement at the close of Amgen's case, the Court decided in TKT/HMR's favor as to literal infringement but denied the motion under the doctrine of equivalents as to the '080 patent. Amgen I, 126 F.Supp.2d at 101 n. 22. Thereafter, TKT/HMR presented its rebuttal case on equivalents. Thus, the proper motion is in actuality one for judgment under Rule 52(a), one which the Court can summarily decide without further evidentiary support.

[3] In Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., Ltd., 234 F.3d 558, 574-75 (2000) ("Festo I"), the Federal Circuit held that prosecution history estoppel completely bars the patent applicant from asserting any equivalent to a narrowed element. This Court previously has followed, but criticized, the Federal Circuit's approach. See Control Resources, Inc. v. Delta Electronics, Inc., 133 F.Supp.2d 121 (D.Mass.2001).

[4] When dealing with a "true" presumption, if the party with the burden presents evidence to rebut the presumed fact, the presumption simply disappears—the "bursting bubble theory." Young, Pollets & Poreda, Mass. Practice Evidence § 301.3 at 119 (2d ed.1998). Here, however, the patentee does not simply have to present evidence to rebut the "presumption" but actually has to persuade the Court that it did not surrender the equivalent in question. Such imprecision in language is regrettable, for it means that the definition of the word "presumption" is one thing in patent law and quite another in civil rights and employment discrimination cases. See, e.g., Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 142-43, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000) (noting that once respondent met its burden of production, the presumption disappeared); St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502, 510, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993) (stating that the presumption "simply drops out of the picture" once the defendant has succeeded in carrying its burden of production). Usually such imprecision does little more than confuse the bar and raise the already excessive cost of litigation still higher.

In the area of criminal sentencing, however, the judicial embrace of words like "evidence," "finding," and "fact," has the pernicious effect of cloaking an intellectual vacuity with forms of due process that are wholly unwarranted, thus mistakenly propping up a system that today is both morally and intellectually bankrupt. See United States v. Green, No. 02-10054-WGY (D. Mass. filed Jan. 10, 2002) (forthcoming).

[5] The Supreme Court did note that the burden here was similar to that outlined in Warner-Jenkinson Co., Inc. v. Hilton Davis Chemical Co., 520 U.S. 17, 117 S.Ct. 1040, 137 L.Ed.2d 146 (1997), and identical to the approach advocated by the United States as Amicus Curiae. Festo II, 535 U.S. at 740, 122 S.Ct. 1831. Neither of these two sources, however, specify the quantum of evidence required. In Warner-Jenkinson, the Supreme Court merely stated that the burden was on the patent holder to establish the reason for an amendment. 520 U.S. at 33, 117 S.Ct. 1040. The United States, in its Amicus Curiae brief, urged the Supreme Court to apply a burden comparable to that of Warner-Jenkinson, and stated that this burden entailed a "concrete showing." Brief of Amicus Curiae United States at 23, Festo II, 535 U.S. 722, 122 S.Ct. 1831, 152 L.Ed.2d 944 (2002) (00-1543).

[6] Proof by a fair preponderance of evidence is proof that shows something to be more likely true than not true. Clear and convincing evidence is "evidence that produces in the mind of the trier of fact an abiding conviction that the truth of the factual contentions are highly probable." Kevin Casey, et al., Standards of Appellate Review in the Federal Circuit: Substance and Semantics, 11 Fed. Cir. B.J. 279, 323 (2001) (internal quotations and citations omitted).

[7] Of course, there will always be a level of uncertainty in patent law that is inherently case-specific, but this uncertainty is exacerbated by the doctrine of equivalents and would be further exacerbated by an inconsistent application of the standard of proof necessary to rebut the "presumption." See John S. Golian, Without A Net: The Supreme Court Attempts to Balance Patent Protection and Public Notice in Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 36 Creighton L.Rev. 541, 603 (2003) ("[T]he greatest amount of uncertainty is not created by the doctrine itself, but by the court's lack of a consistent application of the rule.").

[8] This burden of proof does not enable the doctrine of equivalents to be applied more broadly than has been its historic scope. See Warner-Jenkinson, 520 U.S. at 29, 117 S.Ct. 1040 ("[T]he doctrine of equivalents when applied broadly, conflicts with the definitional and public-notice functions of the statutory claiming requirement."). It ensures that prosecution history estoppel remains tied to its primary purpose without creating either more uncertainty or a complete bar that could eviscerate the doctrine altogether.

