Eli Lilly and Co. v. SICOR PHARMACEUTICALS, INC. , 705 F. Supp. 2d 971 ( 2010 )

705 F. Supp. 2d 971 (2010)

ELI LILLY AND COMPANY, Plaintiff,
v.
SICOR PHARMACEUTICALS, INC., Teva Pharmaceuticals USA, Inc., Teva Parenteral Medicines, Inc., Defendants.

No. 1:06-cv-238-SEB-DML.

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

March 31, 2010.

Alissa Keely Lipton, Jessica Ruth Underwood, Krista Entrop Bianco, Laura P. Masurovsky, Robert D. Bajefsky, Robert Frederic Shaffer, Ryan James Cudnik, William Barrett Raich, Finnegan, Henderson, Farabow, Garrett, & Dunner, LLP, Washington, DC, Charles Edmund Lipsey, L. Scott Burwell, Finnegan Henderson Farabow Garrett & Dunner

LLP, Reston, VA, Jan M. Carroll, Barnes & Thornburg LLP, Indianapolis, IN, John D. Martin, Nelson Mullins Riley & Scarborough, LLP, Columbia, SC, Robert Francis McCauley, Finnegan Henderson Farabow Garrett & Dunner LLP, Palo Alto, CA, for Plaintiff.

David 0. Tittle, Kandi Kilkelly Hidde, Steven G. Cracraft, Bingham McHale LLP, Indianapolis, IN, Elizabeth J. Holland, Karen Shen, Merri C. Moken, Michael J. Freno, Peter L. Giunta, Sheila Mortazavi, Steven J. Lee, Kenyon & Kenyon LLP, New York, NY, for Defendants.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

SARAH EVANS BARKER, District Judge.

This matter is before the Court for decision following a trial on the issues of validity, enforceability, and infringement of two patents held by Plaintiff Eli Lilly and Company ("Lilly"): United States Patent Nos. 4,808,614 ("the '614 patent") and 5,464,826 ("the '826 patent"). The '614 patent covers the compound gemcitabine, and the '826 patent covers the method of treating susceptible neoplasms in mammals with gemcitabine. Lilly markets gemcitabine for injection under the tradename GEMZAR®, which is approved for the treatment of ovarian cancer, breast cancer, non-small cell lung cancer, and pancreatic cancer. (DTX 39.)^[1]

Lilly is the holder of approved New Drug Application No. 20-509 for the use of Gemzar® as a treatment for these cancers. In January 2006, Defendant Sicor Pharmaceuticals, Inc. ("Sicor") filed two Abbreviated New Drug Applications ("ANDAs") under the Drug Price Competition and Patent Term Restoration Act of 1984, 98 Stat. 1585 (popularly known as the Hatch-Waxman Act), with the Food and Drug Administration ("FDA") seeking approval to market generic versions of Lilly's gemcitabine product in two different dosage forms (200 mg and 1 g) prior to the expiration of the '614 and '826 patents. In August 2008, Defendant Teva Parenteral Medicines, Inc. ("Teva Parenteral")^[2] filed a third ANDA seeking approval to market generic versions of gemcitabine in a third dosage form (2 g). Sicor's (and Teva Parenteral's) ANDAs included "paragraph IV certifications," pursuant to 21 U.S.C. § 355(j)(2)(A)(vii)(IV), asserting that the claims of Lilly's '614 and '826 patents are invalid, unenforceable, or not infringed. As required by the Hatch-Waxman Act, Sicor and Teva Parenteral sent Lilly notice of these ANDA filings and paragraph IV certifications.

On February 15, 2006, after receiving notice of Sicor's ANDA filings and paragraph IV certification relating to the 200 mg and 1 g dosage forms, Lilly filed suit against Sicor and Teva USA alleging infringement of the '614 and '826 patents under 35 U.S.C. § 271(e)(2)(A) and 35 U.S.C. § 271(b). Within forty-five days of receiving Teva Parenteral's notice relating to the 2 g dosage form, on September 26, 2008, Lilly filed suit against Teva Parenteral and Teva USA alleging infringement of the '614 and '826 patents. On January 15, 2009, the Court issued an order consolidating the two previously separate actions.

With the present action, Lilly seeks an order: (1) requiring the FDA to change the effective date of ANDA Nos. 77-961 and 77-983 until after the expiration of the '614 and '826 patents; (2) prohibiting the FDA from approving Teva Parenteral's ANDA No. 090644 prior to the expiration of the patents, in accordance with 35 U.S.C. § 271(e)(4)(A); and (3) enjoining Defendants from the commercial manufacture, use, offer to sell, sale, or importation of their gemcitabine products prior to the expiration of Lilly's exclusivity, in accordance with 35 U.S.C. § 271(e)(4)(B).

Defendants have stipulated that if Claim 12 of the '614 patent and Claim 7 of the '826 patent are valid and enforceable, then Defendants' actions constitute infringement of those patents under 35 U.S.C. § 271(e)(2). (PTX 870, PTX 871).

Defendants challenge the validity of the '614 patent on the following grounds: anticipation under 35 U.S.C. § 102 and obviousness under 35 U.S.C. § 103. Defendants challenge the validity of the '826 patent on the following grounds: obviousness-type double patenting, anticipation under 35 U.S.C. § 102, obviousness under 35 U.S.C. § 103, and lack of enablement under 35 U.S.C. § 112. Defendants further assert that Lilly is collaterally estopped from asserting the '826 patent against them as a result of a prior determination of invalidity in a related case brought by another pharmaceutical company against Lilly in another district court.^[3]

On the final day of trial, a post-trial briefing schedule was set for submission of the parties' proposed findings of fact and conclusions of law, post-trial briefs, and responses. These submissions were due on or before November 30, 2009. [Docket No. 339]. Having now considered the evidence adduced at trial as well as the parties' post-trial submissions, we hold, as detailed below, that: (1) the '614 patent is valid and enforceable and Defendants' proposed commercial gemcitabine product infringes claim 12 of that patent; and (2) although the '826 patent is valid and enforceable in all other regards, Lilly is collaterally estopped from defending against Defendants' claim of obviousness-type double patenting relating to that patent, pending a final determination on appeal.

Findings of Fact^[4]

The Parties

Plaintiff Eli Lilly and Company ("Lilly") is an Indiana corporation with its principal place of business in Indianapolis, Indiana. Lilly is engaged in the business of research, development, manufacture, and sale of pharmaceutical products throughout the world.

Defendants Sicor Pharmaceuticals, Inc. ("Sicor"), now known as Teva Parenteral Medicines, Inc. ("Teva Parenteral"), and Teva Pharmaceuticals USA, Inc. ("Teva USA") (collectively "Defendants") are Delaware corporations. Defendants are engaged in the business of selling generic pharmaceutical products, which they distribute in Indiana and throughout the United States.^[5]

The Patents-in-Suit

The '614 patent, issued on February 28, 1989 to Lilly, names Dr. Larry W. Hertel as the sole inventor. The '826 patent, issued on November 7, 1995 to Lilly, names Dr. Larry W. Hertel and Dr. Gerald B. Grindey as joint inventors. Both patents are owned by Lilly.

The '614 patent discloses and claims a series of 2'2'-difluoronucleoside analogs, including gemcitabine. (DTX 1; D.I. 261 ¶ 23.) Claim 12, the only claim being asserted from the '614 patent, specifically claims the nucleoside analog gemcitabine. (Radtke 432:13-20.) The chemical name of gemcitabine appears as the eighth compound in a list of thirty-six potential 2'2'-difluoronucleosides. (DTX 1, col. 3, II. 55-56.) The '614 patent claims priority to U.S. Patent Application Ser. No. 473,883 ("the '883 Application"), filed on March 10, 1983.

The specification of the '614 patent details the antiviral utility of this class of nucleoside analogs. It also contains one paragraph describing the later-discovered anticancer activity of this class of nucleosides and providing specific dosages. (PTX 1 at col. 17, II. 53-68.) The '883 Application, which was the original application in this patent, does not disclose any anticancer activity in the difluoronucleoside analogs. (Radtke 475:17-19.) The anticancer disclosure of the '614 patent was added by Lilly in a continuation-inpart application ("CIP"), Ser. No. 677,146, filed on December 4, 1984 ("1984 CIP Application"). (PTX 7 at 42; D.I. 261 ¶ 30.)

Also on December 4, 1984, the same day of the filing of the 1984 CIP Application related to the '614 patent, Lilly filed a completely separate patent application, which eventually issued as the '826 patent. The '826 patent claims the benefit, under 35 U.S.C. § 120, of a series of applications, the earliest having been filed on December 4, 1984, U.S. Patent Application Ser. No. 677,783 ("the '783 Application"). (PTX 9 at 3-46.)

The '826 patent discloses and claims methods of treating susceptible neoplasms with gem-difluoronucleosides, including gemcitabine. Claim 7 of the '826 patent, the only claim of that patent being asserted here, claims a method of treating susceptible neoplasms in mammals, where the susceptible neoplasm is selected from the group consisting of leukemias, sarcomas, carcinomas, and myelomas, and comprising the administration of a therapeutically effective amount of gemcitabine as treatment for human patients. (PTX 2 at col. 23, 1.42-col. 24, 1.64; Green 220:25-221:6.)

Definition of Person of Ordinary Skill in the Art for the '614 Patent

At trial, Defendants' expert, Dr. Piet Herdewijn, and Lilly's expert, Katherine Radtke, testified that the person of ordinary skill in the art to which the '614 patent pertains would have a Ph.D. or equivalent degree in chemistry or the pharmaceutical sciences, with additional and considerable experience designing and synthesizing nucleoside analogs for biological purposes. (Radtke 432:25-433:15; Herdewijn 980:19-982:2.) Drs. Kathlyn Parker and Larry Overman held a similar view but did not include in their definition a particular expertise in nucleoside chemistry. (Overman 599:6-17; Parker 1137:21-1138:10.) The definition of a person of ordinary skill in the art does not appear to be a matter of major dispute between the parties. We find that the definition must recognize the particular complexity of the science involved in the invention of the asserted claim, as well as the formative nature of the field of nucleoside synthesis at the time of the issuance of the patent.

Accordingly, based on the testimony of these experts, we find that a person of ordinary skill in the art for the '614 patent would have a Ph.D. in chemistry, postdoctoral training in nucleoside chemistry, and experience in the design and synthesis of nucleoside analogs for biological purposes.

Definition of a Person of Ordinary Skill in the Art for the '826 Patent

The invention of the '826 patent relied upon clinical trial experience, and Claim 7 of the '826 patent is directed to treating cancers, which are typically treated by clinical oncologists. As Defendants' expert, Dr. Kenneth Tew explained, in 1984, clinical oncologists administered cancer drugs to patients and were clinical trial investigators. (Tew 1273:9-19, 1277:17-20.) Because of the role of such experimentation in the development of the asserted claim, contrary to the opinion asserted by Defendants' experts that clinical oncologist expertise would not be part of the "ordinary" knowledge in the art, the definition of a person of ordinary skill in the art must consider the role of designing, implementing, and conducting clinical trials necessary for a complete understanding of the skill required in this field. (Green 222:2-16.) Furthermore, Lilly's expert, Dr. Mark Green, testified that the '826 patent, which is a method of use patent, does not require medicinal chemistry.^[6] (Green 222:17-21.)

Accordingly, we hold that, for the '826 patent, a person of ordinary skill in the art would have a medical degree (M.D.) or a Ph.D. degree in biology, pharmacology, or a similar degree in a related field of science, and would have experience in developing chemotherapeutic agents for the treatment of cancer. That person would have knowledge encompassing the design, implementation, and execution of clinical trials. (Green 221:21-222:10.)

Gemcitabine and Nucleoside Analogs Generally

The Food and Drug Administration ("FDA") has approved gemcitabine, sold by Lilly as Gemzar®, for use in treating patients with neoplasms, including pancreatic, lung, ovarian, and breast cancers. (PTX 443 at 10-11.) Gemcitabine has been used to treat approximately 1.5 million patients worldwide since its first approval in 1996. (D.I. 287, Grabowski Rep. ¶ 45.) Currently, gemcitabine is the primary therapy given to patients with advanced pancreatic cancer, as experimental data indicates that it offers improved oneyear survival nine times better than the previous standard of care. (Green 200:11-201:4; PTX 436 at 2403.)

Gemcitabine is a "nucleoside analog," or a modified version of a class of compounds called nucleosides, which are the building blocks of DNA. (Shewach 770:3-14; LDX 8.24.) Nucleosides consist of a 5-member "sugar" ring joined to a "base." (Radtke 426:14-16; LDX 7.2.) By convention, the atoms at each of the corners of the base are numbered 1 through 6, and the carbon atoms at each corner of the sugar are numbered 1' ("1 prime") through 5'. Gemcitabine differs structurally from the natural nucleoside 2'-deoxycytidine in that it has twin fluroine atoms at the C-2' position of the sugar (hence, the "gem" in its name, which stems from "geminal," for twin). (Radtke 433:23-434:4.)

Nucleoside analogs can be used as anticancer or antiviral agents because they are similar enough in structure to naturally occurring nucleosides that cells are tricked into accepting them, but are different enough to disrupt cell functioning and replication once inside the cell. (Rustum 1537:6-1538:6.) Because of these attributes, since at least the 1960s, chemists specializing in nucleosides have explored the implementation of structural changes to naturally occurring nucleosides in an effort to produce nucleoside analogs with therapeutic properties. (Radtke 441:15-24; Shewach 792:19-793:2; PTX-602.) This process and the results it engenders have been regarded as highly unpredictable; as one leading researcher stated in a 1973 review article, "[O]nly Zeus sitting on top of Mount Olympus could predict which modifications would produce clinically effective antiviral agents. Unfortunately, we poor mortals are not privy to such knowledge." (Radtke 430:15-431:15; PTX 647 at 881; see also PTX 692 at 3-5).

Even today, scientists in the field regard nucleoside analog design for therapeutic anticancer purposes as very unpredictable. (Radtke 431:12-432:5.) This unpredictability stems from the profound change in biological properties that can be created by minor changes in the structure of a nucleoside compound. Slight structural modification can drastically alter the biological activity of nucleoside analogs. This is the result of the effect structural modification typically has on the three-dimensional shape of the compound, which alters the analog's ability to interact with biologically significant enzymes. (Radtke 427:15-430:14.) An example given in this litigation was that the building blocks of RNA and DNA, two compounds that are processed by the human body in remarkably different ways, differ physically in only one respect: that the RNA nucleosides have a hydroxyl group (OH) at the C-2' position, whereas DNA nucleosides do not. (Radtke 427:15-428:5; LDX 7.3.)

The key modification in gemcitabine involves the element fluorine. In the early 1980s, fluorine chemistry as applied to the nucleoside field was relatively new, and little was known about how one might synthesize difluorinated nucleosides. (Overman 596:7-601:17; 615:13-616:25.) Indeed, years after the invention of gemcitabine, Dr. Herdewijn wrote that "the synthesis of fluorinated nucleosides is still a difficult task." (PTX 688 at 67.) Furthermore, the properties of fluorine, which is the most electronegative atom of all known elements,^[7] are known to lead to particularly unpredictable results, with corresponding unpredictable effects on biological activity. (Overman 615:13-616:1; Radtke 434:12-20.) Because of this electronegative potency, a geminal fluorine substitution, like the one in gemcitabine, was a particularly unpredictable addition to nucleoside analogs. (Bergstrom 935:15-936:22; PTX 533 at Bergstrom 21; see also Rustum 1605:14-22.)^[8] This unpredictability remained even after Lilly's development of gemcitabine; as Defendants' expert, Dr. Herdewijn wrote in a 1989 review article, "[I]t seems very difficult to predict [a] molecule's biological behaviour when a C-H or a C-OH has been displaced by a C-F group." (PTX 688 at 66; Herdewijn 1097:2-13.)

