Forest Laboratories, Inc. v. Ivax Pharmaceuticals, Inc.

501 F.3d 1263, 41 A.L.R. Fed. 2d 697, 84 U.S.P.Q. 2d (BNA) 1099, 2007 U.S. App. LEXIS 21165, 2007 WL 2482122

• Court: Court of Appeals for the Federal Circuit
• Decided: September 5, 2007
• Docket: 2007-1059
• Status: Published

Opinions

Opinion for the court filed by Circuit Judge LOURIE. Opinion concurring as to parts I through IVA and dissenting as to part PV.B filed by Circuit Judge SCHALL.

LOURIE, Circuit Judge.

Ivax Pharmaceuticals, Inc. (“Ivax”) and Cipla, Ltd. (“Cipla”) appeal from the order of the United States District Court for the District of Delaware entering judgment upholding the validity of United States Reissue Patent 34,712 (“the '712 patent”) in favor of Forest Laboratories, Inc., Forest Laboratories Holding, Ltd., and H. Lundbeck A/S (collectively “Forest”) and enjoining Ivax and Cipla from infringing the '712 patent. We affirm the district court’s entry of judgment on validity and its entry of an injunction as to both Ivax and Cipla, but we modify the injunction to apply only to escitalopram oxalate.

BACKGROUND

Ivax filed Abbreviated New Drug Application 76-765 (“the ANDA”) at the Food and Drug Administration, pursuant to 21 U.S.C. § 856® (§ 505© of the Federal Food, Drug, and Cosmetic Act), for approval to market generic tablets containing 5, 10, or 20 milligrams of escitalopram oxalate (“EO”). The ANDA certified, pursuant to 21 U.S.C. § 355(j)(2)(A)(vii)(IV), that the claims of the '712 patent are invalid and/or not infringed by the manufacture, use, or sale of the products for which approval was sought. Cipla is the intended supplier of EO for Ivax and contributed information for the filing of the ANDA. Forest filed suit on September 22, 2003, alleging that Ivax’s filing of the ANDA infringed the '712 patent under 35 U.S.C. § 271(e)(2)(A).¹ Ivax filed its answer on October 15, 2003, denying infringement and counterclaiming for invalidity of the '712 patent. Forest later amended its complaint to add Cipla as a defendant on May 27, 2004.

The '712 patent issued on August 30, 1994 and relates, inter alia, to a substantially pure (+ )-enantiomer of citalopram (also referred to as “escitalopram”) and nontoxic acid additional salts thereof. Stereoisomers are compounds that contain the same constituent atoms and the same bonding between those atoms but have different spatial arrangements. En-antiomers are stereoisomers that are non-superimposable mirror images of one another. Enantiomers accordingly exhibit different optical activity; the enantiomer that rotates a plane of polarized light in the clockwise direction is the (+)enan-tiomer; the enantiomer that rotates a plane of polarized light in the counterclockwise direction is the (-)-enantiomer. Enantiomers may also be designated as the S-enantiomer and the R-enantiomer according to a different criterion relating to the location of the chiral centers. In the case of citalopram, the (+ )-enantiomer is also the S-enantiomer. A mixture of equal amounts of two enantiomers is called a racemic mixture or a racemate, and separating the two enantiomers from a racemate is referred to as resolving the compound. Forest also owned the now expired U.S. Patent 4,136,193 on the ra-cemic form of citalopram and U.S. Patent 4,650,884 that claims a method for making racemic citalopram using an intermediate racemic 1,4 — diol.

EO, which is the oxalate salt form of eseitalopram, is one of the compounds encompassed by the claims of the '712 patent. It is an antidepressant by virtue of being a selective serotonin reuptake inhibitor and is the active ingredient in Forest’s Lexapro® branded drug. Forest has alleged that Ivax and Cipla infringed claims 1, 3, 5, 7, 9 and 11 of the '712 patent by filing the ANDA. Independent claim 1 of the '712 patent reads as follows:

A compound selected from substantially pure (+)-l-(3-Dimethylaminopropyl)-l-(4’-fluorophenyl)-l,3-dihydroisoben-zofuran-5-carbonitrile and non-toxic acid addition salts thereof.

