Paul H. Berghoff, Banner & Allegretti, Ltd., Chicago, Illinois, argued for plaintiff-appellant. With him on the brief were Edward W. Remus and James C. Gumina. Of counsel were Edward P. Gray and Richard B. Murphy, Eli Lilly and Company, Indianapolis, Indiana.

Robert L. Baechtold, Fitzpatrick, Cella, Harper & Scinto, New York City, argued for defendant-appellee American Cyanamid Company. With him on the brief were Thomas H. Beck and David F. Ryan. Of counsel was Lawrence Alaburda, American Home Products Corporation, Madison, New Jersey.

William L. Mentlik, Lerner, David, Littenberg, Krumholz & Mentlik, Westfield, New Jersey, argued for defendant-appellee Zenith Laboratories, Inc. With him on the brief were Arnold H. Krumholz and Roy H. Wepner. Of counsel was Jay B. Shapiro, Ivax Corporation, Miami, Florida.

James Galbraith and Maria Luisa Palmese, Kenyon & Kenyon, New York City, were on the brief, for defendant-appellee Biochimica Opos, S.p.A.

Before CLEVENGER, RADER, and BRYSON, Circuit Judges.

Opinion for the court filed by Circuit Judge BRYSON. Concurring opinion filed by Circuit Judge RADER.

BRYSON, Circuit Judge.

The ongoing struggle between "pioneer" drug manufacturers and generic drug distributors has once more come before our court. Eli Lilly and Company (Lilly), the "pioneer" drug manufacturer in this case, has filed suit for patent infringement against the appellees, who are involved in various ways in the distribution of a particular generic drug. Lilly sought a preliminary injunction, arguing that the importation and sale of the generic drug in this country infringed Lilly's patent on a process for making a related compound. After a hearing, the United States District Court for the Southern District of Indiana denied Lilly's request for a preliminary injunction. The court found that Lilly had failed to show that it was likely to prevail on the merits of its infringement claim and had failed to show that it would suffer irreparable harm in the absence of preliminary injunctive relief. Eli Lilly & Co. v. American Cyanamid Co., 896 F. Supp. 851, 36 USPQ2d 1011 (S.D.Ind.1995). Because Lilly has failed to overcome the substantial hurdle faced by a party seeking to overturn the denial of a preliminary injunction, we affirm.

* The pharmaceutical product at issue in this case is a broad-spectrum antibiotic known as "cefaclor." Cefaclor is a member of the class of cephalosporin antibiotics, all of which are based on the cephem nucleus. Although there are many different cephem compounds, only a few have utility as antibiotic drugs. Each of the known commercial methods for producing cefaclor requires the production of an intermediate cephem compound known as an enol. Once the desired enol cephem intermediate is obtained, it is then subjected to several processing steps in order to produce cefaclor.

* Lilly developed cefaclor and patented it in 1975. Until recently, Lilly has been the exclusive manufacturer and distributor of cefaclor in this country. In addition to its product patent on cefaclor, Lilly obtained several patents covering different aspects of the manufacture of cefaclor, including processes for producing enol cephem intermediates. Many of those patents have now expired.

In 1995, Lilly purchased the patent at issue in this case, U.S. Patent No. 4,160,085 (the '085 patent). Claim 5 of that patent defines a method of producing enol cephem compounds, including what is called "compound 6," an enol cephem similar to the one Lilly uses in its process for manufacturing cefaclor. The '085 patent will expire on July 3, 1996.

Compound 6 differs from cefaclor in three respects. Although both compound 6 and cefaclor are based on the cephem nucleus, compound 6 has a hydroxy group at the 3-position on the cephem nucleus, a para-nitrobenzyl carboxylate ester at the 4-position, and a phenylacetyl group at the 7-position. Cefaclor has different groups at each of those positions: it has a chlorine atom at the 3-position, a free carboxyl group at the 4-position, and a phenylglycyl group at the 7-position. Each of those differences between compound 6 and cefaclor contributes to the effectiveness of cefaclor as an orally administered antibiotic drug. The free carboxyl group at the 4-position is believed important for antibacterial activity; the chlorine increases cefaclor's antibiotic potency; and the phenylglycyl group enables cefaclor to be effective when taken orally.

To produce cefaclor from compound 6 requires four distinct steps. First, the hydroxy group is removed from the 3-position and is replaced by a chlorine atom, which results in the creation of "compound 7." Second, compound 7 is subjected to a reaction that removes the phenylacetyl group at the 7-position, which results in the creation of "compound 8." Third, a phenylglycyl group is added at the 7-position, which results in the creation of "compound 9." Fourth, the para-nitrobenzyl carboxylate ester is removed from the 4-position, which results in the creation of cefaclor.

On April 27, 1995, defendants Zenith Laboratories, Inc., (Zenith) and American Cyanamid Company (Cyanamid) obtained permission from the Food and Drug Administration to distribute cefaclor in this country. Defendant Biocraft Laboratories, Inc., (Biocraft) had applied for FDA approval to manufacture and sell cefaclor in the United States but had not yet obtained that approval. All three have obtained large quantities of cefaclor that were manufactured in Italy by defendant Biochimica Opos, S.p.A. (Opos).

