From Casetext: Smarter Legal Research

ELI LILLY COMPANY v. ZENITH GOLDLINE PHARMACEUTICALS

United States District Court, S.D. Indiana, Indianapolis Division
Oct 12, 2001
Cause No. IP 99-38-C H/K (S.D. Ind. Oct. 12, 2001)

Opinion

Cause No. IP 99-38-C H/K

October 12, 2001


FINDINGS OF FACT AND CONCLUSIONS OF LAW


In this action for patent infringement, plaintiffs Eli Lilly and Company and Reliant Pharmaceuticals, LLC have sued defendant Zenith Goldline Pharmaceuticals, Inc. for infringing U.S. Patent No. 4,375,547. The '547 patent claims a chemical compound called nizatidine, which is the active agent in plaintiffs' anti-secretory anti-ulcer drug, AXID. Infringement is conceded. The two contested issues are (a) whether the '547 patent is invalid for obviousness, and if not, (b) whether Zenith has willfully infringed the '547 patent so as to make this an "exceptional case" in which attorneys' fees should be awarded to plaintiffs under 35 U.S.C. § 285. On both issues, the case is strikingly similar to Yamanouchi Pharmaceutical Co. v. Danbury Pharmacal, Inc., 231 F.3d 1339 (Fed. Cir. 2000), in which the Federal Circuit affirmed findings of validity and willfulness regarding one of the three other patented drugs of the same class, called "H2 receptor antagonists."

The act of infringement was Zenith's filing of an amended Abbreviated New Drug Application (ANDA) with the Food and Drug Administration seeking permission to manufacture and sell nizatidine as a generic drug before the '547 patent expires. Zenith filed its amended ANDA under the Drug Price Competition and Patent Term Restoration Act of 1984, which is better known as the Hatch-Waxman Act. Relevant portions are codified in 21 U.S.C. § 355(j) and 35 U.S.C. § 271(d)-(e). In the amended ANDA, Zenith included what is called a "Paragraph IV" certification asserting that the '547 patent is invalid as obvious from the prior art. See 21 U.S.C. § 355(j)(2)(A)(vii)(IV). Pursuant to the Hatch-Waxman Act, Zenith also notified Lilly of its filing. Lilly promptly responded by filing a timely suit for infringement. See 21 U.S.C. § 355(j)(5)(B)(iii).

Federal question jurisdiction exists under 28 U.S.C. § 1331 and 1338(a). The case was tried to the court September 10-13, 2001. The court now states its findings of fact and conclusions of law. The substance of a statement shall govern whether it is treated as a finding of fact or conclusion of law.

As explained in detail below, the court finds that the '547 patent is valid. Zenith's obviousness argument is based on pure hindsight and ignores the critical properties of nizatidine, which Zenith's own expert witness conceded could not be predicted from the compound's structure. There is also considerable objective evidence that nizatidine is not obvious, including commercial success, unexpected results, and the failure of others in the field.

The court also finds that Zenith's infringement was willful, making this an exceptional case in which Lilly should recover its attorneys' fees. Before its act of infringement, Zenith obtained from outside patent counsel an oral opinion of invalidity, but Zenith could not reasonably rely upon it. A written opinion provided five weeks after the act of infringement contained key errors and omissions, and the circumstances in which both the oral and written opinions were provided show that Zenith did not reasonably rely on advice of counsel.

I. The '547 Patent and H2 Receptor Antagonists

The '547 patent for nizatidine was issued to Dr. Richard Pioch on March 1, 1983. Dr. Pioch worked for Lilly and had previously assigned his rights to Lilly. The date of invention for the subject matter of claim 1 of the '547 patent is November 27, 1979. The '547 patent will expire on April 12, 2002.

After this lawsuit was filed, Lilly assigned its rights in the '547 Patent to Reliant Pharmaceuticals, LLC, which has joined as a plaintiff. Under its agreement with Reliant, Lilly retains the right to maintain the present suit. This entry treats Lilly and Reliant jointly under the name Lilly.

The '547 patent has only one claim, for nizatidine. The chemical name for nizatidine is N-methyl-N'-2-([2-(dimethylaminomethyl)-4-thiazolyl] methylthio)ethyl 2-nitro-1, 1-ethenediamine. (The structure is shown below at page 19.)

Nizatidine is an H2 receptor antagonist useful for treating peptic ulcers. The enzyme histamine is known to cause parietal cells in the gastric glands in the lining of the stomach to secrete gastric acid. The biological effect of gastric acid secretion results from interaction between histamine and structures in the parietal cells known as "H2 receptors." Although the sequence of proteins in H2 receptors is now known, the shape of the physical structure of the H2 receptors (resulting from the folding of the chain of proteins) still is not known, nor do scientists understand just how histamine causes the effect of gastric acid secretion.

Excessive gastric acid secretion can cause illnesses ranging from mild discomfort to extremely serious conditions. In the late 1960s, scientists at several major pharmaceutical companies were trying to develop compounds to treat these conditions by suppressing the gastric acid secretion that would otherwise be caused by histamine. Such compounds are called "H2 receptor antagonists" because they suppress the biological action of the H2 receptors.

The first successful H2 receptor antagonist was cimetidine, developed by SmithKline French laboratories in Great Britain. In 1988, one recipient of the Nobel Prize for Physiology or Medicine was Sir James W. Black, who led the effort that resulted in cimetidine.

The Nobel Prize was awarded for Black's work in developing receptor antagonists and the broadly applicable principles useful in developing drugs for a wide variety of conditions. The official announcement of the Nobel Prize specifically described the development of cimetidine:

Starting from the structure of the histamine molecule Black developed a series of substances which effectively blocked the H2-receptor mediated effects, in particular the secretion of gastric acid. In 1972 Black and coworkers for the first time defined the H2-receptors by using agonists and antagonists. One of the first synthesized substances, metiamide, was found to heal peptic ulcer but it produced agranulocytosis on rare occasions. Subsequently (1975) Black succeeded in developing another substance, cimetidine, which was found to have a marked effect in the treatment of peptic ulcer but without this side-effect. Blocking of the H2-receptors introduced a new principle in the treatment of peptic ulcer, and a series of new drugs with the same mechanism or action has later been developed. As a consequence the need for surgical treatment of peptic ulcer has decreased significantly.

See www.nobel.se/medicine/laureates/1988/press.html.

Cimetidine was approved by the FDA in 1977. It remains a commercial product today with the brand name TAGAMET. Generic versions have been available since 1994, when the patent on cimetidine expired. Cimetidine turned out to cause some negative side effects, including metabolic blockage through the interference with particular liver enzymes that caused the buildup of certain drugs to toxic levels, and impotence and breast swelling in men when administered in high doses. Also, cimetidine had to be taken four times a day.

These disadvantages spurred a massive effort by virtually all major drug companies to find better second-generation H2 receptor antagonists. As of 1979, when Dr. Pioch invented nizatidine, cimetidine was still the only H2 receptor antagonist approved by the FDA for treating peptic ulcer disease in humans and was still the only H2 receptor antagonist commercially available in the United States.

Since 1979, the FDA has approved only three more H2 receptor antagonists for consumer use in the United States: ranitidine, famotidine, and nizatidine. In 1983, the FDA approved ranitidine, which is also known as ZANTAC. In 1986, the FDA approved famotidine, which is also known as PEPCID. The FDA approved Lilly's nizatidine in 1988. Cimetidine, ranitidine, famotidine, and nizatidine were originally available only by prescription, but all four have since been approved for over-the-counter sale, although at lower doses. Generic versions of cimetidine, ranitidine, and famotidine were approved by the FDA in 1994, 1997, and 2001 respectively. Regardless of the outcome of this lawsuit, generic versions of nizatidine should be available beginning April 12, 2002, when the '547 patent will expire.

II. Zenith's Act of Infringement

Defendant Zenith Goldline Pharmaceuticals, Inc., which is now known as IVAX Pharmaceuticals ("Zenith" in this entry), is a wholly-owned subsidiary of IVAX Corporation. Zenith produces and markets generic drugs.

On September 15, 1998, Zenith filed an Abbreviated New Drug Applications (ANDA) 75-461 with the Food and Drug Administration (FDA) seeking approval to market generic nizatidine. That original ANDA 75-461 contained what is called a "Paragraph III" certification. Zenith's Paragraph III certification stated that Zenith would not market nizatidine until the expiration of the '547 Patent. See 21 U.S.C. § 355(j)(2)(A)(vii)(III).

Three weeks later, on October 6, 1998, Zenith amended ANDA 75-461 to include a "Paragraph IV" certification, which asserted that the '547 patent is invalid. Zenith's Paragraph IV certification stated:

In accordance with section 505(j)(2)(A)(vii)(IV) of the Federal Food, Drug, and Cosmetic Act, and 21 C.F.R. § 314.94(a)(12)(k)(A)(4), Zenith Goldline certifies that the following patents are invalid, unenforceable, or will not be infringed by the manufacture, use or sale of Zenith Goldline's Nizatidine Capsules USP, 300 mg, for which this application is submitted.
• US Patent No. 4,375,547 — Expiry Date: April 12, 2002
We will comply with the notification requirements defined in Section 505(j)(2)(B)(i), Paragraphs I and II of the Federal Food, Drug and Cosmetic Act, (Codified at 21 C.F.R. § 314.95(a)), and the content of our notice will conform to the requirements defined in 505(j)(2)(B)(ii), (Codified at 21 C.F.R. § 314.95(c)).

On November 19, 1998, Zenith sent Lilly a Patent Certification Notice Letter, which is required when a Paragraph IV certification is filed. Ex. 1070. The notice letter informed Lilly of Zenith's Paragraph IV certification and asserted that the sole claim of the '547 Patent is obvious over certain prior art references:

With respect to the '547 patent, certification has been made to the FDA as called for by the Act, § 505(j)(2)(A)(vii)(IV), that claim 1 thereof is invalid as being obvious in view of, inter alia, Durant et al., U.S. Patent 4,046,907; Price et al., U.S. Patent 4,128,658; Crenshaw et al., U.S. Patent 4,375,248; Tobias O. Yellin et al. ICI 125,211: A New Gastric Antisecretory Agent Acting on Histamine H2-Receptor, 25 LIFE SCIENCES 2001 (1979); and J. Bradshaw et al., Ranitidine (AH-19065); a new potent, selected histamine H2-Receptor antagonist, PROCEEDINGS of B.P.S. 464P (1979). The composition of matter, nizatidine, is highly structurally similar to compounds disclosed in Durant et al., Crenshaw et al. and to ranitidine, and any differences there between are suggested by the prior art. In addition, the properties resulting from both ranitidine and nizatidine are highly analogous.