'712 patent col.10 11.31-34. Dependent claim 11 recites a method of preparing the compound of claim 1 by cyclizing an intermediate diol.

The parties stipulated to a specific claim construction for the primary disputed term in the '712 patent, and, based on that agreement, the parties further stipulated that the proposed products included in the ANDA infringe claims 1, 3, 5, 7, and 9 of the '712 patent and that the proposed process for making those products infringes claim 11. Thus, the district court was only required to reach a determination with respect to the counterclaims, including those asserting that the claims are invalid for anticipation and obviousness, and that they were improperly broadened through reissue.

After a bench trial, the district court issued its decision on July 13, 2006, concluding that Ivax and Cipla had failed to prove that the '712 patent is invalid as anticipated. Specifically, the court found that an article by Donald F. Smith (“Smith”) entitled The Stereoselectivity of Serotonin Uptake in Brain Tissue and Blood Platelets: The Topography of the Serotonin Uptake Area (“Smith reference”) did not anticipate claim 1 of the '712 patent because it did not disclose “substantially pure” eseitalopram as claimed in claim 1 and it did not enable a person having ordinary skill in the art to obtain that compound. The court found that chiral High Performance Liquid Chromatography (“HPLC”) was a relatively new and unpredictable technique at the time of the invention and that Smith had worked with the founder of the field of chiral HPLC to separate the enantiomers of citalopram near the time of the invention, but had failed in his efforts. The court also found that a team of chemists at Lundbeck had unsuccessfully attempted to use chiral HPLC to resolve citalopram for two years and that Dr. Danishefsky, Forest’s medicinal chemistry expert, had unsuccessfully tried to resolve compounds with HPLC in the mid-1980’s. The court also found that an inventor of the '712 patent, Dr. Bogeso, had conducted numerous experiments attempting to resolve ra-cemic citalopram through the method of diasteriomeric salt formation, but had also failed. Finally, the court found that Dr. Bogeso only attempted to resolve citalo-pram using a diol intermediate, as recited in claim 11 of the '712 patent, as a last resort and that others of skill in the art would have similarly hesitated because there was a real possibility that the resolved intermediate would re-racemize during the attempt to convert it from the diol intermediate enantiomer to the desired citalopram enantiomer. Thus, the court found that attempting to separate the enantiomers of citalopram based on the knowledge of one of ordinary skill in the art would have required undue experimentation and that the Smith reference was therefore not enabled.

Next, the district court concluded that Ivax and Cipla had failed to prove by clear and convincing evidence that any of the asserted claims of the '712 patent were obvious. The court found that one of ordinary skill in the art at the time of the invention would generally have been motivated to develop new compounds rather than undertake the difficult and unpredictable task of resolving a known racemate. The court further found that a person of ordinary skill attempting to resolve racemic citalopram would have had no reasonable expectation of success for reasons similar to those discussed with respect to enablement of the Smith reference. With respect to the method of claim 11 of the '712 patent, the court found that none of the articles relied upon by Ivax and Cipla described the particular types of reactions claimed (viz., “a Mosher ester serving as a leaving group for a ring closure of a Diol Intermediate,” “an enantioconserving ring closure of a diol containing a tertiary amine to form a tetrahydrofuran,” or “an enantioconserving cyclization reaction of the type needed to convert a tertiary amine like any enantiomerically pure Diol Intermediate into substantially pure ( + )- citalopram”). The court also found that secondary considerations of commercial success, unexpected results, and copying by others supported the validity of the claims.

In addition, the district court found that claim 11 of the '712 patent was not invalid for impermissible broadening during reissue. During the reissue proceeding that resulted in the '712 patent, claim 11 was corrected to claim a method of converting a(-)-diol intermediate to (+ )-citalopram, rather than using a( + )-diol intermediate as shown in the original patent claim. The court found that, given the specific description of the process in the specification, this change amounted to correction of a typographical error. In other words, the court found that the mistake would have been clear to one of ordinary skill in the art reviewing the patent, and therefore that it did not constitute a change in scope from the original claim.