On the same day that Zenith and Cyanamid obtained FDA approval to sell cefaclor in this country, Lilly obtained the rights to the '085 patent and filed suit against Zenith, Cyanamid, Biocraft, and Opos. In its complaint, Lilly sought a declaration that the domestic defendants' importation of cefaclor manufactured by Opos infringed Lilly's rights under several patents, including the '085 patent. Lilly also requested a preliminary injunction, based on the alleged infringement of claim 5 of the '085 patent, to bar the defendants from importing or inducing the importation of cefaclor manufactured by Opos.

The district court held a three-day hearing on the motion for a preliminary injunction. Following the hearing, the court denied the motion in a comprehensive opinion. The court devoted most of its attention to the question whether Lilly had met its burden of showing that it was likely to prevail on the merits of its claim that the defendants were liable for infringing claim 5 of the '085 patent.

Based on the evidence presented at the hearing, the district court concluded that Lilly had shown that it was likely to prevail on the issue of the validity of the '085 patent. With respect to the infringement issue, however, the court held that Lilly had not met its burden of showing that it was likely to prevail.

The district court correctly framed the issue as whether, under the Process Patent Amendments Act of 1988, Pub.L. No. 100-418, §§ 9001-07, the importers of cefaclor infringed claim 5 of the '085 patent, which granted U.S. patent protection to the process that Opos used to make compound 6. The Process Patent Amendments Act makes it an act of infringement to import, sell, offer to sell, or use in this country a product that was made abroad by a process protected by a U.S. patent. 35 U.S.C. § 271(g). The Act, however, does not apply if the product made by the patented process is "materially changed by subsequent processes" before it is imported. 35 U.S.C. § 271(g)(1).

The district court found that compound 6 and cefaclor differ significantly in their structure and properties, including their biological activity. Citing the Senate Report on the Process Patent Amendments Act, the district court found that, because the processing steps necessary to convert compound 6 to cefaclor " 'change the physical or chemical properties of the product in a manner which changes the basic utility of the product,' " 896 F. Supp. at 857, 36 USPQ2d at 1016 (citing S.Rep. No. 83, 100th Cong., 1st Sess. 50 (1987)), Lilly was not likely to succeed on its claim that the defendants infringed Lilly's rights under claim 5 of the '085 patent by importing and selling cefaclor.

The district court also found that Lilly had failed to prove that it would suffer irreparable harm in the absence of a preliminary injunction. The presumption of irreparable harm that is available when a patentee makes a strong showing of likelihood of success on the merits was not available here, the court held, because of Lilly's failure to make such a showing on the issue of infringement. In addition, the court was not persuaded by Lilly's arguments that it faced irreparable economic injury if it were not granted immediate equitable relief. Under the circumstances of this case, the district court found that an award of money damages would be an adequate remedy in the event that Lilly ultimately proves that the importation of cefaclor made by the Opos process infringes the '085 patent. In light of Lilly's failure to establish either a likelihood of success on the merits or irreparable harm, the court found it unnecessary to articulate findings regarding the other factors bearing on the propriety of preliminary injunctive relief--the balance of the hardships and the effect of the court's action on the public interest.

The Process Patent Amendments Act of 1988 was enacted to close a perceived loophole in the statutory scheme for protecting owners of United States patents. Prior to the enactment of the 1988 statute, a patentee holding a process patent could sue for infringement if others used the process in this country, but had no cause of action if such persons used the patented process abroad to manufacture products, and then imported, used, or sold the products in this country. In that setting, the process patent owner's only legal recourse was to seek an exclusion order for such products from the International Trade Commission under section 337a of the Tariff Act of 1930, 19 U.S.C. § 1337a (1982). By enacting the Process Patent Amendments Act, the principal portion of which is codified as 35 U.S.C. § 271(g), Congress changed the law by making it an act of infringement to import into the United States, or to sell or use within the United States "a product which is made by a process patented in the United States ... if the importation, sale, or use of the product occurs during the term of such process patent."

A concern raised during Congress's consideration of the process patent legislation was whether and to what extent the new legislation would affect products other than the direct and unaltered products of patented processes--that is, whether the new statute would apply when a product was produced abroad by a patented process but then modified or incorporated into other products before being imported into this country. Congress addressed that issue by providing that a product that is "made by" a patented process within the meaning of the statute "will ... not be considered to be so made after--(1) it is materially changed by subsequent processes; or (2) it becomes a trivial and nonessential component of another product." 35 U.S.C. § 271(g).

That language, unfortunately, is not very precise. Whether the product of a patented process is a "trivial and nonessential component" of another product is necessarily a question of degree. Even less well defined is the question whether the product of a patented process has been "materially changed" before its importation into this country. While applying that statutory language may be relatively easy in extreme cases, it is not at all easy in a closer case such as this one.

