Geneva Pharmaceuticals v. GLAXOSMITHKLINE PLC., 213 F. Supp. 2d 597 (E.D. Va. 2002)

*598 Conrad Moss Shumadine, Wilcox & Savage, Norfolk, VA, Leslie Morioka, Dimitrios Theodore Drivas, Louis Sebastian Silvestri, White & Case, LLP, New York City, for Geneva Pharmaceuticals, Inc., plaintiffs.

Alan Dale Albert, Troutman Sanders Mays & Valentine, LLP, Norfolk, VA, William Michael Merone, William K. Wells, Jr., Thomas J. Meloro, Steven Jeffrey Lee, Jeremiah Francis Manning, Larissa Ann Soccoli, Kenyon & Kenyon, New York City, for Teva Pharmaceuticals USA, Inc., plaintiffs.

Beth Hirsch Berman, Hofheimer Nusbaum, P.C., Norfolk, VA, Christy Lynn Green, Joseph Francis Jennings, Knobbe Martens Olson & Bear, LLP, Newport Beach, VA, for Ranbaxy Pharmaceuticals, Inc., plaintiffs.

Liam O'Grady, Finnegan, Henderson, Farabow, Garrett & Dunner, Washington, DC, Kelly Ann Casey, Richard Lawrence Rainey, Walter Wayne Brown, Russell Leon Sandidge, Finnegan Henderson Farabow Garrett & Dunner, LLP, Atlanta, GA, Stephen Edward Noona, Kristan Boyd *599 Burch, Kaufman & Canoles, PC, Norfolk, VA, for Glaxosmithkline PLC, defendants.

MORGAN, District Judge.

The parties in this matter come before the Court for a non-jury trial. Geneva Pharmaceuticals, Incorporated ("Geneva"), Teva Pharmaceuticals, Incorporated ("Teva"), Ranbaxy Pharmaceuticals, Incorporated, and Ranbaxy Laboratories Limited he last two are collectively referred to as ("Ranbaxy") challenge three patents held by Defendant Glaxosmithkline, Incorporated. Defendant and predecessor corporations are collectively referred to as ("GSK") for the pharmaceutical drug Augmentin®. The three patents at issue are U.S.Patent 4,525,352^[1] (the "Cole '352 patent"), U.S.Patent 4,529,720^[2] (the "Cole '720 patent"), and U.S.Patent 4,560,552^[3] (the "Cole '552 patent") (collectively the "Cole Patents"). The Plaintiffs' challenge is based on the theory that these three patents are double patented versions of patents already held by GSK, which illegally extend the Defendant's protection for Augmentin®. Ruling from the bench at the conclusion of the trial, which began May 20, 2002, the Court found that GSK had been granted an impermissible extension of its protection for Augmentin®, and invalidated the three patents at issue. This opinion and order further explains the Court's May 23, 2002 ruling.

As early as the 1940s, bacteria were beginning to develop a resistance to penicillins. A resolution to this problem was not easily developed, as resistance to this important class of drugs grew at an alarming rate. By the late 1950s-early 1960s, scientists identified β-lactamase enzymes as the primary bacterial response which neutralized the effectiveness of penicillins' antibiotic activity.

After identifying the major cause of the problem, scientists began working on penicillins that were themselves resistant to the β-lactamase enzymes, but this work was not particularly effective.^[4] Another approach, where penicillins were combined with a compound that was designed to protect the antibiotic from β-lactamase enzymes, did hold out much more promise in the eyes of the scientific community although meeting only limited success initially.

A lead researcher in the field to develop a β-lactamase inhibitor that could serve as a first line of defense for penicillin was a British scientist, Dr. Martin Cole, whose work was conducted for a predecessor corporation of GSK's, the Beecham Group. Dr. Cole's work in the late 1960s included testing numerous substances for their ability to produce a β-lactamase inhibitory effect. In 1968, Dr. Cole's team discovered a promising compound that was potent in defending penicillins against β-lactamase enzymes; unfortunately, its properties *600 were not suitable for wide-scale use in human and animal subjects.

As research continued into 1972, Cole had become aware of work conducted by researchers at Eli Lilly & Company, another pharmaceutical producer. When comparing the structure of his earlier discovered promising compound with that of a substance first identified in a paper published by Eli Lilly & Company, he noticed that its structure looked to portend a similarly potent ability to inhibit β-lactamase enzymes, but with greater usefulness in human and animal subjects. After extensive testing, Cole and his group of researchers determined that this compound could be effectively used in combination with penicillin to block β-lactamase enzymes neutralizing the antibiotic, so that the penicillin could in turn work to defuse the bacteria.^[5] Dr. Cole had discovered that this substance was a revolutionary β-lactamase inhibitor, and it became known as clavulanic acid.^[6] The discovery was announced publicly to the scientific community in 1974.

Another research group, led by Dr. Ian Fleming^[7] of Glaxo Laboratories, Inc., was also working in this field during the same time period as Dr. Cole, in a wholly independent effort to discover a β-lactamase inhibitory compound. Shortly after Dr. Cole made his discovery, Dr. Fleming isolated clavulanic acid in a highly purified form. Glaxo Laboratories, Inc., never itself created a commercial product containing clavulanic acid. Ten months after a patent application was filed to protect Dr. Cole's research, however, an application was filed for Dr. Fleming's highly purified clavulanic acid. A series of patents were issued^[8] relating to clavulanic acid in the names of Dr. Cole and Dr. Fleming, but all problems of these competing patents were resolved when they came under the common ownership of one of GSK's predecessor corporations, and ultimately GSK.^[9]

The U.S. Food and Drug Administration (FDA) first approved GSK's pharmaceutical *601 combination of the penicillin derivative amoxycillin and a potassium salt of clavulanic acid, potassium clavulnate, in 1984. By combining the two substances, the clavulanic acid countered the attack of the β-lactamase enzymes before they had the opportunity to destroy the antibacterial effect of the amoxycillinthe amoxycillin was then able to destroy the harmful bacteria. GSK chose to market the commercial pharmaceutical under the trademark Augmentin®, to reflect the fact that the clavulanic acid was augmenting the natural antibacterial properties of amoxycillin by preventing the β-lactamase enzymes from neutralizing it.

