The question in this declaratory judgment action is whether a drug compound, which in its manufactured state does not infringe the patent in suit, becomes infringing as a result of transitory chemical changes that occur in vivo, that is, as a result of ingestion by the patient. There are fact issues--what changes actually occur when the drug is ingested, and how do we know; there are legal issues--does a compound that changes in the human body into the patented product on its way to becoming the active ingredient in the drug (the original patent on the active ingredient having earlier expired) literally infringe, infringe under the doctrine of equivalents, or infringe by inducement.

The District Court for the District of New Jersey, Civil Action No. 91-3423, entered judgment, following a bench trial, that plaintiff/appellant Zenith Laboratories, Inc.'s (Zenith) commercial sale of cefadroxil DC induces infringement of U.S. Patent 4,504,657 (the '657 patent),1 the patent at issue. Zenith Lab. Inc. v. Bristol-Myers Squibb Co., 24 USPQ2d 1652, 1992 WL 340761 (D.N.J.1992) (Zenith II ). The '657 patent is owned by Bristol-Myers Squibb Co. (Bristol). For the reasons set forth, we reverse.

The chemical compound cefadroxil, an antibiotic of the cephalosporin family effective against bacteria that are resistant to penicillin, was described and claimed in United States Patent No. 3,489,752 (the '752 patent), which issued in 1970; the owner was Bristol. The claims of the '752 patent covered any and all forms of cefadroxil--they described the chemical compound per se. The '752 patent expired in 1987.

In the meantime, Bristol set to work to find a commercially useful crystalline form of the compound that would overcome problems related to manufacture of the product so that it would be usable by humans. The task of developing a commercial production process was assigned to certain of Bristol's chemists. They developed what became known as the Bouzard monohydrate, named after one of the discoverers--a new crystalline form of cefadroxil. Unlike prior forms of cefadroxil, Bouzard monohydrate possesses certain characteristics in its pre-ingested, powdered form, related to bulk density, solubility, and stability (manufacturing-related characteristics), which make it particularly suitable for packaging into capsules.

The Bouzard monohydrate was the subject of the single claim in the '657 patent:

1. Crystalline 7-[D--amino--(p-hydroxyphenyl-)acetamido]-3-methyl-3-cephem-4-carboxylic acid monohydrate exhibiting essentially the following x-ray diffraction properties: [a 37-line table of relative intensities exhibited by Bouzard monohydrate at various scan angles].

In July 1988, Zenith, the plaintiff in this declaratory judgment action, contracted with a Spanish company, Gema, S.A. of Barcelona (Gema), to become the exclusive United States distributor of the form of cefadroxil manufactured by Gema and known as cefadroxil DC. Cefadroxil DC is a hemihydrate form of cefadroxil and thus differs structurally from Bouzard monohydrate.2 Zenith and Gema then undertook to obtain approval from the Food and Drug Administration (FDA) for commercial marketing in the United States of cefadroxil DC. Pursuant to FDA regulations, Zenith and Gema sought abbreviated approval on the grounds that cefadroxil DC was bioequivalent to a form of cefadroxil monohydrate (other than Bouzard monohydrate) which had already been approved by the FDA for commercial sale. In October 1990, the FDA granted approval.3

In June 1991, after learning of Zenith's and Gema's efforts to obtain FDA approval, Bristol filed suit against Gema (and Gema's affiliates) in the District Court for the District of Maryland, Civil Action No. HAR 91-1765 (the Maryland action), alleging infringement of the '657 patent. Subsequently, when Gema announced it was abandoning its plans to manufacture cefadroxil DC, that suit was dismissed upon stipulation of the parties.

Zenith then, in August 1991, filed the present declaratory judgment action against Bristol in the District Court for the District of New Jersey. Zenith's complaint contained four counts. In count one, Zenith sought a declaration that cefadroxil DC does not infringe the '657 patent, and in count two, a declaration that Bristol is equitably estopped from asserting that cefadroxil DC infringes the '657 patent. In count three, Zenith alleged that Bristol, by initiating the Maryland action, violated a consent decree previously entered in settlement of prior litigation between the parties; and in count four, that Bristol had engaged in unfair competition. (Zenith subsequently added a fifth and still pending count alleging that Bristol had committed antitrust violations.)