* Lilly argues that the "materially changed" clause of section 271(g) must be construed in light of its underlying purpose, which is to protect the economic value of U.S. process patents to their owners. Prior to the enactment of the Process Patent Amendments Act, the value of a U.S. process patent could be undermined by a manufacturer who used the process abroad and then imported the product into this country. Because the purpose of the process patent legislation was to protect against such subversion of protected economic rights, Lilly argues that the statute should be read to apply to any such scheme that undercuts the commercial value of a U.S. process patent. In Lilly's view, the product of a patented process therefore should not be considered "materially changed" if the principal commercial use of that product lies in its conversion into the product that is the subject of the infringement charge. Because cefaclor is the only product of compound 6 that is sold in the United States market, Lilly argues, the change in compound 6 that results in cefaclor--no matter how significant as a matter of chemical properties or molecular structure--is not a "material change" for purposes of section 271(g).

Although we are not prepared to embrace Lilly's argument, we acknowledge that it has considerable appeal. Congress was concerned with the problem of the overseas use of patented processes followed by the importation of the products of those processes, and a grudging construction of the statute could significantly limit the statute's effectiveness in addressing the problem Congress targeted. That is especially true with respect to chemical products, as to which simple, routine reactions can often produce dramatic changes in the products' structure and properties.

Nonetheless, while the general purpose of the statute informs the construction of the language Congress chose, purpose cannot displace language, and we cannot stretch the term "materially changed" as far as Lilly's argument would require. The problem is that the language of the statute refers to changes in the product; the statute permits the importation of an item that is derived from a product made by a patented process as long as that product is "materially changed" in the course of its conversion into the imported item. The reference to a "changed" product is very hard to square with Lilly's proposed test, which turns on the quite different question of whether the use or sale of the imported item impairs the economic value of the process patent.

The facts of this case demonstrate how far Lilly's test strays from the statutory text. While Lilly notes that there are only four steps between compound 6 and cefaclor, and that all four steps involve relatively routine chemical reactions, Lilly does not suggest any limiting principle based on the structure of the intermediate product or the nature of the steps necessary to produce the imported product. Thus, even if there were ten complex chemical reactions that separated compound 6 from cefaclor, Lilly's test would characterize the two compounds as not "materially" different as long as the primary commercial use of compound 6 in this country was to produce cefaclor.

Besides not responding to the natural meaning of the term "changed," Lilly's construction of the "materially changed" clause would create a curious anomaly. Lilly's value-based construction of the clause turns in large measure on Lilly's contention that the only commercial use for compound 6 in this country is to produce cefaclor; that is, Lilly views compound 6 and cefaclor as essentially the same product because compound 6 has no commercial use in the U.S. market except to produce cefaclor. Under that approach, however, the question whether compound 6 was "materially changed" in the course of its conversion to cefaclor would depend on whether and to what extent other derivative products of compound 6 are marketed in this country. Thus, under Lilly's theory compound 6 would become materially different from cefaclor if and when compound 6 came to have other commercial uses in the United States, even though the respective structures and properties of the two compounds remained unchanged.

That is asking the statutory language to do too much work. We cannot accept the argument that the question whether one compound is "materially changed" in the course of its conversion into another depends on whether there are other products of the first compound that have economic value. We therefore do not adopt Lilly's proposed construction of section 271(g). We look instead to the substantiality of the change between the product of the patented process and the product that is being imported.

In the chemical context, a "material" change in a compound is most naturally viewed as a significant change in the compound's structure and properties. Without attempting to define with precision what classes of changes would be material and what would not, we share the district court's view that a change in chemical structure and properties as significant as the change between compound 6 and cefaclor cannot lightly be dismissed as immaterial. Although compound 6 and cefaclor share the basic cephem nucleus, which is the ultimate source of the antibiotic potential of all cephalosporins, the cephem nucleus is common to thousands of compounds, many of which have antibiotic activity, and many of which are dramatically different from others within the cephem family. Beyond the cephem nucleus that they have in common, compound 6 and cefaclor are different in four important structural respects, corresponding to the four discrete chemical steps between the two compounds. While the addition or removal of a protective group, standing alone, might not be sufficient to constitute a "material change" between two compounds (even though it could dramatically affect certain of their properties), the conversion process between compound 6 and cefaclor involves considerably more than the removal of a protective group. We therefore conclude that the statutory text of section 271(g) does not support Lilly's contention that it is likely to prevail on the merits of its infringement claim.

In aid of their differing approaches to the issue of statutory construction, both sides in this dispute seek support for their positions in the legislative history of the 1988 statute. As is often the case, there is something in the legislative history for each side. On Lilly's side, for example, are characterizations of the legislation as creating process patent protection that is "meaningful and not easily evaded," H.R.Rep. No. 60, 100th Cong., 1st Sess. 13 (1987), and as excluding products only if they "cease to have a reasonable nexus with the patented process," S.Rep. No. 83, 100th Cong., 1st Sess. 36 (1987). On the other side are directions for applying the statute to chemical intermediates--directions that suggest a narrower construction of the statute than Lilly proposes. On balance, while we do not find the legislative history dispositive, we conclude that it does not unequivocally favor Lilly's position and thus does not raise doubts about the district court's statutory analysis as applied to the facts of this case.