Within 45 days after receiving the notice letter, Lilly filed this suit against Zenith alleging infringement of the '547 patent under 35 U.S.C. § 271(e)(2)(A), and willful infringement under 35 U.S.C. § 285.

Zenith has admitted that its act of amending ANDA 75-461 for nizatidine to add the Paragraph IV certification was an act of infringement of the '547 Patent under 35 U.S.C. § 271(e). See Eli Lilly Co. v. Medtronic, Inc., 496 U.S. 661, 676 (1990) (§ 271(e) defined new and "somewhat artificial" act of infringement for ANDA process).

III. The Validity of the '547 Patent

A. The Law of Obviousness for Chemical Compounds

The only issue on the merits of this case is whether the '547 patent is invalid on the theory that nizatidine would have been obvious to a person of ordinary skill in the art when it was invented in November 1979. A claimed invention is not patentable if the differences between it and the prior art at the time of the invention "are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art." 35 U.S.C. § 103(a); Graham v. John Deere Co., 383 U.S. 1, 13-14 (1966).

The ultimate determination on the issue of obviousness is treated as a question of law, but it is based on factual inquiries that include: (1) the scope and content of the prior art; (2) the level of ordinary skill in the art; (3) the differences between the claimed invention and the prior art; and (4) objective evidence of non-obviousness. Graham, 383 U.S. at 17-18; Yamanouchi Pharmaceutical Co. v. Danbury Pharmacal, 231 F.3d 1339, 1343 (Fed. Cir. 2000); WMS Gaming Inc. v. International Game Technology, 184 F.3d 1339, 1355 (Fed. Cir. 1999).

Such objective evidence may include evidence that the patented invention has enjoyed commercial success, that it met a long-felt but unmet need, that others in the field tried and failed to solve the problem solved by the invention, that the invention produced unexpected and superior results, that others copied the invention, and that others in the industry acquired licenses showing their respect for the invention. In re Rouffet, 149 F.3d 1350, 1355 (Fed. Cir. 1998); accord, e.g., Ruiz v. A.B. Chance Co., 234 F.3d 654, 662 (Fed. Cir. 2000), citing Graham, supra, and Miles Labs., Inc. v. Shandon, Inc., 997 F.2d 870, 877 (Fed. Cir. 1993); W.L. Gore Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1555-56 (Fed. Cir. 1983).

A patent that has been issue is presumed to be valid. 35 U.S.C. § 282. The party asserting invalidity based on obviousness must prove invalidity by clear and convincing evidence. E.g., WMS Gaming, 184 F.3d at 1355; Lindemann Maschinenfabrik GmbH v. American Hoist and Derrick Co., 730 F.2d 1452, 1459 (Fed. Cir. 1984) (reversing finding of obviousness).

The Federal Circuit has often said that it may be easier to satisfy the burden if the party asserting invalidity can show that the relevant prior art was not presented to or considered by the patent examiner. E.g., WMS Gaming, 184 F.3d at 1355, citing Applied Materials, Inc. v. Advanced Semiconductor Materials Am., Inc., 98 F.3d 1563, 1569 (Fed. Cir. 1996); Uniroyal, Inc. v. Rudkin-Wiley Corp., 837 F.2d 1044, 1050 (Fed. Cir. 1988). Defendant Zenith has not made such a showing in this case, although it attempted to do so in summary judgment briefing. The prior art compounds upon which Zenith relies here were not only before the examiner, they were described in detail in the '547 patent itself.

Chemical compounds present special issues of obviousness because of the limited number of elements, recurring groups or substituents in complex molecules, the structural similarities within classes of related compounds, and the ability of chemists to undertake systematic experiments modifying known compounds.

In Yamanouchi, the Federal Circuit upheld the patent on another of the H2 receptor antagonists — famotidine — against a defense of obviousness. The court explained: "For a chemical compound, a prima facie case of obviousness requires `structural similarity between claimed and prior art subject matter . . . where the prior art gives reason or motivation to make the claimed compositions.'" 231 F.3d at 1343, quoting In re Dillon, 919 F.2d 688, 692 (Fed. Cir. 1990) ( en banc); accord, In re Papesch, 315 F.2d 381, 391 (C. C. P. A. 1963).

To prevent the distortions of hindsight from invalidating genuine inventions, close attention to the supposed reason or motivation for making the claimed compound is critical. As a leading treatise explains:

Because of the unpredictable nature of chemical reactions, a newly-synthesized compound may be very similar in structure to known and existing compounds and yet exhibit very different properties. Further, many such new compounds are obvious in the sense that any competent chemist could have synthesized them if requested or motivated to do so.

2 Donald S. Chisum, Chisum on Patents § 5.04[6] at 5-429 (2000) (emphasis added). In deciding obviousness under § 103(a), the focus is not on the ability of chemists to imagine a compound, nor on their ability to synthesize a molecule to order, but on whether the prior art provided apparent reason or motivation to take the steps that led to synthesis of the new compound. The unpredictable nature of chemical reactions is especially pronounced, of course, when dealing with medicinal chemistry, where the biological effects of chemical reactions may be exceedingly difficult to predict from the chemical structure of a compound.

To show obviousness, the reason or motivation offered by the prior art need not offer "absolute predictability" of the results, but it requires at least a "reasonable expectation of success." Yamanouchi, 231 F.3d at 1343, quoting In re Longi, 759 F.2d 887, 896 (Fed. Cir. 1985); accord, In re Vaeck, 947 F.2d 488, 495 (Fed. Cir. 1991) (reversing PTO rejection of claims as obvious where prior art offered no "reasonable expectation of success"), citing In re O'Farrell, 853 F.2d 894, 903-04 (Fed. Cir. 1988).

If the prior art makes a particular experiment or modification only "obvious to try," that does not support a finding of obviousness. See In re Eli Lilly and Co., 902 F.2d 943, 945 (Fed. Cir. 1990), citing In re O'Farrell, 853 F.2d at 903. The Federal Circuit explained in O'Farrell that the "obvious to try" admonition has been directed mainly at two kinds of error. As explained below, one applies here: "In some cases, what would have been `obvious to try' would have been to vary all parameters or try each of numerous possible choices until one possibly arrived at a successful result, where the prior art gave either no indication of which parameters were critical or no direction as to which of many possible choices is likely to be successful." Id.

Obviousness cannot be determined by chemical structure alone. As applied to chemical compounds, "a compound and all of its properties are inseparable" and must be considered in determining obviousness. In re Dillon, 919 F.2d 688, at 697, citing In re Papesch, 315 F.2d at 391; accord, Gillette Co. v. S.C. Johnson Son, Inc., 919 F.2d 720, 725 (Fed. Cir. 1990) (affirming rejection of obviousness defense: "An analysis of obviousness of a claimed combination must include consideration of the results achieved by that combination.").

B. Zenith's Argument from Structural Similarity

The first three Graham factors address the prior art, the differences between nizatidine and the prior art, and the level of skill in the relevant art.

The obviousness inquiry in this case must focus on the state of the art in late November 1979, when Dr. Pioch invented nizatidine. See 35 U.S.C. § 103(a). Defendant Zenith argues that nizatidine would have been obvious to one skilled in the art of synthetic organic chemistry at that time. Zenith bases this contention on structural similarities between nizatidine and three other H2 receptor antagonists known at that time. Zenith called on Dr. Steven W. Baldwin to explain this defense. Dr. Baldwin is an expert in synthetic organic chemistry and has served for many years as a professor of chemistry at Duke University.

1. Dr. Baldwin's Analysis

The analysis begins with the histamine molecule itself. Histamine has the following structure:

[diagram 1]

The diagrams in this entry use certain conventions from organic chemistry. The unidentified angles represent carbon atoms, for which any otherwise unspecified bonds are assumed to be taken up by hydrogen atoms. Thus, the histamine molecule is based on the imidazole ring made up of five atoms — three carbons and two nitrogens. Attached to a carbon atom in the ring is the "side chain."

Dr. Baldwin described the effort to develop an effective H2 receptor antagonist as beginning with the histamine molecule itself, making various modifications to it, and seeing whether those modifications had desirable biological effects. See Tr. 79. The SmithKline French laboratories tried "a huge number of compounds." Some actually promoted gastric acid secretion while others inhibited it. Two early candidates as effective H2 receptor antagonists — burimamide and metiamide — turned out to have problems that precluded their use as safe and effective drugs. After several years of effort, however, the SmithKline French team discovered cimetidine. Like histamine itself, each of those three compounds is based on an imidazole ring, three carbons and two nitrogens. Cimetidine has the following structure:

[diagram 2]

Cimetidine was a breakthrough invention. It launched a revolution in the treatment of peptic ulcers without surgery. It became the number one selling drug in the United States. Its invention was significant enough to figure prominently in the 1988 Nobel Prize for Physiology or Medicine. As noted above, however, cimetidine also produced some undesirable side effects. It also had to be taken four times a day. The SmithKline French team and virtually every major drug company began searching for second-generation H2 receptor antagonists that would be at least as effective as cimetidine but without the side effects.

When Dr. Pioch invented nizatidine in late November 1979, the literature in the field disclosed many thousands of candidate compounds that could be expected to have some level of H2 antagonist activity. According to Dr. Baldwin, the literature at that time had identified two other compounds that showed the most promise as second generation H2 receptor antagonists: ranitidine and tiotidine. For purposes of discussing Dr. Baldwin's opinion, it is useful to show the structures of cimetidine, ranitidine, and tiotidine together:

[diagram 3]

Ranitidine is built on a furan ring, which has five atoms — four carbons and one oxygen. (Another furan-ring compound, called SKF 93479, has also been tested in humans.) Tiotidine is built on a thiazole ring, which has five atoms — three carbons, one sulfur, and one nitrogen, with one carbon between the sulfur and nitrogen.