[9] During the July 28, 2003, motion hearing, the Court told the litigants that it was inclined to apply this burden of proof and both litigants agreed. 7/28/03 Mot. Hr'g Tr. at 58, ll. 5-8 and 59, ll. 14-21.

[10] This approach ensures that Festo issues need be addressed only in that small minority of patent cases in which they make a genuine difference. Since under Dairy Queen, Inc. v. Wood, 369 U.S. 469, 82 S.Ct. 894, 8 L.Ed.2d 44 (1962), jury issues must constitutionally be first resolved, the normal patent case to be tried to a jury ought proceed roughly as follows: Once the case is sufficiently mature, the court will hold a Markman hearing and construe the claims. This done, the court will normally consider a motion for summary judgment. See, e.g., EMI Group N. Am., Inc. v. Intel Corp., 157 F.3d 887, 889 (Fed.Cir. 1998).

There are four possible outcomes:

1) The court rules there is no infringement, either literally or under the doctrine of equivalents. The Festo issue is gone.

2) The court rules that there is literal infringement. Normally, any lurking Festo issue now drops out of the case.

3) The court rules that there is no literal infringement but infringement may exist under the doctrine of equivalents. In this one circumstance, judicial economy may counsel proceeding directly to a Festo hearing before the court. If the outcome vitiates the claim under the doctrine of equivalents, the case is over. If, however, upon hearing, the "presumption" appears to have been rebutted, the jury trial will follow and the matter need finally be determined by the court only if the jury finds liability based upon the doctrine of equivalents. If the case is jury-waived, however, the court should decide the matter before determining whether there is infringement under the doctrine of equivalents.

4) There remain genuine issues of material fact concerning literal infringement. In this instance, the best course is to defer any consideration of Festo issues until after the jury trial. Not only does the Constitution command it, Dairy Queen, 369 U.S. at 471-72, 82 S.Ct. 894, it works smoothly. That is, in these circumstances knotty Festo issues need never be addressed unless the jury finds liability under the doctrine of equivalents—today, a finding unlikely to be upheld given the stringency with which the Federal Circuit applies this doctrine. See e.g., Read Corp. v. Powerscreen of Am., Inc. 44 Fed. Appx. 502, 505-10 (Fed.Cir.2002) (unpublished).

[11] As the intensely fact specific discussion in the text here makes clear, however, this is an unworkable model. Despite the well grounded criticism of Markman (see section II. B.3.a), at least claim construction bears a rough analogy to statutory interpretation. Not so here. If these aspects of prosecution history estoppel are transmuted by fiat into pure matter of law reviewable de novo on appeal, every case with an amendment during the prosecution history will require its Festo hearing, even more lengthy and involved than the Markman hearings of today. This simply will not do.

Sadly, this is precisely the result predicted in the months before the Festo III decision issued:

Now [I] suppose the [Festo III] opinion can come down and say these are all legal, these are all just legal questions. If they do ... it reminds me of the story of Lincoln, great trial lawyer, he's arguing before a jury and ... he looks over at the other lawyer and says ``how many legs does a sheep have?' The fellow says it has four. He says, ``now if I call a tail a leg, how many will it have?' And the fellow says it will have five. Lincoln turns to the jury and says ``no it won't, it still will have four. Calling the tail a leg doesn't make it a leg.' [Joseph F. Anderson, Jr., The Lost Art: An Advocate's Guide to Effective Closing Argument 120 (2002) (citing Victoria L. Eslinger's closing argument reciting this incident)].

....

I leave you with what's happening. Because of the complexity, because of the hostility to juries, what in fact you're doing is marginalizing yourselves, diminishing the District Courts, and the law is pretty much in the Federal Circuit.

Remarks of William G. Young, "Festo: Views from the Bench One Year Later" at BIO 2003, July 10, 2003.

[12] Amgen argues that TKT/HMR's purported standard is a paraphrase of this third situation. Amgen's Mem. in Opp'n at 3.