Congress considered proposals for process patent reform for many years before the Process Patent Amendments Act was finally enacted in 1988. The 1966 report of the President's Commission on the Patent System recommended extending patent protection to the imported products of patented processes, and bills proposing such a measure were introduced in several sessions of Congress thereafter. See "To Promote the Progress of ... Useful Arts", Report of the President's Commission on the Patent System 35-36 (1966); S.Rep. No. 642, 94th Cong., 1st Sess. 5-7 (1976) (surveying patent law reform proposals in the 90th through the 94th Congresses).

In 1983, efforts to enact process patent legislation resumed with the introduction of a number of bills in both Houses of Congress. The bills varied in many respects, but with respect to the issue presented in this case, they were largely alike. The language contained in one of the bills was typical: it gave the patentee an action for infringement against anyone who, without authority, used or sold in the United States "a product produced by [a patented] process," even if the product in question was manufactured abroad. S. 1841, 98th Cong. 1st Sess. (1983). None of those bills became law, in part because of disputes concerning procedural issues over which various advocacy groups differed sharply.

The question of how to define the required relationship between the product of the patented process and the imported product came to the fore during the 99th Congress's consideration of process patent legislation. In 1985, the Commissioner of Patents and Trademarks told a Senate subcommittee that the proposed legislation should specifically provide that the product must be the "direct" product of the patented process in order "to lay to rest arguments that, for example, a product resulting from a series of processes would be covered by a process patent for an intermediate step." Process Patents: Hearing on S. 1543 before the Subcomm. on Patents, Copyrights & Trademarks of the S. Comm. on the Judiciary, 99th Cong., 1st Sess. 9 (1985). The Commissioner pointed out that inserting the word "directly" in the proposed statute would make U.S. law consistent with the laws of many foreign countries, which protect only the "direct" products of patented processes. Id. at 15. The Pharmaceutical Manufacturers Association agreed that the proposed limiting language would be appropriate, as long as it was understood that "immaterial and minor later steps such as salt formation or removal of protective [groups] should not be construed in such a way that importation of the final product of that added step would not be an infringement." Id. at 71. Other interest groups likewise offered support for the proposed limiting language, subject to similar qualifications. See id. at 63 (statement of the Intellectual Property Owners, Inc.); id. at 186 (statement of the American Intellectual Property Law Association).

The following year, the Administration again suggested adding the term "directly" to the proposed statute, this time in a hearing before a House committee. See Intellectual Property and Trade: Hearings before the Subcomm. on Courts, Civil Liberties, and the Admin. of Justice of the H. Comm. on the Judiciary, 99th Cong., 2d Sess. 58-59 (1986). The National Association of Manufacturers (NAM) agreed that the scope of process patent protection should be limited, but regarded the word "directly" as too restrictive; instead, the NAM suggested that the statute "not apply to products materially changed chemically by subsequent steps or processes from the product resulting from the patented process." Id. at 275-76 (emphasis in original). That language, according to the NAM representative, would cover the case of "a chemical intermediate made abroad by a patented process," but then "subjected to a common chemical reaction" and converted into "a salt or amino-derivative." Id. at 275; see also id. at 280.

The drafters of subsequent process patent bills embraced the Administration's suggestion to restrict the scope of the statute, but they did so by using the language suggested by the NAM. Thus, the term "materially changed" was adopted to exclude from the reach of the proposed statute those products that were significantly altered before their importation.

The House report on the 1986 version of the process patent legislation was the first committee report to discuss the meaning of the "materially changed" clause. It explained that if the patented process is for chemical X, and "subsequent modifications change the basic structure of chemical X so that a clearly different chemical Y results, the connection between the patented process and the product, chemical Y, is broken." H.R.Rep. No. 807, 99th Cong., 2d Sess. 21 (1986). The report noted, however, that chemical X would not be "materially changed" within the meaning of the statute if the subsequent modifications of chemical X were only "trivial or conventional in nature" such as "modifications which result in the formation of simple derivatives, including salts or esters, or the removal of impurities," or if the subsequent processing steps "only change [the] shape, size or form" of the product, such as by being diluted or put in tablet form. Id. at 21-22.

Efforts to enact process patent legislation continued the following year; those efforts ultimately bore fruit in 1988 with the enactment of the Process Patent Amendments Act. Although several persons during the 1987 Senate hearings called attention to the need to clarify the "materially changed" clause in light of the difficulty of applying it to chemical intermediates, see Process Patent Legislation: Hearing on S. 568, S. 573, and S. 635 before the Subcomm. on Patents, Copyrights & Trademarks of the S. Comm. on the Judiciary, 100th Cong., 1st Sess. 114 (1987) (statement on behalf of the Intellectual Property Owners, Inc.); id. at 146, the bills considered during 1987 and 1988 continued to employ the prior language without modification.

