Pfizer, Inc. v. Teva Pharmaceuticals USA, Inc. (Fed. Cir. 2008)

    By Kevin E. Noonan --

On Friday, the Court of Appeals for the Federal Circuit handed Pfizer a mixed decision on defendant Teva's appeal of a District Court judgment that Teva was liable for infringing a trio of Pfizer patents (U.S. Patent Nos. 5,466,823; 5,563,165; and 5,760,068) relating to the pain medication Celebrex® (celocoxib).  While keeping Teva's generic celocoxib off the market for now, the Federal Circuit's ruling advances the date when the generic version of Celebrex® will become available by about 12 months.

Pfizer initiated its lawsuit upon notification by Teva of its Paragraph IV certification filed in Teva's Abbreviated New Drug Application (ANDA) with the FDA.  Although Pfizer asserted a number of claims against Teva, the claims relevant to the issues on appeal were claim 9 of the '823 patent:

Compound of claim 8 where the compound is [celocoxib], or a pharmaceutically-acceptable salt thereof.

Claim 17 (dependent on claim 15) of the '165 patent:

A pharmaceutical composition comprising a therapeutically-effective amount of a compound and a pharmaceutically-acceptable carrier or diluent, said compound selected from a family of compounds . . . where the compound is [celocoxib], or a pharmaceutically-acceptable salt thereof.

And claim 4 (dependent on claim 1) of the '068 patent:

A method of treating inflammation or an inflammation-associated disorder in a subject, said method comprising administering to the subject having or susceptible to such inflammation or inflammation-associated disorder, a therapeutically-effective amount of a compound . . . where the compound is [celocoxib], or a pharmaceutically-acceptable salt thereof.

(Where "celocoxib" is 4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-
yl]benzenesulfonamide.)

The District Court found the asserted claims infringed, the claims of the '068 patent not invalid for obviousness-type double patenting, claim 9 of the '823 patent and claim 17 of the '165 patent not invalid for violation of the best mode, and the asserted claims not unenforceable due to alleged inequitable conduct.

The Federal Circuit, in a unanimous opinion by Judge Dyk (joined by Chief Judge Michel and District Judge Matthew Kennelly of the Northern District of Illinois, sitting by designation) reversed the District Court's determination that the claims of the '068 patent were not invalid for obviousness-type double patenting but affirmed Teva's failure to carry its burden of a best mode violation or inequitable conduct on the other asserted patents.  The basis for the Federal Circuit's decision that the '068 patent was invalid for obviousness-type double patenting was that the designation of this patent as a "continuation-in-part" was not merely semantic but had substantive repercussions regarding whether the '068 claims were entitled to the "safe harbor" provisions of 35 U.S.C. § 121.  These provisions entitle a divisional application filed as the result of a restriction requirement (a Patent Office determination that an applicant has filed claims directed to more than one invention) to be exempt from obviousness-type double patenting.  These provisions illustrate a Congressional intention, from the legislative history of the 1952 Patent Act, to prevent the unfairness of forcing an applicant to be required to file a divisional application based on claiming "independent inventions" through restriction while permitting the original application to be used as "prior art" in a obviousness-type double patenting rejection.  The Federal Circuit noted, however, that Congress also evinced the intention to restrict the safe harbor to divisional applications (which have an identical specification to the originally-filed application) and keep any claims in such a divisional application strictly within the bounds of what had been determined to be a separate invention.  This intention meant, for example, that an applicant would not be able to include claims in a divisional that were within the scope of any claims previously elected for examination.  In this way applicants would be precluded from inequitably increasing patent term while not being penalized for electing one invention and pursuing additional inventions in divisional applications.

The Federal Circuit found that the '068 application did not fall within the safe harbor because it was a continuation-in-part application, not a divisional.  Such a "CIP" application by definition contains some of the disclosure of an earlier-filed application in addition to new disclosure.  Because of this distinction, the Federal Circuit found that CIP applications could not fall within the safe harbor because they could encompass additional information (and patentable inventions) that were not a part of the original, restricted claims.

In this case, however, the CAFC has applied the rubric to the application rather than to the claims.  A review of the claims of the '165 patent, as originally filed, shows that the application contains a claim (claim 29) that recites the same method as the '068 patent and a list of species of the generic formula, expressly reciting celecoxib.  This claim was subjected to a restriction requirement and the applicants elected to forego prosecution of these method claims until filing the CIP application that resulted in the '068 patent.  However, the specification of the '068 patent is much more extensive than the earlier-filed '823 and '165 patents; in particular, the '068 patent specifically discloses celecoxib synthesis, and contains biological data relating to the specificity of celecoxib for COX2 over COX1 (the biological basis for its advantages as an anti-inflammatory agent).  Without specifying any rationale other than the noted distinction between divisional and CIP applications, it appears that these differences in the specification may provide additional support for the Federal Circuit's determination.