[13] As will be discussed infra, it is important to note that the Supreme Court incorporated reasonableness and expectations into the standard. If the question was merely whether the patentee could have drafted a claim to encompass the equivalent then the test would simply be a foreseeability test. While some scholars have interpreted the Supreme Court's ruling to collapse the scenarios into one test about foreseeability, see, e.g., Caliendo, supra, at 321, this Court declines so to read Festo II. If the standard were that simple, the Supreme Court would not specifically have called out the tangential example. Moreover, it would not have injected the notoriously equivocal concepts of "reasonableness" and "expectations" into what could have otherwise been a straightforward standard. See, e.g., C. Alan Fu, Prosecution History Estopel: Festo Corp. v. Shoektsu Kinzoku Kogyo Kabushiki Co., 18 Berkeley Tech. L.J. 117, 137 (2003) (noting that the reasonableness test "softens the harsh standard that may be inherent in the foreseeability test" and that it is an "equitable idea").

[14] Alternatively, Festo II might be interpreted to describe three distinct circumstances under which the "presumption" is rebutted: two specific and one more general. This interpretation is supported by the Federal Circuit's remand to the Court, in Amgen II, to interpret if any of the "narrow ways" of rebutting the "presumption" had been met. The Court, however, based on its interpretation of Festo II and the subsequent case law, views the three situations not as distinct, but as linked under a single, broad, overarching standard. Yet even if the Court did adopt the alternative interpretation, it would not substantively change this Court's analysis.

[15] Many of the scholarly articles written since the Supreme Court issued the Festo II decision also support this interpretation. See, e.g., Daniel H. Shulman & Donald W. Rupert, "Vitiating" the Doctrine of Equivalents: A New Patent Law Doctrine, 12 Fed. Cir. B.J. 457, 462 (2003) ("The Supreme Court held that prosecution history estoppel would apply to additional subject matter not expressly relinquished if the patentee could have drafted a claim to cover the additional subject matter, but did not. The Supreme Court identified a number of reasons why a patentee might not have been able to draft such a claim."); John S. Golian, supra, at 543 ("The Court held that in order to rebut this presumption, a patentee must show that the amendment did not surrender a particular equivalent by demonstrating that ``one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent.'"); Leading Cases, 116 Harv. L.Rev. 402, 407, 410 n. 73 (2002) [hereinafter "Leading Cases"] (stating that the "the patentee may rebut the presumption by demonstrating that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent," outlining the two specific examples and noting that the Supreme Court "plainly did not intend these examples to be exclusive, adding that ``some other reason' may suffice to rebut the presumption of estoppel") (internal citations and quotations omitted).

[16] It should be noted, however, that in its analysis of the case, the California district court considered first whether the amendment was tangential and next whether it met the over-arching standard. It then concluded, based on the facts, that the patentee had failed to demonstrate both. Bosch, 223 F.Supp.2d at 1172.

[17] This case along with other cases predating Festo II support the Court's reference to extrinsic evidence. At least one other federal court has relied on extrinsic evidence to aid the analysis. See, e.g., Envirco Corp. v. Clestra Cleanroom Inc., 1998 WL 809369, at *3 (N.D.N.Y. Nov.4, 1998) (noting that determining prosecution history estoppel is based on the patent, specification, and the prosecution history, but if the court requires extrinsic evidence to construe the claims, it may rely on dictionaries, treatises, and, if needed, expert opinion), vacated on other grounds by Envirco Corp. v. Clestra Cleanroom Inc., 209 F.3d 1360 (Fed.Cir.2000).

[18] Although the Court holds that consideration of extrinsic evidence is necessary adequately to determine whether the rebuttal standard laid out in Festo II has been met, it recognizes that such consideration, along with other patent law analyses like Markman, blur the line between the role of judge and jury. See section II.C.3, supra.

[19] The Federal Circuit did not completely rule out resort to extrinsic evidence in determining the third criterion. It stated that intrinsic evidence should be used "when at all possible" and that it "need not decide now what evidence outside the prosecution history record, if any, should be considered in determining if a patentee has met its burden under this third rebuttal criterion." Festo III, 344 F.3d at 1370. In Festo III, the issue was not squarely before the Court as it is here.

[20] The parties disagree about whether foreseeability, the first criterion, should be analyzed with respect to what was known by those skilled in the art at the time of the original patent application or at the time the narrowing amendment was made.