The court entered judgment in accordance with its opinion on November 3, 2006 and at the same time enjoined both Ivax and Cipla “from commercially making, using, offering to sell or selling within the United States, or importing into the United States any products that infringe the '712 patent, including the escitalopram oxalate products referred to in the Abbreviated New Drug Application No. 76-765 until such time as the '712 patent expires.” Ivax and Cipla timely appealed. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1).

DISCUSSION

I. Anticipation

On appeal, Ivax and Cipla argue that the Smith reference clearly anticipates claim 1 of the '712 patent because it discusses and analyzes the efficacy of various drug enantiomers and predicts that one citalopram enantiomer will be more potent as a serotonin reuptake inhibitor than the other. Ivax and Cipla further argue that one of ordinary skill in the art would have known at the time of the invention to use diasteriomeric salt formation to resolve ci-talopram. Specifically, a person of skill in the art would have used the method described in the Wilen reference to resolve the racemic intermediate diol into its enan-tiomers and the method in the Jacobus reference (Williamson ether synthesis) to convert the diol enantiomer (by cyclizing the ether ring) to (+ )-citalopram. Ivax and Cipla add that Dr. Bogeso’s ability to resolve citalopram on his first try after starting with the diol intermediate is further compelling evidence that only routine experimentation was required to separate the enantiomers.

In response, Forest argues that the Smith reference does not disclose “substantially pure” (+ )-citalopram. Forest also argues that the testimony of the experts and the repeated failures of Dr. Bogeso and others to resolve citalopram into its enantiomers support the district court’s determination that the Smith reference was not enabled for (+ )-citalopram. Forest also states that the court was correct to conclude that a person of ordinary skill would have viewed the difficulty in resolving the diol intermediate rather than citalopram itself as significant and a deterrent. In addition, Forest argues that even if a person of skill in the art were to consider using the diol, the Wilen and Ja-cobus references do not involve compounds with structures similar enough to the cital-opram diol intermediate so that a person of ordinary skill would rely upon them to predict the results of a reaction with that compound. More specifically, Forest argues that neither reference discloses a cy-clizing reaction involving a compound, like the citalopram diol, that has a resident tertiary amine or a benzylic alcohol.

Anticipation is a question of fact that we review for clear error following a bench trial. Amgen Inc. v. Hoechst Marion Roussel, Inc., 457 F.3d 1293, 1304 (Fed.Cir.2006). “Under the clear error standard, the court’s findings will not be overturned in the absence of a definite and firm conviction that a mistake has been made.” Impax Labs., Inc. v. Aventis Pharm. Inc., 468 F.3d 1366, 1375 (Fed.Cir.2006) (quotation omitted). “Whether a prior art reference is enabling is a question of law based upon underlying factual findings.” Id. at 1382 (quoting Minn. Mining & Mfg. Co. v. Chemque, Inc., 303 F.3d 1294, 1301 (Fed.Cir.2002)).

We agree with Forest that the district court’s factual findings relating to enablement of the Smith reference are not clearly erroneous, and, based upon those findings, we find no error in the district court’s conclusion that the Smith reference is not enabled with respect to (+ )-citalo-pram. The Smith reference is a pharmacology paper, not a chemical paper. It describes the effects of various enantiom-ers of particular drugs (not including (+ )- citalopram) on the uptake of serotonin in brain tissue and/or platelets. It mentions racemic citalopram (“also of interest”) and shows its structure, but predicts, incorrectly, that the R-enantiomer (the (-)-enan-tiomer for citalopram, not the one claimed in the '712 patent) should be far more potent as a serotonin reuptake inhibitor. Because a racemate does encompass its two enantiomers, it in effect does state that there is a( + )-enantiomer of citalo-pram, but it does not tell how to obtain it. A reference that is not enabling is not anticipating. Elan Pharm., Inc. v. Mayo Found. For Med. Educ. & Research, 346 F.3d 1051, 1054 (Fed.Cir.2003). The Smith reference, as a pharmacology paper, thus does not enable the preparation of the (-t-)-enantiomer of citalopram.