The unpredictability in this area of chemistry is, in large part, the reason that gemcitabine remains the primary treatment for particularly intractable neoplasms. Prior to Lilly's synthesis of this compound, no other research team had synthesized a nucleoside analog with geminal fluorine atoms. (Radtke 473:12-474:5.). Furthermore, between that synthesis and the year 2000, roughly forty similar nucleoside analogs (with geminal fluorine atoms at the 2'-position) had been reported, yet none had proved effective for treating cancer in humans. (Radtke 528:14-529:6; PTX 620 at 87-88.) More broadly, out of nearly 300,000 compounds that had been evaluated as anticancer agents by 1979, shortly before Lilly's synthesis of gemcitabine, only forty were deemed useful in treating neoplasms. (Shewach 744:8-745:8; PTX 692 at 3.)^[9]

Nucleoside Analogs in the Prior Art

Defendants point to three compounds they contend were contained in the prior art and were significant to the development of gemcitabine: Ara-C, which has an OH group "up" at the 2' position; 2'-Fara-C, which has one fluorine "down" at the 2' position; and 2'-F-cytidine, which has one fluorine "down" at the 2' position. These three, according to Defendants, presaged Lilly's discovery.

In the 1970s, the first of these compounds, Ara-C, was a widely used anticancer drug. (Bobek 1300:18-13-1:3.) Although it was relatively effective, the compound was rapidly deactivated in the body by a process called "deamination." (Bobek 1301:8-16.) Because of problems stemming from this lack of effectiveness, researchers at the Fox Laboratory at Memorial Sloan-Kettering Cancer Center ("Sloan-Kettering" or "Fox Laboratory") and elsewhere had undertaken efforts to improve on Ara-C.

The two fluorinated nucleoside analogs, 2'-F-ara-C and 2'-F-cytidine, had been synthesized in the 1960s by Sloan-Kettering. (Radtke 441:15-442:17; Shewach 750:20-751:6.) But neither compound was reported in the prior art to have been advanced to testing in animals, as compounds with promise for medical activity typically are. (Shewach 752:25-753:10.) Nor was either considered a lead compound for further exploration in the late 1970s and early 1980s. (Radtke 440:22-441:6.)

2'-F-cytidine was first reported in 1967 and found to be twenty times less cytotoxic in vitro than Ara-C. (Radtke 441:15-24; PTX 602.) The fact that 2'-F-cytidine did not effectively inhibit DNA synthesis in intact cells indicated that it also did not significantly inhibit the enzyme ribonucleotide reductase ("RNR")^[10]in vivo. Moreover, although this nucleoside analog was discovered in 1967, Fox Laboratory never tested it for biological activity in animals or in the clinic, nor did any other laboratory so test it. (Shewach 752:25-753:10; Radtke 448:2-20.)

In 1970, researchers at Sloan-Kettering reported that they had synthesized 2'-F-ara-C. (Radtke 448:21-449:14; PTX 603.) However, similar to 2'-F-cytidine, this analog posed no improvement on the in vitro cytotoxicity of Ara-C, and the prior art reported no in vivo activity for 2'-F-ara-C. (Radtke 454:13-18.)^[11]

In contrast to these two compounds, which the prior art largely deemed of low antiviral and anticancer activity, two nucleoside analogs referred to as FIAC and FMAU were identified by researchers as the "most potent and selective" of the fluorinated Sloan-Kettering analogs. (Radtke 454:19-456:9; PTX 584 at 30.)^[12] Dr. Herdewijn, Defendants' expert, noted in a 1989 review that FIAC and FMAU were "important" nucleoside analogs for further study, omitting any mention of 2'-F-cytidine or 2'-F-ara-C. That the FIAC and FMAU compounds were more prominent in the prior art than 2'-F-cytidine and 2'-F-ara-C is further supported by a presentation prepared by Dr. Donald Bergstrom in 1980, which mentioned neither 2'-F-ara-C and 2'-F-cytidine but did describe FIAC. (PTX-541 at Bergstrom 3257.)

In addition to providing information regarding the properties of these compounds, the research referenced in the prior art indicated to those skilled in the art that a nucleoside analog's in vitro cytotoxicity does not necessarily correlate with in vivo anticancer activity. (Shewach 795:23-797:7; PTX 725 at 1163; Rustum 1586:13-15.) Thus, any in vitro data reported for 2'-F-ara-C and 2'-F-cytidine was not necessarily indicative of how those compounds would act in vivo. (Rustum 1583:25-1584:4.) As the prior art revealed with considerable clarity, because viruses produce different enzymes than mammalian cancer cells, (Herdewijn 1105:6-15; Shewach 857:14-21; PTX 937 at 486), not all molecules exhibiting antiviral activity also have anticancer activity.

Dr. Miroslav Bobek's Attempts to Synthesize Gemcitabine

Dr. Miroslav Bobek, now a retired medicinal chemist, spent the majority of his career developing modified nucleosides for anticancer purposes at Roswell Park in Buffalo, New York. (Bobek 1299:1-7.) In connection with this work, in June 1977, Dr. Bobek submitted a manuscript to Tetrahedron Letters setting forth proposed synthetic routes for making gemdifluorosaccharides, including a method used in his laboratory for making an intermediate of gemcitabine. (DTX 95; Bobek 1322:6-1323:8, DTX 873 at SICOR/TEVA-Bobek 472.) In March 1978, Dr. Bobek carried out a series of chemical reactions with the intention of creating gemcitabine, but he never succeeded in synthesizing the compound. (Bobek 1324:9-1330:10.)

Dr. Bobek reports that he publicly presented his work on gemcitabine in the 1970s during at least one seminar held at Roswell Park. (Bobek 1317:8-1319:25; DTX 220 at SICOR/TEVA-Bobek 133.) During that time period, Dr. Gerald Grindey, a named inventor on the '826 patent, was a colleague in the same department as Dr. Bobek. (DTX 78.) Dr. Grindey left his position with Roswell Park in December 1980 to join Lilly.^[13]

Dr. Bobek's laboratory notebook contains a series of attempted reactions leading to an intermediate of gemcitabine, but none of them resulted, even hypothetically, in gemcitabine itself. (Bobek 1324:9-23; 1325:11-19; PTX 508 at 61.) His first relevant experiment, in which he attempted to make a difluoronucleoside using DAST,^[14] failed. (Bobek 1358:7-9, 1359:8-9; Radtke 459:12-436:3; PTX 492.) Similarly, an alternative synthesis which started with a 6-member ring sugar that was reported in his grant application was reported to have failed. (PTX 499 at Bobek 202; PTX 500 at Bobek 154.)

Although Dr. Bobek claimed at trial that his experiments and proposed reactions were capable of producing gemcitabine, no other evidence in the record supports this claim. None of Defendants' experts testified that Dr. Bobek actually synthesized gemcitabine, or even that his synthetic route was capable of doing so. Moreover, experimental evidence demonstrated clearly that the synthetic route purportedly used by Dr. Bobek could not have produced gemcitabine. (Radtke 485:11-25; Overman 623:13-625:5; PTX 509; PTX 516.) Eventually, Dr. Bobek abandoned his attempts to synthesize the compound and focused instead on the pursuit of other nucleoside analogs. (Bobek 1330:20-25; 1394:1-6.)

Dr. Donald Bergstrom's Attempts to Synthesize Gemcitabine

Dr. Donald Bergstrom, who was, prior to March 1983, a professor at the University of North Dakota and is now a professor at Purdue University, also had proposed the synthesis of difluoronucleosides. (DTX 265; DTX 206 at SICOR/TEVA-Bergstrom 22.) However, like Dr. Bobek, Dr. Bergstrom also never successfully produced gemcitabine. (Bergstrom 920:18-20.)

Dr. Bergstrom utilized several routes in attempting to synthesize 2',2'-difluoro-2'deoxynucleosides. (Bergstrom 946:23-948:5; PTX 534.) He set forth his rationale for conducting these experiments in grant applications that he submitted to the National Institute of Health ("NIH") in May and October of 1982. (DTX 206 at SICOR/TEVA-Bergstrom 4, 22; DTX 207 at SICOR/TEVA-Bergstrom 47. 63.) His May 1982 grant application stated that "of particular interest to us is the 2'-deoxy-2',2'-difluoro substitution pattern." (DTX 206 at SICOR/TEVA-Bergstrom22.)

Based on this work, Dr. Bergstrom submitted an abstract entitled "Synthesis of Gem-Difluoro Substituted Nucleoside Analogs" (the "Bergstrom Abstract") to the American Chemical Society before a March 1983 meeting at which he was scheduled to make a presentation ("1983 ACS Meeting"). (DTX 265; DTX 264 at 41.) The Bergstrom Abstract, a copy of which Dr. Hertel received on or about March 1, 1983, (PTX 6 at 52), reported that Dr. Bergstrom's laboratory was attempting to synthesize 2' and 3' gemdifluoronucleosides, which Dr. Bergstrom anticipated would include gemcitabine. (DTX 265.)^[15] Dr. Bergstrom publicly presented his research attempts at the 1983 ACS Meeting, and the slides he used in that presentation depicted the structure of gemcitabine. (Bergstrom 914:6-9, 915:1-917:3; DTX 455.)

Although Dr. Bergstrom made a 3',3'-gemdifluoronucleoside, he never successfully prepared the compound gemcitabine. (Radtke 466:13-467:15; PTX 518; Bergstrom 954:5-14.) Furthermore, while Dr. Bergstrom proposed several different de novo synthetic routes, he did not utilize a Reformatsky reaction, the route utilized by Dr. Hertel to synthesize gemcitabine. Because Dr. Bergstrom was never able to synthesize either a 2',2'-difluoronucleoside or the precursor of gemcitabine, he eventually abandoned his efforts and concentrated instead on other nucleoside analogs. (Bergstrom 923:11-19, 921:25-923:7; PTX 534 at Bergstrom 68-69.)

Merrell Dow's Synthesis of Gemcitabine

Dr. Robert Farr and Dr. Brian Metcalf at Merrell Dow Pharmaceuticals, Inc. also sought to synthesize gem-difluoronucleosides. (DTX 851, Metcalf Dep. Tr. 11:16-12:6, 16:14-18; Radtke 512:20-513:3.) Dr. Metcalf's involvement centered largely on envisioning the use of a nucleoside having a 2' gem-difluoro sugar coupled to a natural base to inhibit RNR, which he hoped would prove to have an anticancer utility. (DTX 151-153; DTX 851, Metcalf Dep. Tr. 33:11-14, 20-21, 23-4, 34:2-10; Farr Dep. Tr. 25:6-9, 11-13.)^[16] Dr. Farr, who worked under Dr. Metcalf's supervision until Dr. Metcalf's departure from Merrell Dow in December 1983, proposed a synthetic route that included a Reformatsky reaction and a Sharpless epoxidation to produce the compound. However, this route also proved unsuccessful. (PTX 552 at SA 2212-13; PTX 554 at SA 2200; Radtke 514:17-515:25, 516:3-5.)

Dr. Farr next attempted an approach derived from an article published in 1984, which also proved unsuccessful. (PTX 557 at SA 2182; PTX 566; PTX 6; Farr Dep. 47:18-51:17.) Dr. Farr did not, in fact, succeed in producing gemcitabine until, at the earliest, December 1984, and he did not test it for anticancer utility until May 1985, after a review of Dr. Hertel's synthetic method reported in Lilly's 1984 published British patent application. (DTX 162 at SA 2861, 2865; PTX 16; PTX 557 at SA 2185; PTX 568, 17:47-51, 18:49-30:39; PTX 16, 4:43-9:47; Farr. Dep. 59:12-17, 59:20-60:2, 91:19-93:25.) Merrell Dow filed a patent application on the use of gemcitabine as an antitumor agent in July 1985. (DTX 895.)

Lilly's Invention of Gemcitabine and Discovery of its Utilities

I. The '614 Patent: Difluoronucleosides and Antiviral Utility

Lilly's interest in developing nucleoside analogs as antiviral agents manifested itself in the early 1980s, when Lilly sought to obtain rights from Sloan-Kettering to develop FIAC and FMAU for antiviral utility. (Hertel 270:16-25, 271:23-272:15; PTX 22 at GM 197972.) Lilly was, however, unsuccessful in obtaining a license to these compounds. (Pearce 146:21-24, 147:13-14.) Thereafter, Dr. Larry Hertel, an employee at Lilly, became responsible for searching out ways to synthesize monofluorinated nucleoside analogs like FIAC and FMAU, in order to provide quantities of the compounds for further testing. (Hertel 271:1-13.).

Dr. Hertel proposed a new strategy called "de novo" or "total" synthesis as a method of producing the compounds, despite the fact that this method had been dismissed by other researchers as a "method of limited usefulness." (Hertel 283:13-19, 352:25-353:24; PTX 19 at GM 153104; PTX 22 at GM XXXXXX-XX; PTX 648 at 237.) At this time, he also proposed the preparation of a "key intermediate" he believed could be used to make geminal difluoronucleosides with two fluorine atoms at the C-2' position, hoping that these new compounds would produce antiviral utility. (Hertel 273:13-275:21, PTX 19 at GM 153127; PTX 22 at GM XXXXXX-XX.)

Dr. Hertel's synthesis of geminal difluoronucleosides was not immediately successful, as many of his attempted synthetic methods failed and he encountered unexpected impediments during his research. (Hertel 285:25-287:10; PTX 24 at GM XXXXXX-XX.) Included in his failed attempts were an intensive effort to synthesize the compounds using a "Sharpless epoxidation." (Hertel 282:XX-XXX-XX. 287:18-288:8). Overall, Dr. Hertel spent four months (May 1981 to September 1981) attempting to prepare the necessary Sharpless intermediate. Dr. Hertel tried at least four categories of reaction to accomplish this goal, and he contacted Dr. Sharpless, a Nobel Prize-winning synthetic organic chemist, and Dr. Evans, another prominent synthetic organic chemist, for advice. (Hertel 284:12-14; 287:18-20.) Both of those prominent chemists believed the route Dr. Hertel proposed would be successful. (Hertel 283:4-284:11, 287:11-289:14; PTX-19 at GM 153104.) Nonetheless, all of these attempts failed. (Hertel 285:25-292:13, 302:24-303:8.) During the time period from July to December 1981, Dr. Hertel shifted his focus by attempting a more direct synthetic strategy using the DAST reagent. (PTX 24 at 199; Hertel 301:15-302:23.) Like the previous strategy, this attempt also failed. (Hertel 302:9-10.)

In the face of these setbacks, Dr. Hertel proposed a different de novo approach using a "Reformatsky" reaction. Applying this method, in September 1981, he succeeded in synthesizing the carbon backbone of the 2' gem-difluoro sugar, which was a precursor of gemcitabine. (Hertel 292:14-294:9, DTX-126 at 57; PTX 24 at 193.)

According to Dr. Hertel, despite this success, the process thereafter was quite rigorous and complex, consistent with what was known in the prior art about the unpredictability of fluorination:

And what we discovered very rapidly is once you introduce fluorine into an organic molecule, it significantly alters the chemistry of that molecule by deactivating some centers or activating others to reactions in that in a normal organic molecule wouldn't be involved. And so essentially, you had to kind of learn organic chemistry all over again when you started working with a fluorinated organic.

(Hertel 297:13-20.) Dr. Hertel's laboratory initially attempted to couple the sugar intermediate he had produced with the base thymine, rather than cytosine, the base in gemcitabine. (DTX 129, 130, 421; PTX 19, 24, 26, 31.)

In June 1982, after many more attempts, the laboratory successfully synthesized the first difluoronucleoside in the series, 2',2'-difluorothymidine. (Hertel 311:17-313:4; PTX 26 at 162, 164.) Gemcitabine was the second nucleoside analog made by Dr. Hertel's laboratory. (PTX 31 at GM 161073, 161075; Hertel 327:17-329:7.) Lilly later tested Dr. Hertel's difluoronucleosides, and, by June 15, 1982, the first 2',2'-difluoronucleoside had exhibited promising activity against viruses. (Hertel 316:1-319:7; PTX 26 at GM 162163; PTX 228.) Dr. Hertel's difluoronucleosides demonstrated antiviral activity against both DNA and RNA viruses. (Hertel 316:1-319:7, 330:8-24; PTX 181 at GM 152170; PTX 228; PTX 229; PTX 242.)