[21] A continuation-in-part application generally is one that claims some subject matter common to the parent application but also adds new matter that is not common to and not supported by the parent application. See e.g., Reynolds Metals Co. v. Continental Group Inc., 525 F.Supp. 950, 970 (N.D.Ill.1981); Chisholm-Ryder Co. v. Lewis Mfg. Co., 398 F.Supp. 1287, 1289 (W.D.Pa.1975); Chemithon Corp. v. Procter & Gamble Co., 287 F.Supp. 291, 298 (D.Md.1968). The original filing date, however, applies only to the subject matter that is common to both the parent application and the later-filed application: "[A]ny claims therein that include new matter are only entitled to the actual filing date of the later-filed application, and not the earlier parent application." Reynolds, 525 F.Supp. at 970; Application of Van Langenhoven, 59 C.C.P.A. 934, 458 F.2d 132, 136 (Cust. & Pat.App.1972) ("[S]ubject matter which is first disclosed in a continuation-in-part application is not entitled to the filing date of the parent application."); Chisholm-Ryder, 398 F.Supp. at 1289.

[22] Amgen's argument that this is new matter suggests that an amendment was not the correct vehicle to make its desired changes but instead that the more appropriate action would have been for Amgen to have filed a continuation-in-part application. Cf. Application of Hafner, 56 C.C.P.A. 1424, 410 F.2d 1403, 1406-07 (1969) (noting that a fatal defect cannot be cured by an amendment). Of course, if it had done this, it would not have had the benefit of the filing date of the earlier application. Id.

[23] The Court proceeds directly to determining whether Amgen has rebutted the "presumption" of estoppel. Normally, a court would first analyze whether a narrowing amendment had been made for reasons related to patentability (thereby invoking the "presumption" of prosecution history estoppel) and then proceed to determine if the "presumption" was adequately rebutted. Pioneer Magnetics, 330 F.3d at 1356-57 (dividing the analysis into three parts: (1) whether the amendment was a narrowing one; (2) whether the amendment was made for reasons related to patentability; and (3) whether the presumption can be overcome); SoundTube, 233 F.Supp.2d at 195-97 (separating the analysis into two parts: "Narrowing Amendment" and "Surrender"). Here, however, the Federal Circuit upheld the Court's determination that the amendment was a narrowing one made to overcome a double patenting objection. Further, the Federal Circuit held that this amendment was therefore related to patentability such that the presumption of prosecution history estoppel would apply. Amgen II, 314 F.3d at 1345.

[24] It also changed the dependency of claim 65 from "according to claim 61, 62, 63 or 64" to "according to claim 64 or 68." Amgen's Post-Hr'g Mem. at 12; December 20, 1995, Second Preliminary Amendment, Trial Ex. 3, Tab 3 at 119-20.

[25] As discussed, supra, consideration of extrinsic evidence is entirely proper under this criterion. Festo III, 344 F.3d at 1368-70.

[26] In Amgen I, Amgen argued that the amendment should be construed to mean "the fully processed form of the protein secreted by a cell ... when it transcribes and translates the DNA in Figure 6." Amgen I, 126 F.Supp.2d at 86. TKT/HMR contended that the phrase means "the 166 amino acid sequence of human EPO shown in Fig. 6." Id. The Court agreed that the term "mature" meant the fully processed form of EPO secreted by the cell, id., but found that the Figure 6 limitation constricted the meaning of the claim terms to the entire 166 amino acid sequence disclosed in the figure. Id. at 100. As will be discussed in greater length, infra, Amgen has now submitted extrinsic evidence indicating that Amgen intended the amendment to cover human EPO with a 165 amino acid sequence and that the Patent Office and those skilled in the art interpreted the amendment to do just that. While this evidence is important to the Court's analysis under the third criterion, it is not relevant with respect to whether the equivalent itself was foreseeable.

[27] This is not to say that Amgen's inability to predict the Court's claim construction is wholly irrelevant to whether Amgen can rebut the "presumption." On the contrary, as will be discussed below, there is a reasonableness element to the third criterion of Festo II. The fact that Amgen and one skilled in the art would have construed the amendment to cover human EPO with a 165 amino acid sequence may suggest that it is not reasonable to expect Amgen to have drafted the claim otherwise. That being said, this argument is not relevant with respect to whether the equivalent was foreseeable, and, therefore, Amgen cannot rebut the "presumption" based on foreseeability.

