Glaxo, Inc., and Glaxo Group Limited v. Novopharm, Ltd.

LOURIE, Circuit Judge. Glaxo, Inc. and Glaxo Group Limited (collectively “Glaxo”) appeal from the judgment of the United States District Court for the Eastern District of North Carolina, dismissing Glaxo’s claims that Novopharm Ltd. and its wholly owned subsidiary, Granutec, Inc., (collectively “Novopharm”) (1) infringed U.S. Patent 4,521,431 under 35 U.S.C. § 271(e) by filing an abbreviated new drug application (ANDA), (2) would infringe U.S. Patent 4,672,133 under 35 U.S.C. § 271(g) by importing materials following the approval of the ANDA, and (3) violated state unfair competition law by violating a protective order. Glaxo, Inc. v. Novopharm Ltd., 931 F.Supp. 1280 (E.D.N.C.1996). Because the district court did not err in finding that Glaxo failed to prove infringement of the ’431 and ’133 patents or in determining that Glaxo could not prohibit Novopharm from using publicly available information, we affirm. BACKGROUND This appeal involves the parties’ ongoing dispute over Novopharm’s efforts to market *1564 a generic version of Glaxo’s highly successful anti-ulcer medication, Zantac®. The active ingredient in Zantac® is ranitidine hydrochloride (RHC1), a salt that may occur in at least two distinct crystalline forms. Glaxo owns three patents that are relevant to this dispute: U.S. Patents 4,128,658; 4,521,431; and 4,672,133. The ’658 patent discloses a method of making RHC1 and claims the compound per se, both specifically and generieally. This patent application was apparently filed before Glaxo scientists knew that RHC1 could exist in more than one crystalline form. It makes no reference to crystalline form, but the crystalline form of RHC1 first obtained by the procedures described in the ’658 patent has come to be known as Form 1 RHC1. The ’658 patent will expire on July 25, 1997. After Glaxo scientists discovered that a different crystalline form of RHC1 could be made using the process described in the ’658 patent and that this new form had physical characteristics preferable to those of the first known crystalline form, they filed a patent application for the new form, which they named Form 2 RHC1. This crystalline form of RHC1 is currently the active ingredient in Zantac®. The ’431 and ’133 patents issued from this application, and they claim, respectively, Form 2 RHC1 and a process for making Form 2 RHC1. The claims in both of these patents characterize Form 2 RHC1 by means of a specific, 29-peak infra-red (IR) spectrum. Several dependent claims in these patents also define Form 2 RHC1 as characterized by a 32-intensity x-ray powder diffraction pattern. The ’431 patent will expire in 2002 and the ’133 patent will expire in 2004. In 1991, Novopharm Ltd., a Canadian-based manufacturer, formulator, and seller of generic pharmaceuticals, filed an ANDA with the United States Food and Drug Administration (FDA) seeking permission to sell Form 2 RHC1. See 21 U.S.C. § 355(j) (1994); 21 C.F.R. § 314 Subpart C (1996). In the course of an infringement suit brought by Glaxo under 35 U.S.C. § 271(e)(2), Novop-harm admitted infringement, but challenged the validity of the ’431 patent, alleging that the ’658 patent anticipated the ’431 patent’s claims to Form 2 RHC1. The district court found that none of the methods disclosed in the ’658 patent invariably yielded crystals of Form 2 RHC1, but could yield crystals of either crystalline form. Glaxo, Inc. v. Novopharm Ltd., 830 F.Supp. 871, 876-77, 29 USPQ2d 1126, 1130-31 (E.D.N.C.1993) (Glaxo 1). Based on this finding, the court concluded that the ’658 patent did not inherently disclose Form 2 and therefore did not anticipate the claims of the ’431 patent. Id. at 876-77, 881, 29 USPQ2d at 1131, 1135. We affirmed, inter alia, on the grounds that the district court’s factual findings were not clearly erroneous and that, based on those findings, the court properly rejected Novop-harm’s anticipation defense. Glaxo Inc. v. Novopharm Ltd., 52 F.3d 1043, 1047-48, 34 USPQ2d 1565, 1567 (Fed.Cir.), cert. denied, — U.S. -, 116 S.Ct. 516, 133 L.Ed.2d 424 (1995) (Glaxo 2). Not discouraged by its inability to use Form 2 to bring RHC1 to the market, Novop-harm filed another ANDA on April 25, 1994, this time seeking approval to market Form 1 RHC1. As part of this ANDA, Novopharm filed a so-called “paragraph IV certification,” indicating that its product would not infringe the ’431 patent and that it did not intend to market Form 1 RHC1 before the expiration of the ’658 patent. See 21 U.S.C. § 355(j)(2)(A)(vii)(IV) (1994). The ANDA, as initially filed, specified that the marketed product be approximately 99% pure Form 1 RHC1 (with impurities that may include Form 2 RHC1) as determined by an IR spectroscopic analysis. Amended ANDAs filed by Novopharm would have permitted the marketed product to have a Form 1 RHC1 purity as low as 90%. In response, on July 22, Glaxo sued No-vopharm, asserting infringement of the ’431 patent under § 271(e)(2) and seeking a declaratory judgment that Novopharm would infringe the ’133 patent under § 271(g) if and when it imported the product following FDA approval of