On March 10, 1983, Lilly filed its original patent application related to Dr. Hertel's difluoronucleosides, the '883 application, which listed these compounds, including the first documented reference to gemcitabine. The '883 application also described the method of manufacturing the compounds and explained their antiviral utility. (PTX 6 at 9.) Claim 12 of the patent that issued from this original application, the '614 patent, covers the nucleoside gemcitabine. (Radtke 423:13-20.)^[17]

The '614 patent is entitled "Difluoro Antivirals and Intermediate Therefor." (PTX 1.) Consistent with its title, the patent is directed primarily to using the listed difluoronucleosides as antiviral agents. See id. However, the patent also contains a paragraph describing the later-discovered anticancer activity of these compounds. This paragraph was added on the same day that Lilly filed the original patent application ultimately leading to the '826 patent, the second patent-in-suit. Specifically, on December 4, 1984, Lilly filed the 1984 CIP Application, a continuation-inpart the '883 application that ultimately led to the issuance of the '614 patent. (PTX 7.) In accordance with the "best mode" requirement of 35 U.S.C. § 112, Lilly included in the 1984 CIP Application one paragraph discussing the later-discovered anticancer activity of the difluoronucleosides. That later discovery as contained in a separate patent application and discussed immediately below led to the issuance of the '826 patent.

II. The '826 Patent: Gemcitabine's Anticancer Utility

Soon after he had successfully synthesized difluoronucleosides and had them tested for antiviral activity, Dr. Hertel submitted several of the compounds to Dr. Gerald Grindey, a Senior Research Scientist at Lilly, who was seeking compounds for testing in anticancer screens. (Hertel 330:25-331:20, 332:25-333:10; PTX 31 at 195; Pearce 158:9-15.)

Although few of the compounds demonstrated anticancer utility in the tests conducted by Dr. Grindey, gemcitabine exhibited significant cytotoxicity, orders of magnitude better than the other difluoronucleosides. (PTX 173 at GM 154671; PTX 181 at GM 152170.) Dr. Grindey described it as "singular among all difluoronucleosides tested to date." (Pearce 100:9-19; PTX 179 at GM 258388.) Dr. Grindey's laboratory first screened gemcitabine for cytotoxicity in human cancer cells on November 1, 1983. (PTX 118 at GM 157170; Hertel 334:1-21.) The compound was first tested in vivo on leukemia beginning on February 22, 1984. (Pearce 90:16-91:9; PTX 126.) Gemcitabine showed a very significant improvement in the life span of mice, used in the in vivo testing, over a very broad range of doses in a test initiated on May 9, 1984 and completed on June 6, 1984. (PTX 173 at GM 154671; Pearce 89:20-90:8, 91:22-92:5; PTX 131.)

By August 13, 1984, Dr. Grindey's laboratory had collected and reviewed nearly all of the antitumor data for gemcitabine that appears in the '826 patent. (PTX 2; PTX 131; PTX 132; PTX 129; PTX 136; PTX 128; PTX 134; Pearce 90:16-92:5, 93:20-95:3, 104:20-105:6.) Ultimately, Dr. Grindey's in vivo testing of gemcitabine's antitumor activity demonstrated the compound's activity against a wide array of solid tumors, a "very unusual feature" for a nucleoside analog and the "broadest spectrum of any agent ever tested at Lilly." (PTX 173 at GM 154672; PTX 179 at GM 258390; Pearce 101:15-102:6.)

On December 4, 1984, based on Dr. Grindey's findings, Lilly filed a patent application on their discovery of the anticancer utility of gemcitabine, naming Drs. Grindey and Hertel as inventors. This filing came more than twenty months after the filing of '883 application, which was the original application of the '614 patent. The '826 patent, which resulted from the '783 application, was entitled "Method of Treating Tumors in Mammals with 2',2'-Difluoronucleosides" and contained in vitro and in vivo experimental data supporting the method of using various difluoronucleosides, including gemcitabine, to treat susceptible neoplasms in mammals. (PTX 2.) Claim 7 of the '826 patent is limited to using gemcitabine to treat different types of cancer. (PTX 2; Green 220:25-221:6.)^[18]

The goal of the '826 patent is to treat cancer in humans. (Tew 1273:9-12.) One of ordinary skill in the art reviewing the patent and its prosecution history, along with the relevant art in 1984, would know that the claimed invention dealt with the treatment of cancer in human patients, not all mammals.^[19] The '826 patent provides extensive information as to how to make and use the claimed invention. (PTX 2.) It describes the synthesis of gemcitabine and demonstrates the compound's cytotoxic activity in vitro and in vivo. Using standard mouse tumor model data, the patent discloses a broad range of activity, which would have led a person of ordinary skill in the art using valid methods and appropriate models to expect gemcitabine to have broad efficacy in humans against a variety of cancers. (Green 223:11-16.) Further, the patent discloses that physicians will be responsible for determining therapeutic doses of gemcitabine in humans. (PTX 2 at col. 16, ll. 2-8; Green 220:6-22.)

In order to determine the types of cancers for which gemcitabine would demonstrate efficacy in humans, one skilled in the art would have understood that the compound should be tested in the clinic in phase I and/or phase II studies, which were routine to a person of ordinary skill in 1984. (Green 223:17-25, 258:25-259:7; Tew 1254:1-4.) As Defendants' expert, Dr. Tew, explained, phase I, II, and III clinical trials are "what pharmacologists and clinical oncologists do." (Tew 1286:6-11.) Undertaking this experimentation would have been relatively straightforward to a person of ordinary skill in the art because the approach to designing clinical trials was known and well documented. (Tew 1284:9-25, 1285:1-17.)

III. The Unexpected Properties of Gemcitabine

Gemcitabine exhibited a number of properties that would have been unexpected to the person of ordinary skill in the art, including a unique mechanism of action, broad activity in treating susceptible neoplasms, and increased survival and higher clinical benefit response in the treatment of pancreatic cancer when compared to 5-fluorouracil ("5-FU").^[20]

A number of gemcitabine's mechanistic properties would have been unexpected, such as greater transport into cells and higher accumulation and retention inside cells, both of which provide an improved capacity to kill cancer cells over Ara-C. (Shewach 754:4-755:14, 759:12-763:18, 764:2-6; PTX 695 at 4028, PTX 696 at 100; LDX 8.7-8.8, 8.13-8.16.) Gemcitabine also proved to have superior effects on DNA synthesis, increased incorporation into DNA, and remarkable ability to inactivate the enzyme RNR. (Shewach 754:4-755:14, 763:19-768:21, 769:15-781:11; PTX 712 at 6110, 6116; PTX 705 at 567, 571; PTX 703 at 3220; LDX 8.21-8.29.)

As a result of these properties, gemcitabine exhibited unprecedented activity against a broad spectrum of cancers. (Shewach 754:21-23.) Gemcitabine's activity against tumors, including solid tumors, was unexpected in view of and in comparison to the compounds in the prior art. (Pearce 95:22-96:14; Green 178:16-22, 217:5-19.) Furthermore, at least one study^[21] concluded that patients with advanced pancreatic cancer treated with gemcitabine alone demonstrated a statistically significant, unexpected survival advantage and higher clinical benefit response when compared to those treated with 5-FU. (Green 178:5-15, 200:11-201:4; PTX 436 at 2403, 2406, 2409, 2412; LDX 5.12-5.13.)^[22]

IV. Gemcitabine's Impact on the Market

For decades prior to the development of gemcitabine, researchers experienced significant difficulty developing effective treatments for solid tumors, despite success in treating other forms of cancer. (PTX 460 at 12; PTX 461 at IV-50; Rustum 1581:15-21; Pearce 81:15-82:1.)^[23] During this period, patients suffering from advanced pancreatic cancer consistently faced a particularly negative prognosis because no chemotherapy was available that was effective as a treatment. (Green 183:19-184:5; PTX 452 at 2061.) Prior to gemcitabine, only two drugs had been approved by the FDA for use in patients with this disease: 5-FU and mitomycin. (Green 182:25-183:18.) 5-FU in particular held some promise for survival benefit, based on early stage studies, but it did not ultimately demonstrate a significant increase in survival or pain relief. (Green 184:6-21, 202:13-23; PTX 452; PTX 455 at 1033; PTX 436 at 2403.)

Thus, despite the existence of 5-FU, the need for more effective chemotherapeutic agents for the treatment of solid tumors like those present in pancreatic cancer still existed in the early 1980s, at the time of the invention of gemcitabine and its anticancer utility. The use of gemcitabine to treat patients with advanced pancreatic cancer therefore satisfied this need and, although not a cure, became the accepted standard of care, which it remains today. (Green 178:23-179:11, 204:25-205:17, 205:18-23.)

Because of this effectiveness, gemcitabine, marketed as the commercial product Gemzar®, which embodies the inventions disclosed and claimed in the patents-insuit, became and remains a commercial success.^[24] Gemzar® is now Lilly's fourth-largest selling product worldwide, with net sales growing from $175 million in 1997 to $1.592 billion in 2007. (Grabowski Rep. ¶¶ 14, 35, Ex. 2.)^[25] The evidence adduced at trial indicates that Lilly's marketing expenditures for Gemzar® equal about one-half of one percent of the net sales of Gemzar®, which is a marketing-to sales ratio lower than the average ratio for Gemzar®'s direct competitors as well as other relevant oncology drugs. (Id. at ¶ 18, 48-50, Ex. 6A-6B.)

Conclusions of Law

I. Standard of Review

Federal Circuit precedent governs matters of substantive patent law in this Court pursuant to 28 U.S.C. § 1295(a). Southwire Co. v. Essex Group, Inc., 220 U.S.P.Q. 1053, 1056 n. 6 (N.D.Ill.1983) ("The law that controls this action . . . is that law of the Federal Circuit[, which] has declared that the patent decisions of [the C.C.P.A.] will be considered binding . . . .") (citing S. Corp. v. United States, 690 F.2d 1368, 1370 (Fed.Cir.1982) (en banc)).

Patents are presumed to be valid. Moreover, each claim within a patent is independently presumed to be valid, even if other claims within that patent are held invalid. 35 U.S.C. § 282. Accordingly, the burden of proving invalidity rests with the patent challenger, here, Defendants, who must prove invalidity by clear and convincing evidence. Id.; Golden Blount, Inc. v. Robert H. Peterson Co., 365 F.3d 1054, 1061-62 (Fed.Cir.2004). This burden "is constant and remains throughout the suit on the challenger" and "does not shift at any time to the patent owner." TP Labs., Inc. v. Prof'l Positioners, Inc., 724 F.2d 965, 971 (Fed.Cir.1984). Pursuant to Defendants' obligation to carry this burden, "if the fact trier of the issue is left uncertain, the party with the burden loses." Tech. Licensing Corp. v. Videotek, Inc., 545 F.3d 1316, 1327 (Fed.Cir.2008) (citation omitted).

If the patent challenger alleges invalidity based on prior art that was considered by the U.S. Patent and Trademark Office ("PTO") during the prosecution of the patent, the challenger has the "added burden of overcoming the deference that is due to a qualified government agency presumed to have properly done its job." Ultra-Tex Surfaces, Inc. v. Hill Bros. Chem. Co., 204 F.3d 1360, 1367 (Fed.Cir.2000) (quoting Am. Hoist & Derrick Co. v. Sowa & Sons, 725 F.2d 1350, 1359 (Fed.Cir. 1984)); see also Impax Labs., Inc. v. Aventis Pharms, Inc., 468 F.3d 1366, 1378 (Fed.Cir.2006) ("When the prior art was before the examiner during prosecution of the application, there is a particularly heavy burden in establishing invalidity.") (citation omitted). The rationale underlying the presumption of validity is, however, "much diminished" when the prior art relied upon was not considered by the PTO during prosecution. KSR Int'l Co. v. Teleflex, Inc., 550 U.S. 398, 426, 127 S. Ct. 1727, 167 L. Ed. 2d 705 (2007). That said, if a reference was not before the PTO examiner but was cumulative of others that were before the examiner, the heavy burden of proof nonetheless applies. Metabolite Labs., Inc. v. Lab. Corp. of Am. Holdings, 370 F.3d 1354, 1368 (Fed.Cir.2004).

II. Effective Filing Dates of the Patents

Under 35 U.S.C. § 120:

An application for patent for an invention disclosed in the manner provided by the first paragraph of section 112 of this title in an application previously filed in the United States, . . . which is filed by an inventor or inventors named in the previously filed application shall have the same effect, as to such invention, as though filed on the date of the prior application, if filed before the patenting or abandonment of or termination of proceedings on the first application or on an application similarly entitled to the benefit of the filing date of the first application and if it contains or is amended to contain a specific reference to the earlier filed application. . . .

35 U.S.C. § 112, in turn, provides:

The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make the use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.

In accordance with these statutory provisions, the effective filing date of the '614 patent is March 10, 1983, the filing date of Dr. Hertel's original patent application containing a description of gemcitabine and a disclosure of a method of utility. (PTX 1.) The effective filing date of the '826 patent is December 4, 1984, the filing date of Dr. Hertel's and Dr. Grindey's original patent application containing a description of a method of using gemcitabine to treat susceptible neoplasms. (PTX 2.)

III. Defendants' Claim of Double Patenting

In August 2009, summary judgment was entered against Lilly in the Eastern District of Michigan, in a factually similar case to the case before this Court, in which it was held that obviousness-type double patenting rendered the '826 patent invalid. Sun Pharm. Indus., Ltd. v. Eli Lilly & Co., 647 F. Supp. 2d 820 (E.D.Mich.2009). In handing down that ruling, the court addressed and resolved the exact same double patenting issue presented to us here. Defendants, accordingly, contend that Lilly is estopped from asserting in this litigation that the doctrine of obviousness-type double patenting does not apply.

Defendants invoke the rule promulgated by the Supreme Court in Blonder-Tongue v. University Foundation, 402 U.S. 313, 91 S. Ct. 1434, 28 L. Ed. 2d 788 (1971), which bars Plaintiff from relitigating the validity of the '826 patent. The Blonder-Tongue Court held that once a court determines that a patent is invalid in a proceeding where the patent owner had a full and fair opportunity to adjudicate the issue, the patent owner is collaterally estopped from relitigating the issue in a future case. Blonder-Tongue, 402 U.S. 313, 91 S. Ct. 1434. Collateral estoppel limits patent owners to "one full and fair opportunity for judicial resolution of the same [invalidity] issue." Id. at 328, 91 S. Ct. 1434. Blonder-Tongue also held that the application of collateral estoppel is governed by "the trial courts' sense of justice and equity." Id. at 334, 91 S. Ct. 1434.

Whether collateral estoppel applies is determined in accordance with Seventh Circuit precedent, rather than the Federal Circuit precedent which ordinarily controls in patent cases. Bayer AG. v. Biovail Corp., 279 F.3d 1340, 1345 (Fed.Cir.2002). Thus, under Seventh Circuit precedent, for collateral estoppel to apply, "four elements must be met: `(1) the issue sought to be precluded must be the same as that involved in the prior litigation, (2) the issue must have been actually litigated, (3) the determination of the issue must have been essential to the final judgment, and (4) the party against whom estoppel is invoked must be fully represented in the prior action.'" Meyer v. Rigdon, 36 F.3d 1375, 1379 (7th Cir.1994).

Further, "[c]laim preclusion (res judicata), as Rule 8(c) of the Federal Rules of Civil Procedure makes clear, is an affirmative defense," Rivet v. Regions Bank of Louisiana, 522 U.S. 470, 476, 118 S.Ct. 921, 139 L. Ed. 2d 912 (1998), which "must be pleaded." Blonder-Tongue, 402 U.S. at 350, 91 S. Ct. 1434. "The purpose of such pleading is to give the opposing party notice of the plea of estoppel and a chance to argue, if he can, why the imposition of estoppel would be inappropriate." Id.