[28] TKT/HMR also asserts that Amgen could have drafted a claim that covered human EPO with a 165 amino acid sequence by claiming a fragment or portion of Figure 6 as it did in other claims. TKT/HMR's Mem. in Support at 7. In claim 4 of the '933 patent, "the human erythropoietin amino acid sequence set out in FIG. 6 or a fragment thereof" was claimed. Similarly, the original claim 26 of the 1984 application for the '080 patent sought coverage of "part or all of" the 166 amino acid sequence in Figure 6:

26. A synthetic polypeptide having part or all of the amino acid sequence set forth in Table VI, other than the residues within the sequence numbered 1 through 20, and having the immunological properties of naturally-occurring human erythropoietin.

Trial Ex. 2411, TKT/HMR's App., Tab C at 64 (emphasis added).

Amgen makes various counter-arguments as to why amendments with these exact words were not possible with respect to the product claims at issue. The Court declines to consider issues that are not determinative here. Amgen does not dispute that it could have claimed "mature human EPO" instead of delineating a sequence. Simply put, if it had merely stated "mature human EPO" instead of specifying Figure 6, arguably it would have served Amgen's purported purpose of distinguishing '080 as "only human" (and not "human and nonhuman") and it would have encompassed the alleged equivalent. Any argument that these words would not have been successful in avoiding a double-patenting objection is irrelevant to whether the equivalent was foreseeable and within the patentee's ability to be included in the amendment. See SmithKline, 214 F.Supp.2d at 592 n. 23 (noting that "whether the examiner would have accepted the hypothetical amendments" was not relevant to whether the patentee had the scientific ability to draft the claims to encompass the alleged equivalent).

[29] Amgen argues that the relevant inquiry is whether Amgen could have claimed the 1-165 amino acid sequence of Figure 6 specifically —not whether Amgen could have drafted a claim that would have encompassed it. Amgen's Post-Hr'g Mem. at 8-9. The Court disagrees. While it is true that the Supreme Court stated that a patentee could overcome the "presumption" by showing "some other reason suggesting that the patentee could not reasonably be expected to have described the insubstantial substitute in question," this must be read in the context of the entire opinion. Elsewhere the Supreme Court used the word "embrace" and "encompass." Festo II, 535 U.S. at 734, 741, 122 S.Ct. 1831. This is not a judgment about the strength of Amgen's arguments that it could not specifically call out a 165 amino acid sequence because it is new matter. Instead, the Court views these arguments as irrelevant to the analysis because Amgen simply could have employed the more general description noted above, that is, "mature human EPO" or "human erythropoietin."

[30] Neither the Supreme Court, nor the Federal Circuit has yet provided explicit guidance as to what qualifies as a tangential relationship. The few district court cases and the Federal Circuit's Festo III decision that address the issue are not particularly helpful here. For example, in SoundTube it was clear that the purpose of the amendment was directly related to the equivalent in question because the patentee had "trumpeted the benefits of a constant radius" in its design and "argued that this feature distinguished his invention from the prior art and satisfied section 112 concerns." SoundTube, 233 F.Supp.2d at 198. In Bosch, the prosecution history itself showed that the purpose of the amendment was related to the equivalent in question. Bosch, 223 F.Supp.2d at 1172. There, the patentee had argued that the purpose of the amendment was to narrow the location of a certain aspect of the invention, not its shape. Id. The Patent Office, however, rejected a final version of the claim that specified the location but not the shape. Id. In Festo III, one of the amendments was made in response to a rejection and both of the amendments were made to distinguish prior art patents based, partially, on an aspect of the invention relating to the equivalent. Festo III, 344 F.3d at 1371-74. Here, the intrinsic evidence is not so plain. No prior art disclosed any sequence of human EPO let alone one with a 1-165 amino acid sequence. Therefore, no claim can be made that Amgen was trying to distinguish its claims over such equivalent. Moreover, there was no rejection indicating that it was the depiction of 166 amino acids, as opposed communicating "human," that made the difference in patentability.

[31] C. Alan Fu, in his article in the Berkeley Technology Law Journal, noted that the Supreme Court did not specify what the tangential relationship was but that this was the most logical reading of the test. Fu, supra, at 136.