As of November 1979, cimetidine was on the market as TAGAMET. Ranitidine was not yet on the market, but it had been disclosed in a patent that had issued on December 5, 1978, for compounds using furan rings that showed H2 antagonist activity. That patent, U.S. Patent No. 4,128,658, disclosed thousands of different furan compounds. Ranitidine is example 15 in the patent, though the name is not used. See Ex. 1075. Tests had shown ranitidine is about four times as potent as cimetidine in terms of its H2 receptor antagonist activity.

Tiotidine was disclosed in U.S. Patent No. 4,165,378, which issued on August 21, 1979. Ex. 2054. The '378 patent disclosed guanidine derivatives of imidazoles and thiazoles. The '378 patent also disclosed many compounds with H2 antagonist activity. Tiotidine was example 21. By August 1979, it was public knowledge that tiotidine was undergoing some trials and had been described as "the most potent and selective H2-blocker described to date." Ex. 2063 (abstract 635). Later tests with mice showed that tiotidine caused precancerous lesions in the stomach, so the compound was abandoned for therapeutic use. Those effects were not yet known in November 1979. As Judge Owen wrote in Yamanouchi: "So much for predictability in the field." 21 F. Supp.2d 366, 372 n. 12 (S.D.N.Y. 1998), aff'd, 231 F.3d 1339.

Zenith contends that nizatidine was obvious based on Dr. Baldwin's testimony that a synthetic organic chemist of ordinary skill in November 1979 would have recognized that the compound now called nizatidine was part of a special set of just 15 compounds that could be made by mixing and matching segments of cimetidine, ranitidine, and tiotidine. He testified that the hypothetical researcher "would have every reason to expect that these 15 compounds would have H2 receptor antagonist activity." Tr. 109. Dr. Baldwin recognized that the search for second-generation H2 antagonists involved hundreds of thousands of compounds. Tr. 107. However, for purposes of his analysis in this case, he focused on cimetidine, ranitidine, and tiotidine. He explained this focus because cimetidine was already a successful drug and because ranitidine and tiotidine appeared at that time to be the most promising candidates in the development pipeline.

The basis for Dr. Baldwin's opinion is best illustrated visually. He looked at these three successful or promising compounds, and he then divided each into four portions: an aryl side chain, an aryl ring, a connector, and an unsaturated side chain. Dr. Baldwin illustrated the division as follows:

[diagram 4]

In Dr. Baldwin's method, the "connector" segment in all three compounds is identical, while there are three different aryl side chains (one of which is only a hydrogen atom, for cimetidine), three different aryl rings, and two different unsaturated side chains.

Using this division of the three known compounds into these four segments, Dr. Baldwin concluded that a skilled synthetic organic chemist could mix and match those four segments to create 18 different compounds, including the three known compounds. (Without introducing other variables, such as the locations where the chains bond with the rings, there are 18 possible permutations of the three aryl side chains, three aryl rings, one connector, and two unsaturated side chains. 3 x 3 x 1 x 2 = 18.) Dr. Baldwin compared this analysis to picking dishes for separate courses from a restaurant menu.

Nizatidine is the one of the 15 new compounds that uses the dimethyl-aminomethyl group as the aryl side chain, the thiazole ring as the aryl ring, the common connector, and the nitroethylene group as the unsaturated side chain. Nizatidine has the following structure:

[diagram 5]

From this mix-and-match approach, Dr. Baldwin concluded that a researcher of ordinary skill in synthetic organic chemistry who had studied the public literature on H2 receptor antagonists in November 1979 would have expected all of the 15 compounds to have H2 receptor antagonist activity. Tr. 126. He based that expectation on the common structural features with the three compounds that were known to be H2 receptor antagonists. From that "reasonable expectation," Zenith argues, nizatidine would have been obvious so that the '547 patent is invalid.

2. The Flaws in Dr. Baldwin's Analysis

Dr. Baldwin's analysis fails to show that nizatidine was obvious, for structural similarity alone is not sufficient to establish obviousness. In re Dillon, 919 F.2d 688, 697 (Fed. Cir. 1990) ( en banc). The party claiming obviousness must show that the prior art provided a reasonable expectation that the particular modification of a prior compound would produce a new compound with the desired properties. That is the premise of Yamanouchi. See 231 F.3d at 1345; see also In re Merck, 800 F.2d 1091, 1097 (Fed. Cir. 1986) (upholding PTO finding of obviousness for claim of new use of already known medication; given close structural similarity and similar psychotropic use of two compounds, one of ordinary skill in the art "would have expected amitriptyline to resemble imiprimaine in the alleviation of depression in humans"). It is not enough, however, to show that it would have been merely "obvious to try" the new invention. Id. Dr. Baldwin's analysis fails to meet the test.

First, Dr. Baldwin's analysis considers only whether some level of H2 antagonist activity might be expected. The Federal Circuit rejected that approach in the other H2 receptor antagonist case. In Yamanouchi, the defendant had argued "that an ordinary medicinal chemist would have reasonably expected the resulting compound to exhibit the baseline level of H2 antagonist activity." 231 F.3d at 1345. The Federal Circuit rejected the argument: "The baseline level of activity is a mere 1/165th the activity of cimetidine. This level of motivation does not show a `reasonable expectation of success.'" Id., quoting In re Longi, 759 F.2d at 897.

Similarly here, Dr. Baldwin conceded that his hypothetical researcher in 1979 could not have predicted from the chemical structures the levels of H2 antagonist activity that any of his 15 compounds would be expected to exhibit. Tr. 135. Under the reasoning of Yamanouchi, that concession alone defeats Zenith's theory of obviousness.

To avoid this reasoning, Zenith argued in closing that Dr. Baldwin did not phrase his testimony in terms of "baseline" H2 activity. Tr. 568-69. Dr. Baldwin did not ever specify any level of H2 activity that would have been predictable from chemical structure. Zenith's burden is to show obviousness by clear and convincing evidence. It cannot do so by relying on the ambiguity of its own evidence.

Second, Dr. Baldwin's analysis also failed to consider all the other properties of nizatidine that make it a safe and effective drug. Even if the structure of nizatidine alone could have given Dr. Baldwin's hypothetical researcher a reasonable expectation of a high level of H2 antagonist activity, that would still fall far short of what a researcher needed to expect success with a new compound. The hypothetical researcher would have been looking for a compound that showed not only a high level of H2 antagonist activity compared to cimetidine, but also many other critical properties. Those properties would include the absence of dangerous side effects, the ability to be taken orally, the duration of the action (so that the patient need not take pills as often), the absence of toxicity, and a safe therapeutic ratio (the ratio between a lethal dose and a therapeutic dose).

When considering the alleged obviousness of a chemical compound, the compound, its structure, and all its properties are treated as inseparable. In re Dillon, 919 F.2d at 697; In re Papesch, 315 F.2d at 391. In Yamanouchi, the defendant made an obviousness argument that was remarkably similar to Zenith's argument here, which disregards all those other essential properties of the claimed compound. The Federal Circuit emphatically rejected the argument: "The success of discovering famotidine was not discovering one of the tens of thousands of compounds that exhibit baseline H2 antagonist activity. Rather, the success was finding a compound that had high activity, few side effects, and lacked toxicity." 231 F.3d at 1345.

The same reasoning applies here. Dr. Baldwin conceded that his hypothetical researcher would not have had a reasonable expectation that his method of rearranging the segments from cimetidine, ranitidine, and tiotidine would result in a compound having the safety and efficacy needed for FDA approval. Tr. 136. Dr. Baldwin also conceded that "even minor variations to components of an H2 antagonist chemical structure can have significant effects on biological activity" of the compound. Tr. 170.

Dr. Baldwin's testimony on these points was consistent with the testimony of Dr. Mark Feldman and Dr. Edward C. Taylor, who are both more familiar than Dr. Baldwin is with H2 receptor antagonists and their development, as well as the problems of drug development more generally. For example, the different heterocyclic rings that Dr. Baldwin considered — the imidazole ring in cimetidine, the furan ring in ranitidine, and the thiazole ring in nizatidine and tiotidine — have substantially different chemistries that produce unpredictable different biological effects when they are interchanged for one another. See Tr. 319-20.

Dr. Feldman is a gastroenterologist who has specialized in treatment of peptic ulcer disease. He has been involved in many clinical trials of H2 receptor antagonists and has written extensively on H2 receptor antagonists, including a major review article in 1990 for the New England Journal of Medicine on H2 receptor antagonists. Dr. Taylor is a medicinal chemist who is now a professor emeritus at Princeton University. Among his many professional publications, he is the editor of "The Chemistry of Heterocyclic Compounds," a treatise of more than 70 volumes on the subject. Over the past 50 years, Dr. Taylor has also been involved in new drug development and has invented a number of drugs and other compounds.

Thus, under the reasoning of Yamanouchi and the cases requiring consideration of the properties or results of chemical compounds, Zenith's theory of obviousness falls far short. See also, e.g., Ortho Pharmaceutical Corp. v. Smith, 959 F.2d 936, 943 (Fed. Cir. 1992) (affirming finding of non-obviousness; "when these compounds were being developed, one could not predict the effect of small structural changes on the biological activity of steroid hormones").

Third, even if structural similarity and some level of H2 antagonist activity were enough, Dr. Baldwin's attempt to demonstrate a reason for trying nizatidine in 1979 depends on artificial constraints and pure hindsight. Dr. Baldwin's conclusion depends on starting only with the three compounds — cimetidine, ranitidine, and tiotidine — and without including any of the other H2 antagonists known in 1979.

If one expands the starting compounds, by including burimamide or metiamide or even a few of the thousands of other H2 antagonists described in the literature by November 1979, the class of candidate compounds quickly expands to a very large number. (For example, the same publication from August 1979 that identified tiotidine as "the most potent and selective H2-blocker described to date" also included just before that abstract two other abstracts that identified other promising H2 receptor antagonists. See Exs. 2063 2072 and Tr. 199-200.) One could also start with histamine itself, as the SmithKline French group did at the beginning.

Even when the analysis is limited to the three lead compounds, Dr. Baldwin's conclusion also depends on dividing those three lead compounds into exactly four segments — not two, not three, and not five. Dividing them into two or three segments will not produce nizatidine as one of the permutations. See Tr. 158-59. Dividing them into five or more segments produces a much larger class of candidates. For example, the thiazole ring in cimetidine has a methyl group attached to one of the carbons. Dr. Baldwin chose to treat that methyl group as an integral part of the thiazole ring rather than as a fifth segment that could be varied. If the methyl group were treated as a fifth segment that could be treated as present or not present, that one change would have doubled the number of candidate compounds.