Bristol then moved to dismiss counts one and two, and for summary judgment on counts three and four. Zenith filed cross-motions for summary judgment on these counts. On December 12, 1991, the trial court granted Bristol's motion to dismiss count three (concerning the prior consent decree) on the ground that the count was moot in view of the previous dismissal of the Maryland action. Zenith Lab. Inc. v. Bristol-Myers Squibb Co., 24 USPQ2d 1641, 1651, 1991 WL 267892 (D.N.J.1991) (Zenith I ). As to Zenith's motion for summary judgment on count one (the infringement count), the court stayed that motion pending an oral hearing. Id. at 1648. The court denied all other motions. Id. at 1648, 1650-51.

At the hearing on the infringement count, Bristol conceded that cefadroxil DC, being a hemihydrate, did not literally infringe the '657 patent in its pre-ingested form. Bristol nevertheless argued that Zenith was liable under the patent on two grounds: Zenith's product infringed under the doctrine of equivalents; and Zenith's product converted into the patented compound in the patient's stomach, and thus the sale of cefadroxil DC would induce infringement of the '657 patent under 35 U.S.C. Sec. 271(b) (1988). The trial court concluded that summary judgment was appropriate since Bristol had not raised a genuine factual dispute on these issues; and on March 4, 1992, the court granted Zenith's motion for summary judgment on count one.

However, subsequently, on April 16, 1992, upon Bristol's motion for reconsideration, the court vacated its March 4, 1992 decision. The court denied Zenith's motion for summary judgment on count one on the basis that Bristol, through two declarations submitted with its motion for reconsideration, had demonstrated a genuine dispute on the in vivo conversion issue. Consequently, it ordered that that count be severed from the remaining counts and tried separately.

From May 26, 1992 through June 5, 1992, the court conducted a bench trial on count one. Following the trial, the court rejected Bristol's theory of infringement under the doctrine of equivalents. It found that Bristol, through prosecution history estoppel, had relinquished coverage of cefadroxil DC. However, the court found that cefadroxil DC converts to Bouzard monohydrate in the patient's stomach. Since an act of literal infringement thus occurs in the patient's stomach as a result of ingestion of cefadroxil DC, the court concluded that Zenith's sale of cefadroxil DC would induce infringement of the '657 patent.

On August 6, 1992, the court, in accordance with its findings and conclusions, entered judgment for Bristol. Zenith II, 24 USPQ2d 1652. The court, pursuant to 35 U.S.C. Sec. 271(e)(4)(A) (1988), ordered that the effective date of any approval by the FDA of cefadroxil DC for commercial distribution and sale by Zenith could be no earlier than March 12, 2002, the expiration date of the '657 patent. Id. at 1676. It subsequently certified the finality of that judgment pursuant to Rule 54(b) of the Federal Rules of Civil Procedure. Zenith then filed this appeal.

Zenith presents five separate grounds on which the judgment of the trial court should be reversed. First, the claim of the '657 patent, properly construed, does not cover Bouzard crystals which might form momentarily in a patient's stomach. Second, even assuming conversion in vivo is within the scope of the claim, there was a failure of proof as to whether any crystals, formed in the stomach from ingested cefadroxil DC, literally infringe the '657 patent. Third, even if such crystals formed, there would be no infringing "use" within the meaning of 35 U.S.C. Sec. 271(a) (1988). Fourth, the doctrine of equivalents, applied in reverse, precludes a judgment of literal infringement. And fifth, the court erroneously declined to grant summary judgment that Bristol was equitably estopped from asserting that cefadroxil DC infringed the Bouzard patent.