[32] To be clear, the '080 claims were never rejected for double patenting; rather, the amendments were made voluntarily and preemptively to distinguish the '080 claims from claim 1 of Amgen's '933 patent. Amgen's Post-Hr'g Mem. at 10; Amgen I, 126 F.Supp.2d at 135.

[33] Interestingly, two of these claims did indeed mention Figure 6 and one mentioned "human." Process claim 64 limited the technique to "host cells transformed or transfected with an isolated DNA sequence selected from the group consisting of (1) the DNA sequences set out in FIGS 5 and 6." TKT/HMR's Appendix, Tab E. Process claim 68 recited a technique that involved growing host cells "transformed or transfected with an isolated DNA sequence encoding the human erythropoietin amino acid sequence set out in FIG. 6 or a fragment thereof." (emphasis added). Amgen's Post-Hr'g Mem. at 12; Second Preliminary Amendment, Trial Ex. 3, Tab 3 at 119-120. TKT/HMR does not assert that these two claims limit the claim to human EPO and Amgen argues that it does not so limit it. This may be because these are process, not product, claims. Regardless, because it is undisputed, the Court does not read the emphasized phrases so to limit the claims.

[34] Both parties agree that a patentee can patent a product for human and non-human EPO and also patent a product for only human EPO. This does not run awry of the double-patenting rule because the same invention is not claimed in both patents. An EPO product could literally infringe the claims of one patent without literally infringing the claims of the other. For example, monkey EPO could literally infringe the patent that claims both human and non-human EPO, but it would not literally infringe the patent that claims only human EPO. See In re Vogel, 57 C.C.P.A. 920, 422 F.2d 438, 441 (1970) (noting that 35 U.S.C. § 101 "prevents two patents from issuing on the same invention.... By ``same invention,' we mean identical subject matter. Thus, the invention defined by a claim reciting ``halogen' is not the same as that defined by a claim reciting ``chlorine,' because the former is broader than the latter.").

[35] Interestingly, Amgen did not include the word "human" in the actual third preliminary amendment. Amgen claims it did not do so because it was "superfluous" with the portion of the amendment that referred to the sequence of Figure 6. Amgen's Reply Mem. [Doc. No. 733] at 8. In other words, Figure 6 — which contains 166 amino acids — sufficiently signified human EPO (as opposed to non-human). This, however, is extrinsic evidence to which the Court cannot resort in analyzing this criterion.

[36] The exact remarks are:

Applicant notes that [the amended '080 claims] all differ in scope from glycoprotein claim 1 of [the '933 patent] in specifying that the claimed subject matter comprises the mature human erythropoietin sequence of Figure 6. [Claim 1 of the '080], (like glycoprotein claim 1 [of the '933 patent]), recites carbohydrate differences in comparison to human urinary [EPO] and claim [2] recites a negative limitation with respect to isolation from human urine.

Id.

[37] Recny refers to the article that reveals that the urinary hormone is missing the COOH-terminal Arg 166 amino acid residue. Recny, et al., Structural Characterization of Natural Human Urinary and Recombinant DNA-derived Erythropoietin, J. Biol. Chem., 262(35), 17156-63 (Dec. 15, 1987). This reference is cited in the '080 patent disclosure as one that was reviewed by the Examiner. Amgen's Post-Hr'g Mem. at 11; Trial Ex. 3 ('080 Patent), TKT/HMR's App., Tab A.

[38] This reasonable inference — that the amendment was not made to differentiate a 165 amino acid EPO from a 166 amino acid EPO — would not alone be sufficient to overcome the presumption. Although one scholar deems the tangential standard met when the patentee proves there is no objective intent to disclaim, see J. Jason Lang, The German Resolution: A Proposed Doctrine of Equivalents Analysis and a Flexible Rule of Prosecution History Estoppel For Biotechnology, 52 Emory L.J. 427, 478 (2003), upon close reading of Festo II, it appears such proof is not enough. The prosecution history must affirmatively show that the purpose of the amendment was no more than tangentially related to the equivalent in question — not just a lack of intent to surrender territory. Here, however, proof of lack of intent to surrender is not the only proof supporting Amgen's contentions. As discussed earlier, the prosecution history also affirmatively supports that the purpose of the amendment was unrelated to the equivalent in question.