As of 1979, no researchers in the field of H2 receptor antagonists were describing their work using the four-segment approach used by Dr. Baldwin. Dr. Baldwin has argued that the published research work showed they were actually using something close to his method, relying in particular on U.S. Patent No. 4,128,658, in which ranitidine is Example 15 (see Ex. 1075) and U.S. Patent No. 4,046,907, which claimed a number of imidazole-based H2 receptor antagonists (see Ex. 2056). It appears to the court that the patents Dr. Baldwin relied upon showed that researchers were actually trying a much wider range of variables, dividing the compounds into more than four segments, and generating huge numbers of candidate compounds. Using the actual methods of researchers would have produced (and did produce) a huge number of candidate compounds, from which the selection of nizatidine would not have been at all obvious.

Similarly, even if the hypothetical researcher in 1979 had begun with the three lead compounds but had also considered their isomers and tautomers — compounds in which the same atoms are arranged in different structures — the number of candidate compounds would have multiplied quickly to a much larger number, from which the selection of nizatidine also would not have been at all obvious. Dr. Baldwin's analysis also did not take into account the so-called steric configurations of the compounds, which multiply the possibilities even more.

Zenith's and Dr. Baldwin's response to these criticisms is to invoke the ghost of William of Ockham, whose famous logical razor praised simplicity in philosophical and scientific explanations. Zenith and Dr. Baldwin suggest that there was no need to consider isomers or tautomers or different steric configurations because the promising lead compounds of cimetidine, ranitidine, and tiotidine were so similar in overall structure. Using the lead compounds, they contend, the simplest and most elegant approach, and the natural one to try in the "first pass" at the problem, was the one described by Dr. Baldwin that would have led the ordinary synthetic organic chemist to nizatidine as one of just 15 candidate compounds.

William of Ockham was a fourteenth century English Franciscan philosopher. He formulated his "razor" in different ways. One read: non sunt multiplicanda entia praeter necessitatem — one should not multiply the number of entities unnecessarily. Applied more broadly, his "razor" declares that a simpler explanation is preferable to a more complex one, all other things being equal.

The court has great respect for the power of Ockham's razor in science, philosophy, and law. In this case, however, Zenith's explanation is pure hindsight. To use the method to produce such a small class of candidate compounds, one must artificially impose the constraints on the number and identity of lead compounds, the number of segments into which they are divided, and the exclusion of other variables such as isomers and tautomers. For a researcher working at the time, choosing exactly the right variables in such a method and analysis would not have been obvious. Just as courts "cannot use hindsight reconstruction to pick and choose among isolated disclosures in the prior art to deprecate the claimed invention," Ecolochem, Inc. v. Southern California Edison Co., 227 F.3d 1361, 1371 (Fed. Cir. 2000) (internal quotations marks and citation omitted), it is difficult to build a persuasive case of obviousness based on such a methodology that depends on so many arbitrary and artificial constraints to produce the desired result.

Dr. Baldwin and Zenith contend that the relevant art for this case is synthetic organic chemistry, and that one of "ordinary skill" in that art would have an advanced degree in organic chemistry. Lilly and Dr. Taylor contend that the relevant art for this case is medicinal chemistry and more specifically the development of H2 receptor antagonists. Dr. Taylor testified that one of ordinary skill in that art in 1979 would have been a medicinal chemist with a Ph.D degree in organic chemistry or equivalent training, with two to three years of additional experience in the H2 receptor antagonist field, and would have had an appreciation of the relevant biological events associated with a successful H2 receptor antagonist drug. Tr. 316-17. For purposes of the Graham factor regarding the level of ordinary skill in the art, the court finds Lilly's and Dr. Taylor's position more persuasive. The challenge addressed by Dr. Pioch and others in the field was the invention of a safe and effective drug, not merely the creation of a new organic compound with some level of H2 antagonist activity. The work required the researcher to focus on all biological effects of a compound, including those that determine whether the compound will be safe and effective.

To illustrate the use and benefits of hindsight in Zenith's theory of obviousness, the evidence shows that in the late 1970s, hundreds of highly skilled medical chemists working for virtually every major pharmaceutical company in North America, Europe, and Japan were trying to find second generation H2 receptor antagonists. Their employers recognized the huge potential market for such drugs, with annual sales measured in the billions of dollars. They had every incentive to find a new drug that would be more effective and have fewer side effects than cimetidine. Only three succeeded — ranitidine, nizatidine, and famotidine. Even the gifted team at SmithKline French, whose development of cimetidine contributed to the Nobel Prize, did not see the solution that Dr. Baldwin and Zenith now describe as so simple. Like Yamanouchi, this case "has all the earmarks of somebody looking at this from hindsight." See 231 F.3d at 1345, quoting 21 F. Supp.2d at 370.

Famotidine was not part of Dr. Baldwin's analysis. The invention of famotidine was the subject of litigation in Yamanouchi, discussed in detail in this entry. Among the other companies who tried and failed to develop second generation H2 receptor antagonists were Bristol-Myers, Merck, Glaxo, Pfizer, and ICI. Their research identified millions of potential candidate compounds, but none that turned out to be safe, effective, and successful drugs.

C. Objective Evidence of Non-Obviousness

Objective evidence of non-obviousness has often been described in judicial opinions as addressing "secondary considerations." The law is clear, however, that such evidence is important and must be considered before a court may find a patent claim invalid for obviousness. E.g., In re Rouffet, 149 F.3d at 1355; Demaco Corp. v. F. Von Langsdorff Licensing Ltd., 851 F.2d 1387, 1391 (Fed. Cir. 1988) (district court erred in finding obviousness based on prior art but without considering objective evidence of non-obviousness); Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1380 (Fed. Cir. 1986) (objective evidence must be considered before a conclusion on obviousness is reached, and "is not merely `icing on the cake'").

Such evidence may be used to rebut a prima facie case of obviousness based on prior art references. In fact, such evidence "may often be the most probative and cogent evidence in the record." Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1538-39 (Fed. Cir. 1983). The proponent of the evidence, however, must establish a nexus between the evidence and the merits of the claimed invention. E.g., In re GPAC, Inc., 57 F.3d 1573, 1580 (Fed. Cir. 1995); Demaco Corp., 851 F.2d at 1392.

1. Commercial Success

From May 1988 through September 2000, Lilly's net sales of nizatidine to wholesalers totaled $3.1 billion, with a yearly peak of $401 million in 1995. Ex. 1012. That is strong evidence of commercial success, especially in a field with three other patented competitors (cimetidine, ranitidine, and famotidine) that had higher levels of sales. Strong evidence of commercial success is not surprising in a case under the Hatch-Waxman Act, of course. If the patented drug were not a commercial success, generic manufacturers would have little interest in offering their own versions of the drug.

Zenith has tried to rebut this evidence by suggesting that the commercial success of nizatidine has no nexus with the invention itself, but resulted instead from effective marketing and advertising. Zenith's argument is not persuasive, and Lilly has met its burden of showing a nexus. First, the only active ingredient in AXID is in fact nizatidine. This is not a case in which a patent claims merely one feature of a more complex commercial product, the success of which may derive from other features. Second, the evidence shows that Lilly's marketing and advertising budgets for AXID (nizatidine) have been relatively modest by industry standards. In addition, although nizatidine sales dropped as the competing brands went off-patent and generics entered the market, AXID has continued to have substantial sales. During 1999, when generic versions of cimetidine and ranitidine were on the market, Lilly still sold $239 million worth of nizatidine. Ex. 1012. In short, the later and continued commercial success of nizatidine offers substantial evidence that the invention was not obvious at the time it was made in 1979.

2. Unexpected Results

The Federal Circuit has recognized that unexpected superior results from an invention tend to support a finding that the invention was not obvious to one of ordinary skill in the art. See, e.g., In re Baxter Travenol Labs., 952 F.2d 388, 392 (Fed. Cir. 1991); In re De Blauwe, 736 F.2d 699, 705 (Fed. Cir. 1984).

"The basic principle behind this rule is straight forward — that which would have been surprising to a person of ordinary skill in a particular art would not have been obvious." In re Mayne, 104 F.3d 1339, 1343 (Fed. Cir. 1997), quoting In re Soni, 54 F.3d 746, 750 (Fed. Cir. 1995). "The principle applies most often to the less predictable fields, such as chemistry, where minor changes in a product or process may yield substantially different results." Id.

The principle may apply especially often when dealing with medicinal chemistry. The biological effects of a new compound will often be too complex to predict with any accuracy. Experts for both Zenith and Lilly testified that the biological effects of minor changes in the structure of a candidate for an H2 receptor antagonist simply cannot be predicted with any confidence. Their testimony is well supported by the history of challenges and failures in the field.

The evidence in this case also shows that unexpected results played a critical role in the issuance of the '547 patent. The patent examiner initially rejected the claim of nizatidine as obvious based on two prior art references which, when combined, suggested a class of compounds that included the "4,2 isomer" of nizatidine. Ex. 1003 at 92-95; see also id. at 98-102 (applicant's attorney's response). Nizatidine is the "2,4 isomer," in which the dimethyl-aminomethyl group is attached to carbon atom 2 in the thiazole ring, and the "connector" is attached to carbon atom 4. In the 4,2 isomer, the bonding positions of the dimethyl-aminomethyl group and the connector are reversed.

In response to this rejection, Dr. Pioch submitted a declaration that showed the results of tests of nizatidine and its 4,2 isomer. Dr. Pioch reported that the 4,2 isomer still shows H2 receptor antagonist activity, but the rates are dramatically lower than the rates for nizatidine itself. Ex. 1003 at 103-04. The examiner relied on these unexpected differences in the level of activity exhibited by the two isomers in allowing the nizatidine claim and issuing the '547 patent. Id. at 106-07. Those unexpected results provide substantial objective evidence of non-obviousness.

3. Long-felt Need and Failure of Others

The failure of others to develop alternatives to the invention may also be evidence of non-obviousness. Graham, 383 U.S. at 17-18; In re GPAC, 57 F.3d at 1580. As discussed above, the evidence shows that hundreds of the best medicinal chemists in the world were searching for safe and effective H2 receptor antagonists. After the discovery and initial success of cimetidine, many major pharmaceutical companies all over the world tried to develop improved compounds for the second generation — compounds that would be more effective than cimetidine and have fewer side effects.