We recognize that, in this case, Defendants did not plead collateral estoppel as an affirmative defense. However, it is equally clear that the likely reason for this deficiency is that the Eastern District of Michigan decision subjecting the issue of double patenting as it relates to the '826 patent to collateral estoppel was handed down a mere month prior to the trial in the case at bar. Therefore, Defendants had no meaningful opportunity to plead it. Lilly contends that the pleading requirement is essential and must be upheld because Lilly was not given sufficient notice of the res judicata defense; however, Lilly's contention has been undermined by the fact that Lilly availed itself of an opportunity, in post-trial briefing of the issues before the Court, to argue at length why estoppel should not apply. Accordingly, rigid enforcement of the typical pleading requirement would serve only to undermine "justice and equity" in this case. Blonder-Tongue, 402 U.S. at 334, 91 S. Ct. 1434.

Upon review of the parties' arguments for and against the application of res judicata, we conclude that the four elements of collateral estoppel are clearly met here. Lilly implicitly concedes the first three elements of the collateral estoppel standard have been satisfied: (1) the double patenting issue presented is exactly the same; (2) the issue was actually litigated before the Sun Pharmaceutical Court; (3) final judgment was entered in Sun Pharmaceutical, based on the double-patenting issue. As to the fourth element, Lilly argues only that it was denied a full and fair opportunity to litigate the double patenting issue at trial because the Michigan court's summary judgment decision was "facially and indisputably wrong as a matter of law." (Pl.'s Tr. Br. at 36.) Although we do not here reject Lilly's arguments related to the merits of the Sun Pharmaceutical decision, we do hold that simply disagreeing with the result at the summary judgment phase, which resulted in a final judgment for collateral estoppel purposes, see Smith v. City of Chicago, 820 F.2d 916 (7th Cir.1987), is insufficient to demonstrate the lack of a full and fair representation in the prior litigation. Therefore, Lilly is estopped in this action from defending against Defendants' claim that the '826 patent is invalid for obviousness-type double patenting.^[26]

IV. Anticipation^[27]

A patent claim is invalid under 35 U.S.C. § 102(a) if "the invention was known or used by others in this country, or patented or described in a printed publication in this or a foreign country, before the invention thereof by the applicant for patent." 35 U.S.C. § 102(a). Pursuant to 35 U.S.C. § 102, anticipation requires "the presence in a single prior art disclosure of all elements of a claimed invention arranged as in that claim." Carella v. Starlight Archery & Pro Line Co., 804 F.2d 135, 138 (Fed.Cir.1986) (citation omitted).

Defendants, as challengers of the patents' validity, bear the burden of proving by clear and convincing evidence that the references relied upon to demonstrate anticipation are, in fact, prior art. Norian Corp. v. Stryker Corp., 363 F.3d 1321, 1330 (Fed.Cir.2004). To determine what constitutes prior art for invalidity analysis, the Court must consider the "invention date" for the asserted claim. 35 U.S.C. § 102(a); Mahurkar v. C.R. Bard, Inc., 79 F.3d 1572, 1576 (Fed.Cir.1996). Documents that are not prior art cannot anticipate a patent claim. IMX, Inc. v. Lending Tree, LLC, 2006 WL 47066, at *1 (D.Del. Jan. 10, 2006). In showing that all elements of an invention were anticipated, the challenging party may not rely upon the knowledge of one skilled in the art of the disclosure of another reference to supply missing elements. Structural Rubber Prods. Co. v. Park Rubber Co., 749 F.2d 707, 715 (Fed.Cir.1984).

A. Dr. Bobek's Slides

Defendants attempted at trial to show that a group of slides ("Bobek Slides") produced by Dr. Bobek anticipated the asserted claims of the '614 and '826 patents because those slides contained a drawing of the chemical structure of gemcitabine. (DTX 220 at Bobek 133.) We thus must first determine if these slides did in fact constitute prior art for invalidity purposes. Norian, 363 F.3d at 1330.

As a general matter, public presentations qualify as prior art under 35 U.S.C. § 102(a) and (b). See Ecolochem, Inc. v. S. Cal. Edison Co., 227 F.3d 1361, 1369-70 (Fed.Cir.2000); Mass. Inst. of Tech. v. AB Fortia, 774 F.2d 1104, 1108 (Fed.Cir.1985) (finding that a paper that was presented to persons in the relevant field qualified as a "printed publication" under § 102(b)). Defendants contend that Dr. Bobek's presentations were public, citing Dr. Bobek's testimony to that effect. (Bobek 1318:13-1319:25.) Aside from his testimony, however, no other evidence in the record supports Defendants' contention that these slides were ever disclosed or otherwise made known publicly. In fact, Dr. Rustum and Dr. Bernacki stated that they had no recollection of being present at any meeting where Dr. Bobek purportedly made a presentation disclosing his work on gem-difluoronucleosides, despite being Dr. Bobek's departmental colleagues during that period. (Rustum 1589:21-1590:2; Bernacki Dep. 51:16-23, 52:13-18.) Therefore, Dr. Bobek's testimony on this point is uncorroborated.

Defendants argue that his uncorroborated testimony is sufficient to establish that the public display of the presentation actually occurred, citing Thomson S.A. v. Quixote Corp., 166 F.3d 1172 (Fed.Cir. 1999). The rule of Thomson,—that the testimony of a "disinterested" witness need not be corroborated in certain circumstances—does not, however, apply to the present facts. Not only was Dr. Bobek a paid consultant of Defendants, but his testimony revealed his disappointment and resultant emotional reaction to his not having discovered and invented gemcitabine prior to Lilly's doing so. (Cf. Hertel 325:20-326:3; Farr Dep. 56:20-57:6.) Further, since deciding Thomson, the Federal Circuit has reiterated that "corroboration is required of any witness whose testimony alone is asserted to invalidate a patent, regardless of his or her level of interest." Finnigan Corp. v. Int'l Trade Comm'n, 180 F.3d 1354, 1369 (Fed.Cir.1999); see also Adenta GmbH v. OrthoArm, Inc., 501 F.3d 1364, 1371 (Fed.Cir.2007). Accordingly, because the evidence in the record is insufficient to support a finding that the Bobek Slides were presented publicly in the form advanced by Defendants, we conclude that they do not qualify as prior art by virtue of their public presentation.

Assuming arguendo that Dr. Bobek's uncorroborated testimony established that the presentation was public and thus prior art, the Bobek Slides nonetheless constitute such limited a disclosure as to be entirely inadequate for a showing of anticipation. Regents of Univ. of Cal. v. Howmedica, Inc., 530 F. Supp. 846, 860 (D.N.J.1981), aff'd, 676 F.2d 687 (3d Cir. 1982) ("[T]he projection of the slides at the lecture was limited in duration and could not disclose the invention to the extent necessary to enable a person of ordinary skill in the art to make or use the invention."). Furthermore, Dr. Bobek's presentations were and are insufficient to anticipate the asserted claims because of their scant content. They do not contain any information about antitumor activity, nor are they capable of enabling a person of skill in the art to synthesize gemcitabine. "In order to anticipate a claimed invention, a prior art reference must enable one of ordinary skill in the art to make the invention without undue experimentation." Impax Laboratories, Inc. v. Aventis Pharmaceuticals, Inc., 545 F.3d 1312, 1314 (Fed. Cir.2008).

Defendants contend that Dr. Bobek invented gemcitabine prior to Lilly, but the record before the Court clearly shows that Dr. Bobek never, in fact, invented gemcitabine. Dr. Bobek admitted that he never isolated, characterized, or analyzed any product from the reaction that he asserts resulted in gemcitabine. (Bobek 1324:9-23, 1325:11-19, 1358:7-9, 1359:8-9; Radtke 459:12-463:3; PTX 499 at Bobek 202; PTX 508 at 61.) Moreover, Dr. Bobek never demonstrated any utility for gemcitabine, and therefore never actually reduced the compound he conceptualized in his slides to a practical invention. Dr. Bobek did not disclose his alleged invention of gemcitabine in a patent application, describe it in a published document, or use it publicly, and he eventually abandoned his work on the compound. For these reasons, Dr. Bobek's work cannot be treated as prior art; it cannot support Defendants' theory of prior invention; and it falls far short of anticipating the asserted claims. See Mahurkar v. C.R. Bard, Inc., 79 F.3d 1572, 1577 (Fed.Cir.1996); Lutzker v. Plet, 843 F.2d 1364, 1366-67 (Fed.Cir. 1988).

Defendants also contend that Lilly was able to invent gemcitabine only because of information garnered from Dr. Bobek's work. They assert that Dr. Bobek communicated information relating to gemcitabine, in particular its capacity to function as an anticancer agent, to Dr. Grindey, a named inventor of the '826 patent, while both were employed at Roswell Park. A claim of derivation "requires a showing of both (1) prior conception of the invention by another and (2) communication of that conception to the patentee that is 'sufficient to enable [him] to construct and successfully operate the invention.'" Int'l Rectifier Corp. v. IXYS Corp., 361 F.3d 1363, 1376 (Fed.Cir.2004).

For all of the reasons previously discussed, it is abundantly clear that Dr. Bobek did not conceive of gemcitabine and its utility prior to Lilly's having done so. "Conception requires both the idea of the invention's structure and possession of an operative method of making it." Amgen, Inc. v. Chugai Pharm. Co., 927 F.2d 1200, 1206 (Fed.Cir.1991). Although Dr. Bobek's slides contained a depiction of the structure of gemcitabine, he never devised a successful method for synthesizing the compound. Accordingly, Defendants have failed to prove by clear and convincing evidence that Lilly derived the invention of the '826 patent from Dr. Bobek's work.

Defendants also failed at trial to prove the "communication" required for a derivation claim. The only evidence Defendants offered to support this contention was Dr. Bobek's uncorroborated testimony that he communicated his idea for using gemcitabine as an anticancer agent to Dr. Grindey. (Bobek 1311:11-1312:14.) In addition to being uncorroborated, Dr. Bobek's testimony is unconvincing because of its intrinsic inconsistency. At his deposition, Dr. Bobek alleged that Dr. Grindey had reviewed his 1975 grant proposal related to gemcitabine when Dr. Grindey acted as a grant reviewer for the American Cancer Society. (Bobek 1398:8-1399:5.)^[28] By contrast, when the veracity of this assertion was challenged at trial, Dr. Bobek testified that Dr. Grindey reviewed the grant as an in vivo tester at Roswell Park. Given the inconsistencies inherent in this testimony and the evidence in the record clearly refuting Dr. Bobek's assertion that he communicated his ideas to Dr. Grindey, Defendants also cannot establish the second element of their derivation claim.

The derivation claim advanced by Defendants, which rests entirely on the uncorroborated testimony of Dr. Bobek, is reminiscent of an early observation by the Supreme Court:

The very fact, which courts as well as the public have not failed to recognize, that almost every important patent . . . has been attacked by the testimony of witnesses who imagined they had made similar discoveries long before the patentee had claimed to have invented his device, has tended to throw a certain amount of discredit upon all that class of evidence, and to demand that it be subjected to the closest scrutiny.

Barbed Wire Patent Case, 143 U.S. 275, 284-85, 12 S. Ct. 443, 36 L. Ed. 154 (1892). Dr. Bobek's testimony regarding his earlier conception of gemcitabine and its anticancer utility simply cannot withstand this sort of heightened scrutiny.

For all of these reasons, we hold that Defendants have failed to prove by clear and convincing evidence that Dr. Bobek's work anticipated that patents at issue in this suit.

B. Dr. Bergstrom's Work

1. The Bergstrom Abstract as Prior Art

Defendants also contend that the Bergstrom Abstract is prior art to the '614 patent under § 102(a) because it was published prior to March 10, 1983, the filing date of the first application in the chain that led to the patent. Defendants bear the burden of proving that this is prior art, Mahurkar, 79 F.3d at 1578, and "all the applicant can be required to show is priority with respect to so much of the claimed invention as the reference happens to show. When he has done that he has disposed of the reference." In re Stempel, 44 C.C.P.A. 820, 241 F.2d 755, 759 (1957) see also In re Stryker, 58 C.C.P.A. 797, 435 F.2d 1340, 1341 (1971).

In March 1983, the Bergstrom Abstract disclosed a 3',3'-difluoro-2',3'-dideoxythymidine, a compound with the gem-difluoro substitution at the C-3' position and with a thymine base. This compound clearly was not gemcitabine. (PTX 518; Radtke 466:13-467:12.) By June 1982, Dr. Hertel had already described the structure of 2' ,2'-difluoro-2'-deoxynucleosides; devised a method for their synthesis; actually synthesized a 2',2'-gem-difluoronucleoside; and observed antiviral activity. (E.g., Hertel 311:17-313:4, 316:1-319:7, 330:8-24; PTX 26 at 162, 164; PTX 181; PTX 229.) Therefore, prior to the publication in March 1983 of the Bergstrom Abstract, Dr. Hertel had already invented more of the claimed invention than the Bergstrom Abstract disclosed. Accordingly, that abstract is not prior art and did not anticipate the '614 patent.

2. Failure to Disclose All Elements of the Invention

As discussed previously, the compound disclosed in the Bergstrom Abstract was a 3',3'-difluoro-2' ,3'-dideoxythymidine, a compound with the gem-difluoro substitution at the C-3' position and with a thymine base. Because this compound is not gemcitabine, the disclosure of the compound is insufficient to anticipate the invention of gemcitabine. Carella, 804 F.2d at 138. Moreover, the Bergstrom Abstract did not disclose the operable method of making gemcitabine, which was another element of the claimed invention.

Nor is this a case in which the disclosure in the asserted prior art would cause one of ordinary skill to "at once envisage" the claimed compound. Bristol-Myers Squibb v. Ben Venue Labs, Inc., 246 F.3d 1368, 1380 (Fed.Cir.2001). Defendants contend that the Bergstrom Abstract discloses a small genus of compounds consisting of the natural bases coupled to 2' and 3' gem-difluorosugars, but the evidence clearly demonstrates that Dr. Bergstrom in no way limited his disclosure to the natural bases. (Radtke 471:5-15; Herdewijn 1098:23-1099:5; PTX 518.) Moreover, the evidence showed that Drs. Bobek, Bergstrom and Farr each considered making nucleoside analogs with non-natural bases. (E.g., PTX 533 at Bergstrom 20; PTX 494 at Bobek 18; PTX 498 at 68; Herdewijn 1090:7-16, 1091:5-18; PTX 555.) Thus, viewed in the proper context, the genus the abstract actually encompassed included a limitless number of molecules made with non-natural bases and therefore did not disclose gemcitabine to a person of ordinary skill in the art.

The Bergstrom Abstract at most disclosed the concept of making a similar compound. It did not constitute "the presence in a single prior art disclosure of all elements of a claimed invention arranged as in that claim," Carella, 804 F.2d at 138, and therefore did not anticipate the asserted claims.

3. The Bergstrom Slides

Defendants also contend that slides numbered 3258 and 3261 of Dr. Bergstrom's March 1983 presentation at the Seattle American Chemical Society meeting ("Bergstrom Slides") anticipate both the '614 patent and the '826 patent. While they are correct that a public presentation such as this might qualify as prior art under 35 U.S.C. § 102(a) and (b), Ecolochem, 227 F.3d at 1369-70, the Bergstrom Slides to not qualify as prior art.

The Bergstrom Slides are not prior art to the '614 patent because the presentation occurred on March 23, 1983, after Dr. Hertel invented the claimed compounds and after Lilly filed the '883 application. (PTX 518; PTX 519.) In addition to not qualifying as prior art to the '614 patent, the slides do not contain sufficient information to enable one of ordinary skill in the art to make gemcitabine because they contain no synthetic routes capable of producing the compound, or its 2,2-difluororibose precursor. (Bergstrom 953:9-16.) Accordingly, the Bergstrom Slides do not anticipate claim 12 of the '614 patent. Amgen, 927 F.2d at 1206.