Unfortunately, no significant progress has been made in dealing with bacterial resistance to penicillin since Dr. Cole's pioneering work in the early 1970s, which led to his original U.S.Patent Application Number 05/569,007 filed on April 17, 1975. Indeed, the evidence indicates that additional strains of β lactamase developed as harmful bacteria continue to build defenses against penicillin and clavulanic acid. The result being that clavulanic acid is less effective as a β-lactamase inhibitor in 2002 than it was in 1974. Since 1974, GSK and its predecessors have, through acquisitions and multiple patent applications, sought to present obvious variants of clavulanic acid in a variety of semantically different forms. While some of GSK's multiple patent applications based on the discovery of clavulanic acid as a β tamase inhibitor have been granted, others have been denied. The Cole patents before the Court must fall for double patenting.

Just as its original patents for clavulanic acid were ready to expire, GSK's long pending applications for Patent and Trademark Office "(PTO") approval of four Cole related patents conveniently began to bear fruit. In early 2000, U.S.Patent Numbers 6,031,093^[10] (the "Cole '093 patent"), 6,048,977^[11] (the "Cole '977 patent") and 6,051,703^[12] (the "Cole '703 patent") were issued. The following Spring, U.S.Patent Number 6,218,380^[13] (the "Cole '380 patent") (these *602 four patents are collectively referred to as the "Cole 2000/2001 patents") was issued by the PTO. The practical effect of these patents was that GSK was given patent protection for the pharmaceutical drug Augmentin® until April 17, 2018.

Facing this delay in their ability to market generic forms of Augmentin®, the Plaintiffs in this matter, generic drug manufacturers, filed suit pursuant to 28 U.S.C. §§ 2201-2 and Federal Rule of Civil Procedure 57 seeking a judgment that the Cole 2000/2001 patents, plus the three earlier issued patents at issue in this trial, were invalid. The initial lawsuit was filed in this Court by Geneva^[14] on May 31, 2001, and was consolidated with Teva's later filed action^[15] on October 31, 2001. Ranbaxy's lawsuit^[16], seeking nearly identical relief, was consolidated with the other matters by consent of the parties on January 16, 2002.

Through several summary judgment motions and hearings, the Court has narrowed the issues of the suit. At a December 14, 2001 hearing, the Court invalidated the Cole '380 in light of the Cole '720 patent. At that same hearing, it was ruled that the Cole '552 patent could not be invalidated in light of the U.S.Patent Number 4,525,352, due to 35 U.S.C. § 121.^[17] At the following March 13, 2002 hearing, the Court invalidated the Cole '093, '977 and '703 patents after finding that they were improperly issued by the PTO as obvious variations of the Cole '720 patent.

Following those pre-trial decisions, the issues remaining for trial are the Plaintiffs' challenges to the Cole '552, '352 and '720 patents. At the hearings on the pre-trial motions, the parties relied exclusively upon the patents themselves and the PTO records. At trial, the parties presented relatively few additional exhibits, four expert witnesses and Dr. Cole by deposition. The opening statements and presentation of evidence were completed in two and one half days of trial, albeit the first two days of trial were somewhat longer than average.

In reaching its findings of fact, the Court observes that there are no significant facts in dispute, however, there are considerable conflicts in the opinions of the parties' experts. Geneva relied primarily upon Dr. Christine Sanders, who received a Ph.D in medical microbiology in 1973, joined the faculty of Creighton University medical school also in 1973, and testified as an expert on antibiotics and β-lactamase inhibition. Teva relied upon Dr. Leslie Benet, who earned a Ph.D in pharmaceutical chemistry in 1965, and joined the faculty of the University of California at San Francisco in 1969 in the bio-pharmaceutical sciences department, where he later served as department chair. Dr. Benet testified as an expert in pharmaceutical chemistry and bio-pharmaceutical sciences. GSK relied upon Dr. Christopher Schofield, who earned his D.Phil^[18] in the chemical and enzymatic synthesis of antibiotics in 1985, and while the English equivalent of a tenured professor has done research on clavulanic acid for about ten years. Dr. Schofield's testimony reflects that he first became acquainted with clavulanic acid in 1981 or 1982.

*603 The only other witnesses were Dr. Cole, presented via videotaped deposition by Teva, and Mr. John Dewhirst, called by Geneva. Despite his pre-eminence in the field, the portion of Dr. Cole's deposition presented in evidence was unremarkable. Mr. Dewhirst, a retired patent office attorney, testified regarding patent histories, including conflicts and interferences between patents that were largely resolved by mergers and acquisitions among GSK's predecessors.

Dr. Sanders opined that the Cole '552 patent and the Cole '352 patent were obvious over the Crowley '609 patent. Record pgs. 57 and 109. She also testified as to the natural result of clavulanic acid as a β-lactamase inhibitor, which is crucial to the Court's finding that its function and effect is inherent. The finding of inherency led to the finding that the Cole '552 and '352 patents were anticipated by the Crowley '609 patent. She further opined that claim one of the Cole '720 patent was "essentially the same thing" as claimed in the Fleming '175 patent. Record at 126-127.