As of November 1979, it was too early to know whether anyone else had succeeded. (Ranitidine did not receive FDA approval for use as a safe and effective drug until 1983.) With the benefit of hindsight, it is now clear that although these hundreds of medicinal chemists imagined and tested many thousands of compounds, only three compounds were successful as safe and effective drugs — ranitidine, famotidine, and nizatidine.

The need for improved second-generation H2 receptor antagonists is shown by, among other things, the continued commercial success of nizatidine, ranitidine, and famotidine even after cimetidine went off-patent and cheaper generic versions were available. Physicians with a choice have preferred to use the more expensive second-generation compounds.

In addition, even the SmithKline French team whose development of cimetidine contributed to the Nobel Prize was unable to develop a successful second-generation compound.

This evidence shows that it was very difficult to develop successful alternatives to cimetidine, and it adds support to the conclusion that the '547 patent is not invalid for obviousness.

4. Copying

Copying of the patented invention may also provide evidence of non-obviousness. Diversitech Corp. v. Century Steps, Inc., 850 F.2d 675, 679 (Fed. Cir. 1988). Evidence of copying tends to be more compelling in arts in which there is more room to design around patents or to improve upon them. When the invention in question is a drug that has won FDA approval as safe and effective, the incentive to copy is strong. In fact, the ANDA procedures established by the Hatch-Waxman Act require generic drug manufacturers to copy the approved drug. Variations undermine the FDA's ability to assume that if the patented drug is safe and effective, the generic competitor will also be safe and effective.

The fact that copying is likely to be present in many Hatch-Waxman Act cases does not allow the court to ignore the copying as evidence of non-obviousness, even though it may be entitled to relatively little weight. In this case, in this field of new drug design, the very need for copying results from and emphasizes the unpredictability of medicinal chemistry. (As we now know, for example, tiotidine, which plays such a critical role in Dr. Baldwin's structural analysis, later turned out to be toxic.) To gain FDA approval, therefore, a company in Zenith's position must copy the patented invention as closely as possible. Small changes in chemical structure may have dramatic and unpredictable biological effects. To that extent, the evidence of copying adds a little weight against a finding of obviousness, though it is not essential to this court's ultimate conclusion.

5. Acquiescence of Others

Licenses to a patent may be evidence of non-obviousness if they indicate that others in the industry recognize the strength and validity of the patent in question. See In re GPAC, 57 F.3d 1573, 1580 (Fed. Cir. 1995). Licenses will be entitled to little weight, however, unless the patentee can demonstrate "a nexus between the merits of the invention and the licenses of record." Id., quoting Stratoflex, 713 F.2d at 1539. In this case, the licensing is minimal. Lilly made a recent decision to grant an exclusive license to plaintiff Reliant as Lilly shifted its own business efforts in other directions. Such a licensing arrangement does not demonstrate the validity of the patent. The substitution of one business entity for another adds nothing to the overall equation on the issue of obviousness.

D. Conclusion on Obviousness

After considering all the evidence on the issue of obviousness, the court concludes that Zenith has failed to show that the '547 patent is invalid for obviousness. Zenith failed to make even a prima facie showing of obviousness based on similar chemical structure. Zenith's analysis (1) failed to account for the inability to predict even the level of H2 antagonist activity; (2) failed to account for the inability to predict the other critical properties that have made nizatidine safe and effective as a drug; and (3) was built with hindsight and artificial constraints.

Under the reasoning of Yamanouchi, Lilly was entitled to judgment after Dr. Baldwin conceded that his hypothetical researcher of ordinary skill could not have predicted the level of H2 activity and could not have had a reasonable expectation that nizatidine would be a safe and effective drug. In addition, Lilly has produced substantial objective evidence of non-obviousness. The '547 patent easily withstands Zenith's defense of obviousness under 35 U.S.C. § 103(a). Zenith's ANDA for nizatidine may not be granted on the basis of the Paragraph IV certification asserting invalidity.

In Yamanouchi, the district court had entered judgment as a matter of law under Fed.R.Civ.P. 52(c) at the close of the defendant's evidence on obviousness. 231 F.3d at 1342. In this case, Lilly made a similar motion, not only at the end of Zenith's evidence on the question but earlier, in the very midst of Dr. Baldwin's testimony. As a matter of caution, the court did not grant Lilly's motion but took it under advisement so that a complete record could be developed. That more complete record has also been useful in deciding the issue of willfulness. The fact that Lilly's motion was not granted during trial should not be interpreted as an indication that the issue of obviousness was deemed close or difficult.

IV. Exceptional Case and Willful Infringement

Zenith's only act of infringement here was filing an amended ANDA with a Paragraph IV certification under the Hatch-Waxman Act. Lilly has shown its patent is valid. Any additional relief for the patentee beyond that declaration would require proof of "commercial manufacture, use, offer to sell, or sale within the United States . . . of an approved drug." 35 U.S.C. § 271(e)(4)(C). Lilly contends, however, that the court should find this to be an "exceptional case" under 35 U.S.C. § 285 and award Lilly its attorneys' fees.

Lilly bases its willfulness assertion on the totality of the circumstances, including the weakness of Zenith's invalidity case, Zenith's failure to obtain a written opinion from counsel before the act of infringement, errors and omissions in the written opinion that was provided five weeks after the act of infringement, the inadequacy of Zenith's notice to Lilly of the factual and legal basis of the assertion of invalidity, and Zenith's haste to file its Paragraph IV certification in light of the huge financial gains it might win by being first.

Willful infringement is one form of misconduct that can make a case exceptional for a fee award. Yamanouchi, 231 F.3d at 1346-47; Fromson v. Western Litho Plate and Supply Co., 853 F.2d 1568, 1573 (Fed. Cir. 1988) (after finding willful infringement, district court erred by failing to explain why it did not award attorneys' fees); Avia Group Int'l, Inc. v. L.A. Gear California, Inc., 853 F.2d 1557, 1567 (Fed. Cir. 1988) (finding of willful infringement is sufficient to find case is exceptional). The Hatch-Waxman Act refers to § 285 in 35 U.S.C. § 271(e)(4), indicating that Congress plainly anticipated that an ANDA case could justify an award of attorneys' fees. See Yamanouchi, 231 F.3d at 1346 (affirming finding of willfulness and fee award). The only ground for finding this case exceptional is willful infringement.

An act of infringement may be deemed willful where the infringer is aware of the patent in question and fails to use due care to avoid infringement. E.g., Electro Medical Systems, S.A. v. Cooper Life Sciences, Inc., 34 F.3d 1048, 1056 (Fed. Cir. 1994); Rolls-Royce Ltd. v. GTE Valeron Corp., 800 F.2d 1101, 1109 (Fed. Cir. 1986). One statement of the standard is whether the infringer, "acting in good faith and upon due inquiry, had sound reason to believe that it had the right to act in the manner that was found to be infringing. The law of willful infringement does not search for minimally tolerable behavior, but requires prudent, and ethical, legal and commercial actions." SRI Int'l, Inc. v. Advanced Technology Labs., Inc., 127 F.3d 1462, 1465 (Fed. Cir. 1997).

As the Federal Circuit recognized in Yamanouchi, in ANDA cases involving Paragraph IV certifications challenging the validity of the patent in suit, the alleged infringer will always have been aware of the patent in question. See 231 F.3d at 1347. There is nothing accidental or unintentional about a Paragraph IV certification that claims invalidity. "The joint operation of §§ 271(e) and 285 require the paragraph (2) infringer to display care and regard for the strict standards of the Hatch-Waxman Act when challenging patent validity." Id. However, the fact that the defendant has filed an ANDA with a Paragraph IV certification and lost a lawsuit on the issue of validity is not enough by itself to support a finding of willful infringement.

The plaintiff alleging willful infringement must come forward with clear and convincing evidence that the defendant failed to exercise due care. E.g., Comark Communications, Inc. v. Harris Corp., 156 F.3d 1182, 1190 (Fed. Cir. 1998); SRI Int'l, 127 F.3d at 1465. In deciding whether a case is exceptional for these purposes, the court must consider the totality of the circumstances. Yamanouchi, 231 F.3d at 1347; Kaufman Co. v. Lantech, Inc., 807 F.2d 970, 978-79 (Fed. Cir. 1986).

A. Advice of Counsel to Rebut Willfulness

An important factor in deciding willfulness is whether the defendant sought, obtained, and reasonably relied upon competent advice of counsel as to the validity of the patent the defendant intended to challenge. E.g., Westvaco Corp. v. International Paper Co., 991 F.2d 735, 743 (Fed. Cir. 1993). The mere fact that an opinion was obtained does not necessarily resolve the issue of willfulness. E.g., Minnesota Mining and Mfg. Co. v. Johnson Johnson Orthopaedics, Inc., 976 F.2d 1559, 1580 (Fed. Cir. 1992) (affirming finding of willfulness despite in-house patent counsel's opinion of invalidity).

When an infringer has received an opinion of invalidity from competent counsel before the act of infringement, a finding of willfulness is unusual. However, a court need not take such an opinion at face value. A court presented with an opinion of counsel should review the opinion to determine whether it shows an adequate foundation based on a review of all necessary facts or whether it is merely conclusory. Westvaco, 991 F.2d at 743. "Counsel's opinion must be thorough enough, as combined with other factors, to instill a belief in the infringer that a court might reasonably hold the patent is invalid, not infringed, or unenforceable." Westvaco, 991 F.2d at 743, quoting Ortho Pharmaceutical Corp. v. Smith, 959 F.2d 936, 944 (Fed. Cir. 1992). "To serve as exculpatory legal advice the opinion of counsel is viewed objectively, to determine whether it was obtained in a timely manner, whether counsel analyzed the relevant facts and explained the conclusions in light of applicable law, and whether the opinion warranted a reasonable degree of certainty that the infringer had the legal right to conduct the infringing activity." SRI International, 127 F.3d at 1467. Also, it is important that the opinion of counsel be obtained before the defendant begins the infringing activity. E.g., Underwater Devices Inc. v. Morrison-Knudsen Co., 717 F.2d 1380, 1389 (Fed. Cir. 1983).