We begin with claim construction: whether the claim of the '657 patent, properly construed, covers Bouzard crystals which might form momentarily in a patient's stomach. Zenith argues that, due to statements made by Bristol during the '657 prosecution, the '657 claim is limited to the pre-ingested, powdered form of Bouzard monohydrate. In those statements Bristol particularly emphasized the superior manufacturing-related benefits of the pre-ingested form of Bouzard monohydrate in relation to prior forms of cefadroxil. Bristol, it is argued, thus relinquished coverage of any forms of cefadroxil DC that do not exhibit these manufacturing-related characteristics--that would obviously include any Bouzard monohydrate formed in a patient's stomach.

Zenith has a point. Prosecution history serves as a limit on the scope of claims by excluding any interpretation of the claim language that would permit the patentee to assert a meaning for the claim that was disclaimed or disavowed during prosecution in order to obtain claim allowance. Standard Oil Co. v. American Cyanamid Co., 774 F.2d 448, 452, 227 U.S.P.Q. (BNA) 293, 296 (Fed.Cir.1985). The prosecution history in this case is replete with arguments by Bristol that the importance of this invention was the discovery of a way to manufacture the known compound so as to make it commercially available for medical treatment.

The difficulty with Zenith's argument, however, is in the nature of the sole claim in the patent. The claim as written and allowed simply describes a compound having specified chemical properties. The question before us is not one of validity: whether the claim would be patentable over the prior art if the suggested restriction were not applicable. The question here is one of infringement: whether there is anything in the claim as issued that limits it to the pre-ingested form, keeping in mind the principle that it is improper to read a limitation into a claim "wholly apart from any need to interpret what the patentee meant by particular words or phrases in the claim." E.I. Du Pont de Nemours & Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1433, 7 USPQ2d 1129, 1131 (Fed.Cir.), cert. denied, 488 U.S. 986, 109 S. Ct. 542, 102 L. Ed. 2d 572 (1988).

Zenith nonetheless argues that the term "properties" in the claim is the vehicle for incorporating its proposed restriction into the claim. But that term plainly refers to x-ray diffraction properties. X-ray diffraction properties have nothing to do with the manufacturing benefits which Bristol highlighted.

Zenith's argument suffers from additional infirmities. The '657 claim is a claim for a compound. Its patentability thus derives from the structure of the claimed compound in relation to prior compounds. See In re Spada, 911 F.2d 705, 708, 15 USPQ2d 1655, 1657 (Fed.Cir.1990). The relevance to patentability of the properties or characteristics exhibited by the compound is limited to assessing the significance of the structural distinctions of the claimed compound over the prior art. Second, as we later note in more detail, it is not at all clear that the statements concerning these characteristics played a determinative or even significant role in the Patent and Trademark Office's (PTO's) decision to grant the patent. We conclude, therefore, that while the claim as issued is limited to the crystalline form of cefadroxil exhibiting the specified x-ray diffraction pattern, it is not limited to the compound in its pre-ingested form with the manufacturing characteristics which Bristol emphasized during prosecution.4

We turn next to the question of whether Bouzard monohydrate is actually found in the stomach of patients who ingest cefadroxil DC. One answer is that no one knows--the scientific fact appears to be that there is no known way to actually sample the contents of patients' stomachs at the precise moment and conduct the x-ray diffraction analyses required to ascertain if all 37 lines described in the patent are present.

For its proof Bristol offered instead the testimony of its principal scientific witness, Dr. Harry Brittain, who described various studies and experiments simulating the environment of the human stomach which were conducted under his auspices; Zenith countered with the testimony of its principal scientific witness, Dr. Martha Greenblatt. The trial judge, after hearing all the evidence, concluded that the Bouzard monohydrate in fact does inevitably form in the patient's stomach after ingestion of cefadroxil DC. In deciding this issue, the court specifically resolved the conflicting testimony presented by Dr. Brittain and Dr. Greenblatt:

[T]he Court finds that Brittain has established by a preponderance of evidence, well within his range of expertise, that cefadroxil DC in fact converts in vivo to Bouzard monohydrate.