Teva focused upon the invalidity of the Cole '552 patent in light of the Crowley '609 patent. Its expert witness, Dr. Benet, concurred with Dr. Sanders' opinion that the Cole '552 patent was obvious over the Crowley '609 patent, Record 253-256, and concurred that β-lactamase inhibition was inherent in clavulanic acid. Record 273-287. He did not testify directly regarding the Cole '352 and '720 patents.

Not unexpectedly, GSK's sole witness, Dr. Schofield, had an opinion that differed from those of Drs. Sanders and Benet. While Dr. Schofield was the more knowledgeable regarding the minute details of clavulanic acid and a splendid instructor, the Court found the opinions of Drs. Benet and Sanders more persuasive. A certain degree of partisanship is normal and tolerable in an expert's testimony, but Dr. Schofield went too far. Not only did he cleverly obfuscate his responses to important questions, but his objectivity was often subsumed by his advocacy. For example, after counsel for Teva concluded his cross examination of Dr. Schofield, the doctor reopened his own testimony in a misguided attempt to qualify a previous answer. Record 474-476.

While the Court FINDS each of these three witnesses qualified to testify as experts, it further FINDS that the opinions of Drs. Benet and Sanders more credible and consistent with the evidence than those of Dr. Schofield.

GSK has made much of Dr. Cole's status as the scientist who discovered that clavulanic acid inhibits β-lactamase enzymes, as well as the fact that he was the first to file with the PTO. GSK has proven both of these points, but they are not relevant to the outcome. Both the Crowley '609 patent and the Fleming '175 patent, which the Court FINDS conflict with the Cole patents in issue, were issued prior to the Cole '552, '352 and '720 patents, although Dr. Cole was the first to file with the PTO through his application No. 05/569,007 in 1975. The record is not clear as to why the patents were issued in this order, but the record is clear that each of these patents were issued while commonly owned by predecessors of GSK.

The focus of Teva's evidence, and the principle focus of Geneva's evidence, was upon their claim that GSK's Cole '552 patent is obvious over its commonly held *604 Crowley '609 patent. Congress statutorily limited the duration of a patent-holder's "right to exclude others from practicing a claimed invention" in 35 U.S.C. 154(a) (2) (West 1994).^[20]Eli Lilly & Co. v. Barr Laboratories, et al., 251 F.3d 955, 967 (Fed.Cir.2001). "Obviousness-type double patenting is a judicially created doctrine grounded in public policy, which prevents the extension of the term of the original patent via the patenting of an obvious variation." Georgia-Pacific Corp. v. United States Gypsum Co., 195 F.3d 1322, 1326 (Fed.Cir.1999).

When conducting an obviousness-type double patenting review, a court should conduct a two step analysis.^[21]Eli Lilly, 251 F.3d at 968. First, the court construes the claims in the patents at issue to determine the differences between the two. Id. Next, a court must determine whether the differences in subject matter between the two are sufficient to render the two patentably distinct. Id. Only if the later claim is not patentably distinct from the earlier claim is the commonly owned patent to be ruled invalid due to obviousness-type double patenting. Id. A later patent claim is not patentably distinct from an earlier claim if the later claim is either obvious over, or anticipated by, the earlier claim. Id. Meaning that a finding of obviousness-type double patenting can be founded upon a basis of anticipation or because a patent later is obvious in light of a commonly held earlier patent.^[22]Id.

Plaintiffs claim the Cole '552 patent is obvious over GSK's commonly held United States Patent Number 4,441,609 (the "Crowley '609 patent"). In determining whether the Cole '552 patent is obvious in light of the Crowley '609 patent, the Court must conduct a detailed examination of the claims to determine whether the later merely defines an obvious variation of the earlier. In re Emert, 124 F.3d 1458, 1460 (Fed.Cir.1997). "Without a patentable distinctionbecause the pending claim defines merely an obvious variation of the patented claimthe patentee may overcome the double patenting rejection [only] by filing a terminal disclaimer." Id. (internal citations and quotations omitted). Obviousness, a question of law based on *605 underlying factual underpinnings, is determined from the perspective of a person of ordinary skill in the relevant field; this hypothetical person is presumed to have knowledge of all relevant prior art. Id. at 1462.; Graham v. John Deere, 383 U.S. 1, 17, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966).^[23]

Counsel for Teva presented a detailed analysis of the Crowley '609 patent. Claim 1 of that patent read as follows:

1 A packaged pharmaceutical composition of enhanced storage stability which comprises a closed container containing one or more unit-dose compositions suitable for oral administration each dosage unit of which comprises 20 mg to 1500 mg of amoxycillin trihydrate, 20 mg to 500 mg of potassium clavulanate and a pharmaceutically acceptable carrier with the proviso that the weight ratio of amoxycillin trihydrate to potassium clavulanate is from 6:1 to 1:1 and a desiccant.^[24]

The claims then go on to more fully describe the closed container and the composition of the desiccant.

The Court must compare the above language with that of the relevant language in the Cole '552 patent:

1 A pharmaceutical composition for treating bacterial infections in humans and animals which comprises a synergistically effective amount of clavulanic acid, or a pharmaceutically acceptable salt thereof, and an antibacterially effective amount of a penicillin, or a pharmaceutically acceptable salt or ester thereof.^[25]

Claim 2 of the Cole '552 patent lists the salts of clavulanic acid that are to be used, while Claim 3 contains a list of penicillins. Claim 4 teaches that oral administration is a method of introducing the product into the system of a human or animal subject, which is followed by additional methods of administration in Claims 5-7.

The Crowley '609 patent claims specific dosage ranges of potassium clavulanate and amoxycillin trihydrate, in addition to a closed container, a carrier and a desiccant. Claim 1 of the Cole '552 patent listed its ranges in terms of a "synergistically effective amount" of clavulanic acid and an "antibacterially effective" amount of a penicillin. The Cole '552 patent is silent as to the method of storage.