B. The Oral Opinion

With these standards in mind, the court turns to the evidence on the issue. In 1997, Zenith hired the law firm of Lerner, David, Littenberg, Krumholz and Mentlik ("Lerner David") to evaluate the nizatidine patents, including the '547 patent. On July 11, 1997, Zenith asked Lerner David to order the file histories on the '547 patent and Patent No. 4,760,075 and to determine "whether there is reasonable basis for performing a full invalidity search and opinion on these patents." Ex. 1064. On August 21, 1997, Lerner David sent Zenith a written analysis of the file histories of the '547 patent and the '075 patent (which is not at issue here). Ex. 1028. The project lay relatively dormant until the summer of 1998. See Exs. 1031, 1035. On July 10, 1998, Zenith directed Lerner David to order an American Chemical Society search on nizatidine. Ex. 1036.

On September 15, 1998, Zenith mailed its ANDA on nizatidine with a Paragraph III certification stating its intention to wait until the '547 patent expired before selling generic nizatidine. In the meantime, Lerner David's billing records show that at least one lawyer spent a good deal of time beginning September 10, 1998, considering the validity of the '547 patent on nizatidine. Ex. 1078. The billing records for September show more than 40 hours of attorney time on the nizatidine project.

The critical day came on October 5, 1998. Three attorneys from Lerner David and two patent attorneys for Zenith held a conference call that lasted about 30 minutes. They discussed the validity of the '547 patent and two other nizatidine patents. The participants from Lerner David were Michael Teschner, Arnold Krumholz, and William Mentlik. The participants for Zenith were Michael Keller and Lee Banks.

Four of the participants testified in depositions about the conference call. (Mentlik was Zenith's lead trial counsel in this action and so was not a witness.) None of the participants testified at trial. The designated deposition testimony indicates that the Lerner David attorneys expressed an opinion that the '547 patent was invalid as obvious and that the Zenith attorneys, Keller and Banks, agreed with that view. Banks volunteered that the oral opinion from Lerner David was "in-depth and complete." Banks Dep. at 232. However, with respect to any details that would have allowed Lilly to probe the actual basis for the opinion, the designated evidence from the four witnesses indicates virtually nothing other than the self-serving bottom line.

It is also surprising and more than a little suspicious that, according to the participants, there were and are no contemporaneous notes or other documents reflecting the content of this critical discussion. Banks testified that his testimony about the telephone call was based on his unaided recollection. Banks Dep. at 238. He had already acknowledged that records had shown his recollection of the relevant events was faulty. Id. at 220, 230.

On October 6, 1998, attorneys for Lilly met with Zenith attorney Keller in Florida. The Lilly attorneys disclosed to Keller that no other generic drug manufacturer had filed an ANDA challenging the validity of the '547 patent on nizatidine. That same morning, Keller sent an e-mail to Banks stating: "Lilly was asking questions about this drug. It appears we are first. Brief Eric fully on yesterdays conference call." The reference to "Eric" was to Zenith's Dr. Eric Mittleberg, who did not participate in the telephone conference on October 5, 1998. Dr. Mittleberg was and is Zenith's director of scientific and medical affairs. He made the final decision to file the amended ANDA with the Paragraph IV certification. Later on October 6th, Banks replied to Keller by e-mail: "I have informed Eric of the exact nature of Lerner, David's opinion. Eric wants the amended certification to go in today."

The reference to being "first" is important here, especially in light of Zenith's unusual decision to file the Paragraph IV certification based only on an oral opinion from counsel. Under the Hatch-Waxman Act, the first competitor to file a Paragraph IV certification on a patented drug is entitled (in the absence of a successful suit for infringement) to a 180-day period in which its product will be the only generic competitor. See 21 U.S.C. § 355(j)(5)(B)(iv). Dr. Mittleberg acknowledged that this head-start on other generic competitors is a very lucrative advantage. Industry experience shows that the first competitor tends to be able to keep the lion's share of the business it wins in those first 180 days. Tr. 257-59. At the same time, filing a Paragraph IV certification also has risks. The required notice to the patentee starts a 45-day clock in which the patentee must either surrender the patent's protection or file an infringement suit that is likely to be costly for both sides.

In light of these potential benefits and risks, Zenith's patent attorney Keller wanted to make sure that Dr. Mittleberg was fully aware of the (unspecified) limits of the oral opinion from Lerner David. As Zenith was rushing to prepare the Paragraph IV certification on the afternoon of October 6th, Keller instructed Banks to brief Dr. Mittleberg "fully" on the conference call with Lerner David lawyers. Ex. 1041. Banks did so. Banks also testified that he told Dr. Mittleberg he should call Lerner David's Arnold Krumholz himself to talk about the Lerner David opinion. Banks Dep. at 220-21. There is no evidence from either Dr. Mittleberg or Krumholz that such a call was ever made. See Tr. 278.

Dr. Mittleberg testified at trial. He knows that one of the in-house lawyers told him that Lerner David said the '547 patent was invalid. Dr. Mittleberg did not even remember who told him that, let alone any details of the basis for challenging the patent. Tr. 247. He could not testify about any point except the helpful conclusion — Lerner David and the in-house attorneys had said there was a basis to challenge the patent as invalid. The only witness to testify at trial on the point did not know why or how.

Keller was the patent lawyer for Zenith who was responsible for reaching a conclusion about whether the company had a reasonable basis for concluding that the '547 patent was invalid. Keller Dep. at 54-55, 58-59. The nizatidine ANDA was one of only two cases in which Zenith had filed a Paragraph IV certification based on only an oral opinion from counsel. Id. at 167.

Reliance on oral opinions of counsel is not favored. Minnesota Mining, 976 F.2d at 1580; Bott v. Four Star Corp., 807 F.2d 1567, 1572 (Fed. Cir. 1986), overruled on other grounds, A.C. Aukerman Co. v. R.L. Chaides Const. Co., 960 F.2d 1020, 1038-39 (Fed. Cir. 1992); Shiley, Inc. v. Bentley Lab., Inc., 601 F. Supp. 964, 968 (C.D.Cal. 1985), aff'd, 794 F.2d 1561 (Fed. Cir. 1986). Oral opinions carry less weight, for example, because they must be proved perhaps years after the event, based only on testimony which may be affected by faded memories and the forces of contemporaneous litigation. Minnesota Mining, 976 F.2d at 1580. Nevertheless, reliance on an oral opinion is not necessarily unreasonable as a matter of law. Compare, e.g., Radio Steel Mfg. Co. v. MTD Products, Inc., 788 F.2d 1554, 1558-59 (Fed. Cir. 1986) (affirming finding that infringement was not willful where infringer relied on oral opinion of outside patent counsel), with American Medical Systems, Inc. v. Medical Engineering Corp., 6 F.3d 1523, 1531 (Fed. Cir. 1993) (affirming finding of willfulness; oral opinion of invalidity from counsel was not credible, and infringer did not obtain credible written opinion until 20 months after beginning infringement).

The reasons that oral opinions are disfavored are in full view in this case. The exceedingly limited memories of all participants about the details of the oral opinion on October 5th are not credible. The complete and unexplained absence of contemporaneous notes or documents about the oral opinion is suspicious, especially in light of the Zenith's lawyers' need to brief Dr. Mittleberg "fully" on the "exact nature" of the Lerner David oral opinion and its unspecified "limits." Dr. Mittleberg's failure to follow the recommendation to talk directly with attorney Krumholz about the opinion, and Zenith's decision not to call any participants in the October 5th conference call for live testimony at trial, further undermine Zenith's claim that it reasonably relied on a competent opinion of counsel when it filed the Paragraph IV certification on October 6, 1998.

Both parties in this case presented expert testimony on the competence of the Lerner David opinions in this case, both oral and written.

C. The Later Written Opinion

Lerner David later provided Zenith with a written opinion, on November 12, 1998. Zenith contends that the later written opinion accurately reflects the substance of the oral opinion given on October 5th. Participants in the October 5th conference call testified to that general point in their depositions, but their memories on all other details of the matter are so limited that the general and self-serving conclusions are not credible. None were called to testify at trial.

In addition, billing records from Lerner David show that extensive work continued on the nizatidine issue after the October 5th conference call, including preparation of a written opinion. See Ex. 1078. On October 6th, Teschner's diary entry was: "Further research regarding the nature of amines and dictation of first rough draft of executive summary opinion." On October 7th, another attorney recorded: "Initial review of validity search results of outside consultant." On October 8th, Teschner recorded: "Review and correction to opinion and begin legal research." Later entries in October indicate a "Dialog search for direct comparison between ranitidine and nizatidine and other H-2 antagonist, review of search results." The work on the written opinion continued in early November. The continuing research and analysis after October 5th were extensive. The apparent need for that later work further undermines Zenith's claim that it relied reasonably on the earlier oral opinion.

On November 12, 1998, Lerner David sent the written letter to Zenith's Lee Banks. The opinion letter described the task: "we have examined whether or not sufficient justification exists for the filing of a certification pursuant to 21 U.S.C. § 355(j)(2)(A)(vii)(IV) (a `paragraph IV certification') in an Abbreviated New Drug Application (`ANDA') seeking approval to market a generic version of nizatidine." Ex. 2045 at 1. The letter offered the following opinion:

As we explained to you in a telephone call on October 5, 1998, having completed our study, we are of the opinion that the three patents in question [including the '547 patent] are invalid. Therefore, in our opinion, the filing of a Paragraph IV certification would comply with the statute. However, as we also advised you, while in our opinion the patents are each invalid, we cannot give you any assurance that you will prevail in challenging the validity of these patents in court.
Id.

The fact that the written opinion did not guarantee victory does not undermine Zenith's ability to rely reasonably upon it. See, e.g., Westvaco, 991 F.2d at 744. Also, of course, the fact that a court later disagreed with the opinion does not show that the defendant could not reasonably rely upon the opinion. The issue of willfulness would not even arise unless a court or jury later disagreed with the opinion. See, e.g., Graco, Inc. v. Binks Mfg. Co., 60 F.3d 785, 793-94 (Fed. Cir. 1995).
The written opinion letter also included a detailed analysis of the district court decision in Yamanouchi. The district court issued the opinion on October 1, 1998, but the Lerner David attorneys did not learn of it until mid-October (apparently October 13th). Diary entries for October show time devoted to analyzing the decision. Because Yamanouchi did not figure in the oral opinion on October 5th that Zenith relied upon to undertake the act of infringement, the filing of the Paragraph IV certification, the court has disregarded the later analysis of that case in the written opinion. Based on similar reasoning, the court denied Lilly's earlier motion to compel Zenith to disclose even later advice from its counsel. See Eli Lilly and Co. v. Zenith Goldline Pharmaceuticals, Inc., 149 F. Supp.2d 659, 663 (S.D.Ind. 2001).