The CAFC upheld the District Court's determination that claim 9 of the '823 patent and claim 17 of the '165 patent (directed specifically to celecoxib) were not invalid for failure to satisfy the best mode requirement of 35 U.S.C. § 112, first paragraph.  Teva contended that Pfizer's failure to disclose the COX2 specificity of the claimed compounds was a violation because it prevented the skilled worker from appreciating which of the disclosed compounds exhibited the specificity for COX2.  In Teva's view, this failure of disclosure was significant, because the capacity to block the action of COX2, related to pain, while not inhibiting COX1, related to normal processes such as gastrointestinal integrity, was an important feature of the claimed invention.  The Federal Circuit adopted the view of the District Court that Pfizer was under no obligation to disclose this distinction.  (Presumably, this would not have been the case if the claim was to a method or included the COX enzyme specific limitations.)  Disclosing how to make and use the compounds of the invention was enough to satisfy the best mode requirement.  Interestingly, the CAFC noted that while Teva had raised the defense of invalidity against Pfizer's claims, it had not counterclaimed for a declaratory judgment of invalidity; thus, since claim 9 of the '823 patent and claim 17 of the '165 patent satisfied the best mode requirement and were infringed, it did not need to reach the "difficult issue" raised by the generic claims.  This dicta from the Federal Circuit would seem to raise a caution for similar instances where generic claims are asserted.

Finally, the Federal Circuit found that while the references cited by Teva in its inequitable conduct allegation were material (despite the fact that they would not have constituted invalidating prior art), the CAFC found that Pfizer had rebutted any evidence of intent proffered by Teva.  This rebuttal was evidence that the structure of the core heterocyclic ring in the two prior art references was different from the core heterocyclic ring of Pfizer's patented compounds, and that Pfizer consistently used such distinctions in determining whether prior art was relevant "in hundreds of its other patent applications."  The Federal Circuit also credited testimony on this point by a Pfizer witness, that Pfizer had considered these references and had come to a good-faith determination that they were not material to patentability.  The CAFC cited its own three-prong test for inequitable conduct enunciated in Ferring B.V. v. Barr Labs., Inc. (Fed. Cir. 2006):

(1) the applicant knew of the information;
(2) the applicant knew or should have known of the materiality of the information; and
(3) the applicant has not provided a credible explanation for the withholding.

Here, Teva's failure to establish or rebut the third prong was fatal to its inequitable conduct allegation.

As a consequence of the Federal Circuit's holding the '068 patent invalid, the expiration date for the remaining Orange Book listed patents is now May 30, 2014 (when pediatric exclusivity for the '165 patent expires) rather than June 2, 2015.  Undoubtedly this means millions of dollars in revenue lost, especially since the '823 patent was deemed ineligible for patent term extension under 35 U.S.C. § 156.

Pfizer, Inc. v. Teva Pharmaceuticals USA, Inc. (Fed. Cir. 2008)
Panel: Chief Judge Michel, Circuit Judge Dyk, and District Judge Kennelly
Opinion by Circuit Judge Dyk

Additional information regarding this case can be found at Patently-O and the Orange Book Blog.

Comments

[Louis Ventre]

The court overlooked the last sentence of section 121, which implies that a divisional may be a CIP. That sentence reads, "If a divisional application is directed solely to subject matter described and claimed in the original application as filed, the Director may dispense with signing and execution by the inventor. The validity of a patent shall not be questioned for failure of the Director to require the application to be restricted to one invention." If a divisional is always directed "solely to subject matter described in the original application," then this sentence is irrelevant and has no meaning, which is not preferred in statutory construction.

[Kevin E. Noonan]

Dear Louis:

Yes, we (and others) have considered that sentence, and it does seem to indicate the possibility of an application being both a divisional and a CIP. But another way of parsing the language would be if a CIP was filed with claims solely directed to (and supported by) subject matter disclosed and (originally) claims in the parent, then no new oath/declaration is required.

Looking at it that way, the excess disclosure in the '068 patent, particularly regarding the specific synthesis of celecoxib (which was disclosed but not exemplified in the parent), as well as the Court's understanding that the COX selectivity was an important aspect of the method claims (having decided on the best mode question that selectivity didn't matter for the compound claims) might have made the CIP disclosure of the selectivity of celecoxib a factor in the decision.

Of course, once again there seem to be case-specific facts used to construe the statute, and in a way that is contrary to sound rules of statutory construction (as you point out).

Thanks for the comment.

[Bob]

Perhaps it's more appropriate to say that Judge Dyk references HIS three-prong test for inequitable conduct articulated in Ferring. It's not clear to me that other judges on the Federal Circuit have adopted this quasi-burden-shifting approach to the issue, despite Judge Michel's willingness to join the Dyk opinion.

The double patenting result seems fair, though. Otherwise, section 121 becomes a loophole whereby the patentee can use a CIP to enable MOT claims that were arguably not enabled in the originally filed application.

[Kevin E. Noonan]

Dear Bob:

Well, remember the Supremes didn't grant cert on Ferring, so for now I think the combination of Ferring + Pfizer + no cert means this test is at least in play as precedent, and being a "prong" test it is organizing for the arguments in a way that may make it attractive.

With regard to the Sec. 121 issue, perhaps a better way is to ask the question, no matter what you call it are the claims at issue (ALL the claims) originally filed in the parent that was subject to the restriction requirement, and were those claims supported in the specification satisfying Sec. 112? If the answers are YES, then I think you are entitled to the safe harbor, and the issue you raise does not apply.

Thanks for the comment.