the ANDA. Glaxo also asserted a state law unfair competition claim, based upon its allegation that Novopharm and its expert witness, Dr. Natalie Lazarowyeh, misappropriated Glaxo trade secrets regarding the production of Form 1 RHC1 in violation of a protective order issued by the Glaxo 1 court. Novopharm counter-claimed, assert *1565 ing that Glaxo violated federal antitrust and state unfair competition laws. After conducting a bench trial, the district court first determined that, in order to prevail under § 271(e)(2), Glaxo must prove by a preponderance of the evidence that the product sold by Novopharm pursuant to the approved ANDA “will at least more probably than not read on the [’431] patent.” Glaxo, Inc. v. Novopharm, Ltd., 931 F.Supp. 1280, 1286 (E.D.N.C.1996). The court first construed the ’431 and ’183 claims to be limited to “pure Form 2 [RHC1].” Id. at 1291. The court further found that, even assuming that mixtures of Form 1 and Form 2 RHC1 would infringe the patents, Glaxo had not met its burden of proving infringement of the patents. Id. at 1291-97. Finally, the court concluded that Glaxo’s state law claims were without merit because the confidentiality of all of the relevant proprietary information possessed by Novopharm and Lazarowych was waived by Glaxo at the public Glaxo 1 trial, id. at 1300, or, in the alternative, was independently developed by Novopharm, id. at 1304. Accordingly, the court entered judgment for Novopharm on all of Glaxo’s claims and dismissed the action. Glaxo appealed. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1) (1994). DISCUSSION Glaxo argues that the district court erred by dismissing its claim for infringement of the ’431 patent, its claim seeking a declaration that Novopharm’s post-approval importation of RHC1 would infringe the ’133 patent, and its state law unfair competition claim. We address the issues in that order. A. Infringement of the %31 Patent A determination of infringement requires a two step analysis. “First, the claim must be properly construed to determine its scope and meaning. Second, the claim as properly construed must be compared to the accused device or process.” Carroll Touch, Inc. v. Electro Mechanical Sys., Inc., 15 F.3d 1573, 1576, 27 USPQ2d 1836, 1839 (Fed.Cir.1993). The first step, claim construction, is a question of law which we review de novo; the proper construction of the claims is based upon the claim language, the specification, the prosecution history, and if necessary to aid the court’s understanding of the patent, extrinsic evidence. See Markman v. Westview Instruments, Inc., 52 F.3d 967, 979-81, 34 USPQ2d 1321, 1329-31 (Fed.Cir.1995) (in bane), aff'd, — U.S. -, 116 S.Ct. 1384, 134 L.Ed.2d 577, 38 USPQ2d 1461 (1996). The second step, determining whether a particular device infringes a properly construed claim, is a question of fact which we review for clear error on appeal from a bench trial. See Fromson v. Advance Offset Plate, Inc., 720 F.2d 1565, 1569, 219 USPQ 1137, 1140 (Fed.Cir.1983). In order to prove infringement, a patentee must show that every limitation of the claims asserted to be infringed is found in the accused device. See Dolly, Inc. v. Spalding & Evenflo Cos., 16 F.3d 394, 397, 29 USPQ2d 1767, 1769 (Fed.Cir.1994). Glaxo argues that the district court erred by construing the ’431 (and the ’133) claims to cover only pure Form 2 RHC1 (and a method for making pure Form 2 RHC1). Glaxo argues that the claims are properly construed to encompass products containing any amount of Form 2 RHC1 and methods for making such products. Novopharm responds that the district court did not base its decision on a narrow (“pure Form 2 RHC1”) claim construction. Rather, Novopharm argues that, because the court accepted Glaxo’s broad claim construction for purposes of its infringement analysis, the narrow claim construction was dictum, not binding upon the parties. In fact, at oral argument Novop-harm disavowed the narrow claim construction by admitting that “there is nothing that comes from ... the lower court’s decision that gives Novopharm any entitlement to make a product that contains a mixture [containing Form 2].” We agree that the claims are not limited to pure Form 2 RHC1. In Glaxo 2, we explicitly affirmed the Glaxo 1 court’s finding that Form 2 RHC1 was not reliably obtained by carrying out the procedures of Example 32 of the ’658 patent. Glaxo 2, 52 F.3d at 1047-48, 34 USPQ2d at 1567. That case thus established that Form 2 RHC1 was not inherently produced by the procedures of the ’658 patent and therefore that the validity of the claims to Form 2 RHC1 does not depend on their being limited to pure Form 2 *1566 RHC1. Accordingly, the patent claims directed to Form 2 RHC1 need not be interpreted, nor has Glaxo interpreted them, to exclude mixtures that contain Form 1 RHC1. We therefore hold that the district court erred in stating that the ’431 (and ’133) claims cover only “pure” Form 2 RHC1. However, that error was harmless because the court also found that Glaxo’s proof was not sufficient to prove infringement even if the claims were construed properly, ie., to cover mixtures containing Form 2 RHC1. 