Even if the court assumed that the November 12th opinion letter accurately reflected the analysis given orally on October 5th, that opinion letter still would not justify reliance upon its conclusions. On first reading by someone not familiar with the issues, the opinion appears to reflect some key indicia of reliability. It discusses prior art, the file history, and one case as a source of the applicable law, and it is from reputable outside patent attorneys.

A court errs if it considers only the qualifications of the attorney involved and fails to consider the competence of the opinion itself. Jurgens v. CBK, Ltd., 80 F.3d 1566, 1572 (Fed. Cir. 1996) (reversing and remanding for further consideration after district court declined to follow jury finding of willfulness that necessarily rejected advice of counsel defense).

But upon closer reading — upon the kind of reading that Zenith's own patent attorneys should have given it — several glaring errors of fact and omissions of law come to light. Those errors and omissions weigh heavily against Zenith's claim that it reasonably relied on advice of counsel even if the court were to assume (as Zenith's legal expert did) that the letter accurately stated the basis of the earlier oral advice. Cf. Yamanouchi, 231 F.3d at 1347-48 (affirming finding of willfulness where attorney's written opinion contained just one acknowledged error in chemistry that was critical to its conclusion); Johns Hopkins University v. CellPro, Inc., 152 F.3d 1342, 1364 (Fed. Cir. 1998) (affirming finding of willfulness where written opinions contained shortcomings and errors that should have been obvious to experienced patent attorney who received them).

The substantive part of the written opinion began with a discussion of the law of obviousness based on structural similarity of two compounds, drawn from In re Jones, 958 F.2d 347, 349-50 (Fed. Cir. 1992). The letter discussed the Durant patent, No. 4,046,907 (Ex. 2056), which disclosed thiazole-based compounds that differed from nizatidine only in the existence and/or identity of a group of atoms bonded to the second carbon in the thiazole ring. For nizatidine, that group is a dimethyl-aminomethyl group.

The opinion acknowledged that Durant did not specifically disclose the use of a dimethyl-aminomethyl group in that position. The opinion stated that Durant did

contemplate the possibility of amino-substituted thiazoles. As is known in the art, the term "amino" is used to describe amines generally and both the unsubstituted nitrogen found in compound B and the aliphatic amine dimethyl-aminomethyl group found in nizatidine, are amines. Presuming the term "amino" means something other than just a NH2-group, a conclusion which is reasonable in view of certain established nomenclature practices, the use of a dimethyl-aminomethyl group would appear to be contemplated.

Ex. 2045 at 5. The opinion went on to state: "it is our opinion that the recitation of `amino' groups in Durant et al. would itself contemplate the use of groups such as the dimethyl-aminomethyl group." Id. at 6.

The opinion's treatment of the dimethyl-aminomethyl group as an "amine" or "amino" group was a clear error in chemistry. The evidence at trial from all witnesses on the subject showed a clear and simple difference. An amino or amine group would create a nitrogen-carbon bond with the thiazole ring by bonding a nitrogen atom in the amino or amine group to a carbon atom in the thiazole ring. See, e.g., Tr. 197-98 (Dr. Baldwin); 392-94 (Dr. Taylor). The dimethyl-aminomethyl group in nizatidine creates a carbon-carbon bond between a carbon atom at the base of the group with a carbon atom in the thiazole ring.

The chemical and practical differences between a nitrogen-carbon bond and a carbon-carbon bond are substantial. One skilled in the art would not expect that substituting one for the other would have comparable biological and chemical effects. The erroneous treatment of dimethyl-aminomethyl as having been disclosed by references to amines or amino groups was a basic error that should have been caught by those preparing and reviewing the opinion letter.

Zenith's in-house patent attorneys both had backgrounds in chemistry, and they testified that they had others at Zenith review the technical portions of opinions from outside counsel. Banks Dep. at 161-62, 321-22; Keller Dep. at 145.

The opinion letter then made another serious factual error when it addressed the ranitidine patent, No. 4,128,658. The structure of ranitidine differs from that of nizatidine in one major way. The heterocyclic ring for ranitidine is a furan ring (four carbon atoms and one oxygen atom) rather than the thiazole ring in nizatidine. The opinion also noted that U.S. Patent No. 4,239,769 suggested the use of a thiophene ring (four carbon atoms and one sulfur atom). See Ex. 2045 at 6.

The opinion letter asserted that one of ordinary skill in the art would have considered both the '658 patent and the '769 patent, and would have found the teaching, suggestion, and motivation to modify those compounds so as to form nizatidine. Id. at 7. The opinion letter also discussed a Yellin article comparing cimetidine and tiotidine, as well as a Crenshaw patent, No. 4,374,248.

The opinion letter then offered a different analysis to support its conclusion, based on ranitidine:

Since ranitidine is a commercially successful H-2 antagonist, one of ordinary skill in this area would surely consider various substitutions to potentially improve or modify its efficacy. One substitution which, in our opinion, is suggested by the prior art as a whole and inevitably would be made would be substituting a thiazole ring for the furan ring in ranitidine, as thiazoles already had been shown to be effective as H-2 antagonists.
Id. at 8.

This analysis is based on a fundamental factual error. At the time that nizatidine was discovered in late 1979, ranitidine was not yet a commercially successful H2 antagonist. Ranitidine did not go on the market with FDA approval until 1983. In 1979 it was a promising compound, but its later success was far from obvious in the uncertain world of testing drugs for safety and efficacy (as the later fate of tiotidine also demonstrates). The reference to ranitidine's commercial success was another clear factual error that those who wrote and received the opinion should have caught.

The Lerner David opinion contains a third factual error that is even more important. After discussing the prior art, the authors cautioned that obviousness can be rebutted by a showing of "unexpected and superior results." They wrote: " While we are currently unaware of any basis upon which Lilly could make such a showing, * * * if Lilly were able to demonstrate that nizatidine exhibits unexpected results over structurally-related compounds such as those found in Durant et al. and/or ranitidine, Lilly might be able to overcome a prima facie case of structural obviousness." Id. at 9 (emphasis added).

In fact, Lerner David and Zenith were both aware of such evidence. Lerner David had reviewed the file history of the '547 patent and had sent an analysis of that history to Zenith on August 21, 1997. See Ex. 1028. As discussed above in terms of the unexpected results of nizatidine, the file history showed that the nizatidine patent claim had initially been rejected as obvious based on certain prior art references. The applicant had overcome that rejection by submitting a declaration from the inventor of nizatidine comparing the H2 antagonist activity of nizatidine and its 4,2 isomer. See Ex. 1003 at 103-04. The result of the change in structure was a marked decrease in H2 receptor activity.

These results showed that nizatidine was unexpectedly superior to its own isomer in terms of H2 antagonist activity. Cf. In re Mayne, 104 F.3d 1339, 1343 (Fed. Cir. 1997) (affirming PTO finding of obviousness where one isomer was substituted for another: the structural similarity alone suggested functional equivalence). This information about the different properties of the 2,4 and 4,2 isomers was critical to the patent examiner's conclusion to allow the nizatidine claim over the examiner's earlier conclusion of obviousness.

No one should be shocked, of course, if an opinion of invalidity is later undermined by discovery of unexpected results. That uncertainty is inherent in the enterprise, at least if the unexpected results are not part of the file history. But these unexpected and superior results were part of the file history, and Lerner David already knew of them. Lerner David's Arnold Krumholz, who signed the opinion letter, had also sent the analysis of the file history to Zenith. When asked to reconcile this evidence with the opinion letter's statement, Zenith's Banks was unable to do so. Banks Dep. at 308-09. No other evidence explains the discrepancy.

Zenith has tried to minimize this glaring error in the opinion letter by pointing out that the unexpected and superior results in Dr. Pioch's declaration dealt with an isomer of nizatidine and not with ranitidine or other "structurally-related compounds." The effort is not convincing.

The H2 receptor activity is unexpectedly and dramatically different between isomers that have merely interchanged the bonding positions of the two chains to the thiazole ring. Those results cannot be ignored when trying to argue that nizatidine was obvious based on compounds that differed from it even more substantially. Lerner David's review of the file history surely showed that those results were critical to the examiner's decision to issue the patent on nizatidine.

Moreover, it is not as if the opinion letter acknowledged those unexpected and superior results as compared to the 4,2 isomer and then made a reasoned argument as to why they did not undermine a conclusion of obviousness. The Lerner David opinion letter did not even mention this critical evidence weighing against its conclusion on the issue of obviousness. Nor have any witnesses offered any evidence to support the argument that Zenith has offered at trial.

The Lerner David opinion letter has another important omission. It is based entirely on structural similarity between nizatidine and other H2 receptor antagonists. The opinion fails to consider at all the level of H2 antagonist activity or the other properties that make a compound suitable or unsuitable as a medication for human beings. See Yamanouchi, 231 F.3d at 1347 (affirming finding of willfulness where ANDA notice of Paragraph IV certification was inadequate because it contained "no reference to famotidine's potency, safety, and lack of side effects, among other distinguishing properties accompanying its unusually high activity").

The court explained above in discussing Dr. Baldwin's analysis why mere structural similarity is not enough by itself to support the obviousness defense. The failure to address the real difficulties and unpredictabilities in this field is further evidence that the opinion was not one that Zenith could reasonably rely upon to support its earlier filing of the Paragraph IV certification. There is no evidence that the authors or readers of the opinion letter thought, let alone reasonably thought, that a hypothetical researcher in 1979 could have predicted from nizatidine's structure the key properties that made it a success. Even at trial, Zenith's expert witness testified that no such predictions could have been made from the structure. The Federal Circuit has written: "A good test that the advice given is genuine and not merely self-serving is whether the asserted defenses are backed up with viable proof during trial which raises substantial questions . . . ." Read Corp. v. Portec, Inc., 970 F.2d 816, 829 n. 9 (Fed. Cir. 1992). Zenith's advice fails this test.