The language of the Cole '552 patent compels the Court to look to the specifications to determine what is meant when the dosage range of clavulanic acid is ambiguously described as "synergistically effective amount." Dr. Benet testified that a scientist of ordinary skill in the art would have been required to seek a definition of this term in 1974, Record at 286, and Dr. Sanders testified that the term was not used in its classic sense.^[26] Record at 65. The Federal Circuit has been plain and consistent in its rulings that the patent specifications, otherwise known as the patent *606 disclosure, should not be used as prior art when a trial court is deciding whether an earlier obtained commonly held patent is an obvious variation of a later patent. In Re Vogel, 57 C.C.P.A. 920, 422 F.2d 438, 441 (Cust. & Pat.App.1970); In re Aldrich, 55 C.C.P.A. 1431, 398 F.2d 855 (Cust. & Pat.App.1968); In re Braat, 937 F.2d 589, n. 5 (Fed.Cir.1991). The specifications may, however, be used to serve as a "dictionary" to learn the meaning of ambiguous terms in a patent's claims. In Re Vogel, 422 F.2d at 442.

To properly construe what is meant by this "synergistically effective amount of clavulanic acid," it is necessary for the Court to turn to the specifications. Furthermore, as an element new to even a person of ordinary skill in the art in 1974,^[27] one would need to look to the specifications to determine the basic nature of clavulanic acid. This Court does keep in mind the notion that it is inappropriate to "indiscriminately use all of the generalities" of the specifications, and will only use them as a guide in determining what was meant by a synergistically effective amount of clavulanic acid.^[28]Phillips Petroleum Co. v. U.S Steel Corp., 673 F. Supp. 1278, 1311 (D.Del.1987), aff'd, 865 F.2d 1247 (Fed.Cir.1989).

As the Court looks to the specifications of the Cole '552 patent for elucidation as to what constitutes a synergistically effective amount of clavulanic acid, the Court first looks to Column 1, Lines 51-65:

Combination of a β-lactamase inhibiting substance with a penicillin or cephalosporin might be expected to protect the latter from degradation by bacterial β-lactamase and thereby enhance their antibacterial activity against many infective organisms. This process of enhancement of the antibacterial activity is called synergism when the antibacterial activity of the combination is well in excess of the simple addition of the activities of the two separate substances. The β-lactamase inhibiting component of the mixture is referred to as a synergist and such substances are valuable for increasing the antibacterial activity of penicillins and cephalosporins against resistant organisms. It is one of the objects of this invention to provide such synergists.

To further understand the phrase, the Court next turns to Column 9, Lines 12-18. At that point the specifications disclose:

The synergist used in this treatment will generally be a salt or in-vivo hydrolysable ester of clavulanic acid such as the sodium or potassium salt of clavulanic acid or the acetoxymethyl, pivaloyloxymethyl, phthalidyl or like ester of clavulanic acid. Each unit dose will usually contain from 50 to 500 mg of the synergist, for example, 100 to 250 mg.

The dosage in the disclosure is defined in terms that fit well within the Crowley '609 patent. It is obvious to the Court that the amount to which the patent refers when it describes a synergistically effective amount of clavulanic acid is substantially the same as the amount referenced in claim one of the Crowley '609 patent.

*607 The Crowley '609 patent describes the pharmaceutical composition as "packaged," while the Cole '552 patent describes a pharmaceutical composition. The medicine of the Crowley '609 patent is to be kept in a closed container with a desiccant, and a "pharmaceutically acceptable carrier." Moreover, the Crowley '609 patent speaks of "enhanced storage stability" and a "unit dosage form," while the Cole '552 patent does not. The evidence proves these alleged distinctions to be negligible. As was pointed out to the Court during trial, the closed container, the carrier and the desiccant were well established pharmaceutical tools in 1974. Record at 82. The other differences between the patents are works of semantics.

"The burden is on the party asserting invalidity to prove it with facts supported by clear and convincing evidence." Georgia-Pacific, 195 F.3d at 1329. The Plaintiffs have presented evidence to the Court through Dr. Sanders which demonstrates that the essence of the Crowley '609 patent, truly the only novel invention claimed in the patent, is the clavulanic acid as a β-lactamase inhibitor. Record at 60, 81. After construing both patent's claims, and examining the specifications of the Cole '552 patent for the purpose of defining the term synergistically effective amount of clavulanic acid, the Court FINDS that the Plaintiffs have carried their burden of proof by clear and convincing evidence that the Cole '552 patent is merely an obvious variant of the Crowley '609 patent. (Dr. Benet's opinion on this issue can be found in the Record at 241, and Dr. Sanders' can be found in the Record at 87. To the extent that Dr. Schofield opined to the contrary, the Court FINDS the opinions of Drs. Sanders and Benet more persuasive.)

As the Federal Circuit has stated, "anticipation is the epitome of obviousness." Structural Rubber Prods. Co. v. Park Rubber Co., 749 F.2d 707, 716 (Fed. Cir.1984). References in claims are anticipatory when they disclose every limitation of the claimed invention either explicitly or inherently. Atlas Powder Co. v. Ireco Inc., 190 F.3d 1342, 1346 (Fed.Cir.1999). The existence or function of clavulanic acid was not explicitly known to a person ordinarily skilled in the art in 1974, because it was a new invention. The inherent characteristic of the clavulanic acid as aβ-lactamase inhibitor, as described in the Cole '552 and Crowley '609 patents, is the foundation for finding anticipation a second basis of invalidity.