1 Based on the evidence before it and on the proper claim construction, the district court found that Glaxo failed to prove infringement by a preponderance of the evidence. In fact, the court found that “the preponderance of the [infrared (IR) spectral] evidence indicates that Form 2 [RHC1] is not present in Novopharm’s product.” 931 F.Supp. at 1295. The court also found that the x-ray diffraction evidence before it demonstrated “in clear and convincing fashion that Novopharm’s product would not contain any Form 2 RHC1” and thus would not infringe either the ’431 (or the ’133) patent. Id. at 1298. Glaxo argues that the district court clearly erred in finding a lack of proof. Specifically, Glaxo argues that the only support for the district court’s finding is the following isolated statement of Glaxo’s expert witness, Dr. Byrn: “I would put a probability of 60/40 that [an IR spectrum of Glaxo’s Form 1 RHC1 product is] not Form 2.” Glaxo also argues that because Novopharm’s ANDA initially determined the content of Form 2 RHC1 in its Form 1 RHC1 product based upon a single IR peak at 1045 cm-1, the district court must now engage in an infringement analysis based upon that single peak. Novopharm responds that the district court did not clearly err in finding that Glaxo had failed to prove infringement because the great weight of the evidence adduced at trial indicated that Novopharm will not sell a product containing any identifiable Form 2 RHC1. Specifically, Novopharm argues that the district court properly decided not to rely solely upon the IR peak at 1045 cm-1 because that peak was not necessarily indicative of Form 2 RHC1 and, more importantly, the claims at issue all identify Form 2 RHC1 by reference to a 29-peak IR spectrum. Accordingly, says Novopharm, proof of infringement requires proof that the drug alleged to infringe would exhibit all of those peaks, not a single, potentially meaningless peak. We agree with Novopharm that the evidence supports the district court’s finding that the single peak analyzed was not sufficient to substitute for the claimed 29-peak spectrum. It is elementary patent law that all limitations are material. The single-peak analysis was thus insufficient because, as the district court correctly noted, in order to prove infringement Glaxo was required to establish the presence of each limitation of the asserted claims. See Zenith Lab., Inc. v. Bristol-Myers Squibb Co., 19 F.3d 1418, 1423, 30 USPQ2d 1285, 1289 (Fed.Cir.1994). The court did not clearly err in finding thát Glaxo faded to demonstrate that the accused compositions would exhibit either the claimed characteristic IR or x-ray powder diffraction spectra of Form 2 RHC1, or in concluding that this failure was fatal to Glaxo’s case. Contrary to its arguments before this court, Glaxo was well aware of this basic principle of patent law that to prove infringement, it would be required to show that Novopharm’s product met substantially all of the limitations of the claims. During the prosecution of the ’431 patent, in an argument made to successfully overcome a double-patenting rejection, Glaxo made the following statement: “Claim 2 is dependent upon Claim 1 and in order to infringe Claim 2, the compound must meet all of the limitations of Claim 1 as well as the limitations of Claim 2.” This accurate statement of law demonstrates that Glaxo knew that its decision to claim Form 2 RHC1 according to its IR and x-ray powder diffraction characteristics would later control the nature of the evidence necessary to prove infringement. In this case, Glaxo failed to show that every limitation of the claims it chose to prosecute was present in Novop-harm’s allegedly infringing composition. *1567 Moreover, the court’s finding that the single IR peak was “meaningless” to the identification of Form 2 RHC1 is supported by the testimony of Novopharm’s witness, Dr. Du-rig, as well as by Glaxo’s own witnesses, Drs. Klinkert and Snyder. Durig testified that a sample of Form 1 RHC1 (a sample that Glaxo stipulated to contain no Form 2 RHC1) exhibited a significant peak at 1045 em-1. Similarly, Glaxo’s own witnesses, Klinkert and Snyder, agreed that, given the range of possible chemical impurities, the presence of the single IR peak was insufficient to differentiate Form 1 RHC1 from Form 2 RHC1. These statements, found credible by the district court, not merely the testimony of Byrn, provide ample support for the court’s factual finding that Glaxo failed to prove infringement under a single-peak analysis. Glaxo also argues that the district court erred by applying a conventional infringement analysis, i.e., comparing the allegedly infringing composition to the claims, in an action brought under § 271(e)(2). More precisely, Glaxo argues that the district court erred by focusing on what Novopharm will sell under the ANDA if and when the ANDA is approved, instead of focusing solely on the fact that the scope of approval sought by Novopharm would allow it to manufacture compositions containing Form 2 RHC1. Under Glaxo’s reading of the statute, the ANDA itself, because of its broad scope, was