The Bergstrom Slides arguably do qualify as prior art to the '826 patent, given that they were published before the application leading to that patent. They cannot, however, be viewed to anticipate claim 7 of the '826 patent because the slides contain no information whatsoever about antitumor activity, which is essential to the claimed invention. Carella, 804 F.2d at 138. Finally, and more broadly, Defendants have failed to carry their burden of proving anticipation by clear and convincing evidence, given that Dr. Bergstrom testified that he could not recall which slides he actually showed at the presentation, whether he discussed 2',2'-difluoronucleosides at the presentation, or whether he, in fact, proposed any utility for the compounds he did discuss during the presentation. (Bergstrom 952:15-18. 953:5-8, 917:24-918:6.)

C. The Merrell Dow Research

Defendants further contend that both claim 12 of the '614 patent and claim 7 of the '826 patent are anticipated by the alleged prior invention of those claims by Drs. Farr and Metcalf of Merrell Dow. In patent law, "priority of invention goes to the first party to reduce an invention to practice unless the other party can show that it was the first to conceive the invention and that it exercised reasonable diligence in later reducing that invention to practice." z4 Techs., Inc. v. Microsoft Corp., 507 F.3d 1340, 1352 (Fed.Cir.2007) (citation omitted).

As an initial matter, the facts clearly demonstrate that Lilly reduced the invention to practice prior to any other party, including the researchers at Merrell Dow. Moreover, by the time Drs. Farr and Metcalf actually synthesized and tested gemcitabine, Lilly had already screened it for cytotoxicity and administered it in a wide variety of tests. Indeed, Lilly had filed a patent application disclosing methods of treating cancer using gemcitabine. (PTX 118 at GM 157170; Hertel 334:1-21; PTX 552; PTX 555) Therefore, in order to prove anticipation, Defendants must show that these Merrell Dow researchers first conceived of the claimed inventions and thereafter exercised reasonable diligence in reducing those inventions to practice.

1. The '614 Patent

Defendants concede that Merrell Dow did not actually produce gemcitabine before Lilly did, but contend nonetheless that Dr. Metcalf was the first to conceive of the compound, and Dr. Farr later worked diligently to reduce it to practice. Thus, we first address whether Dr. Metcalf conceived of gemcitabine prior to 1983. "Conception requires both the idea of the invention's structure and possession of an operative method of making it." Amgen, 927 F.2d at 1206. The evidence at trial did not establish that either Dr. Farr or Dr. Metcalf knew how to make gemcitabine at any time prior to Lilly. (PTX 554 at SA 2200; PTX 557 at SA 2182; PTX 566; PTX 6.)^[29] In fact, when Merrell Dow eventually filed a patent application on difluoronucleosides, it disclosed Lilly's synthetic route, which had been previously described in Lilly's published British patent application. (PTX 568, 17:47-51, 18:49-30:29; PTX 16, 4:43-9:47.)^[30]

Thus, Defendants cannot show that the Merrell Dow researchers actually conceived of the claimed invention prior to Lilly, and their claim of anticipation or prior invention with regard to the '614 patent fails.

2. The '826 Patent

Defendants also contend that Merrell Dow has priority of invention with regard to the use of gemcitabine as an anticancer agent in mammals. Once again, in the face of clear evidence demonstrating that Lilly first reduced that utility to practice, Defendants argue that the Merrell Dow researchers conceived of the anticancer utility prior to Lilly and that therefore claim 7 of the '826 patent is invalid for anticipation under 35 U.S.C. § 102(g). The only thing Defendants have successfully proved by clear and convincing evidence, however, was that, after March 1984, Merrell Dow conceived of an anticancer utility for gemcitabine. As a review of the record shows, this is not proof that Merrell Dow was the first to conceive of the anticancer utility or specifically that their discovery preceded Lilly's.

Prior to March 31, 1984, the Merrell Dow documents merely discussed a potential "antiviral activity" for different difluoronucleosides, not a confirmed anticancer activity for gemcitabine. (PTX 552 at SA 2206-07; PTX 553 at SA 2196-97.) By this point in time, Lilly had not only conceived of a potential anticancer utility for gemcitabine, but had tested the compound for such activity in vitro and in vivo. Dr. Hertel sent a sample of gemcitabine to Dr. Grindey for in vitro testing on October 11, 1983, and the sample was tested numerous times in November and December of 1983. (PTX 31 at 195; Hertel 332:25-335:4, 335:13-337:8; PTX 118 at GM XXXXXX-XX.) Dr. Grindey commenced in vivo anticancer testing in February 1984. (Pearce 90:16-91:9; PTX 126.)^[31]

Accordingly, Defendants have failed to prove by clear and convincing evidence that the researchers at Merrell Dow conceived of the anticancer utility prior to Lilly, or worked to reduce that purported conception to practice. Therefore, the Merrell Dow work did not anticipate claim 7 of the '826 patent.^[32]

For all of the foregoing reasons, Defendants have failed to establish that the patents-in-suit were anticipated, either by "the presence in a single prior art disclosure of all elements of a claimed invention arranged as in that claim," Carella, 804 F.2d at 138, or under any alternative theory. Therefore, Defendants' claims of anticipation are unavailing.

V. Obviousness

Defendants also contend that both the '614 patent and the '826 patent are invalid for obviousness. Under the United States Patent Act, a patent cannot be obtained "if the differences between the subject matter sought to be patented and the prior art 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). A party seeking to invalidate a patent based on obviousness must demonstrate "by clear and convincing evidence that a skilled artisan would have been motivated to combine the teachings and prior art references to achieve the claimed invention, and that the skilled artisan would have had a reasonable expectation of success in doing so." Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1361 (Fed.Cir. 2007) (citations omitted).

"[T]he outcome of the trial on this point will depend on whether, in light of all the evidence, the party challenging the patent's validity has carried its burden to prove by clear and convincing evidence that the patent is invalid." Titan Tire Corp. v. Case New Holland, Inc., 566 F.3d 1372, 1376 (Fed.Cir.2009). Moreover, where, as here, the patent challenger alleges invalidity based on prior art that the Patent and Trademark Office ("PTO") considered during prosecution of the patent or cumulative with art considered by the PTO, the challenger has the "[a]dded burden of overcoming the deference that is due to a qualified government agency presumed to have properly done its job." Ultra-Tex Surfaces, 204 F.3d at 1367.

Obviousness is a question of law based on underlying findings of fact, which include: (1) the scope and content of the prior art; (2) the differences between the prior art and the claims at issue; (3) the level of ordinary skill in the art at the time the invention was made; and (4) objective evidence of nonobviousness, if any. Procter & Gamble Co. v. Teva Pharmaceuticals USA, Inc., 566 F.3d 989, 994 (Fed. Cir.2009) (citing Graham v. John Deere Co., 383 U.S. 1, 17, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966)). "[I]ncluded within the [patent's] presumption of validity is. . . a presumption of nonobviousness." Structural Rubber Prods., 749 F.2d at 715.

Obviousness arises when a skilled individual "merely pursues `known options' from a `finite number of identified, predictable solutions.'" In re Kubin, 561 F.3d 1351, 1359 (Fed.Cir.2009) (quoting KSR Int'l., 550 U.S. 398, 421, 127 S. Ct. 1727 (2007)). Section 103 also bars patentability unless "the improvement is more than the predictable use of prior art elements according to their established functions." KSR Int'l, 550 U.S. at 417, 127 S. Ct. 1727. Obviousness does not require absolute predictability; all that is required is a reasonable expectation of success. Eli Lilly and Co. v. Zenith Goldline Pharmaceuticals, Inc., 471 F.3d 1369, 1377 (Fed. Cir.2006) (citing In re Longi, 759 F.2d 887, 896 (Fed.Cir.1985)).

The reference or evidence used to establish obviousness under 35 U.S.C. § 103 must qualify as prior art under one or more sections of 35 U.S.C. § 102. Panduit Corp. v. Dennison Mfg. Co., 810 F.2d 1561, 1568 (Fed.Cir.1987). What a reference teaches is a question of fact addressed to a hypothetical "person of ordinary skill in the art." In re Bell, 991 F.2d 781, 783-84 (Fed.Cir.1993). To prove that a claimed invention was obvious, a patent challenger must demonstrate "that a skilled artisan would have been motivated to combine the teachings of the prior art references to achieve the claimed invention, and that the skilled artisan would have had a reasonable expectation of success in doing so." Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1361 (Fed.Cir.2007).

A. The '614 Patent

1. No Prior Invention

Defendants contend that, prior to the filing of the applications leading to the '614 patent and '826 patent, three different research endeavors—those of Drs. Bobek, Bergstrom, and Metcalf and Farr—proposed gemcitabine and its use as an antiviral or antitumor agent. A finding of contemporaneous or prior invention "can be understood to suggest that one of ordinary skill in the art would have been motivated to combine prior art references to arrive at the claimed invention," and that therefore the invention would have been obvious. Nat'l Steel Car, Ltd. v. Canadian Pac. Ry., Ltd., 357 F.3d 1319, 1337-38 (Fed.Cir.2004).

Defendants' reliance upon the work done by these researchers is, however, misplaced because those scientists' attempts to make gemcitabine were both partial and ultimately unsuccessful. The failures of contemporaneous or prior research to create the claimed invention can be determinative of nonobviousness. Advanced Display Sys. Inc. v. Kent State Univ., 212 F.3d 1272, 1285 (Fed.Cir.2000); Sanofi-Synthelabo v. Apotex, Inc., 550 F.3d 1075, 1088 (Fed.Cir.2008); Forest Labs., Inc. v. Ivax Pharms., Inc., 501 F.3d 1263, 1269 (Fed.Cir.2007).

Indeed, the evidence at trial relating to the efforts of these researchers clearly shows that a person of ordinary skill in the art would not have known how to make gemcitabine. Collectively, Drs. Bergstrom, Bobek, and Farr spent numerous years pursuing various synthetic methods in attempting to make geminally substituted difluoronucleoside analogs. These synthetic approaches included: (1) direct fluorination of a nucleoside; (2) direct fluorination of a sugar followed by coupling to a base; (3) direct fluorination of a six-membered ring sugar, followed by conversion to a five-membered ring sugar, followed by coupling to a base; (4) de novo synthesis involving a Sharpless epoxidation step; and (5) de novo synthesis using D-glyceraldehyde acetonide as a starting material. (E.g., Radtke 496:5-497:4, 498:9-499:2; Bergstrom 905:22-906:2, 911:9-18, 945:6-24; Bobek 1359:5-9; Overman 613:13-621:9; PTX 498 at Bobek 66-68; PTX 494 at Bobek 17-18; PTX 535 at 265; Dep. of Farr 117:13-17.) In light of the fact that all of these methods were attempted and unsuccessfully so in the years leading up to and including the time that Lilly invented gemcitabine, it is abundantly clear that a person of ordinary skill in the art would have been uncertain as to how to synthesize gemcitabine.^[33]

2. Defendants' Attempts to Rationalize Experimental Failures

Through expert testimony, Defendants unconvincingly attempt to discount the failures of Drs. Bobek, Bergstrom, Metcalf, and Farr to create gemcitabine. First, Defendants seek to divert attention from the difficulty of synthesizing gemcitabine by highlighting expert testimony that Dr. Hertel's synthetic route would have been the obvious choice for the person of ordinary skill in the art. Dr. Herdewijn averred that a person of ordinary skill would not have attempted direct fluorination of a nucleoside, (Herdewijn 1028:15-22), and Dr. Kathlyn Parker stated that "only" Dr. Hertel's synthetic route would have been pursued by the person of ordinary skill in the art. (Parker 1169:5-1170:9, 1193:24-1194:9). These opinions ignore the alternate synthetic routes employed by other researchers, however, and are directly contradicted by the facts of the case. For example, the evidence clearly demonstrated that Dr. Bergstrom did himself, in fact, attempt direct fluorination of a nucleoside. (Bergstrom 946:23-948:5; Radtke 509:9-23; PTX-534 at Bergstrom 67, 69.)^[34]

It is equally clear that some of the expert testimony Defendants rely upon was arrived at inappropriately. Dr. Parker admitted that she was aware of Dr. Hertel's successful approach before she conducted her expert analysis; that she never considered the work of Drs. Bobek, Bergstrom, and Farr; that she did not consider any Lilly work other than the successful synthesis reported in the '614 patent; and that she limited her review of the prior art to those references that supported "the story that [she] was telling." (Parker 1168:18-22, 1170:14-23, 1171:15-21, 1178:19-23, 1179:23-1180:18, 1182:16-18). As the Federal Circuit has held, such hindsight-driven analysis cannot be used to refute clear evidence to the contrary. Ortho-McNeil Pharm., Inc. v. Mylan Labs., 520 F.3d 1358, 1364 (Fed.Cir.2008) ("[The] expert . . . simply retraced the path of the inventor with hindsight, discounted the number and complexity of the alternatives, and concluded that the invention . . . was obvious. Of course, this reasoning is always inappropriate. . . .").

Fact finders must avoid "[t]he distortion caused by hindsight bias and must be cautious of arguments reliant upon ex post reasoning." KSR Int'l, 550 U.S. at 421, 127 S. Ct. 1727 (citing Graham, 383 U.S. at 36, 86 S. Ct. 684). In light of this guidance, we find that the more appropriate conclusion was the one drawn by Dr. Overman, who after reviewing all of the attempts by other researchers to create gemcitabine alongside Dr. Hertel's work, concluded that at least five categories of synthetic routes were available to a person of ordinary skill in the art and that such a person would not have known which route held the most promise for the successful synthesis of gemcitabine. (Overman 596:7-597:6, 603:2-605:10, 606:20-607:11, 608:13-23, 649:16-49, 650:21-24, 698:8-13, 711:13-16; LDX 9.4; LDX 9.22.)

Dr. Herdewijn also attempted to explain away the failures of Drs. Farr and Bergstrom by referencing the fact that they utilized the expertise of a general organic chemist, rather than the specialized expertise of a nucleoside chemist. (Herdewijn 983:14-25, 1047:12-19, 1052:22-1053:2.) However, Dr. Herdewijn's explanation of the significance of this expertise is unavailing, given the fact that Dr. Bobek, himself a nucleoside chemist, also repeatedly failed to synthesize gemcitabine using the same methodology. Further, the vast majority of the experimentation failures experienced by Drs. Farr and Bergstrom occurred during steps in the synthetic process that did not require nucleoside chemistry expertise. (Herdewijn 984:6-14 (identifying only one reaction that allegedly involved expertise specific to a nucleoside chemist).)^[35]

Finally, Dr. Herdewijn attempted to rationalize Dr. Farr's failures by testifying that Dr. Farr would have succeeded had he been in possession of an article authored by Poulter and colleagues. (PTX 797). Dr. Herdewijn's opinion in this regard is again unfounded, however, because Dr. Farr was independently aware of the important information contained in that article during the period when he was attempting to synthesize gemcitabine and yet he still failed to make the compound. (PTX 552 at SA 2212-13; PTX 553 at SA 2200; Overman 630:20-631:21.) Accordingly, because Dr. Farr was in possession of this information, the alleged absence of access to the Poulter article cannot rationalize, explain, or supersede, his failure to create gemcitabine.

In the end, Defendants' experts fail in their attempts to explain away the failures of Drs. Bobek, Bergstrom, and Farr. Thus, all that is left in the evidence is the clear fact of those failures. Neither Dr. Parker nor Dr. Herdewijn testified that Dr. Bobek actually synthesized gemcitabine, and the facts adduced clearly show that the route he proposed was incapable of producing such a successful synthesis. (E.g., Radtke 459:12-461:8; Bobek 1358:8-9); PTX 492 at 3435; PTX 494 at Bobek 17. Additionally, Dr. Bergstrom's failure, in which he utilized several different synthetic routes to create gemcitabine, is also well documented in the record. (Bergstrom 920:18-20, 946:23-948:5; PTX 534 at Bergstrom 67-69; Radtke 509:9-23.) And contemporaneous to the invention of the '614 patent, the evidence reveals that Dr. Farr had only attempted a combination synthesis using Sharpless epoxidation, which also failed to create gemcitabine. (PTX 552 at SA 2212-13; PTX 553 at SA 2200; Overman 630:20-631:21.)