The pertinent portion of the 1987 House report on the process patent bill was identical to the portion of the 1986 House report summarized above, except for some new material that was inserted into one paragraph of the 1987 report. The new material appears to have been intended to express the notion that, under certain circumstances, significant changes in the properties or structure of a chemical product do not render the product "materially changed" within the meaning of the statutory language. The principal portion of the added matter explained that a hypothetical chemical product, chemical X, is not "materially changed" if

chemical X is an important intermediate product, such as a polymer, which can materially be changed into an end product, albeit by trivial or conventional processes. In this respect, a product will be considered made by the patented process, regardless of any subsequent changes, if it would not be possible or commercially viable to make that product but for the use of the patented process. In judging the commercial viability, the courts shall use a flexible standard which is appropriate to the competitive circumstances.

H.R.Rep. No. 60, 100th Cong., 1st Sess. 13-14 (1987).

The inserted language is not easy to interpret, in part because it purports to identify some products that can "materially be changed" without being "materially changed." In any event, however, the inserted language appears not to apply to the present case, as it seems to contemplate that when an intermediate that is the product of a patented process undergoes significant changes in the course of conversion into an end product, the end product will be deemed to be made by the patented process if (and only if) it would not be commercially feasible to make the end product other than by using the patented process. In this case, Lilly concedes that it is both possible and commercially viable to make cefaclor by methods that do not include the process defined by claim 5 of the '085 patent. Therefore, even if the explanatory language from the 1987 House report were accorded equal status with the language of the statute itself, the explanatory language would not require that section 271(g) be read to reach the defendants' conduct in this case.

Lilly seizes on the statement in the House report that suggests that a change in "the basic structure" of the intermediate is necessary in order to break "the connection between the patented process and the [final, imported] product." H.R.Rep. No. 60, supra, at 13. Because both compound 6 and cefaclor share the core cephem nucleus, Lilly contends that the process of converting compound 6 to cefaclor does not alter the "basic structure" of compound 6, and that compound 6 is therefore not "materially changed" by the process of converting it into cefaclor.

While Lilly's argument on this point cannot be lightly dismissed, we do not think the use of the term "basic structure" in the House report limits the "materially changed" clause, as applied to a cephem compound, to a change that alters the core cephem nucleus. If adopted, Lilly's argument would mean that there would never be a "material change" resulting from the conversion of one cephem compound to another. Lilly's argument would also leave open the question whether even a change in the cephem nucleus would be sufficiently "basic" if, for example, the initial and end products both contained the beta lactam ring, which is one of the components of the cephalosporin nucleus, and thus were members of the beta lactam "superfamily" of compounds. The effect of Lilly's construction of Section 271(g)(1), both within the family of cephem compounds and within other families of compounds that are based on a common nucleus, would be sweeping. Absent clearer congressional direction, we decline to adopt so broad a principle. We therefore decline Lilly's invitation to find the answer to this case in the House report's reference to changes in the "basic structure" of chemical intermediates.

Like the House report, the 1987 Senate report contains a detailed elaboration on the statutory term "material change." In fact, the Senate report contains what may best be described as a detailed set of instructions to courts called on to construe that term as it applies to particular fields of technology. The report noted that many foreign patent statutes extending process patent protection to a product require that the product in question be made "directly" from the patented process and suggested that the term "materially changed" in section 271(g) was intended to embody a similar but somewhat broader scope of protection; as the Senate Committee explained, the term "directly" was not used, because it might have been read to "exempt too many products that have been altered in insignificant ways after manufacture by the patented process." S.Rep. No. 83, 100th Cong., 1st Sess. 49 (1987).

Acknowledging that the task of determining whether a product was "materially changed" prior to its importation would ultimately be left to the courts, the Committee then set out a "two-phased test" to "give the courts Congressional guidance in what may be a difficult determination." S.Rep. No. 83, supra, at 50. The first part of the test restated the test set forth in the House report, i.e., that a product "will be considered made by the patented process ... if it would not be possible or commercially viable to make that product but for the use of the patented process." Id.

The Senate report provided an analysis of how the first part of the test should be applied in the case of chemical intermediates. The report explained (S.Rep. No. 83, supra, at 51):

If the only way to have arrived at Y is to have used the patented process at some step, e.g., producing X as an intermediate, Y is infringing.

If there is more than one way to have arrived [at] Y, but the patented process is the only commercially viable way to have done so, Y is infringing.

If there are commercially viable non-infringing processes to have arrived at X, the connection between the patented process for producing chemical X and the ultimate product, chemical Y, is broken, and Y would be a non-infringing product having satisfied both phases of the test.

As we noted above, the record makes clear that there is at least one commercially viable process for making cefaclor that does not involve the patented method of synthesizing enol cephems (including compound 6). Opos does not use that non-infringing process, but under the test set forth in the Senate report, it is enough to defeat the claim of infringement that there is another way of producing the intermediate, even if the alleged infringer does not use that alternative process.

The Senate Committee described the second portion of the two-part test for identifying a "material change" as follows (S.Rep. No. 83, supra, at 50):

A product will be considered to have been made by a patented process if the additional processing steps which are not covered by the patent do not change the physical or chemical properties of the product in a manner which changes the basic utility of the product [produced] by the patented process. However, a change in the physical or chemical properties of a product, even though minor, may be "material" if the change relates to a physical or chemical property which is an important feature of the product produced by the patented process. Usually a change in the physical form of a product (e.g., the granules to powder, solid to liquid) or minor chemical conversion, (e.g., conversion to a salt, base, acid, hydrate, ester, or addition or removal of a protection group) would not be a "material" change.