sufficient evidence to prove infringement. Alternatively, Glaxo argues that the ANDA’s broad scope was sufficient to shift to Novop-harm the burden of proving that the ultimately marketed drug would not infringe. Novopharm responds that under § 271(e)(2), the scope of the ANDA is not the sole factor in an infringement analysis and that it does not alter a patentee’s normal burden of proving infringement by a preponderance of the evidence. Novopharm argues that the district court correctly required Glaxo to prove that the product that Novopharm ultimately would put on the market would likely infringe the ’431 patent. We agree with No-vopharm’s view of the statute. We review a district court’s interpretation of statutory language de novo. Astra v. Lehman, 71 F.3d 1578, 1580, 37 USPQ2d 1212, 1214 (Fed.Cir.1995) (citing Glaxo Operations UK Ltd. v. Quigg, 894 F.2d 392, 395, 13 USPQ2d 1628, 1630 (Fed.Cir.1990)). “When ... the terms of a statute are unambiguous, judicial inquiry is complete, except in rare and exceptional circumstances. Moreover, absent a clearly expressed legislative intent to the contrary, a statute’s plain meaning must ordinarily be regarded as conclusive.” Id. (internal quotations and citations omitted). The relevant statute provides, in pertinent part: It shall be an act of infringement to submit — (A) an application under [21 U.S.C. § 355(j)] or described in [21 U.S.C. § 355(b)(2) ] for a drug claimed in a patent or the use of which is claimed in a patent ... if the purpose of such a submission is to obtain approval under [Title 21 of the United States Code] to engage in the commercial manufacture, use, or sale of a drug ... claimed in a patent or the use of which is claimed in a patent before the expiration of such patent. 35 U.S.C. § 271(e)(2) (1994). The plain language of the statute does not alter a paten-tee’s burden of proving infringement, nor does it mandate an infringement analysis limited to the scope of the approval sought. Likewise, Glaxo has not cited any legislative history to indicate that Congress intended to limit the infringement analysis to any particular aspect of the ANDA or to alter a paten-tee’s burden of proving infringement. Glaxo argues that because an action brought under § 271(e)(2)(A) is based solely upon the filing of an ANDA, such an action requires a unique type of infringement analysis. In a case such as this, Glaxo argues, the alleged infringer must disprove infringement if the ANDA permits sale of a composition that may include an infringing product. Glaxo seems to be arguing that the act of infringement consisting of the filing of the ANDA presumptively settles the issue of infringement of the patent when the product is marketed and that anything possibly within the scope of the ANDA must be shown by the applicant not to infringe. Therefore, the question before us is whether the act of infringement created by § 271(e)(2)(A) directs the ultimate infringement inquiry to the ANDA or whether the inquiry should be *1568 focused on the product that is ultimately to be sold. The statute refers to the question whether the purpose of the ANDA is to engage in the commercial manufacture, use, or sale of the patented drug. We conclude that, especially in a ease such as this, involving a compound capable of existing in various forms, the statute requires an infringement inquiry focused on what is likely to be sold following FDA approval. This inquiry must be based on all of the relevant evidence, including the ANDA. As is well-established for infringement actions brought under § 271, see, e.g., Uniroyal, Inc. v. Rudkin-Wiley Corp., 837 F.2d 1044, 1054, 5 USPQ2d, 1434, 1441 (Fed.Cir.1988), a patentee seeking relief under § 271(e)(2) must prove by a preponderance of the evidence that what is to be sold will infringe. That burden is not shifted under § 271(e)(2). Subsection 271(e)(2)(A) was part of the Drug Price Competition and Patent Term Restoration Act of 1984, Pub.L. 98-417, 98 Stat. 1585 (hereinafter “the Hateh-Waxman Act” or “the Act”), legislation designed to benefit makers of generic drugs, research-based pharmaceutical companies, and not incidentally the public. See H.R.Rep. No. 98-857(1), at 14-15 (1984), reprinted in 1984 U.S.C.C.A.N. 2647, 2647-48 (stating that the purposes of the legislation are “to make available more low cost generic drugs [and] to create a new incentive for increased expenditures for research and development of certain products which are subject to pre-market approval”); Statement on Signing S. 1538 into Law, 20 Weekly Comp. Pres. Doc. 1359, 1360 (Sept. 24, 1984) (President’s statement that the Hatch-Waxman Act “will provide regulatory relief, increase competition, economy in government, and best of all, the American people will save money, and yet receive the best medicine that pharmaceutical science can provide”). The Hatch-Wax-man Act, inter alia, allows makers of generic drugs to market generic versions of patented drugs as soon as possible after expiration of the relevant patents, while providing patent holders with limited extensions of patent term in order to recover a portion of the market exclusivity