The strength of Brittain's experiments and results is their totality. Piece by piece, Brittain built an impressive collection of data. The results of each series of experiments strengthen the results of each other set of experimental results. The total effect of the experiments is to prove with sufficient certainty the fact of conversion in vivo.

* * * * * *

[A]lthough she may have significant expertise in fields of inquiry unrelated to the issues in this case, Greenblatt demonstrated little understanding of the problem at issue in this case or the reason for her procedures and thus played no role in the design of her experiments, ..., was lax in her methodology, particularly in the recording of her procedures and observations, and in fact did not even participate in the second series of experiments. The Court finds that Brittain's ... results are simply better supported and vastly more credible. According, those findings are adopted by the Court.

Zenith II, 24 USPQ2d at 1664, 1668.

Courts of law are not the optimal fora for trying questions of scientific truth; even less are appellate courts equipped to choose among conflicting scientific theories. Judges are dependent on the quality and strength of the scientific evidence brought before them. The trial judge who hears the witnesses and does the initial assessment of the evidence deserves considerable latitude in eventually choosing sides based on what is ultimately an educated guess--that is what judging is about.

However, in determining whether a claim in a patent has been infringed, the scientific theories utilized must establish the presence of the limitations recited in the claim. In this case, Dr. Brittain based his scientific conclusions on three kinds of tests--visual observation (obtained through optical microscopy), birefringence comparison, and x-ray diffraction pattern comparison. The trial judge explained:

Brittain first established indisputably through photomicrography that cefadroxil DC crystals are much smaller than Bouzard monohydrate crystals, and are much different in shape. He also established that cefadroxil DC crystals exhibit a much weaker birefringence than do crystals of Bouzard monohydrate, which are highly birefringent. Thus, through microscopy, Brittain could determine with a high degree of certainty whether a sample of wetted cefadroxil DC had changed its crystalline form, and could conclude to a lesser extent that the form to which it had changed was Bouzard monohydrate. Although changes in the physical properties of shape, size and birefringence of the cefadroxil DC crystals could be conclusively determined through microscopy, Brittain could not determine by observation alone that the larger, more birefringent crystals were necessarily Bouzard. To determine conversion to Bouzard with great certainty, Brittain ran x-ray powder diffraction patterns, a technique that in essence takes the "fingerprint" of a crystalline compound. Through x-ray diffraction analysis, Brittain determined with much greater confidence that all of the conversions suspected after optical examination had in fact been to Bouzard monohydrate. (Emphasis added)

Id. at 1665.

The key piece of evidence, then, was the x-ray diffraction pattern analysis. This is also what the claim in the patent requires. The first two kinds of tests are only inferentially relevant given that the claim is drafted in terms of x-ray diffraction lines. Thus they cannot substitute for a proper comparison of those lines.

And here a major problem arises. In order to establish its case, Bristol had to show that the accused compound infringed the claim contained in the patent.5 This required Bristol to show that the diffraction pattern of cefadroxil DC following its conversion in vivo displayed the same diffraction pattern as that of the claimed compound. The district court, instead of requiring the comparison of the accused compound following conversion to be made with the lines specified in the claim, allowed Bristol to make the comparison with the diffraction pattern exhibited by a sample (the reference pattern) of a material considered by Bristol to be the patented compound.

As we have repeatedly said, it is error for a court to compare in its infringement analysis the accused product or process with the patentee's commercial embodiment or other version of the product or process; the only proper comparison is with the claims of the patent. E.g., Martin v. Barber, 755 F.2d 1564, 1567, 225 U.S.P.Q. (BNA) 233, 235 (Fed.Cir.1985).

The difficulty was compounded. The x-ray diffraction pattern exhibited by Bristol's sample (the reference pattern) consisted of a table of only 30 lines of relative intensities. Of this total, the court only compared 22 lines to corresponding lines recited in the claim. Zenith II, 24 USPQ2d at 1665. Based on its comparison, the court concluded the two were sufficiently similar to permit Bristol to use the reference pattern in its infringement analysis. In fact, the number of lines recited in the claim is 37. Thus, 15 of the lines recited in the claim (representing about 40% of the total) were not considered by the court in its comparison. Although the term "essentially" recited in the claim permits some leeway in the exactness of the comparison with the specified 37 lines of the claim, it does not permit ignoring a substantial number of lines altogether. It is the claim that sets the metes and bounds of the invention entitled to the protection of the patent system.