[39] As suggested earlier, the Court reaches this conclusion because the Federal Circuit's Festo III decision directs the Court to conduct its analysis based only on intrinsic evidence. Had the Federal Circuit interpreted the Supreme Court's decision to allow for resort to extrinsic evidence — as this Court so interpreted the decision — Amgen would not have prevailed with respect to this criterion. Two pieces of extrinsic evidence — ironically submitted by Amgen itself — demonstrate that the amendment was indeed more than tangentially related to the equivalent in question.

1) Admissions by Amgen's Counsel

Amgen's counsel stated repeatedly at the July 28, 2003 motion hearing that the purpose of the amendment was not only to limit the EPO claimed in '080 to human (as opposed to human and non-human) but to specify a sequence. "[W]e wanted claims that limited the invention to the sequence so that it specified the sequence of the protein." 7/28/2003 Mot. Hr'g Tr. at 72, ll. 24-25. "The amendment was made to distinguish claims that had no sequence and embraced EPOs of multiple species to claim a protein that has the human sequence." Id. at 73, ll. 14-16 (emphasis added). Although it is likely that Amgen wanted to specify a sequence to avoid potential objections related to the strict enablement and written description requirements of patent law, see Lang, supra, at 464 (stating that "[t]he strict enablement and written description requirements often limit claiming of biotechnological inventions to specific DNA and amino acid sequences"), such a rationale does not negate Amgen's admissions. Notwithstanding any ulterior motive for submitting the amendment, by stating that it wanted specifically to claim a sequence, Amgen admits that at least part of the amendment's rationale was more than tangentially related to any human amino-acid sequence including a 1-165 amino acid sequence, the equivalent in question.

An analogy is appropriate here. Suppose a patentee described, in an amendment, a four-cornered object in detail that happened to depict a square. The patentee may be able to prove that the rationale for the amendment was not related to the shape of the object and, therefore, was not related to an equivalent that specifically claimed a rectangular shape. An assertion by the patentee to the Court, however, that it made the amendment to specify a particular shape suggests that the amendment indeed would be more than tangentially related to the equivalent that depicted a rectangle. Cf. Bosch, 223 F.Supp.2d at 1171 (noting that "a strong argument can be made that the reason JSB used the phrase ``in a semicircle' was to convey that the lever goes around the internal gear, not to convey that the lever must be of a particular shape"). The same logic holds true here, especially when considered in conjunction with another submission by Amgen that is discussed below.

2) Expert Testimony Regarding Interpretation of the Amendment and Amgen's Intentions

Amgen asserts and supports with expert testimony that Amgen intended and those skilled in the art interpreted its amendment to cover human urinary EPO with a 165 amino acid sequence. Amgen plainly stated that it "intended the sequence limitation to cover human EPO compositions having the 1-165 amino acid sequence." Amgen's Post-Hr'g Mem. at 2; see also Amgen's Mem. in Support at 6-10. While a patentee's subjective intent ought not be considered by the Court when unsupported by the prosecution history, Vitronics, 90 F.3d at 1584, this intent is supported by the file wrapper. Amgen disclosed to the Patent Office the reference that reveals that human EPO — Example 10 of the '080 patent — has 165 amino acids; and Amgen explained to the Patent Office that other limitations in the '080 claims — not the Figure 6 limitation — distinguished its claimed EPO from all other human urinary EPO. Third Preliminary Amendment, Trial Ex. 2005, TKT/HMR's App., Tab E at 9. These two pieces of prosecution history support Amgen's assertion that it wrote the amendment in order to cover the recently discovered human EPO with a 165 amino acid sequence and that the Patent Office believed that this amendment so covered it. If the Patent Office did not believe this, it arguably would not have allowed the amendment for it would have made the patent specification incorrect because the patent specification included Example 10, then understood to result in an EPO with a 165 amino acid sequence.

In further support, an expert (and also the author of the amendment) testified that the claim "would cover ... 166, 165, whatever the process format is in the particular [host] cell." Borun Trial Test., Amgen's App., Tab D at 2884-85. This testimony supports Amgen's contention regarding its intent and that those skilled in the art would have interpreted the amendment to cover human EPO with a 165 amino acid sequence.