lost during the lengthy process of development and FDA review. See Eli Lilly & Co. v. Medtronic, Inc., 496 U.S. 661, 669-74,110 S.Ct. 2683, 2688-90,110 L.Ed.2d 605, 15 USPQ2d 1121, 1126-28 (1990); Telectronics Pacing Sys., Inc. v. Ventritex, Inc., 982 F.2d 1520, 1524-25, 25 USPQ2d 1196, 1199-1200 (Fed.Cir.1992). Under the portions of the Act relevant to this appeal, a generic drug manufacturer may seek expedited approval to market a generic version of an already-approved drug by submitting an ANDA. See 21 U.S.C. § 355® (1994); 21 C.F.R. § 314.94 (1996). While performing development work and seeking such approval, a generic drug manufacturer is free from liability for patent infringement based solely upon acts necessary to prepare the ANDA See 35 U.S.C. § 271(e)(1) (1994) (“It shall not be an act of infringement to make use, offer for sale, or sell ... a patented invention ... solely for uses reasonably related to the development and submission of information under the Federal law which regulates the manufacture, use, or sale of drugs____”) (overruling Roche Prod., Inc. v. Bolar Pharm. Co., 733 F.2d 858, 221 USPQ 937 (Fed.Cir.1984), in which this court held that use for commercial development purposes of a patented drug was not within any experimental use exception and was therefore an act of infringement even if such use was necessary to obtain FDA approval); see also H.R.Rep. No. 98-857(11), at 27 (1984), reprinted in 1984 U.S.C.C.A.N. 2686, 2711 (“The provisions of § 202 of the bill [amending 35 U.S.C. § 271 to add new subsection (e)] have the net effect of reversing the holding of the court in Roche Prod., Inc. v. Bolar Pharm. Co.”). The Act also benefits the patentee (aside from patent term extension). If a patent relevant to the ANDA has not expired, the generic drug manufacturer must certify either that the generic drug will not enter the market before the patent’s expiration date or that the patent is “invalid or will not be infringed by the manufacture, use, or sale of the drug for which the [ANDA] is submitted.” 21 U.S.C. § 355®(2)(A)(vii)(III)-(rV) (1994). If the ANDA contains the latter certification, the generic drug manufacturer must notify the patentee, who, if it disagrees with the certification, will then have forty-five days to sue the ANDA applicant for infringement under 35 U.S.C. § 271(e)(2). *1569 See 21 U.S.C. § 355(j)(4)(B)(iii) (1994). Such a suit suspends FDA approval until the earliest of the expiration of the patent, judicial resolution of the correctness of the ANDA applicant’s certification, or thirty months from the receipt of notice. Id. Thus, § 271(e)(2) provided patentees with a defined act of infringement sufficient to create case or controversy jurisdiction to enable a court to promptly resolve any dispute concerning infringement and validity. Pat-entees were given a jurisdictional basis for bringing suit in federal district court under 35 U.S.C. § 271(e)(2) when, in light of § 271(e)(1), the ANDA applicant was not making, using, or selling the patented product, the traditional statutorily-defined acts of infringement. By bringing such a suit, a patentee could also suspend FDA approval of the ANDA pending judicial determination of the correctness of the applicant’s certification filed under 21 U.S.C. § 355(j)(2)(A)(vii)(IV). See 21 U.S.C. § 355(j)(4)(B) (1994); Bristol-Myers Squibb Co. v. Royce Lab., Inc., 69 F.3d 1130, 1131-32, 36 USPQ2d 1641, 1642-43 (Fed.Cir.), cert. denied, — U.S.-, 116 S.Ct. 754, 133 L.Ed.2d 701 (1996). Notwithstanding this defined act of infringement, a district court’s inquiry in a suit brought under § 271(e)(2) is the same as it is in any other infringement suit, viz., whether the patent in question is “invalid or will not be infringed by the manufacture, use, or sale of the drug for which the [ANDA] is submitted.” 21 U.S.C. § 355(j)(2)(A)(vii)(IV) (emphasis added). The only difference in actions brought under § 271(e)(2) is that the allegedly infringing drug has not yet been marketed and therefore the question of infringement must focus on what the ANDA applicant will likely market if its application is approved, an act that has not yet occurred. The occurrence of the defined “act of infringement” does not determine the ultimate question whether what will be sold will infringe any relevant patent. Since the compound for which approval is sought is that which is expected to be marketed, the purpose of the submission of the ANDA is to sell that well-defined compound and the ultimate question of infringement is usually straightforward. However, in this case, where the subject matter is a compound capable of existing in multiple crystalline forms, or mixtures thereof, the ultimate question of infringement is not so simple. The FDA’s interest in fixing the exact nature of such a product to be sold, in discharging its own responsibility to ensure the purity, efficacy, and safety of the product, may cause the nature of the product originally applied for to differ somewhat from that ultimately approved. The act of infringement that gives rise to a case or controversy under section 271(e)(2) has been stated to be “artificial,” Eli