A reference includes an inherent characteristic if that characteristic is the natural result flowing from the reference's explicitly defined limitations. Eli Lilly & Co., 251 F.3d at 970. The natural result that flows from the use of clavulanic acid is its inhibition of β-lactamase enzymes. GSK argues that the Cole '552 patent was not anticipated by the Crowley '609 patent because clavulanic acid does not effectively inhibit β-lactamase enzymes one hundred percent of the time. The Court does not adopt GSK's attempt to raise the bar by redefining what is meant by inherent as that term is used in deciding anticipation.

It is true that every time a person is medicated with Augmentin® the pharmaceutical drug will not be 100% effective, and not every dosage of clavulanic acid can effectively inhibit β-lactamase enzymes to the point that the penicillin can be completely effective. These facts do not mean that clavulanic acid is not performing its inherent function. Record at 103-107. The problem is that new strains of resistant β-lactamase enzymes have evolved, and previously known strains have developed defenses to clavulanic acid. The clavulanic acid does continue to naturally inhibit a certain percentage of β-lactamase *608 enzymes, but it is sometimes overwhelmed by the quantity or nature of the enzymes present. Unfortunately, clavulanic acid, and therefore Augmentin®, is less effective today than it was when first patented. This has been caused in part by harmful bacteria developing defenses to clavulanic acid, just as such bacteria developed defenses to penicillin. Perhaps this result has also been caused in part by clavulanic acid resistant forms of β-lactamase enzymes, but it is not clear from the evidence if such enzymes existed prior to 1974 or have evolved since then. What is inherent is that clavulanic acid naturally inhibits β-lactamase enzymes, and the only reason for the inclusion of clavulanic acid in any of the patented formulations at issue is its β-lactamase inhibitory nature.^[29]

The Court FINDS that the Cole '552 patent is anticipated by the Crowley '609 patent. The Cole '552 patent is both obvious in light of and anticipated by the Crowley '609 patent. "[A] later patent claim that fails to provide novel invention over an earlier claim is not patentably distinct from the earlier claim." Eli Lilly & Co., 251 F.3d at 970. Accordingly, the Court FINDS that the Cole '552 patent is invalid on the grounds of obviousness-type double patenting.

Geneva also challenges the Cole '352 patent on the grounds of obviousness-type double patenting in light of the Crowley '609 patent. The Court will use a legal analysis similar to that used in construing the Cole '552 patent to decide whether to invalidate the Cole '352 patent.

Claims 1 and 2 of the Cole '352 patent read:

1. A pharmaceutical composition useful for treating bacterial infections in humans and animals which comprises a synergistically effective amount of clavulanic acid and an antibacterially effective amount of amoxycillin, in combination with a pharmaceutically acceptable carrier.

2. A pharmaceutical composition useful for treating bacterial infections in humans and animals which comprises a synergistically effective amount of a pharmaceutically acceptable salt of clavulanic acid and an antibacterially effective amount of amoxycillin, a pharmaceutically acceptable salt thereof or a pharmaceutically acceptable ester thereof, in combination with a pharmaceutically acceptable carrier.

The claims following describe different variations of these two claims.

Once again, the term synergistically effective amount of clavulanic acid is used in the patent's claims. For the aforementioned reasons, the Court must look to the specifications for the limited purpose of clarifying what was meant by use of this term. The specifications describe the amounts to be used in terms that are reflective of the Crowley '609 patent, as illustrated in these passages from the patents specifications:

Combination of a β-lactamase inhibiting substance with a penicillin or cephalosporin might be expected to protect the latter from degradation by bacterial β-lactamase and thereby enhance their antibacterial activity against many infective organisms. This process of enhancement of the antibacterial activity is called synergism when the antibacterial activity of the combination is well in excess of the simple addition of the activities *609 of the two separate substances. The β-lactamase inhibiting component of the mixture is referred to as a synergist and such substances are valuable for increasing the antibacterial activity of penicillins and cephalosporins against resistant organisms. It is one of the objects of this invention to provide such synergists.^[30]

The synergist used in this treatment will generally be a salt or in-vivo hydrolysable ester of clavulanic acid such as the sodium or potassium salt of clavulanic acid or the acetoxymethyl, pivaloyloxymethyl, phthalidyl or like ester of clavulanic acid. Each unit dose will usually contain from 50 to 500 mg of the synergist, for example, 100 to 250 mg.^[31]

As with the comparison of the Cole '552 patent, this comparison of the Crowley '609 patent with the Cole '352 patent demonstrates that the later issued patent is an obvious variant of the earlier patent.

Under the same anticipation analysis used with the Cole '552 patent, the Cole '352 patent is anticipated in light of the Crowley '609 patent. The Federal Circuit has ruled that earlier composition claims render obvious a later claim to the method of use claims of the claimed composition for its inherent biological effect. Eli Lilly & Co., 251 F.3d at 968. As explained, the inherent characteristic of clavulanic acid, as described in both these patents, is that it inhibits β-lactamase enzymes.

The invention described in the claims of the Cole '352 patent is not patentably distinct from the core invention of the Crowley '609 patent. Both of these patents' claims describe various combinations of clavulanic acid and penicillin. One is described with specific amounts; the other is described in general terms, but the specifications defining synergistically effective amount of clavulanic acid in the Cole '352 patent make clear that the similarity in amount is too closely related to render the later patent distinct.

After construing the Cole '352 patent and Crowley '609 patent, then construing the differences, the Court FINDS that the two patents are not patentably distinct. The Cole '352 patent is both obvious in light of, and anticipated by the Crowley '609 patent. Accordingly, the Court INVALIDATES the Cole '352 patent.