On the basis of this evidence the trial court concluded that when cefadroxil DC is ingested Bouzard monohydrate is created in a patient's stomach, that that constitutes an infringing use, and that therefore the sale of cefadroxil DC by Zenith would constitute inducement of infringement under 35 U.S.C. Sec. 271(b). Since the finding of infringement was based on testimony which incorporated an improper comparison, and since that comparison was an essential element in the conclusion that infringement occurred, the conclusion that Zenith by selling cefadroxil DC would engage in inducement of infringement is insupportable. Zenith is correct that there was a failure of proof as to whether any crystals, assumed to form in the stomach from ingested cefadroxil DC, literally infringe the '657 claim. In the absence of evidence comparing the '657 claim with the cefadroxil DC after ingestion Bristol has failed to establish any infringing use and therefore we must reverse the district court's conclusion that Zenith's sale of cefadroxil DC induces infringement of the '657 patent.6

On the question of infringement under the doctrine of equivalents, the trial court decided for Zenith, finding that Bristol, through statements it made during the '657 prosecution, was estopped via prosecution history estoppel from asserting that cefadroxil DC infringed under the doctrine of equivalents. According to Bristol, this was error, because the statements relied on by the district court do not support a finding that there was an estoppel; thus the doctrine of equivalents, properly applied, provides an alternative ground on which the judgment for Bristol could be affirmed.

Bristol's point has merit. The essence of prosecution history estoppel is that a patentee should not be able to obtain, through litigation, coverage of subject matter relinquished during prosecution. See Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 870, 228 U.S.P.Q. (BNA) 90, 96 (Fed.Cir.1985). The legal standard for determining what subject matter was relinquished is an objective one, measured from the vantage point of what a competitor was reasonably entitled to conclude, from the prosecution history, that the applicant gave up to procure issuance of the patent. See Prodyne Enters., Inc. v. Julie Pomerantz, Inc., 743 F.2d 1581, 1583, 223 U.S.P.Q. (BNA) 477, 478 (Fed.Cir.1984); Kinzenbaw v. Deere & Co., 741 F.2d 383, 389, 222 U.S.P.Q. (BNA) 929, 933 (Fed.Cir.1984), cert. denied, 470 U.S. 1004, 105 S. Ct. 1357, 84 L. Ed. 2d 379 (1985). Here, we do not believe a competitor would have been reasonably justified in concluding that Bristol, through the statements in question, gave up coverage of those forms of Bouzard monohydrate other than the pre-ingested form.

There is nothing in the record indicating that the statements about manufacturing benefits played any role in the procurement of the '657 patent. As noted earlier, in these statements Bristol argued the superior manufacturing-related characteristics exhibited by Bouzard monohydrate in relation to prior forms of cefadroxil. Yet, these prior forms of cefadroxil were never asserted as prior art by the examiner, and the examiner expressly indicated that Bristol's assertions were not a factor in determining patentability. Moreover, according to the examiner's statement of reasons for allowance, the sole basis upon which allowance was challenged, and upon which the '657 patent was eventually issued, was related to a single prior art reference, Garbrecht, U.S. Patent No. 3,781,282,7 and Bristol's successful showing that the claimed compound was not inherent in that reference. There is no indication the examiner ever relied on the statements regarding manufacturing-related characteristics in allowing the patent to issue.8

Zenith asserts that the statements in question forestalled an obviousness rejection of the claim on the basis of the prior forms of cefadroxil. That, however, is at best highly speculative; it is unpersuasive in view of the examiner's statement of reasons for allowance. Thus, the court's conclusion, that the statements in question gave rise to prosecution history estoppel, is insupportable.