Lilly & Co., 496 U.S. at 675, 677, 110 S.Ct. at 2691, 2692, 110 L.Ed.2d 605, 15 USPQ2d at 1129, 1130, in the sense that a specific infringing composition has not yet been made, used, or sold, and is thus not necessarily available for a court to compare to the claims. This future aspect of the potential infringement and, in this case, the fact that RHC1 may exist in multiple crystalline forms, are precisely what required the district court to engage in the hypothetical inquiry advocated by Novop-harm. Under § 271(e)(2)(A), a court must determine whether, if the drug were approved based upon the ANDA the manufacture, use, or sale of that drug would infringe the patent in the conventional sense. As we have stated: [S]ection 271(e)(2)(A) makes it possible for a patent owner to have the court determine whether, if a particular drug were put on the market, it would infringe the relevant patent. If the court determines that the patent is not invalid and that infringement would occur, and that therefore the ANDA applicant’s paragraph IV certification is incorrect, the patent owner is entitled to an order that FDA approval of the ANDA containing the paragraph IV certification not be effective until the patent expires. See 21 U.S.C. § 355(j)(4)(B)(iii)(II); 35 U.S.C. § 271(e)(4)(A). Bristol-Myers Squibb Co., 69 F.3d at 1135, 36 USPQ2d at 1646. Of course, this hypothetical inquiry is properly grounded in the ANDA application and the extensive materials typically submitted in its support. 2 *1570 Thus, contrary to Glaxo’s arguments, the patentee’s burden of proving ultimate infringement is not met by the filing of the ANDA. The relevant inquiry is whether the patentee has proven by a preponderance of the evidence that the alleged infringer will likely market an infringing product. What is likely to be sold, or, preferably, what will be sold, will ultimately determine whether infringement exists. The district court correctly chose to determine whether Novop-harm would likely sell an infringing composition pursuant to an approved ANDA. In conducting this infringement analysis, the district court properly considered the ANDA itself, the materials submitted by Novop-harm to the FDA, and other pertinent evidence provided by the parties. Thus, the court did not clearly err in finding that Glaxo did not prove infringement of the claims to Form 2 RHC1 by a preponderance of the evidence and in dismissing Glaxo’s claim for infringement of the ’431 patent. B. Infringement Of The’133 Patent Glaxo also sought a declaration that No-vopharm, after receiving FDA approval of its ANDA, would violate 35 U.S.C. § 271(g) (1994) by importing RHC1 after producing it via the process claimed in the T33 patent. Glaxo alleged that Novopharm, by communicating to Glaxo that it intended to market RHC1 before the expiration of the ’133 patent pursuant to FDA approval, had created a reasonable apprehension of imminent and irreparable harm. The district court dismissed Glaxo’s declaratory judgment action on the merits, based upon the same analysis it applied to the § 271(e)(2) infringement claim. While the § 271(e)(2) provides the federal courts with jurisdiction to hear infringement eases regarding claims directed to drugs or to methods of using drugs, it does not provide jurisdiction to hear infringement cases regarding claims directed to methods for making drugs. 35 U.S.C. § 271(e)(2) (1994) (pertaining only to “a drug claimed in a patent or the use of which is claimed in a patent”). Therefore, Glaxo’s claim that No-vopharm would infringe the ’133 patent, which claims a method of making Form 2 RHC1, was necessarily based upon the Declaratory Judgment Act, 28 U.S.C. § 2201 (1994). A patentee may seek a declaration that a person will infringe a patent in the future. See Lang v. Pacific Marine & Supply Co., 895 F.2d 761, 763, 13 USPQ2d 1820, 1821 (Fed.Cir.1990). As is the case when an accused infringer brings a declaratory judgment action, the exercise of jurisdiction over such an action is within the discretion of the district court. See Telectronics Pacing Sys., Inc. v. Ventritex, Inc., 982 F.2d 1520, 1526-27, 25 USPQ2d 1196, 1201 (Fed.Cir.1992); Spectronics Corp. v. H.B. Fuller Co., 940 F.2d 631, 634, 19 USPQ2d 1545, 1547 (Fed.Cir.1991). Accordingly, we will reverse a decision to grant or to deny declaratory relief if “(1) the court’s decision was clearly erroneous, arbitrary, or fanciful; (2) the decision was based on an erroneous conclusion of law; (3) the court’s findings were clearly erroneous; or (4) the record contains no evidence upon which the court rationally could have based its decision.” Minnesota Mining & Mfg. Co. v. Norton Co., 929 F.2d 670, 673, 18 USPQ2d 1302, 1305 (Fed.Cir.1991). Notwithstanding this deferential standard of review, if there is no actual controversy the district court is without jurisdiction to hear a claim for declaratory relief. See Skelly Oil Co. v. Phillips Petroleum Co., 339 U.S. 667, 671-72, 70 S.Ct. 876, 879, 94 L.Ed. 1194 (1950) (stating that the Declaratory Judgment Act “enlarge[s] the range of remedies available in the federal courts but does not extend their