Geneva further challenges the Cole '720 patent in light of GSK's commonly held U.S.Patent Number 4,367,175 (the "Fleming '175 patent") under a double patenting theory. The opening claim of the Cole '720 patent states:

1. A method of effecting β-lactamase inhibition in a human or animal in need thereof arising from a β-lactamase producing bacteria which comprises administering to said human or animal a β-lactamase inhibitory amount of clavulanic acid or a pharmaceutically acceptable salt thereof.^[32]

That is then followed by claims to different variations of clavulanic acid's salts and methods of administration. The Fleming '175 patent uses a chemical formula to define a form of clavulanic acid with 98% purity. The Fleming '175 patent claim does not define a function for this claim, it simply defines the formula.

Utility is an essential requirement of every patent, without it the patent should not issue. 35 U.S.C. § 101 (2001); Application of Bremner, 37 C.C.P.A. 1032, 182 F.2d 216, 217 (Cust & Pat.App.1950) ("It was never intended that a patent be granted upon a product, or a process producing *610 a product, unless such product be useful."). A patent possesses utility "if it will operate to perform the functions and secure the results intended, and its use is not contrary to law, moral principles, or public policy." Callison v. Dean, 70 F.2d 55, 58 (10th Cir. 1934); see, e.g., Standard Oil Co. v. Montedison, S.p.A., 494 F. Supp. 370, 381 (D.Del.1980) (The utility requirement is satisfied when an inventor has learned enough about the product to justify the conclusion that it is useful for a specific purpose.); Esteé Lauder Inc. v. L'Oreal, 129 F.3d 588, 593 (Fed.Cir.1997). To determine the specific utility claimed in the Fleming '175 patent and compare it with the claims of the Cole '720 patent, the Court necessarily must look to the specifications, because the claims themselves offer no insight into what the patent claims it does.

The Court looks to the Fleming '175 patent Column 1, lines 8-9, where the specifications state:

This invention relates to a novel antibiotic and its salts and to a process for their production in pure form.

The next column explains:

Clavulanic acid and its salts also possess the ability to inhibit β-lactamase enzymes produced by gram-positive organisms, for example those produced by strains of Staphylococcus aureus and Bacillus cereus and also the enzymes classified in classes II-V from gramnegative bacteria produced by organisms such as strains of Proteus mirabilis, Escherichia coli, Proteus morganii, Klebsiella aerogenes, Salmonella typhimurium, Shigella sonnei and Haemophilus influenzae.^[33]

The only difference between the claims of these two patents is that the Fleming '175 patent claims pure salts of clavulanic acid, the utility of which is the inhibition of β-lactamase enzymes in humans, while the Cole '720 patent claims a method of using clavulanic acid to inhibit J-lactamase enzymes in humans and animals. Furthermore, the pharmaceutically acceptable salts of clavulanic acid listed in claims 6-11 are obvious variations of what is already claimed. This, as with all of these patents, appears to be an attempt to place a new face on clavulanic acid so that more patent protection can be attained.

The Defendant seeks to divorce the inherent properties of the compound described in the Fleming '175 patent from the substance itself. The natural result flowing from the compound described with its chemical formula in the Fleming '175 patent is the inhibition of β-lactamse enzymes. The patent protection that the Defendant received in the Cole '720 patent was not for a novel invention. The Court FINDS that all of the claimed characteristics described in the Cole '720 patent also flow inherently from the Fleming '175 patent. To allow the Cole '720 patent to continue in force would mean the Court would be allowing a timewise extension of the Fleming '175 patent, which is not allowed. Accordingly, the Court FINDS the Cole '720 patent is invalid because it is anticipated by the Fleming '175 patent.

GSK attempts to distinguish between patents that are not patentably distinct by three general means. First it attempts to break down the Crowley '609 patent into 5 elements, and the Cole '352 and '552 patents into three elements, thereby attempting to create a patentable distinction. By breaking down the Crowley '609 patent into elements, however, GSK has classified a glass container, a carrier and a desiccant as elements of the invention. In attempting to justify this position, at one point GSK's attorney went so far as to argue *611 that a generic form of Augmentin® could have been marketed after the Crowley '609 patent was granted, but before the Cole '552 patent was granted, so long as the product omitted the carrier, was sold in a blister pack in lieu of a closed container with a desiccant and contained a form of clavulanic acid that was less than 98 percent pure (to avoid the Fleming '175 patent). The Court FINDS that the addition of a closed container with desiccant and the inclusion of a carrier are inconsequential variables known to those versed in the art in 1974, and therefore not a basis for finding a patentable distinction between the Crowley '609 patent on the one hand and the Cole '352 and '552 patents on the other. Dr. Sanders' testimony on this point can be found in the Record at 81, and Dr. Benet's can be found in the Record at 253-255. Again the Court FINDS the opinions of Drs. Sanders and Benet more persuasive than the opinion of Dr. Schofield to the contrary.

GSK's second argument is that clavulanic acid is not 100 percent effective as a β-lactamase inhibitor, and therefore its properties are not inherent. The Court FINDS that the natural result of the introduction of clavulanic acid into a compound also containing a penicillin is the inhibition of the efforts of the β-lactamase enzymes to neutralize the anti-bacterial effects of the penicillin. Because it is the natural result, it is the inherent function of clavulanic acid as an element of each of the patents under consideration. During the 28 years since Dr. Cole first made an application for a patent that included clavulanic acid as a β-lactamase inhibitor, research has uncovered a number of β-lactamase enzymes that are not effectively inhibited by clavulanic acid. Certain classes of these β-lactamase enzymes may have always been largely, or at least partially, immune from the inhibitory effects of clavulanic acid, while other β-lactamase enzymes appear to have developed an immunity over the years of its use. The evidence is somewhat murky upon the distinction. No matter, clavulanic acid's only value in combination with a penicillin is as a β-lactamase inhibitor. GSK's argument that clavulanic acid also possesses antibacterial qualities, or that it is described in the Crowley '609 patent as storage life prolonging, do not mean it is not inherently a β-lactamase inhibitor.