However, any error in the court's application of prosecution history estoppel was harmless; the court's conclusion that the doctrine of equivalents is inapplicable to this case is correct as a matter of law. To prove infringement under the doctrine of equivalents, Bristol bore the burden of showing that cefadroxil DC was the equivalent of Bouzard monohydrate. See Corning Glass Works v. Sumitomo Elec., U.S.A., Inc., 868 F.2d 1251, 1259, 9 USPQ2d 1962, 1967 (Fed.Cir.1989). A necessary part of the function/way/result equivalency analysis is the function of the substituted element as seen in the context of the patent, the prosecution history, and the prior art. Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 609, 70 S. Ct. 854, 856, 94 L. Ed. 1097 (1950), reh'g denied, 340 U.S. 845, 71 S. Ct. 12, 95 L. Ed. 620 (1950).

In view of Bristol's numerous statements in the prosecution history regarding the superior manufacturing-related benefits of the Bouzard crystal in relation to the prior forms of cefadroxil, it is clear that the primary, if not the only, function of the Bouzard crystal form of the drug as compared to other forms is to facilitate pre-ingestion manufacturing. No other intended function is described or suggested.9 Since any unanticipated production of the Bouzard crystal in the patient's stomach as a result of ingesting cefadroxil DC does not even remotely perform that function, the "function" part of the function/way/result test of equivalency is not met. As a matter of law, there can be no infringement under the doctrine of equivalents.

Bristol cites a number of cases to demonstrate that the doctrine of equivalents has been applied to find infringement in situations involving in situ or in vivo conversions. However, these cases do not support Bristol's argument. In two, Broadview Chemical Corp. v. Loctite Corp., 159 U.S.P.Q. (BNA) 80, 1968 WL 8397 (D.Conn.1968), aff'd in relevant part, 406 F.2d 538, 160 U.S.P.Q. (BNA) 449 (2d Cir.), cert. denied, 394 U.S. 976, 89 S. Ct. 1472, 22 L. Ed. 2d 755 (1969) and Studiengesellschaft Kohle, M.b.H. v. Dart Industries, Inc., 549 F. Supp. 716, 216 U.S.P.Q. (BNA) 381 (D.Del.1982), aff'd, 726 F.2d 724, 220 U.S.P.Q. (BNA) 841 (Fed.Cir.1984), the doctrine of equivalents was not the issue in the case.10 In the others, Atlas Powder Co. v. E.I. DuPont de Nemours & Co., 588 F. Supp. 1455, 221 U.S.P.Q. (BNA) 426 (N.D.Tex.1983), aff'd, 750 F.2d 1569, 224 U.S.P.Q. (BNA) 409 (Fed.Cir.1984); Chemical Cleaning Corp. v. Dow Chemical Co., 379 F.2d 294, 155 U.S.P.Q. (BNA) 49 (5th Cir.1967), cert. denied, 389 U.S. 1040, 88 S. Ct. 777, 19 L. Ed. 2d 829 (1968); Ortho Pharmaceutical Corp. v. Smith, 18 USPQ2d 1977, 1990 WL 121353 (E.D.Pa.1990), aff'd, 959 F.2d 936, 22 USPQ2d 1119 (Fed.Cir.1992); and Beecham Group Ltd. v. Bristol Lab. Ltd., [1978] R.P.C. 153 (House of Lords 1977), the substituted compound or ingredient converted in vivo or in situ to the compound or ingredient called for by the claim, and performed the same function as that of the claim.11 The infringer was thus using the equivalent element to perform the same function as the claimed compound, a typical case for doctrine of equivalents application.

Even assuming that cefadroxil DC converts in vivo to Bouzard monohydrate in measurable amounts, that particular crystalline form does not perform the function of facilitating encapsulation. On the record before us, the trial court was correct in his judgment that Bristol had failed to establish infringement under the doctrine of equivalents.

For the reasons given, the judgment of the trial court in favor of Bristol on count one is reversed.

REVERSED.