jurisdiction”); Aetna Life Ins. Co. v. Haworth, 300 U.S. 227, 240, 57 S.Ct. 461, 463, 81 L.Ed. 617 (1937) (stating that the Declaratory Judgment Act “is opera *1571 tive only in respect to controversies which are such in the constitutional sense”). Accordingly, we must first determine if an actual controversy exists. See Loveladies Harbor, Inc. v. United States, 27 F.3d 1545, 1555 n. 26 (Fed.Cir.1994) (“[CJourts must attend to their jurisdiction even if the parties do not.”) (citing Louisville & Nat’l R.R. Co. v. Mottley, 211 U.S. 149, 152, 29 S.Ct. 42, 43, 53 L.Ed. 126 (1908)). Only if such a controversy exists, will we consider whether the court abused its discretion in dismissing the declaratory judgment action on the merits. The requirement that there be an actual controversy “is met by a sufficient allegation of immediacy and reality.” Lang, 895 F.2d at 764, 13 USPQ2d at 1822. Accordingly, when a patentee seeks a declaratory judgment against an alleged future in-fringer, the patentee must demonstrate that two elements are present: (1) the defendant must be engaged in an activity directed toward ... an infringement charge ... or be making meaningful preparation for such activity; and (2) acts of the defendant must indicate a refusal to change the course of its actions in the face of acts by the patentee sufficient to create a reasonable apprehension that a suit will be forthcoming. Id. Under this standard, the district court properly exercised its jurisdiction to consider Glaxo’s declaratory judgment claim. Glaxo’s complaint was based in part on a letter dated June 8, 1994, in which Novop-harm asserted that it intended to market its Form 1 RHC1 product after December 5, 1995 (the then expiration date of the ’658 patent) but before the expiration of the ’431 patent. Novopharm also indicated that it had submitted an ANDA accompanied by data sufficient to make FDA approval imminent. Thus, unlike T'electronics, in which we affirmed the dismissal of a declaratory judgment for lack of jurisdiction, the threat of Novopharm entering the U.S. market was not “years away” nor was there doubt that Novopharm wished to sell some form of RHC1. See 982 F.2d at 1527, 25 USPQ2d at 1201. Rather, Novopharm was systematically attempting to meet the applicable regulatory requirements while preparing to import its product. Some of Novopharm’s acts that form the basis of the declaratory judgment action are of course protected from liability for infringement under § 271(e)(1) (“It shall not be an act of infringement to ... import into the United States a patented invention ... solely for uses reasonably related to the development and submission of information under a Federal law which regulates the manufacture, use, or sale of drugs or veterinary biological products.”). Nevertheless, the protected status of Novopharm’s activities leading to its submissions to the FDA does not by itself prevent the district court from considering Glaxo’s request for declaratory relief because such relief is directed to the time after the ANDA is approved, when § 271(e)(1) no longer provides a shelter against infringement liability. Accordingly, declaratory relief is available to the patentee asserting a “method of making” claim if, as here, sufficient facts are alleged to create an actual case or controversy. Such allegations may include, as here, imminent FDA approval and actual threats of future infringement. Therefore, the district court properly exercised its discretion to hear Glaxo’s declaratory judgment action, even though that action was premised in part on actions protected under § 271(e)(1). On the merits, as discussed earlier, the court did not clearly err in finding that Glaxo failed to prove that Novopharm’s product would contain Form 2 RHC1. If its product does not contain Form 2 RHC1, then Novop-harm cannot be using a process that provides Form 2 RHC1. Thus, Glaxo also failed to prove that Novopharm was using a process claimed in the ’133 patent. Accordingly, the district court did not abuse its discretion in dismissing Glaxo’s declaratory judgment action. C. Unfair Competition Claim For Violation Of The Protective Order Finally, Glaxo argues that the district court erred by finding that the Glaxo 1 court’s Protective Order was not violated. Specifically, Glaxo argues that Dr. Lazarow-ych, after gaining access to relevant, sensitive information as an expert witness in Glaxo 1, violated the Protective Order and *1572 its related affidavit by working with Novop-harm and by participating in the development of its process for making Form 1 RHC1. Glaxo argues that both Novopharm and Lazarowych should be found hable even if the information they used was publicly available because Lazarowych swore, without qualification, that she would “never use the information [in any document marked CONFIDENTIAL UNDER PROTECTIVE ORDER], directly or indirectly, in competition with [Glaxo].” Novopharm responds that neither it nor Lazarowych acted in contempt of the Protective Order or the related affidavit. Novopharm argues that the fact that the information was publicly available means that the use of that information is not