Ivax and Cipla acknowledge that the Smith reference itself does not teach one of ordinary skill how to make ( + )-citalo-pram, but their arguments that it is enabled by other references are largely a recounting of the testimony favorable to their theory of the case without explanation as to why we should have a definite and firm conviction that mistakes were made by the district court in its fact-finding. In other words, they do not inform us why the district court was not entitled to rely on the evidence favorable to Forest or demonstrate that the evidence favorable to them heavily outweighed the evidence favorable to Forest. Such evidence includes the failures of various scientists to resolve citalopram as recited above. Given Ivax and Cipla’s failure to disturb the detailed and thorough factual findings underlying the district court’s decision, we see no error in the finding that the Smith reference does not enable one of ordinary skill to make (+ )-citalopram and hence that the Smith reference does not anticipate claims to (+ )-citalopram.

II. Obviousness

Ivax and Cipla argue that (+)-citalo-pram was obvious in light of racemic cital-opram and descriptions of techniques available to separate enantiomers from their racemates. Further, they argue that the general expectation in the art that one enantiomer would be more potent than the other provided reason for a person of ordinary skill in the art to isolate the enan-tiomers. For reasons similar to those discussed with respect to their argument that the Smith reference was enabled, Ivax and Cipla contend that a person of ordinary skill in the art would have had a reasonable expectation that one could separate the enantiomers of citalopram. Ivax and Cipla also argue that Lexapro’s® commercial success was due to aggressive marketing rather than any alleged superiority of the drug to alternatives and that it did not possess unexpectedly superior properties. Ivax and Cipla also argue that claims 3, 5, 7, and 9 represent only obvious variations on claim 1 with no elements that are not standard in medicinal chemistry and that claim 11 represents the obvious way of resolving citalopram in light of the teaching of the Wilen and Jacobus references.

In response, Forest argues that any pri-ma facie obviousness based on racemic ci-talopram was rebutted by the evidence demonstrating the difficulty of separating the enantiomers and the unexpected properties of (+)-eitalopram. Forest argues that it was unexpected that all of the therapeutic benefit of citalopram would reside in the (+ )enantiomer, resulting in escitalo-pram having twice the potency of racemic citalopram. Forest also argues that the district court was entitled to credit evidence that a person of ordinary skill in the art would not easily have turned to the diol intermediate to attempt resolution of ra-cemic citalopram both because of the uncertainty involved and because Wilen and Jacobus describe compounds less complex than those necessary here to resolve the diol intermediate and then convert the (-)- diol enantiomer to escitalopram.

“Obviousness is a question of law, reviewed de novo, based upon underlying factual questions which are reviewed for clear error following a bench trial.” Leapfrog Enter., Inc. v. Fisher-Price, Inc., 485 F.3d 1157, 1160 (Fed.Cir.2007) (quoting Alza Corp. v. Mylan Labs., Inc., 464 F.3d 1286, 1289 (Fed.Cir.2006)).

We agree with Forest that the district court’s key factual findings underlying its conclusions on obviousness are not clearly erroneous, and, based upon those findings, we find no error in the court’s conclusion that the asserted claims of the '712 patent are not invalid for obviousness. As with their arguments on anticipation, Ivax and Cipla mainly emphasize the evidence that is favorable to their desired outcome without addressing the evidence favorable to Forest. The latter includes the failure of the inventors and others to resolve citalopram without undue experimentation and the testimony of Forest’s experts. The district court applied the Graham factors to conduct a thorough analysis of the evidence, and we find no clear error on facts and no error of law. See Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966). These findings fully support the conclusion that the claimed subject matter would not have been obvious to one of ordinary skill in the art.

III. Broadening Reissue

Ivax and Cipla argue that claim 11 of the '712 patent is invalid because it represents a broadening of original claim 11. They argue that the change in the optical rotation sign of the diol intermediate in claim 11 during reissue was clearly a broadening of the claim because the claim now covers a process beginning with a different enantiomer. Ivax and Cipla also argue that a typographical error may nonetheless be broadening and that a typographical error must be evident to the general public in order to serve the public notice function of patents. In response, Forest argues that the district court correctly determined that the reissue application corrected a typographical error that was readily apparent to one of ordinary skill in the art reviewing the patent and therefore did not result in any change in the scope of the patent.