Accordingly, Defendants have failed to carry their burden to establish that a person of ordinary skill in the art could have used any of the methods other than that used by Dr. Hertel to synthesize gemcitabine, and have likewise failed to show that his method would have been known or obvious to, or anticipated by, a person of ordinary skill in the art. Titan Tire Corp., 566 F.3d at 1376-77. As these researchers' repeated unsuccessful attempts to synthesize gemcitabine demonstrate, the person of ordinary skill in the art did not and would not have known how to synthesize gemcitabine. Compare Advanced Display Sys., 212 F.3d at 1285 ("Zhou's deposition furnishes persuasive evidence that the West patent is nonobvious by describing ADS's repeated failures to design the claimed invention.") to In re Merck & Co., 800 F.2d 1091, 1098 (Fed.Cir.1986) (finding obviousness where four other groups independently and successfully invented the claimed invention). Defendants, therefore, have not proved by clear and convincing evidence that the invention of the '614 patent would have been obvious. Interconnect Planning Corp. v. Feil, 774 F.2d 1132, 1143 (Fed.Cir.1985).

3. Obviousness Based on Structural Similarity

In challenging the validity of the '614 patent, Defendants also attempt to establish "obviousness based on structural similarity," which "may be proven by the identification of some motivation that would have led one of ordinary skill in the art to select and modify a known compound in a particular way to achieve the claimed compound." Altana Pharma AG v. Teva Pharms. USA, Inc., 566 F.3d 999, 1007 (Fed.Cir.2009). According to Defendants, one of ordinary skill in the art would have been motivated to create gemcitabine by the promising characteristics of three structurally similar compounds: Ara-C, 2'-F-ara-C, and 2'-F-cytidine.

Defendants assert that a "sufficiently close" relationship between gemcitabine and those three compounds would have "create[d] an expectation that" gemcitabine would have similar, perhaps improved antiviral and anticancer properties over those compounds. In re Dillon, 919 F.2d 688 (Fed.Cir.1990). Defendants first contend that Ara-C, an FDA-approved drug used to treat patients with various leukemias, would have been a lead compound for the person of ordinary skill, and that the person of ordinary skill would have been motivated to modify Ara-C with a gem-difluoro substitution at the 2'position.^[36] However, although Ara-C was "good as an anticancer drug" at the time of the development of gemcitabine, researchers were looking for alternatives to Ara-C that might exhibit improved anticancer activity. (Rustum 1545:10-1546:19; Radtke 542:15-19; PTX 693.) Included in those alternatives were 2'-F-ara-C and 2'-Fcytidine, which Defendants assert also had promising anticancer properties that would have motivated a person of ordinary skill to arrive at gemcitabine.^[37]

Assessment of the obviousness of a chemical compound cannot be based merely on comparisons between that compound's chemical structure and structures in the prior art. In re Papesch, 50 C.C.P.A. 1084, 315 F.2d 381, 391 (1963). There must be "some reason that would have led a chemist to modify a known compound in a particular manner" to establish obviousness by structural similarity. Takeda Chem. Indus., Ltd. v. Alphapharm Pty., Ltd., 492 F.3d 1350, 1357 (Fed. Cir.2007).

Upon review of the record of evidence adduced at trial, we conclude that Defendants' have failed to prove obviousness by virtue of gemcitabine's structural similarity to those three compounds because: (1) the prior art lacked any meaningful precedent for gemcitabine; (2) the fluorination required to make gemcitabine was too unpredictable to create a reasonable, motivating expectation; and (3) the purported "promise" of compounds in the prior art was too insignificant to motivate one of ordinary skill in the art.

The prior art contained no clear indication that a difluorinated nucleoside analog would exhibit the activity demonstrated by gemcitabine. Indeed, no geminally substituted difluorinated nucleoside analogs existed in the prior art, and the biological activity of a geminal substitution at the C-2' position had never, in fact, been explored. (Radtke 435:15-25; 466:13-467:15, 470:13-25.) Moreover, because the monofluorinated nucleoside analogs that were contained in the prior art exhibited less in vitro cytotoxicity than Ara-C, which has no fluorine atoms, fluorination in general remained of questionable desirability. (E.g. PTX 587; Shewach 789:12-790:20)

Viewed more broadly, the record clearly indicates that a person of ordinary skill in the art would have known that the category of nucleoside analog into which gemcitabine falls is notoriously unpredictable. As the Federal Circuit has held, the suggestion that one chemical compound would have been obvious from a different but structurally similar compound generally rests on the assumption that small changes would not materially alter the properties of the parent compound. In re Soni, 54 F.3d 746, 750 (Fed.Cir.1995). However, where the evidence shows that small changes alter the resulting properties significantly, that assumption is refuted, and a case for obviousness by structural similarity is greatly diminished. Papesch, 315 F.2d at 391.

At the time of Lilly's invention of gemcitabine, it was a common perception that fluorination often had drastic effects on the chemical properties of nucleoside analogs. (Radtke 434:5-435:2; Bergstrom 935:15-936:22.) Dr. Bobek wrote in a grant application prepared contemporaneous to Lilly's invention of the claims in the '614 patent that "it is very difficult to predict the biological behavior of e.g. 2',2'-gem-difluorocytidine." (PTX 498 at Bobek 69.) Dr. Bergstrom corroborated this view, testifying that one "can't predict" whether a CF2 group would have the positive effect on biological activity that nucleoside researchers were hoping to attain. (Bergstrom 892:3-5.) Dr. Bergstrom further wrote that the activity of a nucleoside modified by fluorination "may be significantly different" and admitted that "the CF2 group may behave quite different in its effect on the structure and ultimately on biological activity." (PTX 533 at Bergstrom.) Additionally, Dr. Rustum acknowledged his surprise at the effectiveness of gemcitabine, given the unpredictable effects of fluorination. (Rustum 1617:9-18.)

Based on this evidence, it is clear that one of ordinary skill in the art at the time would have known that even minor changes in the structure of a nucleoside analog were capable of unpredictable effects on biological activity. (Radtke 434:5-435:14, 473:16-22; PTX-533 at Bergstrom 21.) Therefore, Defendants cannot prove obviousness simply by claiming that gemcitabine, by virtue of its structural similarity with other compounds, would have had reasonably expected properties. Soni, 54 F.3d at 750; see also Takeda Chem., 492 F.3d at 1356.

Finally, the record demonstrates that Defendants have overstated the promising antiviral and anticancer attributes offeredand thus the motivation engendered-by Ara-C, 2'-F-cytidine, and 2'-F-ara-C. Although Ara-C was "good as an anticancer drug," its usefulness was limited to leukemia, and not to the solid tumors that gemcitabine is used to treat. (Radtke 541:20-22.) Moreover, although Ara-C had demonstrated antiviral activity in some regard, this activity was never successfully shown in clinical tests, and a 1973 clinical study clearly discouraged the use of the compound as an antiviral agent. (PTX 473 at 1281.)

Defendants' contentions that 2'-F-cytidine and 2'-F-ara-C were promising compounds is similarly misplaced. Prior to the invention dates of the asserted claims, 2'-F-ara-C and 2'-F-cytidine had been reported in the prior art literature. Dr. Rustum testified that 2'-F-cytidine was reported to have properties that could lead to improved antitumor activity over Ara-C. (Rustum 1533:21-1534:3.) However, both compounds were significantly less cytotoxic in vitro than Ara-C. (Shewach 789:12-790:4; Radtke 445:5-446:11; PTX-583 at 205; PTX-587 at 23.) Furthermore, neither compound showed promising antiviral activity, and neither was reported in the prior art to have been tested in vivo. (Shewach 752:25-753:10; Radtke 440:25-441:6, 451:1-452:8.) Because of such deficiencies, these compounds were bypassed in favor of other compounds in research conducted in the years prior to the filing of the patents in suit. (Radtke 44:25-441:6, 454:24-455:6.) Therefore, any motivation engendered by those compounds was, at best, minimal.

"Patents are not barred just because it was obvious to explore a new technology or general approach that seemed to be a promising field of experimentation, where the prior art gave only general guidance as to the particular form of the claimed invention or how to achieve it." Procter & Gamble Co. v. Teva Pharms. USA, Inc., 566 F.3d 989, 997 (Fed.Cir.2009) (quotation omitted). Dr. Hertel was clearly following the "general guidance" of the field. Defendants have adduced no evidence, however, that he was motivated by those three structurally similar compounds to create gemcitabine, nor does the evidence adduced support a finding that one of ordinary skill would have been so motivated.

For all of these reasons, Defendants have failed to establish "obviousness by structural similarity" with regard to the '614 patent.

B. The '826 Patent

Defendants also contend that the invention of the '826 patent would have been obvious based on gemcitabine's structure and known antiviral properties, as well as its similarity to Ara-C, 2'-F-ara-C, and 2'-F-cytidine. They stress that, whereas obviousness in the context of the '614 patent depended upon the motivation to create gemcitabine, the question of obviousness as it relates to the '826 patent focuses on whether the person of ordinary skill in the art would have been motivated to use gemcitabine, which had already been invented, as an anticancer agent.^[38]

1. The Difficulty of Discovering Antitumor Activity

As an initial matter, it is important to note the context of Defendants' obviousness contention, i.e., the particularly challenging field of designing and discovering anticancer drugs. As Dr. Rustum acknowledged, such drug design is "a very difficult process."^[39] This is illustrated by the experience of researchers at Roswell Park, which has never advanced a new chemical beyond early phase testing in an anticancer trial. (Rustum 1576:2-1578:11; see also Bobek 1414:24-1415:12.) Indeed, a 1979 article indicated that the "diligence of numerous investigators in cancer research as well as the difficulty of the problem are attested to by the fact that out of the nearly 300,000 compounds that have been evaluated as anticancer agents 40 or so are today considered useful...." (PTX 692 at 3; Shewach 744:8-745:8.) This background is critical to an understanding of what the "reasonable expectations" of a person of ordinary skill in the art would have been in relation to gemcitabine's potential anticancer utility.

Defendants' obviousness claim relating to the '826 patent is divided into two parts: First, Defendants argue that the antiviral properties reported for Ara-C and 2'-F-ara-C would have led a person of ordinary skill to expect anticancer activity in gemcitabine, which had previously shown antiviral activity. Second, they contend that the prior art data on 2'-F-ara-C and 2'-F-cytidine pointed directly toward gemcitabine's potential for antitumor activity.

2. The Relationship Between Antiviral and Antitumor Activity

Defendants first contend that the antitumor activity of gemcitabine would have been obvious in light of its known antiviral activity. (Rustum 1536:14-19.) However, the prior art taught that antitumor activity could not necessarily be predicted from antiviral activity. (E.g. Shewach 749:18-750:14; PTX 721 at 567.) As the Fox Group at Sloan-Kettering discovered, often high antiviral activity correlated with low cytotoxicity. (Shewach 751:14-752:111; PTX 587 at 22.)

At certain points, Defendants concede that a general "inverse relationship" exists between antiviral activity and cytotoxicity. Relying on the Watanabe reference,^[40] however, they contend that this relationship did not apply with regard to 2'-F-ara-C, which Watanabe reported to have both antiviral and anticancer activity (in other words, a "direct relationship" between the two). According to Defendants, a person of ordinary skill in the art would have read Watanabe as creating a motivating expectation that a "direct relationship" between antiviral and anticancer activity would be present in gemcitabine because of that compound's similarity to 2'-F-cytidine.

Defendants far overstate the significance of this Watanabe data. Considering the entirety of the reference in the proper context, it is clear that Watanabe's findings would not have actually motivated a person of ordinary skill to develop gemcitabine for anticancer activity because 2'-F-ara-C was not the most promising antitumor agent of the compounds tested in that study. (PTX 587 at 22.) In fact, the prior art, including Watanabe and other studies, contained no significant data for 2'-F-cytidine's anticancer activity. Not only does the prior art not report any in vivo activity for 2'-F-ara-C, (Radtke 453:3-10, 454:13-18; Shewach 752:25-753:10; Herdewijn 1067:17-22.), but also Table II of Watanabe reported that this analog was two and one-half times less cytotoxic than Ara-C.

Indeed, the text of Watanabe does not suggest that 2'-F-ara-C was the most promising compound for any purpose. Watanabe instead selected FIAC and FMAU as the most "potent and selective" compounds and proposed them for in vivo testing. (PTX 584 at 27, 30, 36, 38; Radtke 454:19-455:6.) Defendants have not adduced any evidence suggesting why a person of ordinary skill would have disregarded the textual teaching of Watanabe and the mediocre cytotoxicity data for 2'-F-ara-C contained therein.^[41] Defendants' reliance on Watanabe is clearly an attempt to reconstruct the data available to the person of ordinary skill in the art at the time with the benefit of hindsight, which is inappropriate. Interconnect Planning, 774 F.2d at 1143.

Because no evidence suggests that a person of ordinary skill in the art would actually have had a reasonable expectation of gemcitabine's antitumor properties based on Watanabe or other prior art in 1984, Defendants have failed to establish the obviousness of gemcitabine's antitumor activity based on its established antiviral activity.

3. The Cytotoxicity of 2'-F-ara-C and 2'-F-cytidine

Defendants also attempt to prove that 2'-F-ara-C and 2'-F-cytidine had antitumor activity, including the capacity to inhibit RNR and the potential for decreased deamination, that would have motivated the person of ordinary skill in the art to use gemcitabine as an antitumor agent.

The Stubbe paper^[42] taught that 2'-F-cytidine was an RNR inhibitor, giving it potentially improved antitumor activity over Ara-C. (DTX 252; Rustum 1535:16-20.) Thus, according to Defendants, because, like 2'-F-cytidine, gemcitabine has a fluorine atom in the 2'-down position, a person of ordinary skill in the art would have been motivated to use gemcitabine to inhibit RNR. (Rustum 1546:5-19; Herdewijn 1023:20-1024:4.)

In advancing this argument, however, Defendants mischaracterize the results reported in the Stubbe paper. Dr. Stubbe reported that another compound, 2'-deoxy-2'-fluoroadenosine, was fifty times more effective as an inhibitor than 2'-F-cytidine. (Shewach 782:9-783:1; PTX 611 at 5511.) Furthermore, Defendants ignore another prominent study, the Brox paper,^[43] which provided evidence against 2'-F-cytidine's capacity to inhibit RNR to a significant degree. (Shewach 792:3-15; PTX 593 at 1841; LDX 8.38.) Indeed, when Dr. Bergstrom later created a slide entitled "Biologically Significant Fluronucleosides," he included FIAC and 2'-deoxy-2'-fluoroadenosine, but he made no mention of 2'-F-cytidine or 2'-F-ara-C. (PTX 541 at 3257; Herdewijn 1118:24-1119:11.)

Therefore, a person of ordinary skill in the art would not have concluded, based on the Stubbe paper, that 2'-F-cytidine was an especially potent inhibitor of RNR.^[44] Rather, a person of ordinary skill would have drawn a conclusion similar to the one drawn by Dr. Bergstrom, who expressed a motivation to develop 2'-deoxy-2'-fluoroadenosine 5'-diphosphate, the molecule clearly emphasized in the Stubbe paper. (PTX 534 at Bergstrom 60; PTX 541 at Bergstrom 3257; Bergstrom 901:19-22, 940:25-941:18; Herdewijn 1118:24-1119:11; PTX 611.)

Aside from the Stubbe reference, the record at trial clearly shows that researchers at the time were not motivated by 2'-F-cytidine to propose difluoronucleoside analogs like gemcitabine as anticancer agents. For example, the researchers at Merrell Dow as well as Dr. Bergstrom never expressed any expectation that 2',2'-difluoronucleosides would inhibit RNR, only that they "hoped" this activity, among a list of many possible activities, would be present. (Metcalf Dep. 34:11-12; Farr Dep. 27:6-13; PTX 534 at Bergstrom 47.)^[45] Defendants have, at most, successfully demonstrated by clear and convincing evidence that a person of ordinary skill would have been uncertain as to what to expect of a compound like gemcitabine based on the results in the prior art. The record contains no clear prior or contemporaneous teaching toward the use of gemcitabine to treat tumors.