GSK's third attempt to establish a patentable distinction between the Crowley '609 patent and the Cole '352 and '552 patents is based upon their contention that the Court must confine its examination of the patents to the claims alone. While their argument is generally correct, it is not an absolute. Dr. Cole uses the term synergistically effective amount in the patent, which was a novel phrase in 1974. Dr. Cole may certainly act as his own lexicographer, but where he uses an unfamiliar term or a familiar word in an unfamiliar context, the Court must look to the specifications for the definitions of such terms. Synergy is generally defined as the result obtained by adding two elements which form a compound whose effect exceeds the sums of the effect of its two component parts. Thus the use of a synergistically effective amount to describe the dosage of one element of a compound creates an ambiguity requiring further definition.

In its final analysis, the Court rejects GSK's three pronged defense and FINDS its eight point factual presentation tangential to the basic issue. The only patentable invention contained in any of the three Cole patents challenged at trial is clavulanic acid as a β-lactamase inhibitor.^[34]

Title 37 C.F.R. § 1.78^[35] gives the PTO the power to force a party prosecuting two patent applications claiming nearly identical inventions to state which invention was the first, or prior art. When the one invented first is claimed, the second application is effectively abandoned. Geneva cited prior art as separate grounds for invalidating one or more of the Cole patents. This ground, however, has not been fully explicated and must therefore be rejected. In view of the Court's findings of obviousness and anticipation, it does not rule upon Geneva's other claims of invalidity.

The Court has construed the differences between the Cole '552 patent and the earlier issued Crowley '609 patent, and FINDS that there are no patentable distinctions between the two. The '552 patent is both obvious in light of and anticipated by the Crowley '609 patent. Similarly, the Cole '352 patent is both obvious in light of and anticipated by the Crowley '609 patent, and accordingly the Court FINDS the Cole '352 patent is not patentably distinct from the earlier issued Crowley '609 patent. Further, the Court FINDS that the characteristics of the composition described in the earlier issued Fleming '175 patent are inherent and identical to the explicit characteristics claimed in the Cole '720 patent. An extension of protection for the inherent qualities of the Fleming '175 patent is impermissible. Each of the three Cole patents in issue here are attempts to make a patentably distinct variation of the same core substance, clavulanic acid as a β-lactamase inhibitor. GSK has had its full term of patent protection for clavulanic acid as an element of Augmentin® and related antibiotics, and may not have more than the statutorily prescribed limit. Accordingly, the Court INVALIDATES GSK's patent protection for the Cole '552, Cole '352 and Cole '720 patents.

The Clerk is REQUESTED to send a copy of this order to all counsel of record.

It is so ORDERED.

[1] Issued to GSK on June 25, 1985, the patent expired June 25, 2002.

[2] Issued to GSK on July 26, 1985, the patent expires July 26, 2002.

[3] Issued to GSK on December 24, 1985, the patent expires December 24, 2002.

[4] While scientists were able to develop penicillin compounds that were resistant to specific β-lactamase enzymes, they were not able to develop a penicillin that would itself combat the broad range and variety of the β-lactamase enzymes produced by bacteria. Without a penicillin with the ability to fight off the wide range of β-lactamase enzymes that any given bacteria was producing, these medications were of little use to treating physicians who lacked the tools to determine with complete specificity what β-lactamase enzymes were present.

[5] Later work by GSK researcher Patrick Crowley discovered that a potassium salt of clavulanic acid made the drug even more useful in human and animal subjects by increasing the compound's stability. His work was developed and later issued as United States Patent Number 4,441,609, and is commonly referred to as the '609 Crowley Patent.

[6] The scientists at Eli Lilly & Company initially brought this substance to the public consciousness, but they were apparently unaware of what the substance could do, or how important it would eventually turn out to be. Dr. Cole's groundbreaking research essentially filled in the gaps left by Eli Lilly & Co. researchers, creating one of the most successful pharmaceuticals of the late 20th century.

[7] Dr. Fleming's work was conducted on behalf of another predecessor of GSK, Glaxo Incorporated. While originally held by two different entities, the patents of Dr. Cole and Dr. Fleming came into common ownership through the merger of corporations that ultimately became part of GSK.

[8] As of December 25, 2001, the four patents that issued as a result of the Fleming Application have expired.

[9] These patents were the subject of U.S.Patent and Trademark Office Interference No. 100,451. When two applications claim nearly identical inventions, the PTO issues what is known as an interference. 35 U.S.C. § 135; Godtfredsen v. Banner, 503 F. Supp. 642 (D.D.C.1980) (The issue to be determined by Board of Patent Interferences is priority of invention.). During that process, an intensive investigation is mounted to determine which party is actually entitled to patent protection as the first inventor. Id. Interference No. 100,451 was initiated in this case due to the similar nature of the Fleming and Cole patents. Rather than contest the issue through the processes of the PTO, Beecham Group purchased the rights to the Fleming '175 patent from Glaxo Laboratories Limited.

[10] Filed with the PTO in application No. 08/417,628 on April 6, 1995, that application was a continuation of application No. 07/749,482 (the "'482 application") filed on August 15, 1991. The '482 application was a continuation of application No. 07/210,339 (the "'339 application"), filed on June 23, 1988, and eventually abandoned. The '339 application was itself a continuation application of the original application from which all of the Cole patents originate, application number 05/569,007, filed with the PTO by GSK on April 17, 1975, and later abandoned.