It seems fairly clear that under this second part of the test, the change from compound 6 to cefaclor would be regarded as a material change. The chemical properties of the two compounds are completely different, the "basic utility" of the products is different, and the chemical structure of the two products is significantly different. The changes between compound 6 and cefaclor go far beyond the minor changes that the report described as not material, such as the conversion to a salt, base, acid, hydrate or ester, or the removal of a protective group.

Lilly interprets the references in the Senate report to the "basic utility" and "properties" of the product of the patented process in a quite different manner. The "basic utility" and principal "property" of compound 6 in the U.S. market, according to Lilly, is to serve as an intermediate for the production of cefaclor. Because the defendants have "exploited" that utility or property, Lilly argues, compound 6 cannot be regarded as undergoing a "material change" in the course of its conversion to cefaclor.

Lilly's argument distorts the terms "utility" and "properties" beyond recognition. The chemical and biological properties of compound 6 are plainly different from those of cefaclor, and the utility of the two compounds, as that term is conventionally used, is quite different. Cefaclor is a powerful oral antibiotic, with a set of chemical and biological properties that give it great utility in that regard; compound 6 has no such properties, and it has no significant utility as an antibiotic. Moreover, the premise of Lilly's argument--that compound 6 has "utility" only as an intermediate in the preparation of cefaclor--is flawed. As the district court noted, compound 6 can be used to produce a variety of cephalosporin antibiotics, of which cefaclor is only one. While Lilly claims that cefaclor was the only derivative of compound 6 that was on the commercial market in the United States at the time of the district court's decision in this case, other cephalosporin antibiotics that are producible from compound 6 were on sale in other countries, and proceedings were pending to obtain authorization to market at least one of those antibiotics in this country. Thus, despite Lilly's creative effort to draw support from the references to "utility" and "properties" in the Senate report, the two-part test in the Senate report appears to offer no aid to Lilly's statutory argument.

The conference report on the process patent bill contained only a short discussion of the "material change" issue, but that report reiterated the test found in both the House and Senate reports: that infringement can be found, even in the case of a significant change in the imported product, if "it would not have been possible or commercially viable to make the different [product] but for the patented process." H.R.Conf.Rep. No. 576, 100th Cong., 2d Sess. 1087 (1988). See also Bio-Technology Gen. Corp. v. Genentech, Inc., 80 F.3d 1553, 1562-63 (Fed.Cir. 1996) (citing the same test, in the biotechnology context, from the Senate report). Once again, because there is at least one known commercial method for making cefaclor that does not use the patented process, the language in the conference report on the Process Patent Amendments Act does not help Lilly.

At the end of this Long March through the legislative history, we cannot claim that the legislative background of the 1988 Act provides a conclusive answer to the question of how the "materially changed" clause should be construed in general, and how it should be applied to the facts of this case in particular. What we are able to say, however, is that the legislative history does not compel adoption of Lilly's proposed analysis of the statute. In fact, to the extent that Congress intended the courts to look to the committee reports for guidance in construing the "materially changed" clause, the reports support the conclusion that the district court reached and that we uphold: that compound 6 is likely to be found to have been "materially changed" in the process of its conversion into cefaclor, and that the importation and sale of cefaclor is therefore not likely to be held to infringe Lilly's rights under claim 5 of the '085 patent.

Lilly also challenges the district court's conclusion that it failed to show that it would suffer irreparable harm if the district court did not grant a preliminary injunction in this case. We conclude that the court did not commit clear error in finding that Lilly failed to prove irreparable harm; the court therefore acted within its discretion in denying Lilly's request for preliminary relief. See New England Braiding Co. v. A.W. Chesterton Co., 970 F.2d 878, 882, 23 USPQ2d 1622, 1625 (Fed.Cir.1992); H.H. Robertson Co. v. United Steel Deck, Inc., 820 F.2d 384, 391, 2 USPQ2d 1926, 1930 (Fed.Cir.1987).

Because Lilly has not made a strong showing on the issue of infringement, it is not entitled to a presumption of irreparable harm. See High Tech Medical Instrumentation v. New Image Indus., Inc., 49 F.3d 1551, 1556, 33 USPQ2d 2005, 2009 (Fed.Cir.1995). Nor did the district court commit clear error in rejecting Lilly's arguments that, apart from the presumption, it would suffer irreparable harm in the absence of a preliminary injunction. In particular, the district court found that, under the specific circumstances of this case, "an award of money damages would be an adequate remedy in the event that Lilly ultimately establishes" infringement. 896 F. Supp. at 860, 36 USPQ2d at 1019. In light of the structure of the cefaclor market, the court found that calculating lost profits would be a relatively simple task. The court also found that the two distributors who have been authorized by the FDA to sell cefaclor in this country have adequate assets to satisfy any judgment likely to be awarded if Lilly were to prevail on the merits of its infringement claim.