Finally, the unpredictability of difluoronucleoside chemistry further undermines Defendants' obviousness claim. As the prior art taught, the presence of two fluorine atoms on the same carbon atom materially alters the structure and therefore the biological activity of the molecule. (Bergstrom 936:24-937:11; PTX-533 at Bergstrom 21.) Thus, even assuming arguendo that 2'-F-ara-C or 2'-F-cytidine showed the promise urged by Defendants, the biological effects produced by these monofluorinated molecules would not have provided a reasonable example of what to expect in a difluorinated molecule like gemcitabine. (Bergstrom 935:15-936:22.)^[46]

For the foregoing reasons, Defendants have failed to demonstrate that either the antiviral activity or the anticancer activity of compounds contained the prior art would have created a reasonable, motivating expectation of such activity in gemcitabine. Therefore, Defendants' claim of obviousness based on this theory fails.

C. Objective Evidence of Nonobviousness

A court is obligated to consider objective evidence of nonobviousness when such evidence is present. Custom Accessories, Inc. v. Jeffrey-Allan Indus., Inc., 807 F.2d 955, 960 (Fed.Cir.1986); Knoll Pharm. Co. v. Teva Pharms. USA, Inc., 367 F.3d 1381, 1385 (Fed.Cir.2004). In the case at bar, Lilly has demonstrated a nexus between the claimed inventions and various secondary considerations, which we therefore consider. Brown & Williamson Tobacco Corp. v. Philip Morris, 229 F.3d 1120, 1130 (Fed.Cir.2000).^[47]

1. Gemcitabine's Unexpected Properties

An invention's unexpected positive properties support the conclusion that the invention would not have been obvious. Procter & Gamble, 566 F.3d at 997. Gemcitabine showed numerous capabilities that were not hinted at in the prior art, including: (1) superior accumulation and retention in cancer cells; (2) enhanced interference with DNA synthesis; and (3) enhanced inhibition of RNR. (Shewach 754:4-764:8, 785:1-786:10.) Furthermore, gemcitabine was particularly effective against solid tumors, a surprising characteristic in light of the prior art, and the compound demonstrated a superior survival advantage for patients with advanced pancreatic cancer when compared to 5-FU, the previous standard of care. (Green 178:5-15; PTX 436 at 2403.)

Defendants' expert, Dr. Isacoff, testified that gemcitabine is no better than 5-FU in treating patients with advanced pancreatic cancer and that Lilly's findings that gemcitabine presented an improved treatment were based on faulty experimentation. (Isacoff 1433:5-7.) His testimony is directly refuted, however, by the review of the FDA and the Oncologic Drug Advisory Committee, which approved Lilly's experimental method. (Green 203:8-21; Pearce 115:12-21, 119:8-10). Further undermining his assertion is the fact that the National Comprehensive Cancer Network Guidelines in Oncology determined that gemcitabine monotherapy is the standard of care, whereas 5-FU is merely "salvage therapy." (Green 206:15-207:8, 227:5-10; PTX 480 at MS-14.)^[48]

Furthermore, Defendants' contention that gemcitabine does not offer survival benefits and various other surprising properties is unconvincing in light of their admitted desire to market and sell gemcitabine as a pancreatic cancer treatment. The evidence adduced at trial clearly shows that gemcitabine exhibited many surprising and superior characteristics in view of the prior art, which support a finding that the invention would not have been obvious. See Knoll Pharm., 367 F.3d at 1385.

2. Long-Felt Need

Evidence of a long-felt and unsolved need for the solution offered by the patented invention supports a showing of nonobviousness. Tec Air., Inc. v. Denso Mfg. Mich., Inc., 192 F.3d 1353, 1361 (Fed. Cir.1999). Based on the previously discussed opinion of Dr. Isacoff that gemcitabine offered little or no improvement over 5-FU, Defendants contend that gemcitabine did not meet any need. Once again, the choice by the National Comprehensive Cancer Network to make gemcitabine, and not 5-FU, the standard of care for the treatment of pancreatic cancer seriously undermines Defendants' position. Although gemcitabine did not present a cure for pancreatic cancer, it was the first drug in thirty years to produce an improvement in overall survival. (Green 202:7-12; PTX 436 at 2412.) Indeed, gemcitabine remains the standard of care at present and is generally regarded as the most effective drug for the treatment of advanced pancreatic cancer. (Green 179:12-20; PTX 456 at 267-73; PTX 480 at MS-14.) Accordingly, it is clear that gemcitabine met a long-felt need, a finding that lends strong support to the conclusion that the anticancer invention claimed in the '826 patent would not have been obvious.

3. Commercial Success

The commercial success of an invention is evidence of its nonobviousness. Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275, 279, 64 S. Ct. 593, 88 L. Ed. 721 (1944); Al-Site Corp. v. VSI Int'l, Inc., 174 F.3d 1308, 1325-26 (Fed. Cir.1999). Although the patentee generally bears the burden of establishing a nexus between commercial success and the merits of the patented invention, this nexus is presumed where, as here, "the marketed product embodies the claimed features, and is coextensive with them." Brown & Williamson, 229 F.3d at 1130 (citing J.T. Eaton & Co. v. Atl. Paste & Glue Co., 106 F.3d 1563, 1571 (Fed.Cir.1997); Demaco Corp. v. F. Von Langsdorff Licensing Ltd., 851 F.2d 1387, 1392-93 (Fed.Cir.1988)); (see PTX 443; Green 217:9-19 (explaining that Gemzar® has been approved solely as an anticancer agent and that gemcitabine is its sole active ingredient).)

Worldwide sales of Gemzar®, the commercial name of gemcitabine, have exceeded $1 billion annually since 2003. (Grabowski Rep. ¶¶ 14, 35, Ex. 2.) The drug has captured over 90 percent of the market of patients with advanced pancreatic cancer, as well as over 50 percent of patients with invasive bladder cancer. Defendants contend that Lilly's marketing expenditures rebut the nexus between commercial success and the benefit of the claimed invention. However, Gemzar®'s ratio of marketing expenditures to sales revenues is not excessive relative to its direct competitors. (Grabowski Rpt. ¶¶ 18, 48, 49, Ex. 6A, 6B.) Moreover, those expenses do not exceed the average ratio for other oncology drugs. (Id.) Accordingly, Lilly's marketing expenditures cannot rebut the nexus between Gemzar®'s benefit and its commercial success.

The trial record makes clear that the singular benefits made available by gemcitabine have led to commercial success, a fact further supported by Defendants' current attempts to market the drug themselves. Eli Lilly & Co. v. Zenith Goldline Pharms., Inc., 2001 WL 1397304, at *12 (S.D.Ind. Oct. 29, 2001) ("If the patented drug were not a commercial success, generic manufacturers would have little interest in offering their own versions of the drug.") "[P]articularly telling" in the case at bar is the fact that Defendants hope to sell Gemzar®, despite already selling a generic version of 5-FU, the former standard of care for pancreatic cancer. (Lee 73:10-11); see Forest Labs. v. Ivax Pharmaceuticals, Inc., 438 F. Supp. 2d 479, 496 (D.Del.2006). Accordingly, the commercial success of Gemzar® provides additional objective evidence of the nonobviousness of gemcitabine.

4. The Failure of Other Nucleoside Analogs

Evidence that others in the art failed despite concerted efforts to develop inventions similar to the patentee's invention is also recognized as objective evidence of nonobviousness. Knoll, 367 F.3d at 1385. As of the time of the filing of the application in 1984, thousands of nucleoside analogs had been subjected to research and testing, and only one, 5-FU, had been approved by the FDA for use against solid tumors. (Green 179:21-180:5.) Moreover, that single drug had not been approved for the primary utilization of Gemzar®. (Green 179:16-20.) This track record of failure prior to Lilly's success with gemcitabine provides clear support for a finding of nonobviousness.

5. Copying of Gemcitabine

A showing of nonobviousness is also supported by evidence of copying. Diamond Rubber Co. of N.Y. v. Consol. Rubber Tire Co., 220 U.S. 428, 441, 31 S. Ct. 444, 55 L. Ed. 527 (1911). By the year 2000, forty nucleoside analogs had been reported in the literature imitating the geminal difluoro substitution at the C-2' position found in gemcitabine. (PTX 620 at 87.) This evidence of copying also objectively supports a finding of nonobviousness.

The secondary, objective evidence of nonobviousness adduced at trial—including gemcitabine's unexpected properties, its capacity to fill a long-felt need, its commercial success, similar compounds' failure to demonstrate successful properties, and other researchers' endeavors to copy gemcitabine—reinforces the Court's conclusion that the claimed inventions would not have been obvious.

VI. Enablement

The statutory basis for the enablement requirement is found at 35 U.S.C. § 112, which provides in relevant part:

The specification shall contain a written description of the invention and of the manner and process of making and using it, in such full, clear, concise and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same.

The "enablement requirement is satisfied when one skilled in the art, after reading the specification, could practice the claimed invention without undue experimentation." Sitrick v. Dreamworks, LLC, 516 F.3d 993, 999 (Fed.Cir.2008) (quoting AK Steel Corp. v. Sollac, 344 F.3d 1234, 1244 (Fed.Cir.2003)). "[A] considerable amount of experimentation is permissible, if it is merely routine, or if the specification in question provides a reasonable amount of guidance with respect to the direction in which experimentation should proceed." PPG Indus. Inc. v. Guardian Indus. Corp., 75 F.3d 1558, 1564 (Fed.Cir. 1996) (quoting Ex Parte Jackson, 217 U.S.P.Q. 804, 807 (1982)). Enablement is determined as of the filing date of the patent application. Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384 (Fed.Cir.1986).

The Federal Circuit has explained that "the how to use prong of section 112 incorporates as a matter of law the requirement of 35 U.S.C. § 101 that the specification disclose as a matter of fact a practical utility for the invention." In re Cortright, 165 F.3d 1353, 1356 (Fed.Cir. 1999) (quoting In re Ziegler, 992 F.2d 1197, 1200 (Fed.Cir.1993)). There is a lack of utility under § 101 "when there is a complete absence of data supporting the statements which set forth the desired results of the claimed invention." 165 F.3d at 1356 (quoting Envirotech Corp. v. Al George, Inc., 730 F.2d 753, 762 (Fed.Cir. 1984)).

With the expert testimony provided by Dr. Kenneth Tew, Defendants contend that the disclosure of the '826 patent does not enable claim 7. According to Dr. Tew, because undue experimentation, in the form of clinical trials, is necessary to determine which tumors would be susceptible to gemcitabine, claim 7 is not enabled. (Tew 1264:3-9; 1282:15-1283:22.) Dr. Tew testified at length about the extensive work that must be done to design a clinical trial, stating that such a process is "invariably complex" and can cost "tens of millions of dollars." (Tew 1254:23-1256:12, 1261:4-1263:22, 1265:3-1266:2.) Dr. Tew did not, however, cite any references to support many of his assertions, which casts doubt upon those averments. See Upjohn, 225 F.3d at 1311.

Furthermore, the fact that experimentation may be complex does not necessarily make it undue. If the art typically engages in such experimentation, it is not undue, and the enablement requirement is met. In re Wands, 858 F.2d 731, 737 (Fed.Cir.1988); PPG Indus., 75 F.3d at 1564. Although Dr. Tew is correct in asserting that clinical trials were necessary to determine which neoplasms would be susceptible to gemcitabine, such trials were standard and routine to a person of ordinary skill in the art in 1984. (Green 223:17-25.) During his testimony, Dr. Tew admitted as much. (Tew 1286:6-11 (stating, with regard to the design of clinical trials, "that's what pharmacologists and clinical oncologists do").)

Moreover, as the evidence at trial established, and Dr. Tew admitted, the approach to designing clinical trials is "well documented" in the literature. (Tew 1284:9-25, 1285:1-17.)^[49] Where, as here, it requires only routine experimentation to determine whether a given embodiment falls within the scope of the claim, it is axiomatic that the claim is enabled. PPG Indus., 75 F.3d at 1564; In re Angstadt, 537 F.2d 498, 504 (C.C.P.A.1976). Therefore, Defendants have failed to show by clear and convincing evidence that claim 7 of the '826 patent was not enabled.

VII. Conclusion

For all of the reasons explicated above, the Court hereby declares that:

(1) Because Lilly is collaterally estopped from defending against Defendants' claim of obviousness-type double patenting, claim 7 of the '826 patent must be declared invalid on that basis, subject to a final determination of the companion case out of the Eastern District of Michigan now pending before the Federal Circuit.

(2) Defendants have failed to prove by clear and convincing evidence that either claim 12 of the '614 patent or claim 7 of the '826 patent is invalid as anticipated under 35 U.S.C. § 102; invalid as obvious under 35 U.S.C. § 103; or invalid for lack of enablement under 35 U.S.C. § 112; and unenforceability due to inequitable conduct has neither been alleged nor proved by Defendants. The asserted claim of the '614 patent is, therefore, neither invalid nor unenforceable.

(3) Defendants stipulated that if the asserted claims of the '614 patent and '826 patent are not invalid or unenforceable, then its actions regarding its gemcitabine products that are the subject of ANDA Nos. 090644, 77-961, and 77-983 constitute infringement of claim 12 of the '614 patent. The Court therefore finds that Defendants have infringed and threatened to infringe in the future claim 12 of the '614 patent.

Accordingly, it is hereby ORDERED THAT:

(1) Pursuant to 35 U.S.C. § 271(E)(4)(A), the effective date of any approval of the gemcitabine products that are the subject of ANDA Nos. 090644, 77-961, and 77-983 SHALL NOT BE a date that is earlier than the latest date of expiration of Lilly's '614 patent.

(2) Pursuant to 35 U.S.C. § 283 and 35 U.S.C. § 271(e)(4)(B), Defendants and their officers, agents, servants, employees, privies, and other acting for, on behalf of, or in concert with them or either of them are PERMANENTLY ENJOINED from the commercial manufacture, use, offer to sell, or sale within the United States, or importation into the United States, of the gemcitabine products that are the subject of ANDA Nos. 090644, 77-961, and 77-983, or any gemcitabine product that is not colorably different therefrom, prior to the date of expiration of Lilly's '614 patent. This permanent injunction order is effective immediately upon the entry of this ruling. Final judgment shall enter accordingly.

IT IS SO ORDERED.

NOTES

[1] When used in this order, "PTX" refers to Lilly's Trial exhibits; "DTX" refers to Defendants' trial exhibits; "LDX" refers to Lilly's demonstrative exhibits; and "TDX" refers to Defendants' demonstrative exhibits.

[2] After the present action was originally filed, Sicor Pharmaceuticals, Inc. changed its name to Teva Parenteral Medicines, Inc.

[3] That decision, Sun Pharm. Indus., Ltd. v. Eli Lilly & Co., 647 F. Supp. 2d 820 (E.D.Mich. 2009), is currently awaiting a decision on appeal.

[4] Defendants have moved to Admit into Evidence certain Deposition Designations and Exhibits Referenced Therein [Docket No. 355]. In order to streamline burdensome discovery, throughout pretrial and trial the parties negotiated agreements as to what evidence would be offered for admission. Outside of these agreements, on the final day of trial and thereafter, Defendants offered various additional exhibits for admission, (DTX F), but Lilly objects to the admission of certain of these submissions: DTX 55, 86, 149, 165, 845, and PTX 295-305, 307. Lilly does not object to the admission of the following: DTX 56, 145, 160, 180, 840, 843, 844, and 847. Defendants contend that Lilly's objections were belated, but the record makes clear that the late timing of the objections was the result of the late filing of the evidence by Defendants. It appears from the record that the parties did not, in fact, agree to admit this evidence. Moreover, this evidence is not, as Defendants contend, independently admissible. It is, by and large, unauthenticated or irrelevant, in particular because it lacks the support of introducing witness testimony. See Fed.R.Evid. 401, 602. Accordingly, Defendants' Motion to Admit is DENIED with regard to DTX 55, 86, 149, 165, 183, 845, and PTX 295-305, 307, but it is GRANTED as to DTX 56, 145, 160, 180, 840, 843, 844, and 847.