an act of unfair competition. We agree with No-vopharm that neither it nor Lazarowych was in violation of the Protective Order or the related affidavit. In reviewing the court’s dismissal of Glaxo’s unfair competition claim, which raises issues not within our exclusive jurisdiction, we are guided by the law of the regional circuit. See Baldwin Hardware Corp. v. Franksu Enter. Corp., 78 F.3d 550, 556, 39 USPQ2d 1090, 1093 (Fed.Cir.1996); Molins PLC v. Quigg, 837 F.2d 1064, 1066, 5 USPQ2d 1526, 1527 (Fed.Cir.1988). Under Fourth Circuit law, the party asserting the violation of a protective order, which is a form of civil contempt, must prove that violation by clear and convincing evidence. See In re General Motors Corp., 61 F.3d 256, 258 (4th Cir.1995) (“The burden is on the complainant to prove civil contempt by clear and convincing evidence.”). A district court’s finding that a protective order has not been violated will be reversed only if clearly erroneous. See Glosser v. A.H. Robins Co., 950 F.2d 147, 149 (4th Cir.1991). The court did not clearly err in determining that neither Novopharm nor Laza-rowych violated the Protective Order. At paragraph 14(a), the Protective Order clearly states that it does not restrict the dissemination of information in “documents that are used as exhibits in Court (unless such exhibits were filed under seal).” Glaxo does not contest the district court’s findings that “all of the purported ‘secrets’ are to be found in documents submitted as exhibits without seal,” 931 F.Supp. at 1303, and that Glaxo therefore “openly exposed the bulk of its confidential documents before the entire industry at the previous trial.” Id. at 1306. Therefore, dissemination or use of the allegedly sensitive information cannot violate the Protective Order. The Glasser case, 950 F.2d at 149, on which Glaxo relies, does not support interpreting the Protective Order to cover publicly-available information. ' Likewise, dissemination of the information did not contradict Lazarowych’s obligation under the affidavit. As Novopharm accurately points out, the affidavit merely summarized the obligations under the Protective Order without adding additional obligations. It served to notify each affiant of his or her obligations under the Protective Order and to memorialize those obligations. Thus, La-zarowych’s affidavit must be interpreted consistently with the Protective Order. Accordingly, neither the Protective Order nor the related affidavit encompasses the information allegedly used by Lazarowych; her alleged use of this information cannot form a valid basis for Glaxo’s unfair competition claim. 'The district court did not clearly err in finding that Novopharm did not violate the Protective Order or that Lazarowych did not violate any specific obligation under the Protective Order or the related affidavit. Because the unfair competition claim was based solely on Novopharm’s alleged violation of the Protective Order, the district court did not err in dismissing Glaxo’s unfair competition claim. CONCLUSION The district court correctly required Glaxo to prove by a preponderance of the evidence the likelihood that Novopharm would infringe the ’431 patent, i.e., that it would manufacture, use, or sell an infringing composition when the ANDA is approved. The district court did not clearly err in finding that Glaxo failed to prove that Novopharm would infringe the ’431 patent by selling RHC1 pursuant to its ANDA. Based upon the same factual finding, the court did not abuse its discretion in dismissing Glaxo’s declaratory judgment action asserting infringement of the ’133 patent under § 271(g). Finally, the district court did not clearly err in finding *1573 that neither Novopharm nor its employees violated the Protective Order. Accordingly, the judgment of the district court is AFFIRMED. Footnotes [1] The district court did not decide whether small amounts of Form 2 RHC1 in a mixture with Form 1 RHC1 would infringe the '431 patent. For that reason and also in view of our disposition of the infringement issue infra, whether small amounts of Form 2 RHC1 would infringe and, if so, what amount of Form 2 RHC1 would be required are not before us on this appeal. [2] As Novopharm points out, access to actual samples and the extensive technical data required by the FDA generally removes much of the uncertainty from a court’s otherwise hypothetical inquiry. For example, pursuant to 21 *1570 U.S.C. § 355(j)(2)(vi) (1994) (incorporating the requirements of § 355(b)(l)(B)-(F)), the FDA may require an applicant for an ANDA to submit actual samples of the drug. Also, pursuant to 21 C.F.R. § 314.94 (1996), the ANDA must include information that shows the drug product is bioe-quivalent to an already-approved drug, see § 314.94(a)(7), and must describe in detail the composition, manufacture, and specification of the drug substance and drug product. See § 314.94(9)(i). In this case, Novopharm provided Glaxo with actual samples of its Form 1 RHC1 composition. Glaxo could have further informed the court’s infringement analysis by offering analyses of these samples at trial, but it chose not to do so.