The reissue statute reads as follows:

Whenever any patent is, through error without any deceptive intention, deemed wholly or partly inoperative or invalid, by reason of ... the patentee claiming more or less than he had a right to claim in the patent, the Director shall ... reissue the patent for the invention disclosed in the original patent ... for the unexpired part of the term of the original patent.... No reissued patent shall be granted enlarging the scope of the claims of the original patent unless applied for within two years from the grant of the original patent.

35 U.S.C. § 251.

The '712 reissue patent resulted from an application filed more than two years after the grant of the original patent, and the claims of a reissue patent filed after that date are invalid if they enlarge the scope of the original claims. See 35 U.S.C. § 282; Quantum Corp. v. Rodime PLC, 65 F.3d 1577, 1583 (Fed.Cir.1995). However, a change in a reissue application that is only clerical does not necessarily broaden the scope of the claims and so does not render the patent invalid. The question before us is whether the change effected in the reissue application here broadened the scope of claim 11 or merely clarified or corrected the original claim.

Comparison of the scope of the reissue claims with the claims of the original patent is a matter of claim construction, and it is performed from the perspective of one having ordinary skill in the art. See Pannu v. Storz Instruments, Inc., 258 F.3d 1366, 1370 (Fed.Cir.2001); Phillips v. AWH Corp., 415 F.3d 1303, 1313 (Fed.Cir.2005) (en banc). Whether the claims of a reissue patent violate 35 U.S.C. § 251 is a question of law that we review de novo based on underlying facts reviewed for clear error. Medtronic, Inc. v. Guidant Corp., 465 F.3d 1360, 1373 (Fed.Cir.2006).

We agree with Forest that the change in the optical rotation sign for the diol intermediate in claim 11 of the '712 patent did not broaden the scope of the claim. The patent specification supports, even compels, this conclusion. In Reaction Scheme I, the process begins with a racemic mixture of the diol intermediate. See '712 patent col.5 11.41-42. The diagram of Reaction Scheme I reinforces that conclusion because it includes a notation next to the diagram of the diol intermediate that reads “(+) and (-).” ’712 patent (emphasis added). The patent then describes a reaction sequence that results in production of “the ester as a diastereomeric mixture.” Id. at col.5 11.47-48. It is the diastereomeric mixture of an ester (of a monoalcohol) that is then subjected to HPLC to produce an enantiomerically pure compound that can be converted into the desired (+ )-citalopram end product. Id. at col.5 11.47-59. Further, as found by the district court, because Reaction Scheme I begins with the racemate of the diol intermediate, the patent is ambiguous as to which enantiomer of the diol intermediate is actually converted to ( + )-cital-opram.

In contrast, while the description of Reaction Scheme II also begins with the racemate of the diol intermediate, it is the diol intermediate itself that is resolved to produce an enantiomerically pure product compound: Id. at col.6 11.8-33. The description also specifically describes using the (-)-diol intermediate to produce ( + )ci-talopram. Id. at col.6 11.29-42. Further support is again provided by the diagram of Reaction Scheme II, which includes a notation that reads “(-) or (+)” next to the structural diagram of citalopram. '712 patent (emphasis added). Thus, the enan-tiomeric labeling for the end product is reversed from that of the starting compound. Given this plain reading and the additional supporting expert testimony also relied upon by the district court, we see no error in the district court’s finding that a person of ordinary skill in the art reviewing the patent would find the error in claim 11 relating to the optical sign of the diol intermediately apparent. The diagram of Reaction Scheme II makes clear that it is the (-)-diol that is converted to (+)-eitalopram and that the correction in the claim corresponds to the disclosure in the specification. We therefore agree that the change in the optical sign during reissue does not represent a change of claim scope, but merely a correction of the claim to be consistent with the disclosure of the specification.

IV. The Scope of the Injunction

Ivax and Cipla argue that the language of the injunction is overly broad in extending to “any products that infringe the '712 patent.” Also, they arg-ue that the artificial act of infringement created by 35 U.S.C. § 271(e)(2)(A) is narrow and that Cipla’s provision of information to Ivax for its filing of the ANDA is not an act of infringement. Ivax and Cipla further argue that the injunction granted by the district court violates our holding in International Rectifier Corp. v. IXYS Corp., 383 F.3d 1312, 1316 (Fed.Cir.2004). Forest responds that the injunction is sufficiently narrowly defined because of the infringement stipulation of the parties and the detailed record. Forest also argues that because Cipla will manufacture and import the EO products described in the ANDA if it is approved, Cipla may properly be enjoined for inducement of infringement.