Lilly contends that the loss of profits on sales of cefaclor because of competition from the appellees will result in irreparable injury to Lilly's overall pharmaceutical research efforts. As the district court pointed out, however, that claim of injury is not materially different from any claim of injury by a business that is deprived of funds that it could usefully reinvest. If a claim of lost opportunity to conduct research were sufficient to compel a finding of irreparable harm, it is hard to imagine any manufacturer with a research and development program that could not make the same claim and thus be equally entitled to preliminary injunctive relief. Such a rule would convert the "extraordinary" relief of a preliminary injunction into a standard remedy, available whenever the plaintiff has shown a likelihood of success on the merits. For that reason, adopting the principle that Lilly proposes would "disserve the patent system." Illinois Tool Works, Inc. v. Grip-Pak, Inc., 906 F.2d 679, 683, 15 USPQ2d 1307, 1310 (Fed.Cir.1990).

We have examined Lilly's other claims of irreparable injury based on the effects of generic competition in the cefaclor market during the remaining life of the '085 patent, and we do not find them sufficiently compelling to justify overturning the district court's conclusion on the issue of irreparable harm. Nor are we persuaded by Lilly's argument that the district court erred in finding that both Zenith and Cyanamid would be able to satisfy a monetary judgment against them if Lilly were to prevail on the merits. While there is some ambiguity in the record about whether Zenith would have to call on its parent corporation in the event of a judgment against it, and whether the parent would be obligated to respond to such a judgment, the President of Zenith testified that the parent corporation would stand behind Zenith with respect to a judgment entered against it for patent infringement. The district court's finding on the "sufficient assets" issue therefore cannot be held to be clearly erroneous.

In sum, we do not agree with Lilly that the district court erred in finding an absence of irreparable harm in this case. For that reason, and because we agree with the district court that Lilly has failed to establish that it is likely to succeed on the merits of its infringement claim, we hold that the district court did not abuse its discretion in declining to enter a preliminary injunction pending the resolution of this case on the merits.

AFFIRMED.

RADER, Circuit Judge, concurring.

I concur in the result reached by the majority because Lilly did not show irreparable harm. On that basis, the district court did not abuse its discretion. I depart from the court's reasoning and conclusion about the "material change" standard under 35 U.S.C. § 271(g).

* The court's majority places great emphasis on the legislative history to resolve the meaning of "material change"--a curious approach given its recognition that the legislative history contains "something ... for each side." The enactment history is far from dispositive in this case. The record of the enactment of this provision evinces a bitter battle between the pharmaceutical industry and its generic industry competitors.

In the first place, neither combatant could convince either house of Congress to enact a statutory standard clearly favorable to their segment of the industry. The hearings before the various subcommittees considering intellectual property components of the bill highlight some of the positions taken by industry and trade groups. See A Bill to Protect Patent Owners from Importation into the United States of Goods Made Overseas by use of a United States Patented Process: Hearings on S. 1543 Before the Subcomm. on Patents, Copyrights and Trademarks of the Senate Comm. on the Judiciary, 99th Cong., 1st Sess. 9, 15, 71, 186-87 (1985) (advocating various degrees of limitation on initial broad statute:--PMA favoring "direct products" with legislative history allowing for trivial changes; AIPLA--opposing "direct" alone as creating loophole without detailed explanation); Intellectual Property Rights: Hearings on S. 1860 and S. 1869 Before the Subcomm. on International Trade of the Senate Comm. on Finance, 99th Cong., 2d Sess. 175, 182 (1986) (arguing on behalf of the International Anti-Counterfeiting Coalition against "direct" limitation as too narrow and for liability for "obvious use of the innovator's patent"); Intellectual Property and Trade: Hearings Before the Subcomm. on Courts, Civil Liberties, and the Admin. of Justice of the House Comm. on the Judiciary, 99th Cong., 2d Sess. 275-76 (1986) (arguing on behalf of NAM against "direct" limitation and for "materially changed chemically by subsequent steps").

Without a clear resolution in the statutory language, the battleground shifted to the committee reports. On this front, each combatant could find lobbyists to lace the reports with tutorials to the courts about applying the ambiguous provisions of section 271(g) in future litigation.

With a focus on future litigation, these committee reports became particularly unenlightening as an aid to interpret statutory language. These reports surrendered any pretext of informing members of Congress about the meaning of pending bills before a vote on the floor. Instead, these tutorials, by their own admission, addressed judicial officers, not legislative officers. And the legislative officers ignored them. In fact, the floor debates during passage of the Process Patents Act do not refer to any of the relevant portions of the committee reports.

In the House debates preceding passage of H.R. 4848, only two committee reports were even cited. Both were cited in a discussion unrelated to the process patent portion of the omnibus bill. See 134 Cong. Rec. 17880-18063 (1988) (citing H.R.Rep. No. 40, Pt. 1, 100th Cong., 1st Sess. 124 (1987) and S.R. Rep. No. 71, 100th Cong., 1st Sess. 123 (1987)). Moreover, in all of the debate on the floor of the House, no member discussed (or appears to have even mentioned) "material change." The debate in the Senate preceding passage of H.R. 4848 contains even less. No Senator cites to any committee report or mentions "material change." See 134 Cong. Rec. 20104 (1988).