While all of the intrinsic evidence shows by a fair preponderance that the purpose of the amendment was not to differentiate human EPO with 166 amino acids from human EPO with 165 amino acids (as discussed earlier), when considered in combination with the extrinsic evidence, it shows at least part of the rationale underlying the amendment was to depict a sequence that covered the recently discovered human urinary EPO with 165 amino acids. As such, the rationale would be more than tangentially related to the equivalent in question. Although the amendment may have been specifically designed to avoid double patenting by claiming human EPO, it was also specifically designed to cover human urinary EPO, which was then known by Amgen, the Patent Office and those skilled in the art to have the 1-165 amino acid sequence. Thus, when considering the intrinsic and extrinsic evidence, it is apparent that the amendment is more than peripherally related to the "insubstantial substitute," human urinary EPO with 165 amino acids.

[40] The drafter is one skilled in the art. The fact that he interpreted the amendment in this way supports the contention that others skilled in the art would have interpreted the amendment similarly.

[41] Without Example 10, the fact that Amgen was claiming human EPO with 166 amino acids would not itself have been incorrect because, at the time of the amendment, it was not known that no human EPO comprising 166 amino acids existed. According to counsel, at that time there were "reports in the literature that there are expressed by certain cells 166 forms of EPO and there are expressed by other cells 165." 7/31/03 Mot. Hr'g Tr. at 77, ll. 23-25, and 78, l. 1. In other words "what was known was the gene sequence for human EPO encoded 166 amino acids" and "when that gene sequence is put in certain cells it gets 165. For example, when it was put in the CHO cells in Example 10 of Lin's patent it produced 165." Id. at 78, ll. 11-16. To this day, it is actually unknown whether there is a human EPO with 166 amino acids. Id. at ll. 3-4. "EPO has never been purified from human blood. It is so vanishingly small, no one has ever gotten to its actual human form." Id. at ll. 6-8.

[42] In Haynes, the Federal Circuit recognized that the vantage point of one skilled in the art is consistent with the vantage point of a reasonable competitor. Haynes, 8 F.3d at 1578 n. 4.

[43] As mentioned, supra, in Amgen I, Amgen argued that the amendment should be construed to mean "the fully processed form of the protein secreted by a cell ... when it transcribes and translates the DNA in Figure 6." Amgen I, 126 F.Supp.2d at 86. TKT/HMR contended that the phrase means "the 166 amino acid sequence of human EPO shown in Fig. 6." Id. The Court agreed that the term "mature" meant the fully processed form of EPO secreted by the cell, id., but found that the Figure 6 limitation limited the meaning of the claim terms to the entire 166 amino acid sequence disclosed in the figure. Id. at 100.

[44] The record indicates that the Court did not rely on extrinsic evidence to construe the claim. Amgen I, 126 F.Supp.2d at 86, n. 12 (noting that the Court did not receive evidence during the Markman hearing); see also, Amgen I, 126 F.Supp.2d at 81 (commenting on the importance of separating claim construction from issues of infringement to avoid conflating fact and law and noting that during the Markman hearing the Court considered counsel's arguments relating to the specification and prosecution history but that "evidence was neither offered nor admitted").

[45] Importantly, TKT/HMR has not produced any evidence that those skilled in the art interpret the claim differently. In fact, all of the evidence supports Amgen's contentions. Thus, under the preponderance of the evidence hurtle, Amgen has met its burden.

[46] Furthermore, it is clear from the record that claiming any amino acid sequence 165 or 166 is like claiming a moving target. No one knows what is really expressed in humans because, as Amgen's counsel explained, "EPO has never been purified from human blood. It is so vanishingly small, no one has ever gotten to its actual human form." 7/31/03 Mot. Hr'g Tr. at 78, ll. 6-8. To that end, biotechnology is itself an "unpredictable technology." Lang, supra, at 457. This unpredictability impairs a patentee's ability to draft claims to cover the full scope of the patent because the art "generally offer[s] only a handful of linguistic tools for circumscribing an invention." Id. at 477, 485 ("The unpredictability of this art alone should rebut the presumption of prosecution history estoppel."); Borun Trial Test., Amgen's App., Tab D at 2884-85 (explaining that after it became known that the 166th amino acid was chopped off, the Genetic Institute and others challenged the conclusion that a CHO cell product terminates with the 165th amino acid).