A. Scope of Products

We do not agree with the scope of the district court’s injunction that includes products other than escitalopram oxalate. ‘(Although the standard of review for the issuance and scope of an injunction is abuse of discretion, whether the terms of the injunction fulfill the mandates of Federal Rule of Civil Procedure 65(d) is a question of law that this court reviews de novo.” Signtech USA, Ltd. v. Vutek, Inc., 174 F.3d 1352, 1356 (Fed.Cir.1999). In International Rectifier, we held that “the only acts [an] injunction may prohibit are infringement of the patent by the adjudicated [products] and infringement by [products] not more than eolorably different from the adjudicated [products]. 383 F.3d at 1316. In order to comply with Rule 65(d), the injunction should explicitly proscribe only those specific acts.”

Here, the '712 patent covers a range of products beyond those described in the ANDA. The statute, 35 U.S.C. § 271(e)(4)(B), provides that “injunctive relief may be granted against an infringer to prevent the commercial manufacture, use, offer to sell, or sale within the United States or importation into the United States of an approved drug.” Thus, while the injunction may properly extend to the “approved drag,” it should not extend to the remainder of the products covered by the patent. The injunction is therefore modified to delete the language “any products that infringe the '712 patent, including.”

B. Inclusion of Cipla

However, we find that it was not inappropriate for the district court to include Cipla within the scope of the injunction. Section 271(e)(2) may support an action for induced infringement. Allergan, Inc. v. Alcon Labs., Inc., 324 F.3d 1322, 1331 (Fed.Cir.2003). “The only difference in the analysis of a traditional infringement claim and a claim of infringement under section 271(e)(2) is the timeframe under which the elements of infringement are considered.” Id. An inquiry into induced infringement focuses on the party accused of inducement as the prime mover in the chain of events leading to infringement. Here, we do not know if Cipla first approached Ivax or vice versa, but the plan to manufacture, import, market, and sell the EO products described in the ANDA was undoubtedly a cooperative venture, and Cipla was to manufacture and sell infringing EO products to Ivax for resale in the United States. Under the standards for inducement which we apply to 35 U.S.C. § 271(b), Cipla has therefore actively induced the acts of Ivax that will constitute direct infringement upon approval of the ANDA, and it was thus not inappropriate for the district court to include Cipla within the scope of the injunction.

The dissent asserts that § 271(e)(1) exempts Cipla from being enjoined with Ivax. We disagree. Cipla is providing information, and will provide material, that Ivax will use to obtain FDA approval. Up to that point, there is indeed no infringement. And, in fact, Ivax is not currently liable for infringement, as long as it is only pursuing FDA approval, not commercially manufacturing or selling the infringing product. However, just as Ivax will be liable for, and hence is being enjoined from, the commercial exploitation of escitalopram when it is approved by the FDA and during the life of the patent, so should Cipla be enjoined. They are partners. Cipla would be contributing to the infringement by Ivax, so the injunction should cover both partners. It is true that, as the dissent states, § 271(e)(2) defines Ivax’s filing of its ANDA as an infringement, and Cipla did not file the ANDA; however, when the question of an injunction against commercial activity arises, Cipla is as culpable, and hence entitled to be enjoined, as Ivax.

CONCLUSION

For the reasons stated, we affirm the district court’s grant of judgment of no invalidity of the '712 patent and the entry of injunction, as modified herein, as to both Ivax and Cipla pursuant to the stipulation of infringement.

AFFIRMED

1 . 35 U.S.C. § 271(e)(2)(A) provides:

It shall be an act of infringement to submit an application under section 505(j) of the Federal Food, Drug, and Cosmetic Act or described in section 505(b)(2) of such Act for a drug claimed in a patent or the use of which is claimed in a patent if the purpose of such submission is to obtain approval under such Act to engage in the commercial manufacture, use, or sale of a drug or veterinary biological product claimed in a patent or the use of which is claimed in a patent before the expiration of such patent.