Further, it appears that, apart from the Crenshaw patent that did not even surface at trial, the Lerner David opinion was not based on any prior art that was not before the patent examiner. The key compounds upon which the opinion and Zenith's trial presentation relied — cimetidine, tiotidine, and ranitidine — were all before the examiner and are discussed in the '547 patent itself. In one of its first decisions on advice of counsel, the Federal Circuit affirmed a finding of willfulness in the face of similar advice. Noting that 35 U.S.C. § 282 assigns a burden that is "most formidable when the party asserting invalidity relies upon prior art considered by" the PTO, the court concluded: "In short, the attorney's advice, based solely on file history prior art, does not by itself raise an inference of good faith substantial enough to convince us that the trial court's determination of willful infringement was clearly erroneous." Central Soya Co. v. George A. Hormel Co., 723 F.2d 1573, 1577 (Fed. Cir. 1983) (emphasis in original); accord, SRI International, 127 F.3d at 1466 (affirming finding of willfulness where counsel's opinion simply repeated arguments that had been rejected on re-examination).

The Lerner David opinion letter reflects yet another serious omission. As discussed above, the evidence on the objective indicia of non-obviousness weighs substantially against a finding of obviousness in this case. Apart from the erroneous claim that the authors were unaware of any evidence of unexpected results, the opinion included no discussion at all of the objective indicia of non-obviousness or the possibility that such evidence might rebut a prima facie case of obviousness.

In a case arising from a Paragraph IV certification under the Hatch-Waxman Act, certain objective indicia are likely to be present. As discussed above, commercial success and copying will probably be present. Evidence of long-felt need and the failure of others may not always be present in Hatch-Waxman Act cases, but the possibility of such evidence surely must be considered in any reasonable evaluation of obviousness.

To try to explain the failure to mention these objective indicia, Zenith points out that the authors and the recipients were all patent lawyers and did not need to be reminded about these well-known points. As applied to these fact-sensitive factors that directly affect the determination, the explanation is not persuasive. See, e.g., In re Hayes Microcomputer Products Patent Litigation, 982 F.2d at 1543 (affirming finding of willfulness despite counsel's opinion that claims were obvious; letter failed, among other points, to consider "secondary considerations in determining obviousness"); see also Johns Hopkins University v. CellPro, Inc., 152 F.3d 1342, 1364 (Fed. Cir. 1998) (affirming finding of willfulness; opinion letters mentioning inequitable conduct made no mention of necessary "intent to deceive," which is "often difficult to establish"); Sensonics, Inc. v. Aerosonic Corp., 81 F.3d 1566, 1571 (Fed. Cir. 1996) (counsel's opinion was flawed because it made "no mention of Aerosonic's copying and other objective indicia of unobviousness, although precedent requires that these factors be considered," but appellate court affirmed trial court finding of no willfulness based on totality of evidence and deferential review of factual finding). In addition, Zenith has not offered any evidence at all that any of the lawyers involved in the conference call or in writing or reading the opinion letter ever gave any actual consideration to those objective indicia of non-obviousness. Their failure to do so further undermines the ability of Zenith to rely on advice of counsel as a defense.

Lilly also points out that the Lerner David opinion letter failed to address the burden of proof on the issue of obviousness. The burden of proof requires a party challenging a patent's validity for obviousness to produce clear and convincing evidence to overcome the statutory presumption that a patent is valid. E.g., WMS Gaming, 184 F.3d at 1355. Zenith explains the silence on the issue of the burden of proof as having been unnecessary in an opinion letter from one group of patent attorneys to another. The burden of proof is well-known, simple, and general. The court finds Zenith's explanation is plausible as to that rule of black-letter law.

A further factor showing this is an exceptional case is the inadequacy of Zenith's notice under the Hatch-Waxman Act. Zenith was required to notify Lilly that Zenith had filed its ANDA asserting that the '547 patent is invalid. Such a notice "shall include a detailed statement of the factual and legal basis of the applicant's opinion that the patent is not valid . . . ." 21 U.S.C. § 355(j)(2)(B)(ii).

In Yamanouchi, one important factor supporting the Federal Circuit's decision to affirm a finding of willfulness was the failure of the defendant's ANDA notice to satisfy this requirement. The defendant's notice "does not present a prima facie case of invalidity, and makes no reference to famotidine's potency, safety, and lack of side effects, among other distinguishing properties accompanying its unusually high activity." 231 F.3d at 1347. The Federal Circuit even emphasized the testimony from the defendant's expert witness that was strikingly similar to Dr. Baldwin's testimony here: "Moreover, Dr. Loev admitted at trial that, as of 1992, he could not tell from [famotidine's] chemical structure whether it would be toxic nor predict its lack of side effects. He further testified that he could not predict the effects on potency that would be caused by the structural manipulations he claimed to be obvious." Id., quoting Yamanouchi, 21 F. Supp.2d at 376. The Federal Circuit then said: "When Danbury proceeded in the face of these weaknesses, its certification amounted to baseless and unjustified misconduct. In certifying invalidity, Danbury disregarded its duty to exercise due care." 231 F.3d at 1347.

In light of Dr. Baldwin's concession that he also could not predict nizatidine's level of activity, its lack of toxicity, or the lack of serious side effects from its structure, precisely the same reasoning applies here. Zenith's notice to Lilly of the ANDA Paragraph IV certification is as inadequate as Danbury's notice was in Yamanouchi. Zenith's notice to Lilly, which was drafted by Lerner David, cited five publications from prior art and stated its conclusion in two sentences: "The composition of matter, nizatidine, is highly structurally similar to compounds disclosed in Durant et al., Crenshaw et al. and to ranitidine, and any differences therebetween are suggested by the prior art. In addition, the properties resulting from both ranitidine and nizatidine are highly analogous." Ex. 1070. That's it. That does not come close to "a detailed statement of the factual and legal basis" of the assertion of invalidity.

In sum, the evidence here shows clearly and convincingly that Zenith's infringement of the '547 patent was willful. Zenith knew of Lilly's patent, of course, and it knew that its proposed product would infringe the '547 patent. That much will be present in any Paragraph IV cases under the Hatch-Waxman Act. Most important, the evidence shows clearly and convincingly that Zenith did not rely reasonably and in good faith on the advice of counsel. Zenith did not have a reasonable basis for believing the '547 patent was invalid.

Zenith rushed to file its Paragraph IV certification as quickly as possible on the basis of a hurried oral opinion that no participants can describe so as to give anyone confidence that sufficient care had been taken. The financial incentives to file immediately were great, as Zenith noted at the time. Those involved in the October 5th conference call knew that the opinion was highly likely to result in expensive litigation, but also that the potential rewards of success would be measured in many millions of dollars. Yet no one in that conference call testified to any content beyond the bare bottom line. No one in that conference call jotted down even a single note, despite the concern about the "exact nature" of the opinion and its unspecified limits. Dr. Mittleberg failed to follow Zenith's own lawyer's advice to talk directly with Lerner David's Krumholz about the validity issue.

Even when Lerner David took more than a month to clean up the oral opinion by preparing a written opinion, the opinion letter contained glaring errors and omissions that both the authors and the recipients should have recognized. See, e.g., Johns Hopkins University v. CellPro, 152 F.3d at 1364 (recipient of opinion letters was experienced patent attorney and should have recognized obvious shortcomings of counsel's opinions). Those errors and omissions had the cumulative effect of thoroughly undermining the reliability of the conclusion of obviousness.

The court recognizes that reasonable and good faith challenges to the validity of patents are an essential part of the patent law system:

A party who has obtained advice of competent counsel, or otherwise acquired a basis for a bona fide belief that a patent is invalid, can be said to serve the patent system in challenging that patent in a law suit conducted fairly, honestly, and in good faith. Such a party should not have increased damages or attorney fees imposed solely because a court subsequently holds that belief unfounded, particularly when the issues may be fairly described as "close."
Kloster Speedsteel AB v. Crucible, Inc., 793 F.2d 1565, 1581 (Fed. Cir. 1986). The Paragraph IV provision of the Hatch-Waxman Act also invites challenges to drug patents, so long as the challenges are reasonable and in good faith.

This case, however, does not present a close question on the issue of validity. Zenith was not relying reasonably and in good faith on advice of counsel when it decided to go forward with the act infringing the '547 patent. The evidence shows clearly and convincingly that Zenith and its lawyers were going through the motions of preparing an advice of counsel defense. "The evidence shows that the advice of counsel was more of a protective device than a genuine effort to determine before infringing whether the patent was invalid." In re Hayes Microcomputer Products Patent Litigation, 982 F.2d at 1544 (affirming finding of willful infringement despite advice of counsel that patent was invalid). This is an exceptional case for purposes of 35 U.S.C. § 285, and Lilly is entitled to recover its attorneys' fees.

Conclusion

For the reasons detailed in this entry, the court finds that U.S. Patent No. 4,375,547 is not invalid as obvious under 35 U.S.C. § 103(a). The court further finds that Zenith's act of infringement was willful, made without exercising due care, and that this case is an exceptional case under 35 U.S.C. § 285. The court postponed taking evidence on the amount of any fee award until it was determined whether a fee award would be appropriate. Unless the parties agree or the court later orders otherwise, Lilly shall submit a detailed written fee petition no later than November 30, 2001. (Plaintiff Reliant, which voluntarily joined in this suit earlier this year after signing a license agreement with Lilly, is not entitled to its own fee award.) Zenith shall submit any response no later than January 15, 2002. Lilly may file any reply no later than January 31, 2002. The court will schedule an evidentiary hearing or oral argument on written request, but will otherwise rule based on the written submissions.

So ordered.


Summaries of

ELI LILLY COMPANY v. ZENITH GOLDLINE PHARMACEUTICALS

United States District Court, S.D. Indiana, Indianapolis Division
Oct 12, 2001
Cause No. IP 99-38-C H/K (S.D. Ind. Oct. 12, 2001)
Case details for

ELI LILLY COMPANY v. ZENITH GOLDLINE PHARMACEUTICALS

Case Details

Full title:Eli Lilly And Company and Reliant Pharmaceuticals, LLC, Plaintiffs, v…

Court:United States District Court, S.D. Indiana, Indianapolis Division

Date published: Oct 12, 2001

Citations

Cause No. IP 99-38-C H/K (S.D. Ind. Oct. 12, 2001)