Moreover the reports show all the signs of unresolved conflict evident in the statutory language itself. Early versions of the bill were silent on the scope of protection. Senate Bill 1535 refers only to the importation of "a product produced by such process." S.Rep. No. 663, 98th Cong., 2d Sess. 30 (1984). In the analysis, the report indicates that the goal of the bill is to prevent importation of "products produced by a process covered by the patent." Id. at 5. The Senate committee notes: "The principle effect of these changes is to prevent competitors of a patent owner from avoiding the patent by practicing the patented process outside the United States and marketing the resulting product in this country." Id.

By 1986, "material change" had crept into the statute. See H.R. 4899. The committee explains the addition: "Without such limitations, liability could attach, for example, to the seller of a car in which the steel was made from iron ore that was mined by a patented process, or in which the reflective surface used on the rear view mirror was made by a patented process." H.R.Rep. 807, 99th Cong., 2d Sess. 20 (1986). Clearly, the House was concerned with notions akin to proximate cause--that is, how far down the line should we cut off potential liability.

The 1987 report accompanying H.R.1931, sets forth significantly greater and more confusing detail concerning "material change." H.R.Rep. 60, 100th Cong., 1st Sess. 13-14 (1987). The Committee states that it "intends to provide protection to process patent owners which is meaningful and not easily evaded." Id. The report then suggests no liability would attach where the "basic structure" of the product is changed by a subsequent process. Id. at 13. However, the next paragraph notes: "the subsequent processing modifications of chemical X may only be trivial or of a conventional nature even though a material change occurred in chemical X." Id. Such a non-material change would occur, for example, where "chemical X is an important intermediate product ... which can materially be changed into an end product." Id. By this test, the House sets forth a commercial viability test that looks to the efficiency of making the end product absent use of the infringing process. Id. at 13-14. This same report goes on to identify at least two more tests for "material change," an active ingredient test and an integral or important component test. Id. at 14.

Likewise, the related Senate Report accompanying Senate Bill S. 1200, provides immense detail on the definition of "material change." S.Rep. No. 83, 100th Cong., 1st Sess. (1987). In this report, the committee explains that S. 1200 "introduces" a "new phrase, materially changed" to restrict the scope of the bill "to exclude ultimate products that, because of intervening manufacturing steps, cease to have a reasonable nexus with the patented process." Id. at 36. The committee also recognized that "the courts will have to assess the permutations of this issue of proximity to or distance from the process on a case-by-case basis." Id. at 46; see Bio-Technology General Corp. v. Genentech, Inc., 80 F.3d 1553, 1560 (Fed.Cir.1996) (recognizing Congress intended the courts to resolve the question of "proximity to the product of the patented process on a case-by-case basis"). Interestingly, this statement follows the debate over the scope of the Act, suggesting Congress chose an intentionally vague limitation such as "material change" as a compromise between the pharmaceuticals and generics.

The report then sets forth a "two-phased" test that looks to the commercial viability of making the end product absent use of the infringing process coupled with a basic utility test. Id. at 50. This test like others in the enactment history provides little practical utility in assessing material change. As with its House counterpart, the directly conflicting and confusing tests set forth in the section analysis leave little doubt that they were inserted by lobbyists for use in future litigation by their clients. In sum, the enactment history maps all the graves on this inconclusive legislative battleground, but shows no route away from the combat zone. If anything, the enactment history "is more conflicting than the text [of the statute] is ambiguous." Wong Yang Sung v. McGrath, 339 U.S. 33, 49, 70 S. Ct. 445, 453, 94 L. Ed. 616 (1950).

With no real enlightenment about the meaning of "material change" in the enactment history, this court should interpret this language in light of the overriding purpose of the statute. As recognized by this court's majority, the Process Patents Act sought to afford meaningful protection to owners of United States process patents. Before the Act, foreign entities could import goods made from patented processes used abroad without liability for patent infringement. The Act sought to plug this loophole.

Sadly this decision will create another massive loophole in the protection of patented processes. This decision will, in effect, deny protection to holders of process patents on intermediates as opposed to "final" products. This decision denies protection to a patented process anytime it is not the only way to make an intermediate, even if it is the most economically efficient way to produce the intermediate.

In view of the purpose of the statute, compound 6 and cefaclor are essentially the same product. Compound 6 has no commercial use in the U.S. market except to make cefaclor. The patented process is thus in use to make compound 6--a product only four simple, well-known steps from cefaclor. The record shows no other current commercial use of compound 6.

Rather than attempting to distill an elixir from this intoxicating witches brew of enactment history, this court should interpret "material change" consistent with the overriding purpose of the Act--to provide protection to process patent holders. With its eye firmly fixed on the purpose of the Act, this court would avoid eliminating processes for intermediates from the protections of the 1988 Act.