[11] Filed with the PTO on April 6, 1995, in application No. 08/417,625 as a continuation of application No. 07/749,428 (the "'482 application"). The '482 application was a continuation of application No. 07/210,339 (the "'339 application"), filed on June 23, 1988, and eventually abandoned. The '339 application was itself a continuation application of the original application from which all of the Cole patents originate, application number 05/569,007, filed with the PTO by GSK on April 17, 1975, and later abandoned.

[12] Filed with the PTO on April 6, 1995, as application number 08/418,055, which was a continuation of application No. 07/749,428 (the "'482 application"). The '482 application was a continuation of application No. 07/210,339 (the "'339 application"), filed on June 23, 1988, and eventually abandoned. The '339 application was itself a continuation application of the original application from which all of the Cole patents originate, application number 05/569,007, filed with the PTO by GSK on April 17, 1975, and later abandoned.

[13] Filed with the PTO on August 15, 1991, in application No. 07/749,482 as a continuation of application No. 07/210,339 (the "'339 application") filed on June 23, 1988, and eventually abandoned. The '339 application was itself a continuation application of the original application from which all of the Cole patents originate, application No. 05/569,007, filed with the PTO by GSK on April 17, 1975, and later abandoned.

[14] Civil Action Number 2:01cv391.

[15] Civil Action Number 2:01cv677.

[16] Civil Action Number 2:01cv925.

[17] 35 U.S.C. § 121 does not allow a court to invalidate a patent using a reference patent that was created from the same application as the result of a PTO decision to issue a restriction requirement severing the two patents. See the Court's February 25, 2002 opinion for a full explanation of 35 U.S.C. § 121, and its application to the case at bar.

[18] The English equivalent of the American Ph.D.

[19] In the Court's December 14, 2001 ruling, it was decided that the Cole '552 patent could not be reviewed in light of the '352 because of 35 U.S.C. § 121. The current challenge to the Cole '552 patent does not involve 35 U.S.C. § 121 because these patents were not mutually created as the result of a single PTO restriction requirement.

[20] A patent holder has twenty years of exclusivity before the public is allowed to copy the invention for profit. When the patents being considered today were issued, the statutory period of exclusivity was seventeen years.

[21] The Court has used a one-way analysis to determine if the patents at issue are double patented, as is the general practice in such a review. The parties stipulated that the two way test, more specifically discussed in prior orders where it was at issue, was inapplicable to the patents at trial.

[22] In previous stages of this litigation, the Court had to first determine whether 35 U.S.C. § 121 (2001) (" § 121" or section "121") would bar the Court from ruling on the matter of double patenting of the clavulanic acid patents. That is not at issue during the trial.

Section 121 provides in pertinent part:

A patent issuing on an application with respect to which a requirement for restriction under this section has been made, or on an application filed as a result of such a requirement, shall not be used as a reference either in the Patent and Trademark Office or in the courts against a divisional application or against the original application or any patent issued on either of them. (Emphasis added).

For § 121 to come into play, all of the patents have to originate from one patent that is split into multiple patents due to a restriction requirement imposed by the PTO. The Crowley, Clayton and Cole patents were not created from one patent that was ordered split by the PTOthey were created separately, patent applications were initiated for each of them separately, and they were issued separately by the PTO. Meaning those patents are properly subject to a double patenting review by this court applying a one way test.

[23] A person of ordinary skill in the art was persuasively defined by Dr. Sanders. Record at 71-73

[24] Crowley '609 patent, column 8, lines 15-23.

[25] '552 patent column 40, lines 29-34.

[26] Dr. Schofield, on the other hand, testified that this term was known in 1974. Record at 338. He later testified that a synergistically effective amount is the same as a β-lactamase inhibitory amount, Record at 341, which is a specialized use of the term requiring definition. Dr. Schofield, though possessed with an impressive range of knowledge, is at a disadvantage in looking back upon what was known to one with ordinary skill in the art in 1974 since he was only about 14 years old at the time. Drs. Sanders and Benet were experienced scientists by 1974, and the Court FINDS their opinions upon this issue more persuasive.

[27] When the issue of what was known in 1974 arises, the test is what did a person with ordinary skill in the art know in 1974.

[28] Clavulanic acid is used in this Court's analysis to represent this acid as well as its various salts, as it is the acid, not the derivative salts, which was the invention. For similar reasons, penicillin is the term used for penicillin itself, as well as its variants, such as amoxycillin. Changing clavulanic acid to one of its salts or changing one of its salts into another does not create a patentable distinction in the patents under consideration; neither does changing penicillin into one of its salts or esters create such a distinction here.

[29] GSK argues that Claim 5 of the Crowley '609 patent states that clavulanic acid is present only as a storage life enhancer, but potassium clavulanate is merely the most stable salt of clavulanic acid. Thus potassium clavulanate is still the β-lactamase inhibitor in the compound.

[30] '352 patent, Column 1, Lines 49-64.

[31] '352 patent, Column 9, Lines 12-18.

[32] '720 patent, column 40, lines 12-18.

[33] Fleming '175 patent, column 2 lines 26-34.

[34] It is of note that the specifications of the three Cole patents at issue in trialthe Cole '552 patent, Cole '352 patent and Cole '720 patentare identical in substance, however the claims differ.

[35] Title 37 C.F.R. § 1.78: If an application or a patent under reexamination and at least one other application naming different inventors are owned by the same party and contain conflicting claims, and there is no statement of record indicating that the claimed inventions were commonly owned or subject to an obligation of assignment to the same person at the time the later invention was made, the Office may require the assignee to state whether the claimed inventions were commonly owned or subject to an obligation of assignment to the same person at the time the later invention was made, and, if not, indicate which named inventor is the prior inventor.