[5] Defendants have also filed a Motion to Strike, or in the Alternative, Motion for Leave [Docket no. 367], requesting that the Court strike Lilly's Counter-Findings of Fact and Counter-Conclusions of Law [Docket No. 361], or in the alternative, requesting leave to file their own Counter-Findings and Conclusions. At the close of trial, the Court established a schedule for the exchange of post-trial briefs. [Docket No. 339]. That order required the parties to "file their opening briefs on or before Friday, November 6, 2009," and "their response briefs on or before Monday, November 30, 2009." Both parties filed proposed Findings of Fact and Conclusions of Law along with their opening briefs, despite the fact that the scheduling order made no mention of such Findings and Conclusions, thus demonstrating an understanding that, in the case at bar, such a filing would accompany any "brief." Nonetheless, only Lilly filed Findings and Conclusions in accompaniment with its Response Brief. Defendants have not explained their decision not to file Findings and Conclusions with their Response Brief, and it would be unfairly prejudicial to allow them, now, after seeing Lilly's Counter-Findings, to file responsive Findings and Conclusions out of time. Accordingly, Defendants' Motion to Strike is hereby DENIED.

[6] We note that the analysis regarding the '826 patent would be the same if the definition of a person of ordinary skill in the art as proposed by Drs. Rustum or Herdewijn, as opposed to the standard proposed by Drs. Green and Radtke, was adopted. (See Shewach 747:14-18; Rustum 1512:13-20.)

[7] Dr. Radtke explained that fluorine's powerful electronegativity works like a magnet, deactivating the carbon atoms on either side of the carbon to which it is attached. (Radtke 434:5-435:2.)

[8] Drs. Radtke, Herdewijn, and Rustum agreed that the prior art did not contain any examples of biologically active nucleoside analogs with geminal substitutions at the C-2' position, let alone a difluoro substitution at the C-2 position. (Radtke 435:15-25; 437:18-438:6; Herdewijn 1092:15-19; Rustum 1605:2-11.)

[9] Dr. Youcef Rustum, an expert witness offered by Defendants, testified with regard to anticancer agent development that "the success rate is very low. So out of 10,000 or 20,000 compounds tested here and there, you may have one compound make it to the market." (Rustum 1576:7-9.)

[10] RNR is critical to DNA replication and is essential for cell proliferation. If RNR is inhibited, DNA cannot be replicated and cell proliferation is prevented. (Rustum 1529:9-1530:9.)

[11] A subsequent publication reported that Fox Laboratory had, in fact, tested 2'-F-ara-C in vivo but that the compound showed "little antitumor activity." (Shewach 752:25-753; Rustum 1602:3-1603:15 PTX 620 at 90.)

[12] These compounds progressed to studies in animals and were promising enough at that time to draw considerable attention from Lilly. (PTX 584 at 27.)

[13] As discussed in the Conclusions of Law, Defendants attempted to show at trial that Dr. Grindey, not Dr. Hertel, originally proposed using gemcitabine as an antitumor agent at Lilly and that he did so based on knowledge gained from Dr. Bobek.

[14] DAST is a fluorinating agent, which as its name and the evidence provided by both parties' experts suggests, is a compound utilized in fluorine substitution. (DTX 95; Herdewijn 1032:16-22; Overman 619:5-10.)

[15] Lilly informed the PTO that the Bergstrom Abstract "suggests 2',2'-difluoro structures as well." (PTX 6 at 52.)

[16] Dr. Metcalf also conceived, generally, of a proposed structure for the anticancer agent he hoped Merrell Dow would produce. (DTX-151 at SA 2208.)

[17] The following references are cited on the face of the '614 patent: Watamane (1979) (PTX-587); U.S. Patent No. 4,211,733 ("'773 patent"); the Bergstrom Abstract (PTX-518); Adamson (1971) (PTX-581); Fox (1981) (PTX-584); and Penglis (1981) (PTX-648).

[18] The following references are cited on the face of the '826 patent: Brox (1975) (PTX-593); Adamson (1971) (PTX-581); and Goodman and Gilman (1980) (PTX-473).

[19] During the prosecution of this patent, Lilly reported mouse tumor model data, which led Defendants' expert, Dr. Tew, to conclude that the patent was not enabled because determining all neoplasms susceptible to gemcitabine in all mammals would require "undue experimentation." (Tew 1251:23-1252:17.) This opinion is belied by the nature, use, and development of the patent. As Dr. Pearce testified, it was standard practice to investigate potential anticancer agents in mouse tumor models. (Pearce 80:19-81:10). The fact that Lilly so tested gemcitabine is not an indicator that it intended to use gemcitabine to treat mice with cancer, let alone any mammal other than humans with cancer. It merely reflects the fact that Lilly was following standard procedure for the development of a drug to be used in humans.

[20] At the time Lilly was developing gemcitabine, 5-fluorouracil was the accepted standard of care for advanced pancreatic cancer. (Pearce 116:1-10.)

[21] The FDA approved gemcitabine for the treatment of patients with pancreatic cancer based on the results of a multicenter, randomized, blinded, and controlled phase III study, identified at Lilly as B9E-MC-JHAY ("the JHAY study"). (PTX 274.) The results of the JHAY study were published in Burris et al, Journal of Clinical Oncology 15(6):2403-13 (1997). (PTX 436.)

[22] Defendants' expert Dr. Isacoff criticizes the results of this study, (e.g. Isacoff 1433:5-7), but the record of evidence at trial, including other expert testimony, convinces the Court of the soundness of the study. (Pearce 115:12-21, 119:8-20; Green 203:8-204:3, 206:15-207:8, 227:5-10; PTX 480 at PANC-D, MS-14; LDX 14.4-14.5.) Of particular relevance is the fact that Dr. Isacoff himself prescribes gemcitabine to half of his patients diagnosed with advanced pancreatic cancer. (Isacoff 1462:13-16.)

[23] To date, only eleven nucleoside drugs, including gemcitabine, have been approved by the FDA for the treatment of cancer. (Green 179:21-25.)

[24] Defendants have conceded that their proposed product is a generic "copy" of Gemzar® that would infringe both the '614 patent and the '826 patent.

[25] Perhaps because of this commercial success, gemcitabine, the first nucleoside analog with a geminal difluoro substitution at the C-2' position, has engendered a relative boom in attempts to produce such nucleosides, with 40 having been reported in the literature by the year 2000. (PTX 620 at 87.) Of these, only gemcitabine has received FDA approval. (Id. at 88.)

[26] The Court takes judicial notice of the fact that Lilly has appealed the judgment issued against it in Sun Pharm. Indus., Ltd. v. Eli Lilly & Co., 647 F. Supp. 2d 820 (E.D.Mich. 2009), and that the matter is currently pending before the Federal Circuit. We assume that the parties will return for reconsideration of our decision that collateral estoppel applies, should the Federal Circuit overturn Sun Pharmaceutical. Although we make no prediction as to the outcome of that case on appeal, we note preliminarily that the trial record before this Court in our judgment largely supports Lilly's substantive contentions relating to obviousness-type double patenting.

[27] The decision in Sun Pharmaceutical did not address any of the remaining issues presented before this Court, including anticipation, obviousness, and enablement, as they relate to the '826 patent. Because the parties have maintained their dispute over these issues, and have not withdrawn their arguments in light of the Sun Pharmaceutical decision, we address these issues fully on the complete trial record before the Court.

[28] This averment is unconvincing because Dr. Grindey's curriculum vitae indicates that he was not on the American Cancer Society grant committee in 1975. (DTX 78 at GM 194054.)

[29] Further, Defendants' argument regarding the timing of Dr. Metcalf's alleged conception rests solely on his uncorroborated deposition testimony, which is insufficient as a matter of law. Finnigan Corp. v. Int'l Trade Comm'n, 180 F.3d 1354, 1369 (Fed.Cir.1999).

[30] We note that, while the original text of the '614 patent is available for assessing obviousness under 35 U.S.C. § 103 as of its March 1983 effective filing date, it is not evidence of what was generally known in the art as of that date and may not be used as of that date to supplement what are otherwise inadequate disclosures or priority proofs of others. In re Glass, 492 F.2d 1228, 1231-32 (C.C.P.A. 1974).

[31] Moreover, Defendants' reliance upon Dr. Metcalf as the prior conceiver of the anticancer utility is undercut and contradicted by the fact that, when Merrell Dow did finally file a patent application on their difluoronucleoside anticancer work, the patent application did not list Dr. Metcalf as an inventor. (PTX 568; DTX 895 at SA 1448.)

[32] Just as Defendants clearly failed to show at trial that Lilly derived the invention of gemcitabine from Dr. Bobek, they also failed to show that Dr. Bobek's purported disclosure of gemcitabine anticipated Lilly's invention of the compound. See supra § IV.A.

[33] For a compound to have been obvious, a person of skill in the art must have been able to synthesize that compound without undue experimentation. Accordingly, naming or sketching that compound is insufficient. See In re Brown, 51 C.C.P.A. 1254, 329 F.2d 1006, 1010-11 (1964); In re Wiggins, 488 F.2d 538, 543 (C.C.P.A.1973); In re Samour, 571 F.2d 559, 563 (C.C.P.A. 1978). Therefore, although proposed sketches of gemcitabine were contained in contemporaneous research, the experimental failures of all researchers outside of Lilly demonstrate the nonobviousness of the asserted claims.

[34] Casting further doubt on Dr. Herdewijn's opinion regarding the synthesis methods is the fact that he contended that DAST could be used to make the C-2 difluorinated sugar precursor of gemcitabine, (Herdewijn 1032:16-22), ignoring a clear admission to the contrary by Dr. Bergstrom: "Attempts at direct fluorination of 2-ketofuranose derivatives using DAST to afford the corresponding 2,2-difluoro derivatives have not been successful." (PTX 535 at 362.) Indeed, Dr. Herdewijn acknowledged, in a separate portion of his testimony, that the DAST approach failed when attempted at other positions on the sugar, and he offered no explanation as to why the C-2 position would be different. (Herdewijn 1058:10-14, 1059:15-1060:10.) "[T]here must be some factual support for an expert's conclusory opinion." Upjohn Co. v. Mova Pharm. Corp., 225 F.3d 1306, 1311 (Fed.Cir.2000).

[35] Moreover, Defendants' contention that only a nucleoside chemist had the skill to make gemcitabine is directly contradicted by the facts. Drs. Bobek and Bergstrom, both nucleoside chemists, failed to make the compound, whereas Dr. Hertel, who had a background other than that in nucleoside chemistry, was the first to succeed in synthesizing gemcitabine. (Hertel 353:13-15 ("[W]hat I brought to the table, not having a background in nucleoside chemistry, was that I didn't have blinders on how you need to make nucleosides.").)

[36] Like gemcitabine, Ara-C is an analog of the naturally occurring nucleosides cytidine and deoxycytidine.

[37] These compounds differ from ara-C in only one respect: they each have a fluorine atom at the 2' position.

[38] Because the "effective filing date" of the '614 patent is March 1983, its use as prior art for the '826 patent, although acceptable under 35 U.S.C. § 102, is limited to the original application text, which does not include the later-discovered reference to the anticancer activity of gemcitabine. In re Lund, 54 C.C.P.A. 1361, 376 F.2d 982, 988 (1967); (see also Rustum 1610:15-18.)

[39] While Dr. Rustum asserted that one of ordinary skill in the art would have reasonably expected gemcitabine to be successful, he supported this theory by asserting (unconvincingly) that every molecule that Dr. Bobek made at Roswell Park, as well as every molecule made by many other chemists, was created and tested with a "reasonable expectation" of success based on what was known from the prior art. (Rustum 1577:19-1578:7.) This claim is contrary to the other, more convincing evidence adduced at trial.

[40] Kyoichi Watanabe et al., Nucleosides. 110. Synthesis and Antiherpes Virus Activity of Some 2'-Fluoro-2'-deoxyarabinofuransylpyrimidine Nucleosides, 22 J. Med. Chem. 21, 21-24 (1977) (DTX 250).

[41] Furthermore, although the Watanabe reference was not before the PTO during prosecution of the '826 patent, (see DTX 2), the '773 patent was considered by the PTO which contained a comparable disclosure. (DTX 223.) Ultra-Tex Surfaces, 204 F.3d at 1367.

[42] J. Stubbe & J.W. Kozarich, Fluoride, Pyrophosphate, and Base Release from 2'-Deoxy-2'-fluoronucleoside 5'-Diphosphates by Ribonucleoside-Diphosphate Reductase, 255 J. Biol. Chem. 5511, 5511-13 (1980) (DTX 252).

[43] L.W. Brox et al., Studies on the Growth Inhibition and Metabolism of 2'-Deoxy-2'-fluorocytidine in Cultured Human Lymphoblasts, 34 Cancer Res. 1838, 1838-42 (DTX 241).

[44] Notably, 2'-F-cytidine and gemcitabine inhibit RNR by way of completely different mechanisms, a fact that confirms the lack of a reasonable expectation of a relationship between the two compounds. (Shewach 783:16-787:6.)

[45] That this rose only to the level of a "hope" may have been because the only evidence suggesting the possibility that 2',2'-difluoronucleosides would exhibit antitumor activity arose out of in vitro testing of cytotoxicity, which is not indicative of the more important goal, in vivo demonstration of anticancer activity. (Rustum 1586:13-15; Shewach 795:23-797:7.)

[46] Further, Defendants admit that nothing in the original application leading to the '614 patent "singled out" gemcitabine from the other nucleoside analogs that were disclosed. (Lee 1652:4-8; Hertel 383:13-384:2, 385:25-386:15; PTX-6). An invention would not have been obvious if one of ordinary skill could only "try each of numerous possible choices until one possibly arrive[s] at a successful result, where the prior at [gives] ... no direction as to which of many possible choices is likely to be successful." Procter & Gamble Co. v. Teva Pharms. USA, Inc., 566 F.3d 989, 996-97 (Fed.Cir.2009).

[47] Defendants contend that Lilly cannot rely upon secondary considerations to show nonobviousness for the '826 patent because the '614 patent gave Lilly a monopoly on all uses of gemcitabine. This misstates the purpose of requiring the disclosure of a patented invention, which is to give the public an opportunity to learn from and improve upon it. This is why the patent statute expressly provides for patents on new uses of old compositions, as well as the "domination" of subsequent improvement patents by earlier basic patents. See 35 U.S.C. §§ 101; 100(b); In re Kaplan, 789 F.2d 1574, 1577-78 (Fed.Cir.1986). The patentability of such improvement inventions is established by objective evidence of nonobviousness, notwithstanding the "domination" of an earlier patent. Eli Lilly & Co. v. Zenith Goldline Pharms. Inc., 471 F.3d 1369, 1379 (Fed.Cir.2006). Further, Defendants' contention that the '614 patent "blocked" the development of gemcitabine as anticancer agent is unsupported; the case Defendants cite in support, Merck & Co. v. Teva Pharms. USA, Inc., 395 F.3d 1364, 1366 (Fed.Cir.2005), dealt with an improvement patent application filed eleven years after the earlier patent had issued. By contrast, here, the '614 patent did not issue until 1989, four years after the original application leading to the '826 patent.

[48] Moreover, it is worth noting that Dr. Isacoff's contention in this regard is belied by his own testimony elsewhere, in which he stated that he prescribes gemcitabine to half of his patients with advanced pancreatic cancer. (Isacoff 1462:13-16.)

[49] Besides this admission, the foundations for Dr. Tew's contention regarding the "complexity" of the necessary clinical trials are seriously flawed, in that he treats gemcitabine as though it were patented for use in all mammals for all types of cancer. This ignores the prosecution of the patent, which clearly demonstrated that gemcitabine was intended for use in humans alone. (E.g., Green 220:6-22, 223:11-16; PTX 2 at col. 16, ll. 2-8.)